Closed Minds?

Is a ‘Cancel Culture’
Stifling Academic Freedom and
Intellectual Debate in Political Science?
Faculty Research Working Paper Series

Pippa Norris
Harvard Kennedy School

August 2020
RWP20-025

Visit the HKS Faculty Research Working Paper Series at:

https://www.hks.harvard.edu/research-insights/publications?f%5B0%5D=publication_types%3A121
The views expressed in the HKS Faculty Research Working Paper Series are those of the author(s) and do not
necessarily reflect those of the John F. Kennedy School of Government or of Harvard University. Faculty Research
Working Papers have not undergone formal review and approval. Such papers are included in this series to elicit
feedback and to encourage debate on important public policy challenges. Copyright belongs to the author(s). Papers
may be downloaded for personal use only.

www.hks.harvard.edu
 Closed minds?
Is a ‘cancel culture’ stifling academic freedom and intellectual debate in political science?

Pippa Norris

Pippa_Norris@Harvard.edu

Synopsis: Recent years have seen extensive debate in popular commentary about a pervasive ‘cancel
culture’ thought to be taking over college campuses. A progressive orthodoxy, it is argued, has silenced
conservative voices and diverse perspectives. This development, it is claimed, has ostracized contrarians,
limited academic freedom, strengthened conformism, and eviscerated robust intellectual debate. But
does systematic empirical evidence support these claims? After reviewing the arguments, Part II of this
study outline several propositions arising from the cancel culture thesis and describes the sources of
empirical survey evidence and measures used to test these claims within the discipline of political science.
Data is derived from a new global survey, the World of Political Science, 2019, with 2,446 responses
collected from scholars studying or working in 102 countries. Part III presents the results. Part IV
summarizes the key findings and considers their broader implications. Overall the study confirms the
significant impact of Left-Right ideology on reported experience of the cancel culture in political science
– but important contrasts were found in post-industrial and developing societies.

Keywords: academic freedom, sociology of education, cancel culture

Word count: 8,868

Version 1.1 Date: 5 August 2020

1
Heated battles about the so-called ‘cancel culture’ on college campuses have been intensified by recent
controversies surrounding issues of racism and ethnicity, sexual harassment and misogyny, non-binary
gender identities and transphobia. 1 The concept of a ‘cancel culture’ can be defined broadly as attempts
to ostracize someone for violating social norms. The notion has also been understood more narrowly as
“the practice of withdrawing support for (or canceling) public figures and companies after they have done
or said something considered objectionable or offensive.”2 This practice is analogous to the tactic of
consumer-boycotts withdrawing support for perceived unethical brands and corporations, a common
form of political activism.3 The cancelling strategy typically uses social media to shame individuals with
the intention of exerting penalties with different degrees of severity, ranging from limiting access to public
platforms, damaging reputations, and ending careers to instigating legal prosecutions. The process is
exemplified by notorious cause célèbres in American popular entertainment, such as the cases of O.J.
Simpson, Roseanne Barr, and Michael Jackson. The impact is thought to have spread more widely in many
societies, however, with book deals being torn up, editors and journalists demoted or fired, and public
intellectuals attacked. The phenomenon has claimed scalps among well-known media celebrities (like the
comedian Louis C.K.), leading politicians (for example, former-Senator Al Franken), authors (J.K. Rowling),
and corporate executives (such as Roger Ailes at Fox News). Critics warn that the process may have started
with legitimate criticism of cases attracting widespread moral disapproval, but it is a slippery slope. As the
tumbrils have rolled downhill, like revolutions eventually eating their own, the risks are that the process
will ultimately undermine liberal tolerance of contrarians in many fields, including stifling intellectual
debate on college campuses.

Can public shaming be regarded as an appropriate tactic? Debate continues.

On the one hand, this strategy can be justified as an effective tool for achieving social justice by victims
unable to obtain legal redress or public apology. Examples include the MeToo boycotts directed against
powerful sexual predators alleged to have repeatedly committed harassment, rape, and even pedophilia.
Black Lives Matter activists have also used public shaming when calling out the authors of racist textbooks,
fatal cases of police violence in communities of color, racial stereotypes used to sell consumer products,
and university departments lacking diversity. In an era of rapidly changing moral standards and
heightened cultural sensitivities around social identities, it can be argued that powerful figures in the
public eye should try to move with the times, avoid causing unnecessary offence, and thereby be held
accountable for their words and actions (or inactions). From this perspective, public shaming has a
legitimate role through criticizing the use of derogatory and offensive language like racial or homophobic

2
slurs, highlighting the unacceptable abuse of power such as sexual harassment, or criticizing engagement
in practices of cultural appropriation. Feeling comfortable to speak one’s mind in an uninhibited and
robust fashion may appear to be a healthy liberal virtue in an open society. But it can also be regarded as
problematic where blunt talk carelessly hurts others, as reflected in the notion that ‘white lies’, or saying
nothing, may be the more diplomatic strategy to offending. Indeed, those seeking to limit public shaming
can also be seen to violate basic principles of free speech.

On the other, however, critics argue that the movement has gone too far, especially on college campuses,
so that it now threatens classical liberal values at the heart of academic life. As exemplified in On Liberty
by John Stuart Mill, liberalism champions tolerance of non-conformity and freedom of speech, even for,
or indeed especially for, especially for the expression of deeply unpopular and contrarian opinions.4 Only
by questioning dogma and the conventional wisdom can we become aware of our own prior values and
beliefs. In his public writings and speeches, Mill defended many controversial causes of his time, from
Fenians planning an armed uprising to end British rule in Ireland to suffragettes demanding women’s
rights to vote. Conservatives argue that recent years have seen growing silencing of contrarian voices
challenging the liberal hegemony in many cultural spaces, but especially in the academy, thereby limiting
freedom of speech, increasing social pressures for ideological conformity, reinforcing intellectual
exclusion, group-think bubbles, ‘Us-Them’ segregation, academic intolerance, and self-censorship. The
net result, critics claim, is a rush to collective judgment and a new climate of censorship by the ‘mob’,
magnifying even minor errors of judgment. 5 Kukianoff and Haidt regard this trend on campus as a
sanctimonious ‘coddling’ of student minds, where emotional discomfort is seen as equivalent to physical
harm, with colleges failing to cultivate resilience in a hostile world.6 Williams argues that lack of academic
freedom is detrimental for scientific progress, as researchers are unable to investigate all perspectives.7
The U.S. Education Secretary, Betsy DeVos, has accused liberal faculty members of forcing their views
upon students, telling them what to think, with indoctrination replacing education.8 Many Republicans
claim that academic life is now dominated by scholars with progressive liberal or leftwing values.
Intolerance of dissenting views, especially among the progressive ‘far-left’, it is argued, silences
conservative perspectives, brainwashing students into ‘politically correct’ views. As expressed in President
Trump’s Mount Rushmore speech:

“One of their (the left’s) political weapons is ‘cancel culture’ — driving people from their jobs,
shaming dissenters, and demanding total submission from anyone who disagrees. This is the very

3
 definition of totalitarianism, and it is completely alien to our culture and our values, and it has
absolutely no place in the United States of America.”9

But concern about these issues is not confined to conservative Republicans; as expressed in the infamous
letter in Harper’s Magazine, it has also been endorsed by such luminary and diverse public intellectuals
as Martin Amis, Gloria Steinem, J.K. Rowling, Salman Rushdie, Margaret Atwood, and Noam Chomsky:

“Our cultural institutions are facing a moment of trial. Powerful protests for racial and social
justice are leading to overdue demands for police reform, along with wider calls for greater
equality and inclusion across our society, not least in higher education, journalism, philanthropy,
and the arts. But this needed reckoning has also intensified a new set of moral attitudes and
political commitments that tend to weaken our norms of open debate and toleration of differences
in favor of ideological conformity.”10

Theories suggests several factors in modern societies which may help to explain the rise of a culture
11
culture. Perhaps the most important factor in most accounts is the liberal hegemony in the
contemporary academy, which is expected to lead the minority of conservative faculty and students to
feel that their voices have been silenced. 12 Other potentially reinforcing conditions include contemporary
shifts in the cultural values in post-industrial societies, deepening polarization over issues of identity
politics which divide social conservatives and liberals.13 Technology is widely regarded as another culprit,
by expanding opportunities for collective expression via digital media, with the politics of outrage
constantly reinforced by filter bubbles and echo chambers. Social media are thought to have amplified
the reach, and accelerated the pace, of the cancel culture spreading on college campuses and beyond.

The cultural backlash in politics is a further reason. Populist parties and leaders, seeking to stir resentment
and energize their followers, have sought to exploit claims of political bias as a polemical cudgel to attack
intellectuals in colleges and universities. This is illustrated by President Trump’s Mount Rushmore speech,
claiming the mantle of shared conservative victimhood and denigrating the out-of-touch progressive
intellectual elites. Moreover, the populist message is likely to resonate with his followers; Pew Research
Center polls report that in late-2018, only 48% of Republicans expressed confidence in college and
university professors, compared with 84% of Democrats, a larger partisan gap than other groups like the
police or business executives.14

This phenomenon has attracted most attention in the U.S but it is not simply a result of partisan
polarization during the Trump era or another case of American exceptionalism; similar concerns echo
elsewhere. In the UK, for example, in 2019, the Human Rights Commission published new guidelines for

4
universities designed to protect free speech, following concern about censorship ‘deplatforming’
15
speakers, withdrawing invitations out of fear that they may prove controversial. A recent report,
drawing on a YouGov survey of faculty in the social sciences and humanities, concluded that UK
universities face growing threats to academic freedom.16 This called attention to ‘structural
discrimination’, with progressive social norms arising from the predominance of leftwing scholars,
generating a chilly climate of intolerance for the small minority challenging these views. 17 The report
quoted examples of actions restricting unorthodox thinkers and viewpoint diversity, include events
restricting speakers, cases of staff facing disciplinary actions or dismissal, and the abuse or harassment
towards contrarians expressing views on contested subjects, such as the issues of gender, race and
ethnicity. A survey of British students reported similar perceptions of liberal bias in universities and a
reluctance by conservative students to express themselves for fear of disagreement with their peers.18

But despite the wealth of popular commentary, does systematic empirical evidence support the claim of
a pervasive ‘cancel culture’ taking hold of academic life in many countries? Some political complaints can
be dismissed as rhetorical talking points. More persuasive support comes from citing specific cases of
intolerance and silencing on college campuses: speakers shouted down, events cancelled, Twitter outrage
trending, faculty disciplined, and so on. It can be argued, however, that the plural of anecdotal cases is
not data. Highlighting outrage directed against a few high-profile public intellectuals may cherry-pick
cases to confirm prior expectations. Moreover, such actions are far from new, examples can be cited of
similar censorship during previous eras of heated politics, especially during the mid-1960s and early-1970s
culture wars, albeit spread through mimeographed posters and spirit duplicated newsletters before digital
media. It is important use more systematic scientific survey evidence to establish whether specific cited
cases today are anything more than isolated instances or whether, as a general tendency, conservative
scholars report experience of growing restrictions on academic freedom, pressures to be politically
correct, and the silencing of diverse perspectives in higher education.

Accordingly, drawing upon these arguments, Part II of this study proceeds to outline several propositions
arising from the cancel culture thesis and then describes the sources of empirical survey evidence and
measures used to test these claims within the discipline of political science. Data is derived from a new
global survey in the discipline of political science, the World of Political Science, 2019, with 2,446
responses collected from scholars studying or working in 102 countries. Part III presents the results. The
conclusions in Part IV summarize the key findings and considers their broader implications.

5
 II: Hypotheses, evidence and measures

Testing the cancel culture claims about the growing silencing of contrarian speakers, the cancellation of
events, and limits on conservative thought on college campuses is far from straightforward. Difficulties
arise because by their very nature, practices of self-censorship are similar to non-events, akin to Conan
Doyle’s infamous dog which didn’t bark. Much popular media speculation is politically driven. The debate
involves complex normative issues. The logic of the argument suggests several propositions, however,
which are potentially open to testing against empirical evidence from surveys of the professoriate.

The ideological identities of political scientists

The first (albeit weaker) test of the cancel culture thesis concerns claims about the predominate
progressive left-wing ideological values and beliefs of college cultures. Claims of a liberal hegemony
among the professoriate, silencing conservative voices, reflects a longstanding concern. Paul Lazarsfeld’s
pioneering 1958 book The Academic Mind, based on a large-scale representative survey of American social
scientists, was the first to demonstrate that scholars tended to be more sympathetic to liberal or leftwing
values than the general population.19 The radical era of heated college and university politics during the
1960s and early-1970s triggered further research into the political beliefs and values of college and
university professors, with the findings confirming the pervasive liberal tilt of the academy. This was
documented by Everett Carll Ladd and Seymour Martin Lipset’s The Divided Academy (1976)20, reporting
that about 46% of professors in their survey identified themselves as left or liberal, 27% were middle of
the road, while 28% were conservative. Social scientists were found to be further left than most other
disciplines. Moreover this pattern was not confined to American campuses, as similar skews were found
by Albert Halsey and Martin Trow’s The British Academics (1971).21 A decade ago, Gross and Simmons
(2007) updated the evidence with another largescale survey of the American professoriate,
demonstrating that conservatives and Republican identifiers remain relatively rare among faculty in U.S.
universities, especially in the social sciences, although many held moderate middle-of-the-road views.22
More recently, Shields and Dunn reviewed five major U.S. surveys of academics conducted since 2000 and
concluded that the percentage of self-identified conservatives was found to range between 5% and 15%
in the social sciences and 4% to 8% in the humanities.23 In Europe, as well, analysis of the European Social
Survey pooled data suggest that professors in 31 European countries are usually more liberal and left-
leaning than other equivalent professions like lawyers, architects and physicians, although the political
values and attitudes of academics on issues such as economic redistribution and EU integration are far
24
from homogeneous. Therefore, a series of previous studies of the sociology of the academy have

6
consistently reported for many decades that a pervasive liberal-left skew is observable, at least in post-
industrial societies where most research has been conducted. This can be expected to be observed in the
discipline of political science. This proposition can be tested by surveys examining either the values,
attitudes and beliefs of faculty and students, or, as in this study, their self-identified leanings of political
scientists across a Left-Right ideological scale.

Experience of worsening academic freedoms

But a progressive bias on campus doesn’t necessarily imply a lack of tolerance for pluralistic debate, or
that contrarian views are unwelcome, or ideas silenced. The stronger test of the cancel culture thesis
concerns whether experience of this phenomenon is thought to have worsened. These measures also
have important limitations, after all the majority of liberal faculty may well be unaware of any hegemonic
bias, and, even if acknowledged, to downplay that this as a serious problem. But the minority of
conservative scholars can be expected to be more likely to express concern about this phenomenon, if
they have direct experience of a cancel culture, for example, agreeing most strongly from their experience
with indicators of a worsening intellectual climate, such as growing restrictions on academic freedom of
speech, pressures for ideological conformity, and the enforcement of politically correct speech.
Accordingly, this study can compare reported experience of change in these conditions getting better or
worse among self-identified conservative and liberal scholars in political science.

Mediating conditions: cohort and cross-national comparisons

As a related issue, another limitation with the available evidence is that benchmark data is unavailable to
monitor change over time from successive surveys of scholars in political science. If a cancel culture has
gradually intensified in recent years however, as often claimed, this should also be evident through
comparisons of the reported experience of birth cohorts. In particular, generations may be expected to
differ in how they respond to these issues. For example, a Politico/Morning Consult poll of public opinion
reported that a plurality of Americans (46%) believed that a ‘cancel culture’ of group shaming had gone
too far, with 49% believing that it has had a negative impact on society. But the youngest birth cohorts
(Generation Z and Millennials) expressed the most sympathy towards punishing people with views
regarded as offensive. For these reasons, the younger generation of scholars may be expected to prove
both most liberal in their identities and also least concerned about public shaming for socially offensive
speech and acts. By contrast, the older cohorts are predicted to be more conservative and to feel that
academic freedom and intellectual tolerance have worsened. Perceptions of the cancel culture may also
be influenced by gender, particularly if women’s attitudes are shaped by the issues of sexual harassment,

7
abuse, and violence highlighted by the ‘Me Too’ movement seeking to oust and shame sexual predators,
including some notorious cases involving senior faculty in higher education. The effects of gender can also
be controlled in the multivariate models.

Finally, if academia reflects broader shifts in society, associated with a general cultural backlash and
deepening polarization over values, this should be evident through cross-national comparisons. In
particular, a ‘cancel culture’ should be expected to be more prevalent in colleges and universities located
in affluent post-industrial societies - especially the United States, where most studies have been
conducted. These countries have seen intensified polarization and deepening political battles in the
cleavage over conservative v. liberal values, with the cultural backlash exemplified deep divisions over by
the Trump presidency in the US and by battles between Leavers and Remainers over Brexit in the UK. It
remains unclear whether these cleavages vary by level of development. A wealth of survey evidence
suggests that socially liberal values towards cultural issues such as gender equality, secularization, and
homosexuality have advanced most among the younger generation and the most educated social sectors
in affluent post-industrial societies.25 By contrast, many poorer developing societies in the rest of the
world are characterized by a broader consensus about conservative values, which continue to
predominate in more traditional cultures. On this basis, cultural polarization about the cancel culture may
be expected to be most evident in colleges and universities located in the United States, as well as in
similar affluent post-industrial societies in Western Europe and Australasia, compared with those in
developing societies. This can be tested by breaking down the analysis to compare the predictors in the
US, in the pooled sample of 23 post-industrial societies, and in 78 developing countries.

To summarize, for all these reasons, personal experience of a worsening ‘cancel culture’ may be expected
to vary among scholars of political science by their (i) Left-Right ideological identities, (ii) generational
birth cohort and gender, and (iii) the type of post-industrial or developing society in which they live, study
and work.

Survey data and measures

Building upon the Lazarsfeld survey tradition in the sociology of knowledge, empirical data to explore
these issues can be examined from a new global survey of scholars in the discipline of political science,
the World of Political Science, 2019. Previous studies report that the liberal skew among scholars is more
evident among social scientists rather than other disciplines.26 Moreover, the phenomenon of a cancel
culture involves controversies about many politically sensitive issues, especially inequalities of power and

8
status associated with race and gender, so the profession of political science is a particular appropriate
discipline to test the core arguments.

The World of Political Science survey (WPS-2019) provides a representative profile of the political science
profession across the world.27 Invitations asking political scientists to participate were distributed through
social media notifications (Facebook, emails, and Twitter), the ECPR Newsletter list and IPSA lists, and
through several national associations (CPSA, PSA UK, Australian PSA and Russia). Overall, 2,446 responses
were collected from 3 February to 7 April 2019 from respondents studying or working in 102 countries. In
total, 1,245 responses were collected from political scientists studying or working in 23 affluent post-
industrial societies (including the US, Europe, and Australasia).

The World of Political Science survey gathered information through an online Qualtrics questionnaire
about multiple dimensions of professional life and work, including role priorities, social background
characteristics, national origins, qualifications, thematic subfield of expertise, and methodological skills.
The survey also included a battery of 22 items monitoring direct experience of changes in the profession
during the last five years. Respondents were asked:

“Academic life is often thought to be in a state of change. Using the following scale, based on your
experience, please indicate whether you think the quality of the following aspects of academic life
have changed over the last five years.”

Potential responses in this battery of items were measured on a 5-point scale and recoded as follows: Got
a lot better (1), got somewhat better (2), no change (3), got somewhat worse (4), got a lot worse (5).28
The questions are therefore designed to tap into experiential judgments about the direction of change.
The list included both positive and negative types of changes, to avoid potential affirmative response bias.

The concept of a ‘cancel culture’ was defined earlier as attempts to ostracize someone for violating social
norms. The change battery included three items selected as proxy or indirect measures of this underlying
concept, based on the direct experience of respondents. These asked whether, in their experience, each
of the following aspects of academic life had got better, no change, or got worse, using the 5-point scale:

i. “Respect for open debate from diverse perspectives.”

ii. “Pressures to be 'politically correct’.”

iii. “Academic freedom to teach and research.”

9
The items can be seen as valid proxy indicators of experience about several claims in the ‘cancel culture’
thesis, including that college campuses have seen worsening tolerance of pluralistic debate and opposing
viewpoints, growing pressures to conform with prevalent (progressive or liberal) social norms and forms
of expression in higher education, as well as diminished academic freedom and autonomy for individual
academics to pursue their interests through teaching and research.

The correlation of these three items was tested for reliability, generating a moderately strong relationship
(Cronbach Alpha =.598), suggesting that they could be combined to create a single consistent scale. The
recoded variables were summed to create a Cancel Culture Index, standardized to a 100-point continuous
scale for ease of interpretation where a higher score reflect experience that academic freedom, pressures
to be politically correct and respect for open debate had worsened during recent years. The multivariate
analysis examines the impact of ideology on the Cancel Culture Index after adding controls for cohort of
birth and gender, in successive models for the global sample, among those living, studying and working in
post-industrial societies, and in the US. We can look into responses in more detail by data visualization
charts illustrating contrasts and similarities among Leftwing and Rightwing scholars on each of the
separate items in the index. We can also compare the experiences of more Leftwing and Rightwing
political scientists in ten countries where we have at least 70 respondents per nation.

III: Results of the analysis

A leftwing academic culture?

As a first step, is there systematic evidence for a ‘liberal’, ‘progressive’ or a ‘leftwing’ skew in the
ideological identities of political scientists?29 This is commonly claimed in arguments about the ‘silencing’
of conservative or rightwing voices on campus, supporting suspicions about the dismissal of legitimate
challenges from alternative viewpoints and the left-wing indoctrination of students. The underlying
assumption is that political scientists cannot be neutral, so that their personal values and political
attitudes will be reflected in their teaching and research. There are obviously many ways to gauge
ideological values and beliefs, such as by asking about positions on a range of controversial issues, such
as attitudes towards LGBTQ rights, religion, racial justice, or economic inequality. For consistency with the
long tradition established by several previous major studies, the survey sought to monitor ideological
identities. This is also the most consistent approach for cross-national comparisons, where the salient
issues vary from one place to another. Respondents were asked where they would place themselves on
the following question: “Generally speaking, how would you usually describe yourself?” Responses from
1,788 political scientists were collected through a sliding scale in the questionnaire ranging from most

10
leftwing (0) to most rightwing (10). The self-identified scale position was subsequently categorized into
four categories: ‘Far-left’ (0-1), moderate left (2-4), moderate right (5-8), and far-right (9-10). For
subsequent analysis, this was simplified by dichotomizing the scale into Left (0-5) and Right (6-10).

Figure 1 illustrates the distribution of how political scientists worldwide and in the U.S. identified their
own position across the Left-Right ideological scale. Overall, replicating a pattern observed in many
previous studies of social science faculty, the distribution displayed a normal curve with a positive leftwing
skew.30 The mean score for all respondents worldwide was 4.53 on the 0-10 Left-Right scale. In total, the
majority of political scientists (58%) around the globe saw themselves as located in the moderate left of
the ideological spectrum, while another 14% positioned themselves on the ‘far left’. The ideological
imbalance of the discipline should not be exaggerated, however; just over one quarter (27%) of political
scientists worldwide placed themselves as moderate right, although at the same time very few
respondents (only 2%) saw themselves as ‘far right’.

[Figure 1 about here]

In comparison, when the sample was confined to political scientists currently studying or working in the
United States (irrespective of their country of origin or citizenship), the positive skew towards left-wing
identification strengthened. The mean score in the US sample was 3.16 on the 0-10 Left-Right scale. Two-
thirds of American political scientists (65%) saw themselves as moderate left on the ideological scale,
which an additional small group (15%) located themselves as far left. By contrast, overall one fifth (20%)
saw themselves as moderate right, but almost no respondents saw themselves as ‘far right’. To this extent,
the first proposition is confirmed in the WPS data; many surveys since Lazarsfeld’s original 1958 study
have reported a moderate left-wing or liberal skew among the American academy. 31 As one recent study
observed: “Progressives rule higher education. Their rule is not absolute. But conservatives are scarcer in
academic than in just about any other major profession.”32 The new data confirms this pattern among
political scientists working both in US universities and also globally. This underscores the common
complaint by conservatives that their voices are relatively under-represented in political science,
especially in America. Lack of intellectual diversity can be problematic within the discipline, especially in
controversial areas such as the politics of race, gender, power and inequality, limiting alternative
perspectives questioning the prevailing normative values and empirical theories embedded in the
mainstream research agenda. Several explanations of this long-standing phenomenon are possible,
including the socialization effects of higher education on ideological values and the reinforcement effects
of self-recruitment.33

11
This provides only a partial test of the cancel culture thesis, however, as an ideological skew in higher
education (especially a persistent academic culture) does not necessarily imply growing intolerance for
alternative values and beliefs, limiting intellectual debate and pluralism. This leads towards the more
critical propositions for this study: on the basis of their personal experiences, do many political scientists
feel that restrictions on academic freedom of speech, pressures for ideological conformity, and politically
correct speech have worsened in recent years? And, given the ideological imbalance in the discipline, and
their minority status, are conservative political scientists most likely to agree with these statements?

Multivariate regression is used to examine the impact of the 10-point Left-Right ideological identity scale
on the Cancel Culture Index. The simple models controls for age and gender, for the reasons discussed
earlier. To consider whether this phenomenon differs by the type of society, as may be expected by
broader cultural shifts accompanying development, Model A is restricted to 193 respondents studying or
working in the US, Model B includes 1,023 respondents in twenty-three advanced post-industrial societies
(in North America, Western Europe and Australasia), while Model C includes the pooled sample of 487
scholars in 78 developing countries worldwide.

[Table 1 and Figure 1 about here]

Table 1 shows the results of the analysis. The successive models confirm that Left-Right ideological identity
is a significant and consistent predictor of the Cancel Culture Index. The effects of ideology is stronger in
the U.S. than in all post-industrial societies; as illustrated in Figure 2, American scholars on the moderate
right and far right report experiencing worsening pressures to be politically correct, limits on academic
freedom, and in lack of respect for open debate. This reflects progressive predominance and the minority
status of conservative scholars within the discipline, observed earlier. Moreover, this was not simply
another case of American exceptionalism; in the pooled sample across all the post-industrial countries, as
predicted, more rightwing political scientists reported that, in their own experience, the cancel culture
had worsened in recent years. In the pooled sample of political scientists in developing societies, however,
the sign of the coefficient in this relationship reverses to become negative, meaning that those on the left
thought that, in their own experience, the cancel culture had worsened in higher education. The
conclusion speculates about potential reasons for these societal contrasts and their implications for
understanding this phenomenon. In terms of the other background demographic characteristics, in the
U.S., women political scientists were also significantly more likely than men to believe that the cancel
culture had got worse, but there was no different by age. In Model B, comparing scholars in post-industrial

12
societies, however, both men and the older generations were significantly more likely to believe that the
cancel culture had worsened.

Experience of respect for open debate

To understand this further, the analysis can be disaggregated to look at each of the indicators in the
summary Index. Figure 3 illustrates the distribution of assessments that respect for open debate from
diverse perspectives in academic life has changed for better or worse during the prior five years.
Agreement gets to the heart of the cancel culture thesis by implying an erosion of tolerance for intellectual
diversity and more closed minds on campus. Overall the distribution shows a normal curve (Mean=3.16
on the 1-5 scale), with the plurality view suggesting no change in the quality of open debate over time. In
total, almost as many political scientists reported on the basis of their own experience that open debate
had improved as that it had worsened. If the distribution is broken down by the dichotomous measure of
Left-Right ideological identity, however, some modest differences can be observed. Leftwing faculty were
more likely to feel that the quality of debate had not changed in their experience in recent years. By
contrast rightwing faculty were slightly more likely to believe on the basis of their own experience that
open debate has got a lot worse over time. The contrasts by ideology proved modest in size, however,
and differences should not be exaggerated.

[Figures 3 and 4 about here]

Experience of pressures to be politically correct

Figure 4 shows responses to how perceptions that pressures to be ‘politically correct’ in academic life had
altered during recent years. This concept refers to conformity towards predominant social norms, a form
of self-censorship of authentic words and actions designed to fit in and avoid opprobrium. On this issue,
there was clearer evidence for concern – and in this case one shared on both sides of the ideological
spectrum. Again, the largest plurality of respondents (39%) reported that in their own experience there
had been no change over time. But overall one third reported that these pressures had worsened
‘somewhat/a lot’ (36%) in their experience, clearly outweighed the proportion feeling more positively that
these pressures had got ‘somewhat/a lot’ better (15%). Contrasts were also observed when broken down
by Left-Right ideology. More of those on the Right thought that these pressures had got ‘somewhat/a lot’
worse (60%), more than those on the Left (45%). By comparison, 42% of political scientists on the Left
thought that there had been no change in these pressures to conform, in their experience, compared with
20% of their colleagues on the Right.

13
Experience of academic freedom

Figure 5 examines the distribution towards the last indicator under comparison: experience of academic
freedom to teach and research. Of course this does not specify the reasons for any perceived change, with
restrictions which may arise from practices associated with a ‘cancel culture’, or alternatively from
broader structural developments and job dissatisfaction, due to the growth of bureaucratic control and
managerialism in university administration, the casualization of employment and weakening of tenure
security in academic careers, growing pressures to teach, and/or restrictions on academic freedom more
generally in societies experiencing democratic backsliding, such as Hungary and Turkey. Nevertheless, the
question provides an important indicator of the quality of intellectual life and feelings of autonomy in the
profession. Here the largest plurality (47%) thought that, in their experience, academic freedom had
deteriorated ‘somewhat/a ‘lot’. Another third reported no change, while a fifth responded that freedom
had improved ‘somewhat/a lot’ (20%). On this issue, however, contrary to expectations, those on the Left
proved more negative than those on the Right; almost half of those on the Left (49%) thought academic
freedom had deteriorated ‘somewhat/a lot’, compared with 39% of those on the Right.

Cross-national comparisons

How do these patterns vary across countries? Are differences between left and right observed in the U.S
another case of American exceptionalism – or do they reflect general patterns found in comparable post-
industrial societies? In many cases, unfortunately we have too few respondents for reliable analysis. But
there are ten diverse countries with a minimum of at least 70 respondents. As shown in Figure 6, in fact
there are remarkably similar experiences reported in a range of post-industrial societies – especially in the
Anglo-American democracies which share similar historical traditions and systems of higher education.
Thus, the US pattern where those on the right report worse experiences of the cancel culture is clearly
reflected in the cases of Canada, Australia, the UK – but also in the Western European cases of Germany,
Italy and the Netherlands, as well as Russia and (to a lesser extent) Sweden. The main exception is Nigeria
– where there are no significant differences between Left and Right. Not surprisingly, responses to the
question about reported experiences of pressures to be politically correct show a similar distribution, with
greater experience of this reported by rightwing scholars in all the countries except Nigeria. Clearly there
is room for further comparative research into several possible reasons for this, as discussed below, but
larger national samples are needed to analyze this with any degree of reliability.

14
 Conclusions and implications

Debates about the cancel culture have intensified in recent years as part of deepening ideological and
value cleavages dividing progressive liberals and social conservatives. Of course, the public outing of
heretics and dissenters is nothing new, as exemplified by the fanatical prosecution of religious non-
conformists, the medieval punishment by stocks and pillory in the town square, or the notorious Salem
witch trials. Contemporary media speculation about the shaming of celebrities and leading public figures
in the worlds of popular entertainment, publishing, and politics, as well as colleges and universities,
encapsulated in the ‘cancel culture’ label, has often generated more political heat than intellectual light.
Yet there are genuine grounds for concern if a cancel culture has evolved, as critics charge, from the
legitimate criticism of socially offensive words and deeds to become a tidal wave eroding tolerance of
dissent, stifling free speech, and enforcing a progressive of left-wing orthodoxy among professors,
administrators, and students. But is there evidence, beyond some specific anecdotes, that this has actually
occurred?

Several key findings can be highlighted.

Firstly, as predicted from the series of previous surveys of academics in the US and other Western
societies, the WPS survey data confirmed the leftwing skew in the discipline of political science. The extent
of the imbalance should not be exaggerated, as the majority of scholars described their position on the
ideology scale as moderately leftwing on the political spectrum, rather than far left. Overall a substantial
minority - around one quarter of political scientists worldwide – also identified as moderately rightwing.
The leftwing predominance in the discipline was stronger in the U.S., however, than worldwide.

Yet this imbalance in the personal ideological leanings in the profession is far from sufficient, by itself, to
settle the heated debate about the effects of the liberal/left hegemony on issues of academic freedom of
expression and social pressures for contrarians to conform with progressive values. The evidence
presented in this study generated a summary Cancel Culture Index, reflecting experience of growing
restrictions on academic freedom of speech, pressures for ideological conformity, and the enforcement
of politically correct speech. The results of the models confirmed the significant effects of Left-Right
ideology which consistently predicted scores on this index.

Most importantly, however, the effects varied in direction by the type of society under comparison. Models
suggest that in the United States, and in 23 comparable post-industrial societies, self-identified rightwing
political scientists were most likely to report personal experience of a worsening cancel culture. By

15
contrast, among those studying and working in universities and colleges in the 78 developing societies, it
was the self-identified leftwing scholars who reported a worsening cancel culture.

What can explain these contrasts? Further research with macro-level comparative evidence is needed to
investigate the reasons for the cross-cultural differences, since this could be attributed to factors such as
the role of socioeconomic development and modernization, patterns of freedom of expression,
democratization, and the type of regime in each society, the polarization of political partisanship, the role
of long-standing cultural traditions, and structural contrasts in the institutions and policies of higher
education.

One plausible explanation draws upon Elisabeth Noelle-Neumann’s classic concept of a ‘spiral of silence’,
developed almost four decades ago, drawing upon studies in social psychology and interpersonal
communications. This idea describes situations where, for fear of social isolation or loss of status, people
are hesitant to express authentic opinions contrary to prevalent social norms.34 The more that individuals
feel that their opinion reflects majority opinion, however, the more willing they become to voice it in
public discourse.

But the balance of public opinion is far from static, instead it is well documented to evolve over time in
line with processes of societal development. The notion of a spiral of silence can be usefully combined
with the theory and evidence for value change presented in Cultural Backlash.35 A wealth of cross-national
and time-series survey data, including from the European/World Values Surveys, demonstrates that
socially conservative values usually continue to prevail as the majority view in many developing countries.
This includes attitudes, beliefs and values on social issues dividing conservatives and liberals, such as
towards the division of sex roles for women and men, non-binary gender identities, homosexuality and
LGBTQ rights, traditional views of sexuality, family and marriage, the importance of religion, and feelings
of nationalism and nativism. In many developing countries, the traditional values held by conservative
scholars and intellectuals will therefore continue to reflect the majority culture. But liberal college
professors and students holding progressive views in these countries may feel pressures to conform with
the predominant social norms and values in the classroom and common room.

By contrast, the balance of public opinion on a wide range of issues has gradually shifted in a more socially
liberal and progressive direction to become the majority view in public opinion in many affluent post-
industrial societies in Western Europe and North America. The balance of public opinion in society has
been transformed in a more liberal direction towards issues such as LGBTQ rights, secularization, gender
equality, ethnic diversity, and racial inclusion. This shift is gone furthest among the groups who have been

16
at the vanguard of these changes, namely the younger generation (hence students) and college educated
populations (hence professors). 36 As a result, to avoid social isolation on college campuses, and potential
loss of professional status and opportunities, rightwing scholars holding traditional conservative beliefs
and attitudes are likely to feel growing pressures to conform with evolving informal social values both in
the academy and more broadly in postindustrial societies. Academics may also find themselves out of step
with contemporary legal policies governing higher education and evolving norms of expression and
behavior in the workplace, including practices which used to be widely tolerated just a few decades ago.
Well documented processes of long-term generational cultural change in many Western societies means
that the proportion of those holding traditionally socially conservative values has gradually experienced a
tipping point in recent decades, as this group shifts from hegemonic to minority status on college
campuses and in society, heightening ideological and partisan polarization. In this regard, the reported
experience of a chilly climate in academia among rightwing scholars, documented in this survey,
exemplifying the classic spiral of silence phenomenon, seems likely to reflect their reactions to broader
cultural and structural shifts in post-industrial societies.

Finally, when the indicators in the composite index were disaggregated into their component parts, there
were some important observable contrasts. Leftwing faculty are more likely than those on the right to
believe that there has been little or no change in respect for open academic debate and pressures to be
politically correct. Given the predominance of progressive liberalism on college campuses, those on the
left may be simply unaware of the experience of more conservative colleagues – and deny that there is a
problem - or intense polarization over issues of identity politics may make them unsympathetic to these
claims. By contrast, however, a broader consensus can be observed among political science scholars about
experiencing deteriorations in academic freedom to teach and research, which may reflect broader
structural changes in higher education, as much as ideological shifts.

The survey which is the basis of this study builds upon a long tradition developed in the sociology of the
academy. Most previous work on these issues has focused on scholars in higher education working in the
U.S. and in similar affluent Western societies. One advantage of the WPS survey is that this expands the
comparisons worldwide, and thus facilitates insights into different cultures. Nevertheless, the data
considered in this study is limited in several ways. In particular, the survey asks scholars about their
ideological identities and their direct experience of changes in the profession, but the latter cannot be
separated in practice from perceptions of the academy. Perceptions, by themselves, are important for the
social construction of reality. If a scholar says that they feel social pressures to confirm with predominant

17
values, or that they self-censored their authentic words or actions to avoid ostracism, then we should take
them at their word. Further survey measures would be useful, however, to explore experience of different
aspects of the cancel culture in more depth, for example whether conservatives felt that they had failed
to be appointed or promoted because of their political views, whether they had been involved in
organizing any public events which had subsequently been cancelled, or whether they have ever felt
uncomfortable speaking up in faculty meetings or the class room because of their values. Larger national
samples, especially of scholars in varied developing countries, would be invaluable to pursue the cross-
national comparisons. Finally, the survey is restricted to the discipline of political science. It is unclear
whether similar generalizations can be observed in related social science disciplines, such as sociology,
economics, and social psychology, as well as in the humanities and natural sciences, and equivalent survey
data of the professoriate facilitating comparisons across academia could explore these issues further.

18
Figure 1: Distribution of Left-Right identities among political scientists

Worldwide US only

Note: Q30: “Generally speaking, how would you usually describe yourself ideologically?” Scale response from ‘Most leftwing’ (0) to ‘Most rightwing’ (10).

Worldwide N. 1,733. Mean=3.54 Positive Skew=0.495.

U.S. responses only. N. 221. Mean=3.16 Positive Skew=0.737.

Source: World of Political Science survey, 2019

19
Figure 2: Experience of changes in the Cancel Culture Index by Left-Right identity, US only

Source: World of Political Science survey, 2019, US only N. 281.

20
Figure 3: Experience that respect for open debate in academic life has changed for better or worse by Left-Right identity

Notes: Q10-11: “Academic life is often thought to be in a state of change. Using the following scale, based on your experience, please indicate whether you think
the quality of the following aspects of academic life have changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse,
got a lot worse.” ‘Respect for open debate from diverse perspectives’ All countries worldwide.

Source: World of Political Science survey, 2019

21
Figure 4: Experience that pressure to be ‘politically correct’ in academic life has changed for better or worse by Left-Right identity

Notes: Q10-11: “Academic life is often thought to be in a state of change. Using the following scale, based on your experience, please indicate whether you think
the quality of the following aspects of academic life have changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse,
got a lot worse.” ‘Pressures to be 'politically correct’ All countries worldwide.

Source: World of Political Science survey, 2019

22
Figure 5: Experience that academic freedom to teach and research has changed for better or worse by Left-Right identity

Notes: Q10-11: “Academic life is often thought to be in a state of change. Using the following scale, based on your experience, please indicate whether you think
the quality of the following aspects of academic life have changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse,
got a lot worse” ‘Academic freedom to teach and research’ All countries worldwide.

Source: World of Political Science survey, 2019

23
Figure 6: Experience of changes in the Cancel Culture Index by Left-Right identity, 10 societies

Note: The standardized 100-point Cancel Culture Index is created by combining three items using the following question “Academic life is often thought to be in
a state of change. Using the following scale, based on your experience, please indicate whether you think the quality of the following aspects of academic life
have changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse, got a lot worse.” ‘Academic freedom to teach and
research’, ‘Respect for open debate from diverse perspectives’, and ‘Pressures to be politically correct.’ 10 countries worldwide each with 70-281 respondents.
Source: World of Political Science survey,

24
Figure 7: Experience of pressures to be politically correct by Left-Right identity, 10 societies

Notes: Q10-11: “Academic life is often thought to be in a state of change. Using the following scale, based on your experience, please indicate whether you think
the quality of the following aspects of academic life have changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse,
got a lot worse.” ‘Pressures to be 'politically correct’. 10 countries worldwide each with 70-281 respondents.

Source: World of Political Science survey, 2019

25
Table 1: Predicting scores on the Cancel Culture Index

Model A: Model B: Model C:

US only Post-industrial societies Developing societies
B Std. Beta Sig. B Std. Beta Sig. B Std. Beta Sig.
Error Error Error
Left-Right ideological identity (Left 0, Right 10) 2.05 0.57 0.26 *** 0.76 0.24 0.10 *** -1.19 0.38 -0.14 ***
Gender (Women 1/ Men 0)) 1.45 2.15 0.05 * -1.20 0.88 -0.04 ** 0.51 1.75 0.01 N/s
Age (Year of birth) -0.01 0.06 -0.01 N/s -0.08 0.03 -0.08 *** 0.09 0.07 0.06 N/s

(Constant) 72.1 223.1 100

Adjusted R2 0.05 0.02 0.05

N. respondents 193 1023 487

N. countries 1 23 78

Note: Multivariate regression analysis. Dependent Variable: The 100-pt Cancel Culture Index (Low score= Got better, High score=Got worse) Sig P. * 05, **. 01
*** .001 The standardized Cancel Culture Index is created by combining three items using the following question “Academic life is often thought to be in a state
of change. Using the following scale, based on your experience, please indicate whether you think the quality of the following aspects of academic life have
changed over the last five years. Got a lot better, got somewhat better, no change, got somewhat worse, got a lot worse.” ‘Academic freedom to teach and
research’, ‘Respect for open debate from diverse perspectives’, and ‘Pressures to be politically correct.’

Source: World of Political Science survey, 2019.

26
Acknowledgments: The World of Political Science survey could not have been possible without the
generous assistance and support of Martin Bull, Kris Deschouwer, and Rebecca Gethen at the
ECPR, as well as Marianne Kneuer and Mathieu St-Laurent at IPSA, and all colleagues who
participated in the survey.

1
Aja Romani, 30 Dec 2019. Why we can’t stop fighting about cancel culture.’ Vox.
https://www.vox.com/culture/2019/12/30/20879720/what-is-cancel-culture-explained-history-debate
2
Ryan Lizza. 22 July 2020. ‘Americans tune in to ‘cancel culture’ — and don't like what they see.’ Politico
https://www.politico.com/news/2020/07/22/americans-cancel-culture-377412
3
Elizabeth Bradford Lightfoot. 2019. ‘Consumer activism for social change’. Social Work 64(4): 301-9.
4
John Stuart Mill. 1865. On Liberty.
5
David Brooks. 23 July 2020. ‘The Future of Nonconformity.’ New York Times.
https://www.nytimes.com/2020/07/23/opinion/substack-newsletters-writers.html
6
Greg Lukianoff and Jonathan Haidt. 2018. The Coddling of the American Mind: How Good Intentions and Bad
Ideas Are Setting Up a Generation for Failure. New York: Penguin books.
7
J. Williams. 2016. Academic freedom in an age of conformity: Confronting the fear of knowledge. New York, NY:
Palgrave Macmillan.
8
https://www.insidehighered.com/news/2017/02/24/education-secretary-criticizes-professors-telling-students-
what-think
9
President Donald Trump. 4th July 2020. Speech at Mount Rushmore, South Dakota.
https://www.whitehouse.gov/briefings-statements/remarks-president-trump-south-dakotas-2020-mount-
rushmore-fireworks-celebration-keystone-south-dakota/
10
Harper’s Magazine. 7 July 2020. ‘A Letter on Justice and Open Debate.’ https://harpers.org/a-letter-on-justice-
and-open-debate/
11
Neil Gross and Ethan Fosse. 2012. ‘Why are professors liberal?’ Theory and Society, 41(2), 127–168.
12
Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
12
Jon A. Shields and Joshua M. Dunn Sr. 2016. Passing on the Right. Conservative Professors in the Progressive
University. New York: Cambridge University Press.
13
Pippa Norris and Ronald Inglehart. 2019. Cultural Backlash. New York: Cambridge University Press.
14
https://www.pewresearch.org/politics/2019/07/22/how-americans-see-problems-of-trust/
15
Equality and Human Rights Commission. Feb 2019. Freedom of expression: A guide for higher education
providers and students’ unions in England and Wales. www.equalityhumanrights.com
16
Remi Adekoya, Eric Kaufmann and Tom Simpson. 2020. Academic Freedom in the UK. Policy Exchange.
https://policyexchange.org.uk/publication/academic-freedom-in-the-uk-2/
17

https://www.telegraph.co.uk/news/2019/12/05/conservative-students-feel-unable-express-views-british-
universities/

27
19
Carll Everett Ladd and Seymour Martin Lipset. 1976. The Divided Academy: Professors and Politics NY: W.W.
Norton.
20
Everett Carll Ladd and Seymour Martin Lipset. 1976. The Divided Academy: Professors and Politics. New York:
W.W. Norton & Company.
14
Albert Halsey and Martin Trow. 1971. The British Academics. Oxford University Press.
22
Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
23
Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
23
Jon A. Shields and Joshua M. Dunn Sr. 2016. Passing on the Right. Conservative Professors in the Progressive
University. New York: Cambridge University Press. P.2.
24
Herman G. van de Werfhorst. 2019. ‘Are universities left-wing bastions? The political orientation of professors,
professionals, and managers in Europe.’ British Journal of Sociology 71(1): 47-73.
25
Pippa Norris and Ronald Inglehart. 2019. Cultural Backlash. New York: Cambridge University Press.
26
Cf Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
27
For details, see Pippa Norris. 2020. ‘The World of Political Science: Internationalization and its Consequences.’ In
Political Science in Europe: Achievement, Challenges, Prospects Eds. Thibaud Boncourt, Isabelle Engeli, and Diego
Garzia. Essex/New York: ECPR Press/Rowman and Littlefield.
28
The original survey items were recoded for consistency with the thesis, so that a higher score on each indicated
negative evaluations, meaning that respondents perceived a worsening ‘cancel culture’.
29
It should be noted that the terms ‘liberal’, ‘left’ and ‘progressive’ are used interchangeably in this study, as are
‘rightwing’ and ‘conservative’, following traditional American usage, although strictly speaking these are separate
concepts.
30
Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
31
Neil Gross and Solon Simmons. 2007. ‘The Social and Political Views of American Professors.’
https://www.researchgate.net/publication/228380360_The_Social_and_Political_Views_of_American_Professors
32
Jon A. Shields and Joshua M. Dunn Sr. 2016. Passing on the Right. Conservative Professors in the Progressive
University. New York: Cambridge University Press.
33
Neil Gross. 2013. Why are professors liberal and why do conservatives care? Cambridge, MA: Harvard University
Press.
34
Elisabeth Noelle-Neumann. 1974. The Spiral of Silence. Chicago: University of Chicago Press.

35
Pippa Norris and Ronald Inglehart. 2019. Cultural Backlash. New York: Cambridge University Press.
36
Ronald Inglehart. 1997. Modernization and Post-modernization. Princeton University Press; Christian Welzel.
2014. Freedom Rising: Human Empowerment and the Quest For Emancipation. New York: Cambridge University
Press; Ronald Inglehart. 2018. Cultural Evolution: People's Motivations are Changing, and Reshaping the World.
New York: Cambridge University Press; Pippa Norris and Ronald Inglehart. 2019. Cultural Backlash. New York:
Cambridge University Press.

28