Nationalism, Status, and Conspiracy Theories: Evidence from Pakistan

Abstract

Does nationalism increase beliefs in conspiracy theories that frame minorities as subversives? From China to Russia to India, analysts and public commentators increasingly assume that nationalism fuels belief in false or unverified information. Yet existing scholarly work has neither theoretically nor empirically examined this link. Using a survey experiment conducted among 2,373 individuals and 6 focus groups with 6–8 participants each, for a total of 50 individuals, we study the impact of nationalist sentiment on belief in conspiracy theories related to ethnic minority groups in Pakistan. We find that nationalist primes – even those intended to emphasize the integration of diverse groups into one superordinate national identity – increase belief in statements about domestic minorities collaborating with hostile foreign powers. Subgroup analysis and focus groups suggest that nationalism potentially increases the likelihood that one views rights-seeking minorities as undermining the pursuit of national status.

Conspiracy theories that cast rights-seeking minority groups as agents of geopolitical rivals are rampant in a variety of political contexts. The Chinese government, for example, has long sought to undermine the Uighurs, the country's Muslim minority group, by attributing foreign influence as the fundamental driver of Uighur political dissent (Kuo and Mylonas 2022; Roberts 2020). In democratic India, the ruling Bharatiya Janata Party (BJP) routinely paints members of the minority Muslim population as anti-national and as working on behalf of neighbouring Pakistan (Human Rights Watch 2020). In Pakistan, the subject of this study, rights-seeking ethnic minority groups are frequently accused of being ‘foreign agents’ acting at the behest of external powers.

Troublingly, such anti-minority conspiracy theories often exist alongside powerful narratives furthering a commitment to one's nation – what scholars call nationalism (Anderson 1983). In many contexts, enduring nationalist narratives rely on false or unverified information, partly due to concerted efforts by state authorities and political parties. For example, the Chinese government instrumentalized Han-centred nationalist notions in its discourse around the place of ethnic minorities in the Chinese polity. In India, conspiracies about Muslims are commonplace within Hindu nationalist narratives propagated by the BJP.

What is the relationship between nationalism and conspiracy theories about minority groups? Some analysts and journalistic accounts assume that nationalism causes an increase in belief in false information more broadly (Chakrabarti 2018; Hemon 2020). As one report argues, ‘facts often take a backseat when nationalistic tendencies of ordinary citizens kick in’ (Gill 2018). These reports are consistent with findings from international relations, showing that nationalism can affect a range of outcomes related to foreign policy-related preferences (Mylonas and Tudor 2021; Powers 2021). They are also consistent with studies that examine how the creation of the nation-state and the subsequent rise of nationalism was accompanied by a suspicion of minorities, such that the ‘belief that disloyal members of society, backed by outside powers, posed a persistent threat to the territorial integrity of a state has often unified [the rest of] the imagined community’ (Radnitz and Mylonas 2022, 4). However, scholars have not empirically examined the causal connection between nationalism and belief in conspiracy theories, a gap this study aims to fill by exploring this relationship in Pakistan.

First, we consider the possibility that nationalism leads to an increase in belief in conspiracies about minorities, premised on the traditional understanding of nationalism as producing hawkish attitudes against perceived enemies of the nation. Second, we test if nationalism's impact on such a belief is conditioned on the type of nationalism being evoked. Both seminal works (see, Gellner 1983; Greenfeld 1992) and more recent research (Powers 2021; Tudor and Slater 2021) have sought to disaggregate nationalism to capture a variety of appeals, arguing that not all nationalism is alike. In line with this scholarship, we introduce a new typology of nationalism, with each type evoking the country's status through different means: (1) chauvinistic nationalism emphasizes a nation's military and economic strength and superiority; (2) cohesive nationalism emphasizes the integration of diverse groups into one superordinate national identity; and, (3) international legitimacy nationalism emphasizes a nation's standing and prominent place in the international system. Third, we examine possible mechanisms whereby nationalist sentiment may lead to an increase in belief in such conspiracies.

We utilize a mixed-method approach to study these questions. To begin, we carried out a survey with an embedded experiment over the phone on a representative sample of 2,373 cell phone users in Pakistan in December 2020. Additionally, we conducted six focus groups in Islamabad, Lahore, Peshawar, and Karachi – the primary urban centres in the country – with 6–8 participants each, for a total of 50 individuals.

Three main results emerge. First, we find that priming nationalist sentiment increases belief in anti-minority conspiracies relative to a control condition of no prime. Second, of the three types of nationalism we consider, international legitimacy nationalism increases conspiracy belief the most. Contrary to our expectation that a more inclusive version of nationalism would reduce such belief, cohesive nationalism also increases belief in ‘minority-as-subversives’ conspiracy theories. Third, we find suggestive evidence that nationalism leads to higher conspiracy uptake among those respondents more concerned with Pakistan's status globally. Our focus group discussions suggest that nationalism leads individuals to believe such conspiracies due to concerns about these groups' (negative) effects on Pakistan's global reputation, which run contrary to the perceived duty of a citizen to present the state in a positive light. We interpret this argument as falling within the broader category of motivated reasoning processes, explicitly triggered by concerns over international reputation.

This paper contributes to several literatures and policy debates. It expands the rich academic literature on causes of belief in conspiracies and misinformation by studying the effect of nationalism, a hitherto ignored variable in these conversations and debates. Our findings also support the theoretical view that nationalism is a powerful force that exerts a strong influence on belief in conspiracy theories portraying minorities as subversives. The results contrast with scholarship which suggests that more inclusive forms of nationalism can help counter distrust of minorities and reduce conflict. As such, these findings contribute to scholarly and policy debates on minority rights and the health of democracies in the age of populism and social media. They raise concerns that nationalism, by adding to pernicious beliefs about vulnerable groups like minorities, can harm democracy and worsen the quality of human rights and civil liberties in a given polity.

Theoretical Foundations of Nationalism and Conspiracy Theories

We follow Guess and Lyons (2020), who define conspiracy theories as including ‘a belief that a hidden group of powerful individuals exerts control over some aspect of society’ and who act ‘to achieve some malevolent end’ (Barkun 2003, 3). Conspiracy theories are distinct from but adjacent to misinformation, with the important distinction that while misinformation is provably false, conspiracy theories are often unverifiable (Sunstein and Vermeule 2009). However, we expect our findings to speak broadly to the literature on misinformation because conspiracy theories and misinformation coexist in the ecosystem of false information. This study also contributes to research on ‘fifth columns’ or ‘domestic actors [perceived to] work to undermine the national interest, in cooperation with external rivals of the state’ (Radnitz and Mylonas 2022, 3). Anti-minority conspiracy theories are central to narratives surrounding fifth-column charges and accusations in several contexts (Radnitz and Mylonas 2022). This section integrates insights from these disparate literatures to explore whether and how nationalism might impact belief in information that frames minorities as subversives.

International relations scholars have consistently found that nationalism affects a range of preferences, including a propensity for conflict (Powers 2021; Snyder 2000), foreign policy preferences (Ko 2020), and attitudes towards ethnic minorities (Charnysh, Lucas, and Singh 2015; Kalin and Siddiqui 2020). Studies have also shown that nationalism can exert a powerful influence on cognitive processes, alter preferences, and push leaders and citizens towards costly policies (Hafner-Burton et al. 2017). Research on misinformation, meanwhile, finds that respondents are susceptible to believing false information that conforms to their worldview (Greenhill and Oppenheim 2017; Nyhan and Zeitzoff 2018). In the United States, for example, partisanship is a consistent predictor of belief in fake news (Berinsky 2012; Nyhan 2020), and these findings have been replicated in developing countries (Siddiqui 2020).

Given nationalism's ability to alter individual preferences and political misinformation's ability to make inroads into the mass public, it is possible that exposure to nationalism increases the uptake of conspiracy theories about minorities by triggering concerns about a nation's enemies. In this view, nationalism is an ideology of commitment to one's nation, and the defence of the nation against its enemies is vital (Haas 1986). Scholars have argued that nationalism is a driver of militarism because of the centrality of the nation's adversaries to nationalist projects (Van Evera 1994). Public opinion research has found support for the notion that nationalism fuels belief in militaristic preferences (De Figueiredo and Elkins 2003; Federico, Golec, and Dial 2005; Kemmelmeier and Winter 2008).

Modern nationalist projects have viewed minorities as tied to their geopolitical adversaries since the late colonial period (Wimmer 2013). In particular, as the process of decolonization divided ethnic groups across borders, ethnic nationalism challenged post-colonial national boundaries in various regions. In some of these decolonized states, nationalist sentiment developed a decidedly negative view of ethnic minorities, prompting a turn to explicit and implicit appeals that portrayed ‘minority populations as a threat to the majority’ (Radnitz and Mylonas 2022). This tendency was particularly pronounced in states embroiled in rivalries and territorial disputes with neighbouring nations (Sambanis, Skaperdas, and Wolforth 2015; Weiner 1971) and was exacerbated where minority groups had ethnic, cultural, historical, and/or linguistic ties to groups in neighbouring states. In these contexts, where ethnic grievances and conflicts with state authorities were more likely (Cederman, Wimmer, and Min 2010; Cederman, Gleditsch and Buhaug 2013), states frequently channelled this distrust into nationalist narratives, often embedded in educational materials, such as school textbooks, and depicted minority ethnic and sectarian groups as potential threats to the nation's unity.

Yet nationalist projects need not only be concerned with the nation's adversaries. Several classic works on nationalism have pointed to ‘categories of nationalism’ on the premise that they engender distinct nationalistic identities, such as liberal, ethnic, and diaspora national identities (Gellner 1983). International relations scholars have also recognized differences in nationalism and speculated on the varied effects of different types of nationalist imaginings on outcomes of war and peace (Van Evera 1994). A growing body of literature has affirmed these insights, showing that nationalism is not uniform and that variations in nationalist sentiment can have differential effects on key outcomes (Mylonas and Tudor 2021; Powers 2021). Tudor and Slater (2021), for example, argue that nationalism can fall into two broad categories – inclusive or exclusionary in ethnoreligious terms – and that nationalism's effect on key outcomes, such as democracy, depends not on nationalism per se but on its relationship to social groups within the state. Similarly, Powers (2021) argues that inter-group norms of nationalist projects, such as unity, equality, and reciprocity, shape nationalism's effects on hawkish foreign policy attitudes.

Building on these general insights and motivated by our case study of Pakistan (discussed next), we consider the possibility of three distinct types of nationalism (summarized in Table 1): chauvinistic, cohesive, and international legitimacy. Our conceptualization of nationalism focuses on the varied paths to the status of the nation and predicts how that shapes peoples’ views of minority groups, including conspiracy theories about them.

Table 1. Types of nationalism

The first type, which we call chauvinistic nationalism, elevates the nation's status by referring to uniqueness and superiority along a variety of possible dimensions relative to outsiders. This form of nationalism often invokes material capabilities (such as the military or the economy), while downplaying the adversary's same attributes to elevate national status. Such nationalism also delineates who constitutes the ‘in-group’ and who is an ‘out-group’, which can range from foreign adversaries to disloyal citizens or domestic groups (Abdelal et al. 2006; Ko 2020). Such delineation heavily rests on a relational comparison casting the in-group as being distinct and superior in their history, tradition, and behaviour to the nation compared to the out-group (Abdelal et al. 2006).

The second, cohesive nationalism, elevates status by binding heterogeneous groups within a nation and emphasizing commonalities and equality between all social groups. Anderson (1983) originally posited the idea that people could become bound by the sentiment of a shared community and common fate and thus an imagined community – or nation. For him, however, this sense of community was an unintended outcome of the expansion of literacy and print capitalism. This spirit of shared community can also be fostered purposefully. As Barth (1969) argued, in the context of ethnicity, identities are not static; groups of people can choose to emphasize or de-emphasize certain traits or markers, which can lead to the formation of new identities. Indeed, recent work shows that recategorization of oneself and others into a single superordinate national group is possible (Charnysh, Lucas, and Singh 2015; Kalin and Siddiqui 2020). Accordingly, nationalism can tout the nation's ability to be a ‘big tent’, inclusive enough to accommodate a broad spectrum of backgrounds and able to acknowledge linguistic, ethnic, and religious cleavages, and see this as elevating status.

While we hypothesize that cohesive nationalism will decrease belief in conspiracies about rights-seeking minority groups as subversives, it nonetheless runs the risk of alienating those minorities who do not fit the nationalist script. That is, the focus on the ‘good’ minority – those who partake in elevating the country's status rather than bringing attention to its ills – is a key feature of cohesive nationalism. When a particular individual or group does not fall within this category by, for example, demanding greater rights or drawing attention to its problems, it may be perceived as bringing disrepute to the nation, and this form of inclusive nationalism may backfire. The primary audience of such nationalism is domestic, with a focus on how various internal groups work together to enhance a country's status. Thus, the content of such nationalism focuses on arenas where the state is able to showcase a pluralistic and unified front.

The third nationalism, international legitimacy nationalism, ties status to the country's standing in the international system (Larson and Shevchenko 2019; Renshon 2017). In this view, countries’ statuses depend not just on material abilities and resources but on how the country is perceived to contribute to and viewed by other nations in the international system (Larson and Shevchenko 2019). As a result, status may be elevated by a country's contributions to the international sphere, especially in pursuit of peace, providing aid and assistance to other countries bilaterally, or playing a key role in international organizations such as the United Nations. Here, the element of national superiority and status derives from actual or perceived validation by international organizations, major powers, and allies.

Table 1 lays out the three types of nationalism, the source of status under each, and our predictions about how such nationalist sentiment would affect belief in anti-minority conspiracy theories. Under chauvinistic nationalism, belief in conspiracies about subversive minorities may increase by heightening concerns about minorities damaging national status by advancing nefarious designs of geopolitical adversaries. This is especially true if minorities seek rights, as it may be perceived as signalling that the nation is deficient in some fashion. By contrast, under cohesive nationalism, tying status to being inclusive by incorporating minorities into the nation could reduce the willingness of respondents to view minorities negatively, and to believe in conspiracies portraying minority groups as subversives. It is nonetheless possible that cohesive nationalism could also increase distrust of rights-seeking minority groups, as noted earlier, and, in turn, increase belief in conspiracies about them. In particular, social identity theory would suggest that social categorization into a cohesive national community entails condemnation of ‘black sheep’ who do not follow group norms: subversive minorities fall into this category. Finally, under international legitimacy nationalism, we expect that, because minorities have no apparent explicit relevance to the pursuit of status, respondents would be neither more nor less likely to believe conspiracies about minorities as subversive.

In the remainder of this paper, we demonstrate that contrary to our expectations, all forms of nationalism lead to increased belief in conspiracies and explore possible explanations for this finding.

Empirical Context

In this section, we assess the aforementioned theoretical implications in the Pakistani context, which is rife with both nationalism and conspiracy theories.

Nationalism in Pakistan

Since the country's formation in 1947, following its Partition from India, Pakistan's military and political elites have offered – and sought to socialize the country's population into – a particular vision of the nation and state. In its simplest form, this vision sees Pakistan as a home to the Muslims of South Asia. It views Islam – which played a key role in the creation of the country and partition of British India – as a crucial source of internal cohesion and legitimacy. India plays a critical role in the nationalist discourse, both as the symbolic other against which Pakistan defines itself, as well as the perennial, existential threat facing the country and its people (Cohen 2004). India and Pakistan have fought four wars since 1947 and relations remain tense, particularly because of rival claims over Kashmir, with the countries approaching war as recently as 2019.

Ethnolinguistic divisions have shaped this nationalist vision. Pakistan is home to numerous different ethnic groups: Punjabis make up about 45 per cent of the country's population, followed by Pashtuns, who are the second largest ethnic group with 15 per cent of the country's population. Sindhis and the Baloch comprise 14 per cent and 3.6 per cent of the population, respectively. Successive Pakistani governments have struggled to balance ethnic aspirations with their desire for strong central control. Leaders of the newly independent Pakistan, in particular, faced an immediate challenge in governing the country's Eastern wing with a majority of ethnic Bengalis. They responded to political demands from the East with repressive strategies, choosing to define Pakistani national identity in opposition to exclusive ethnolinguistic claims, with rights-seeking ethnic groups framed as traitorous or seditious (Butt 2017). This approach culminated in the separation of East Pakistan – and the formation of Bangladesh – after a brutal civil war in 1971. India's support of the independence movement, which led to the creation of the state of Bangladesh, heightened fears within Pakistan that India was working to break up the country further. This narrative continues to pervade national discourse today and even justifies military budgets and spending (Siddiqa 2007).

After the formation of Bangladesh, ethnic nationalism along the Western border remained a major concern. That the main ethnic groups along the border cross state boundaries – Afghanistan has a large Pashtun population and Iran a sizeable Baloch one – fueled suspicion among Pakistani authorities, who rejected ethnic nationalist claims as unacceptable irredentism (Ahmad 2019). The modern-day insurgency in Balochistan and the revival of Pashtun nationalism in the form of the non-violent protest movement, the Pashtun Tahafuz Movement (PTM), evoke similar concerns among state officials and political elites.

Over the last decade, the pursuit of international status and reputation has become a key component of Pakistani nationalist discourse and has been linked by prominent Pakistani commentators to increased national standing and power. In the words of former diplomat Maleeha Lodhi, ‘Increasingly, a country's reputation and positive image count and play a pivotal role in earning respect and enhancing its diplomatic weight in international affairs’ (2021). The pursuit of status is traceable, at least in part, to the politics and pressures of the war on terror when Pakistan's military ruler General Pervez Musharraf, hoping to curry favour with then President George W. Bush, sought to distinguish Pakistan as an example of an ‘enlightened’ Muslim society. Later, when the US government accused Pakistan of playing a ‘double game’ and pressured it to do more to curb the safe haven of the Taliban insurgency in Afghanistan, Pakistani elites responded by seeking international recognition for their losses in the brutal violence of the Tehreek-e-Taliban Pakistan and Pakistani military sacrifices to mitigate the violence (Siddiqa 2009).

However, as Western criticism persisted due to the war in Afghanistan, and the Indian government assertively framed Pakistan as a backer of terrorist groups in the region, Pakistani elites sought to counter this messaging by projecting a soft and positive image of the country. As summed up by a (later) editorial in the right-of-centre newspaper The Nation, ‘It appears that we have not been able to cope with these diplomatic challenges effectively to rinse our image among the comity of nations and eliminate the label of a terror-sponsoring state through preemptive and robust diplomacy’ (The Nation 2023.) As a result, the state and civil society alike pursued various programmes and campaigns to advance a positive image of the country because ‘it is crucial for a nation like ours to focus on projecting a positive image internationally’ (The Nation 2023). In 2018, the-then director general of the Pakistani military's media wing, the ISPR, announced a campaign to ‘project a positive image of the country’ (Kaifee 2018), and in 2020, Prime Minister Imran Khan built on the Army's information campaign by launching a social media campaign with the similar goal of raising Pakistan's reputation and status (Daily Times 2020).

These themes and narratives of nationalism – since the country's formation to a more contemporary period – are prevalent in Pakistan's textbooks, television screens, and, increasingly, on social media. Yet not all nationalist articulations are the same. Over time different political and social actors have emphasized different elements of superiority and status. While media content sometimes emphasizes multiple facets of nationalism, there are still abundant examples of social and traditional media content that align closely with the three theoretical categories we conceptualize in the previous section. We present three such examples in Table 2.

Table 2. Examples of types of nationalism in Pakistan

^a Islamic R Voice. ‘Power Of Pakistani Taimur Missile |Pakistan ICBM Missiles Technology 2020|New York Times Report.’ 22 October 202. https://youtu.be/HPwbCF7D5Ec?si=PVyrPi48pfCHSlx9. Last accessed 29 January 2024.

^b Brigadier Masud Ahmad Khan – Retd on Twitter. https://x.com/masudakhan6/status/1150707438787727361?s=46&t=9PedQ8OvknCFqvlDJfjimA. Last accessed 29 January 2024.

^c PoliceAwam. https://twitter.com/PoliceAwam/status/736872274532405248. Last accessed 26 January 2024.

Consistent with our conception of chauvinistic nationalism, social media linked to institutions of the Pakistan state – both the powerful Pakistani army as well as governing and opposition political parties – frequently centre content on Pakistan's military prowess and its ability to defeat the country's many state and non-state adversaries, including, for example, jingoistic coverage of the country's nuclear weapons programme. State authorities regularly push nationalist narratives seeking to unify and promote the integration of ethnic and religious groups as loyal and patriotic Pakistanis – which is consistent with our conception of cohesive nationalism. For example, in July 2019, Pakistan military's public relations branch tweeted a press release congratulating the army's first Sikh army officer on his marriage, with the caption ‘Pakistan Army is symbol of national integration and respect rights of minorities’ (see also Table 2).¹ Finally, Pakistan state-linked media outlets frequently elevated the country's contributions to the international community, highlighting, for example, the country's contribution to United Nations peacekeeping forces, as a form of international legitimacy nationalism.²

While these nationalist narratives sometimes rely on false information, at other times they merely frame existing information in a particular light; as such, they function as propaganda, which Tucker et al. (2018, 3) define as based on true information but which, nonetheless, ‘disparage opposing viewpoints’. For the purposes of this paper, we focus on nationalist prompts not grounded in false information to see their effect on conspiracies pertaining to minority groups.

Conspiracy Theories and Rights-Seeking Groups in Pakistan

Conspiracy theories and misinformation are widespread in Pakistan, persisting in the public sphere, often with harmful effects (Siddiqui 2020). Such narratives and falsehoods pervade political and social discourse, cover a wide range of topics, and are often widely believed. They are sometimes deliberately spread but may take on a life of their own at the local level – whether through social media or word of mouth. Militant non-state actors, religious movements, mainstream parties, and state agencies alike frequently rely on the propagation of narratives to shore up their worldview, even when these are not supported by facts. Disinformation campaigns smear analysts and activists who are critical of dominant national security narratives, or who express their support of opposition political parties and movements (Raza 2019). Much of this disinformation is comprised of doctored or fabricated images and videos shared on social media platforms, including Twitter, WhatsApp, and Facebook. Dedicated channels on YouTube also make unsubstantiated allegations, particularly focused on vilifying groups perceived as anti-state. These social media platforms are widely used in Pakistan; in a phone-based survey conducted by the authors across four provinces in Pakistan, 62 per cent of the respondents reported using WhatsApp, 56 per cent on Facebook, 46 per cent on YouTube, and 17 per cent on Twitter.

In this analysis, we focus on conspiracy theories about ethnic minority groups in the country. In recent years, the Pakistani state's attempt at maintaining unity among its disparate ethnic groups has involved a concerted social media campaign often relying on unverified information. This campaign is aimed at making national identities more salient and sometimes has the effect of increasing intolerance towards domestic outgroups, particularly those groups that demand greater rights within the country. One rights-seeking movement that received recent attention is the PTM, a non-violent Pashtun civil rights movement that campaigns against the excesses of the Pakistani military's counterinsurgency campaign in the country's North-Western periphery. The PTM has argued that, in its pursuit of broader foreign policy goals, Pakistan's military has brought conflict and misery to Pashtuns on both sides of the Pakistan-Afghanistan border over the last four decades. A PTM leader alleged in a New York Times op-ed that the military had ‘unleashed thousands of trolls to run a disinformation campaign against the P.T.M., accusing us of starting a “hybrid war”. Almost every day they accuse us of conspiring with Indian, Afghan or American intelligence services’ (Pashteen 2019). In response, the military has publicly charged that foreign spy agencies fund the PTM and that the PTM is itself propagating disinformation about its treatment at the hands of the state of Pakistan to tarnish the country's standing (Syed and Raza 2019).

Similarly, political activists seeking more rights for the minority ethnic Baloch have been targeted in state messaging as disloyal. For years, Baloch student groups, various civic organizations, and tribal leaders have demanded more control over Balochistan's economic resources, complaining that the country's political elite routinely expropriates Baloch resources and provides little in return to the Baloch population (Bansal 2008). An impetus for such messaging is a long-standing low-level insurgency in Balochistan that seeks to impose costs on Pakistani military and state targets to expel it from the region. Figure 1 depicts an example of a meme which circulated on social media showing PTM leader, Manzoor Pashteen, being handed a uniform by an Indian official, with the hashtag #RAWSponsoredPTM (RAW is the Indian intelligence service). The photo also shows a uniform for the Baloch Liberation Army, a Baloch militant group. Our study is interested in how nationalist sentiment increases belief in such claims. We conceive of these claims as conspiracy theories because they are ultimately unverifiable, and because their effect does not entirely ‘hinge on the truth value of the claims being made’ (Guess and Lyons 2020, 10).

Figure 1. Example of a meme on social media alleging that both the Pashtun Tahafuz Movement and Baloch nationalists are supported by foreign – in this case, Indian – actors.³

Research Design

We administered a nationwide survey with an embedded experiment to 2,373 Pakistani respondents. The survey was carried out in December 2020 during the height of the Covid-19 pandemic and was therefore conducted over the phone. It lasted approximately fifteen minutes and was administered using Random Digit Dialing in an effort to gain a representative sample of the cell phone-holding population. The survey was carried out by enumerators from the Pakistan Institute of Public Opinion (an affiliate of Gallup International in Pakistan, hereafter called Gallup Pakistan; see Supplementary Appendix A1 for more on sampling and randomization procedures). It is estimated that over 80 per cent of Pakistanis use mobile phones; however, as Table A3 shows, we had an underrepresentation of women (only 16 per cent of our sample was female). This is perhaps unsurprising given that women are less likely than men to own cell phones, and are also less likely to answer the phone or talk to individuals with whom they are not familiar. The generalizability of the results should be analyzed with these features of the sample in mind.

Respondents were randomly assigned to a control or one of three treatment conditions (Figure 2). In the control, no information was provided to respondents, and the outcome measures – belief in the accuracy of various true, false, and unverified statements – were asked directly. In the treatments, prompts related to each of the three types of nationalisms introduced in the theoretical section – chauvinistic, cohesive, and international legitimacy – were presented (see Table 3). Each of these prompts sought to heighten nationalist sentiment in the respondent. Given high baseline levels of nationalism among segments of the Pakistan populace (see A17), we sought to make the treatments as strong as possible by reiterating in numerous ways different aspects of Pakistani national identity that would evoke nationalist sentiment in the respondents.⁴ Each treatment was collectively composed of three distinct statements that were separated by questions asking respondents if they were aware of a certain piece of nationalistic information. By keeping the statements short and including questions, we sought to ensure that respondents were listening to the treatments.⁵ We ended each set of treatment vignettes with an additional question, asking respondents how proud they were to be Pakistani on the basis of various aspects of the three types of nationalism we theorized. This question allowed us to both measure pride following the treatments and reiterate the nationalist nature of the prime to the respondents.

Table 3. Complete text of treatments

^a ‘U.S. offered $5 billion for refraining from nuclear tests: Nawaz Sharif,’ The Hindu, 28 May 2022, https://www.thehindu.com/news/international/U.S.-offered-5-billion-for-refraining-from-nuclear-tests-Nawaz-Sharif/article16304121.ece. ‘Nuclear Anxiety: In Washington; After an Anguished Phone Call, Clinton Penalizes the Pakistanis’, New York Times, 29 May 1998, https://www.nytimes.com/1998/05/29/world/nuclear-anxiety-washington-after-anguished-phone-call-clinton-penalizes.html.

Figure 2. Research design.

Following the treatment, the respondents were asked their opinion of the accuracy of a number of statements, both true and false.⁶ This paper looks at two outcome measures that portray ethnic minority groups as subversives and allied with external actors. We asked respondents to indicate whether they thought the following conspiracy theories were accurate on a 1–4 scale, with 1 being very accurate and 4 being not at all accurate. The results displayed here have been recoded to allow for ease of interpretation, with higher numbers indicating greater belief in the alleged conspiracy theory.

1. (1) Pashtun Subversive: Pashtun groups, like the Pashtun Tahafuz Movement (PTM), are supported by foreign intelligence agencies and are aimed at breaking Pakistan into pieces.

2. (2) Baloch Subversive: Many people in Balochistan support groups which collude with India to bring harm to Pakistan.

Research on misinformation and conspiracy theories carries with it ethical obligations not to mislead the respondents, especially with potentially malign information. We designed our experimental treatments and outcome measures to manage exposure to divisive information cues in two ways. First, we based all our treatments on information commonly used by state authorities, institutions, and the media to further nationalist narratives in Pakistan. Second, we asked only about statements that were already circulating widely in society (see Figure 1 for an example). The outcome measures were also preceded with the statement, ‘Now we'd like to ask you about some different topics. Some people believe these statements to be accurate; some believe them to be false. Please indicate how accurate you believe the following statements are.’ Third, we pretested and piloted the survey beforehand, and consulted with Gallup Pakistan, which has conducted polling in Pakistan for several decades, on the appropriate language for the context. Finally, we offered a detailed debrief to the respondents through the enumerators at the end of each survey, stating that there was no evidence to support the conspiracy cues.

Findings: Does Nationalism Increase Belief in Anti-Minority Conspiracy Theories?

We first evaluate the main proposition that nationalism leads to an increase in belief in conspiracy theories casting minorities as subversives. We compare the raw averages across the control to those of all treatment groups grouped together. First, we collapse the scale of belief in conspiracy into two categories, whether the respondent believes it to be accurate (3 and 4 on our recoded 1–4 scale) or inaccurate (1 and 2 on the 1–4 scale). This descriptive data (see Table 4) shows a clear pattern: 28 per cent of the respondents in the control group believed the conspiracy theory about the PTM and 42 per cent believed it about the Baloch. After the treatment, belief in foreign support of PTM went up to 34 per cent and belief in foreign support of Baloch increased to 48 per cent. There is also a 4 per cent and 6 per cent accompanying decrease in people who found the PTM and Baloch conspiracies to be inaccurate compared to the control group, as well as a decrease in individuals choosing not to respond to the questions.⁷ Figure 3 depicts the full Likert score distribution by treatment and control for both outcome measures.

Table 4. Belief in conspiracy theories about minorities in control v. treatments

Figure 3. Distribution of belief in accuracy of conspiracy theory, treatment v. control.

Next, we estimate a simple linear probability model. For each of the two outcome measures, we estimate OLS models where belief in each statement (measured on a 4-point scale, with high numbers indicating greater belief) is regressed on a dummy variable indicating if a respondent received any treatment, as well as separate OLS regressions for each individual treatment. Table 5 shows these regression results. We find that nationalism in general (columns 1 and 5) increases belief in both types of anti-minority conspiracies.

Table 5. Effect of nationalism prime on belief in conspiracy theories about minorities as subversive

Robust standard errors in parentheses.

***p < 0.001, **p < 0.01, *p < 0.05, +p < 0.1.

The table shows ordinary least square regressions with robust standard errors. Outcome measures are measured on a 1–4 scale, with greater numbers indicating greater belief in the conspiracy. Models 1 and 5 show the estimate for the pooled treatments, while subsequent columns show the estimates for the individual treatments compared to just the control condition.

Does belief in the accuracy of anti-minority conspiracy theories vary by the type of nationalist prompt to which respondents were exposed? We find that respondents in each of the three treatment groups were more likely to believe that Pashtun groups like the PTM were supported by foreign groups and were aimed at breaking apart Pakistan. While respondents in each of the three treatments were also more likely to believe that the Baloch support groups colluded with India to destabilize Pakistan, this was statistically significant only for those who received the international legitimacy treatment where the coefficient was larger than the chauvinistic and cohesion treatments (see Table A5). Figure 4 plots these results graphically using a coefficient plot of the estimated effect of each treatment individually on belief in conspiracies of minorities as subversives relative to the control.

Figure 4. Effect of different types of nationalism primes on belief in conspiracy theories. The coefficient plot of the estimated effect of each treatment on belief in conspiracies relative to the control.

Who is Most Affected by Nationalism?

Our survey experimental findings suggest that nationalism can act as a powerful force by increasing belief in conspiracy theories that frame minorities as subversives. An important result of the analysis is that international legitimacy nationalism has the greatest effect and that benign variants of nationalism do not decrease belief in such theories. What might explain these findings? In this section, we explore possible explanations for these results by examining how individuals with different pre-existing beliefs responded to the treatments.⁸

Exposure to nationalist sentiment can potentially alter how individuals process information by engaging in motivated reasoning, due to which individuals sort information that suits prior beliefs, or beliefs which they find attractive (Kunda 1990). It is possible, then, that individuals, on being exposed to nationalistic primes, are more likely to believe cues that confirm their pre-existing beliefs. It is also possible that they reject cues that disconfirm their pre-existing beliefs.⁹

First, we consider whether individuals who are more tolerant of religious minorities in Pakistan are more resistant to conspiratorial beliefs about minorities more generally and are, accordingly, less moved by the treatments. Intolerant respondents are coded as those who selected ‘not at all’ on a 1–4 trust scale of how much they trust members of other religious groups. We find no such heterogeneous effect (see Table 6).

Table 6. Treatment and interaction effects for relevant subgroups

Robust standard errors in parentheses.

***p < 0.001, **p < 0.01, *p < 0.05, +p < 0.1.

OLS regressions with interaction effects for four categories of respondents. Intolerant respondents are those who selected ‘not at all’ on a 1–4 trust scale of how much they trust members of other religious groups. India Hawks are those who believe that Pakistan should treat India as an enemy rather than as a friend. High International Esteem includes those who said that ‘Pakistan's contributions to international peace are very well recognized by the international system’ relative to ‘somewhat’ or ‘not well recognized’. ‘Prefer military’ are those who agreed with the statement that military rule is sometimes preferable to democratic rule.

Second, we examine whether individuals who are more concerned with Pakistan's perceived enemies are more likely to be moved by the nationalism primes. The logic may be that those who believe Pakistan should treat India as an adversary and are concerned with its ‘inimical designs’ against the country will swiftly believe conspiracy theories about India supporting subversion through rights-seeking ethnic minority groups in Pakistan. We find no significant interaction effect of the nationalism treatment on Indian hawks.

Third, we examine whether respondents who were more likely to be sympathetic to military rule in Pakistan – and who may, therefore, be more sympathetic to militarized propaganda or less interested in democratic norms – have differential responses to the treatment. We measured this variable by coding as 1 those who said they agreed with the statement, ‘In some circumstances a military government can be preferable.’¹⁰ We find no significant interaction effect of the nationalism treatment on those who prefer military rule.

Finally, we examine whether individuals who believe Pakistan's contributions to peace are well-recognized by the international system (what we call the high international esteem subgroup) were more likely to believe conspiracy theories about minorities. It could be that individuals concerned about the status of the nation worry that rights-seeking minorities undermine the nation by embarrassing it, and thus are more likely to assume the worst about them; that is, purported subversive activities. As shown in Table 6, we find that respondents in the high international esteem subgroup are more likely to be moved by the nationalism prompts to believe the conspiracy theory about the PTM being subversive (p = 0.025). This result appears consistent with our finding that the international legitimacy treatment had the strongest substantive effect on beliefs about Baloch collusion with hostile foreign powers. Together, they suggest that concerns over Pakistan's status are a possible contributor to belief in conspiracy theories about rights-seeking minorities. We explore this mechanism in more detail next.

Concerns with Pakistan's Status and Reputation: Focus Group Evidence

The survey experimental evidence suggests that the international legitimacy treatment had the strongest treatment effect on both outcome measures and that individuals confident about Pakistan's international status were more likely to believe some anti-minority conspiracies than respondents evincing less international status concerns.¹¹ Focus groups conducted in Pakistan in November 2021 support this as a potential explanation linking nationalist sentiment to conspiratorial belief. We focused on respondents in four major urban centres in the country: Islamabad (the capital), Lahore (Punjab), Peshawar (Khyber Pakhtunkhwa), and Karachi (Sindh). Four of the six focus groups were composed of students at universities; the remaining two focus groups were a random sample of residents in the city for a total number of fifty respondents. We recruited respondents from universities due to the ease of recruitment for the focus group format, and because this population is of special interest to our study given its use of social media. To ward off the concern that a single university might provide unrepresentative insights, we recruited from various universities within each city to draw on respondents representing different ethnic, linguistic, and political backgrounds.¹² Having two non-higher education samples allowed us to see the extent to which views and opinions were shared across different strata. Nonetheless, the focus groups are not intended to be representative; rather, we use them primarily to provide further qualitative evidence in favour of our proposed mechanism.

The focus groups, which lasted approximately 1.5–2 hours, discussed several topics related to this study, including participant attitudes towards different ethnic groups, their opinion of rights accorded to various minority groups within Pakistan, and nationalist sentiment. Respondents across our different locations expressed a strong belief in the importance of maintaining or improving the country's status, and how rights-seeking groups – inadvertently or maliciously – serve to harm this effort. One respondent, for example, noted the importance of international perception by saying:

We have to portray Pakistan in light of our Muslim values to show that this is what the Muslim world is. But what we are portraying and what the world is seeing are not those true values that they should be.

Another respondent noted that it was important to conceal internal differences and highlight the positives of the nation:

Negatives are in every country but most countries do not take these negatives internationally. Like if there is a small case, it will go viral around the world that this happened in Pakistan. So if we cover these things, stay united and highlight only the positive aspects then Pakistan will cover it too.

To ground our discussion in a specific case, we asked participants about Malala Yusufzai, a young Pashtun girl who was shot by the Taliban in Swat and has since been awarded the Nobel Peace Prize. While the example of Malala was distinct, topically, from the types of conspiracies covered in the survey experiment, we opted to ask about her as she is well-known in the country and because she has aroused significant controversy in Pakistan (Reeves 2013), and, as such, we considered it a topic with which most individuals would be familiar. Participants were, almost without exception, categorical that they did not believe the story surrounding Malala's shooting. One participant explicitly linked his disbelief to national reputation and status concerns: ‘I think this is a conspiracy to defame Pakistan.’ Another noted that Malala's case could not possibly be accurate in part because women were accorded their full rights in Pakistan and that this coexisted easily with Pakistan's Islamic identity.

Yet another respondent drew equivalences between Malala receiving the Nobel Prize and a Pakistani documentary filmmaker who won an Academy Award for a documentary about acid attack victims in Pakistan. The participant noted that ‘Pakistan has the lowest rate of it [acid attacks], but they [the international community] watched it and gave her the Oscar Award.’ This idea, then, of disbelieving the veracity of claims which bring negative attention to Pakistan was very resonant – even when fellow compatriots were being bestowed with international honours for their work.

Still other focus group participants directly made the connection between reputational concerns and efforts by external forces to hurt the country. One noted:

There are external forces that want to humiliate us. We are not like that. It is not our fault nor it is the fault of our state that the green [Pakistani] passport does not get respect. We are being humiliated on an international level.

Another said, ‘Pakistan is defamed everywhere. So much hype is created everywhere around the world that this happened in our country.’ When asked specifically about the Pashtun Tahafuz Movement, and when shown the headline of a New York Times op-ed written by PTM leader, Manzoor Pashteen, a participant questioned why Pashteen turned to international actors, given that doing so would harm the country's reputation:

I feel bad that if they [the Pashtun community] want their demands to be heard and want to represent themselves, they should first talk to the media here. Why are they directly going to foreign media? Look, even we didn't know [about the atrocities against the Pashtun community], you told us. This is just to portray on international media that this is happening in Pakistan, that is happening in Pakistan.

Much of the evidence, then, suggests that concern with tarnishing a country's reputation makes belief in conspiracies about minorities as subversives more likely. Individuals are more willing to believe that activist minorities are supported by foreign entities or enemies than accept that their fellow citizens are treated poorly by the state because doing so would involve diminishing their view of Pakistan and tarnishing it on the international stage. Theoretically, our proposed logic falls within a larger body of scholarship that examines the relationship of the individual to the nation-state. Hur (2020), for example, highlights the ethical obligations that citizens feel to their nation, which in turn leads to compliance with the state. We propose that nationalism, even in its inclusive varieties, can increase the obligation that citizens feel towards protecting the state's reputation and status, particularly on the global stage. Our finding is consistent with existing literature on how concerns with status affect such outcomes as conflict behaviour (Lebow 2008; Renshon 2017) or the likelihood of leaders using force in militarized disputes (Dafoe and Caughey 2016).

Taken together, our focus group and survey evidence suggest that those particularly concerned with Pakistan's international standing may be more susceptible to belief in conspiracy theories about minorities as subversives. This is consistent with other accounts of Pakistan's national leaders highlighting the need to showcase only a ‘positive image’ of Pakistan and encouraging fellow citizens to do the same on the premise that Pakistan has made important contributions that go unrecognized (Anis 2018). One social media campaign denoted by the hashtag #PakPositive has encouraged users to share affirmative images of the country through news stories and personal experiences. These images and posts have varied from those lauding Pakistan's natural beauty as superior to those of other (especially European) countries to claims that Pakistan has been recognized as one of the most peaceful countries in the world, or that it is a uniquely favourable place for women. Such campaigns tend to be amplified by co-ordinated social media activity lauding Pakistan's culture, politics, and society, ostensibly to dispel negative perceptions of the country and often relying on misinformation. They are sometimes directly aimed at countering the messaging of rights-seeking groups (Shackle 2020). Minority groups, which challenge the status quo as detrimental to them and who demand greater rights, are then seen as working against the state.

Discussion and Conclusion

Our analysis has several major implications. We show that nationalism can increase individuals' willingness to believe in anti-minority conspiracy theories, and even seemingly benign versions of nationalism can be a source of malignant attitudes. We also provide suggestive evidence that this relationship between nationalism and anti-minority conspiratorial belief is due to concerns with Pakistan's international status and reputation, which rights-seeking minority groups are seen as harming. In this section, we outline some possible limitations with our research design – where we do not think they are likely to affect our results, and where they raise avenues for future research.

First, one possible concern is that social desirability bias played a role in how respondents answered the questions after receiving the nationalism treatments, particularly if they believed they were being monitored via the survey by the state or if the treatment itself influenced how respondents expressed their belief in the outcomes. However, it is unlikely that social desirability or other forms of survey bias are responsible for our results. For one, our survey asked about belief in a number of other outcomes (both true and false statements), where we might have expected respondents to offer more socially desirable answers if the treatment was affecting respondents in that way. However, we do not see this to be the case (see A4). Additionally, we see at least one counter-intuitive result: the cohesive nationalism treatment did not make respondents more likely to express positive views of ethnic minority groups, which would potentially be the more socially desirable answer given the content of the treatment. Most significantly, our focus group findings demonstrate concerns about minority rights' advocacy, their alleged support by foreign powers, and help support the link between nationalist sentiment and these beliefs.

Other design choices also have potential limitations. For example, we do not include a placebo treatment, opting instead for a pure control. Second, each treatment contained a bundle of related but distinct ideas based on our best efforts to capture how each nationalist narrative actually appears in Pakistani discourse. As such, our design only permits us to say that the composite bundle of each treatment is responsible for the effects we see. Having established this baseline, future research should seek to disentangle further the underlying components of each type of nationalism.

Future research should also examine how broadly our results generalize to other contexts. One possibility is that these findings apply to nationalist contexts with more exclusionary national identities. Tudor and Slater (2021) disaggregate nationalisms by formative moments of national identities, ranging from exclusive identities, focused by differentiation from an ‘other’, to inclusive ones, which ensure access to equal rights for all. They show that exclusive identity nations – such as Pakistan – have political outcomes distinct from national communities defined as inclusive in ethnoreligious terms. It is plausible, then, that the failure of cohesive nationalism found in the Pakistani context may only replicate in contexts founded on exclusive national identities, but would not generalize to states where inclusive nationalism is the norm or the founding principle.

Another possible scope condition may involve types of internal minority groups: nationalism exposure might disproportionately increase belief in conspiracies about those minorities who have cross-border coethnic kin due to the spectre of ethnic irredentism, which holds for the Baloch and Pashtuns in Pakistan. Similarly, given the background condition of Pakistan's rivalry with India, our findings may only apply to countries that have long-standing rivalries, particularly those which either share in common an identity with the state's minority group or where the majority believes that the rival state will benefit from their state's dissolution. It is also possible that these findings are most applicable where international legitimacy concerns among citizens are high; indeed, scholars note that the level of citizen concern with national status is variable (Renshon 2017). Overall, the extent to which our findings speak to the relationship between nationalism and conspiracies provides an important avenue for future research.

Finally, our analysis has policy implications for various governmental and non-governmental efforts to build inclusive nationalist narratives in an attempt to counter pernicious nationalist attitudes out of concern for their downstream effects on behavioural outcomes. Our results suggest that counter-misinformation work needs to reconsider existing approaches that elevate seemingly benign variants of nationalism, whether emphasizing superordinate identities or a country's contributions to international peace. It is plausible that approaches centred on benign variants of nationalism are doing as much harm as more pernicious versions of nationalism.