What drives modern Diogenes? Individual values and cosmopolitan allegiance

Introduction

Why do some individuals view themselves as citizens of the world? According to the 2005–2008 wave of the World Values Survey (WVS) carried out in 57 countries across the world, over 30% of the 65,269 respondents strongly identify as world citizens. This is an important phenomenon that merits attention. The term ‘cosmopolitan’ derives from the Greek word kosmopolitês, meaning citizen of the world (Miller, 2010). When asked where he was from, Diogenes the Cynic responded that he was ‘a citizen of the world.’ World citizenship anchors the very project of cosmopolitanism. Moral, legal, and political commitments of cosmopolitanism are tied to this very allegiance. Thus, it is crucial to analyze what drives the current generation of Diogenes.

Yet despite a growing body of work on cosmopolitanism, the literature lacks a clear explanation of what moves individuals to see themselves as world citizens. Partly responsible for this lacuna is the normative and abstract nature of the cosmopolitanism scholarship. Theoretical works have examined the moral principles of cosmopolitanism (Dallmayr, 2003; Nussbaum, 1996; Waldron, 2010), the possibility of cosmopolitan justice (Beitz, 2005; Caney, 2008; O’Neill, 2000), and the compatibility between cosmopolitanism and nationalism (Bowden, 2003; Nussbaum, 1996; Rorty, 1994). ¹ Different aspects of cosmopolitan democracy have also been explored (Archibugi, 2004; Cochran, 2002; Franceschet, 2000; Held, 1999; Koenig-Archibugi, 2011). Yet scholars have paid insufficient attention to what makes individuals identify as world citizens in practice.

Only a handful of empirical studies have analyzed cosmopolitanism at the individual level. Scholars have examined the correlates of a general cosmopolitan orientation (Norris, 2000; Norris and Inglehart, 2009), transnational identities (Jung, 2008; Pichler, 2012), and cosmopolitan support for international institutions (Ecker-Ehrhardt, 2012). In the field of International Relations (IR), studies on foreign policy dispositions have established that cooperative internationalism marked by a cosmopolitan sense of obligation toward the wider international community is an important dimension structuring mass and elite attitudes toward international politics (Chittick, Billingsley, and Travis, 1995; Holsti and Rosenau, 1988; Nincic and Ramos, 2010; Rathbun, 2007; Wittkopf, 1990).

Despite important progress, however, a satisfying account of what drives individuals to see themselves as world citizens still eludes us. Cosmopolitan allegiance and a series of cosmopolitan orientations such as positive feelings toward immigrants and a weak sense of nationalism are often lumped together (Norris, 2000; Schueth and O’Loughlin, 2008). Scholars have argued that younger people are more likely to be cosmopolitan (Jung, 2008; Norris, 2000). Although important, by itself, age does not fully explain why individuals identify as world citizens in the first place. IR scholars have examined how a cosmopolitan posture informs individuals’ foreign policy preferences, yet little has been done to uncover the origins of this posture. In sum, our understanding of cosmopolitan allegiance remains limited.

This is a critical void for three main reasons. Cosmopolitan allegiance taps a person’s self-view as a world citizen. This is a voluntary attachment. Unlike socially ascribed ones, this attachment implies a choice (Huddy, 2001; Thoits, 2003). What underlies this self-understanding is not immediately obvious. The world community is ill-suited to offering feelings of stability, security, or positive distinctiveness, factors found to drive psychological group membership (Brewer, 1991; Castano et al., 2002; Hogg and Abrams, 1993). Therefore, the question of cosmopolitan allegiance begs an answer.

Second, world citizenship implicates important political and normative debates about the global order. It is well known that identity generates mass support for political systems. The cosmopolitan order is no exception. A democratic and efficacious cosmopolitan order requires some sense of world citizenship at the mass level (Calhoun, 2007; Ecker-Ehrhardt, 2012; Kuper, 2004; Urbinati, 2003). Empirical evidence supports this intuition. IR scholars have found that a cosmopolitan foreign policy orientation increases support for peaceful conflict resolution, multilateral diplomacy, and international institutions (Chittick et al., 1995; Herrmann and Keller, 2004; Herrmann et al., 1999; Hurwitz and Peffley, 1987; Nincic and Ramos, 2010). Since cosmopolitan allegiance constitutes the backbone of bottom-up cosmopolitanism, understanding the origins of people’s sense of world citizenship is crucial.

The third and final consideration is a normative one. The individual is the chief unit of moral concern in the cosmopolitan project (Brown and Held, 2010b; Pogge, 1992). For a richer account of cosmopolitanism, it should also be the main unit of theoretical and empirical analysis. Those aware of the importance of this matter have urged scholars to investigate the ‘cosmopolitan condition of real people’ (Beck and Sznaider, 2006: 9). This research takes on this task.

In this article, I analyze the psychological logic of cosmopolitan allegiance by bringing in individual values. Macro theories of cosmopolitanism routinely rely on micro-assumptions about world citizens. In particular, existing approaches characterize world citizenship as the manifestation of specific human motives. World citizens are other-regarding humanitarians in normative accounts (Brown and Held, 2010b; Nussbaum, 1996), self-interested elites in critical economic models (Micklethwait and Wooldridge, 2000), open-minded multiculturalists in cultural ones (Hannerz, 1990; Skrbis Kendall and Woodward, 2004), and autonomy-seeking free agents according to liberal perspectives (Bohman, 2007; Delanty, 2006). ² Uncovering the psychological dynamics of cosmopolitan allegiance, therefore, requires tapping into individuals’ core motives.

Values offer one viable means of exploring human motivation. Values are cognitive representations of trans-situational motivational goals that function as reference points for individuals (Rokeach, 1973; Schwartz, 1992). The distinctiveness of values is that they operate at a higher and general level of cognition, are stable across different situations, and serve as evaluative standards. As Rohan (2000: 258) explains, a value is a higher-order ‘meaning-producing cognitive structure.’ Therefore, values are analytically prior to attitudes, preferences, and roles, and thus inform the latter (Feldman, 2003; Roccas et al., 2002; Rohan, 2000; Schwartz, 1992). Specifically, as guiding principles in people’s lives, values influence which roles people find appealing. Individuals take on attachments that are congruent with their value priorities (Feather, 1994; Hitlin, 2003, 2007).

To illuminate the psychological underpinnings of cosmopolitan allegiance, I apply the theory of personal values identified by Shalom Schwartz and derive testable hypotheses. Schwartz’s theory is considered to be the most systematic theory of individual values (Feldman, 2003; Hitlin and Piliavin, 2004; Rohan, 2000). The structure of values theorized by Schwartz has been supported by a myriad of empirical tests in more than 300 empirical samples over the last few decades. ³ Employing the Schwartz Value Theory (SVT) allows me to start with a strong theoretical foundation of values and examine the effect of different values on cosmopolitan allegiance within a single analytical framework (Roccas, Schwartz and Amit, 2010).

Schwartz identifies 10 motivationally distinct values (Schwartz, 1992, 1994, 2012; Schwartz and Bilsky, 1987; Schwartz and Boehnke, 2004). Rooted in human needs, each value expresses a set of desirable motivational goals. The theory posits two main value conflicts pertaining to competing individual needs. Self-transcendence values, which include universalism and benevolence, stand in opposition to self-enhancement values, which include achievement and power. Conservation values composed of conformity, tradition, and security stand against openness-to-change values composed of stimulation and self-direction (Schwartz, 1992, 1994). Hedonism lies between self-enhancement and openness to change.

Multinomial logistic regression analysis using WVS data (2005–2008) shows that self-transcendence, self-enhancement (except power), and openness-to-change values lead to cosmopolitan allegiance, while conservation values hinder this attachment. This finding indicates that world citizenship on the ground is a rich, multifaceted role attractive to individuals with varied outlooks on life. In people’s perceptions, world citizenship is congruent with morality, diversity, and liberty as well as with self-interest. Cosmopolitan identifiers care about themselves and about others, and they endorse diversity. Yet, they resist the subordination of the individual to societal hierarchies and conventions, and the suppression of independent thought and action. The compatibility of world citizenship with egocentric motivational goals at the mass level indicates that the pessimism about individuals’ ability to be cosmopolitan may be misplaced. The emphasis that publics place on openness values suggests that freedom and diversity are critical components of legitimate global governance.

By offering a psychological account of cosmopolitan allegiance, this study adds the missing micro-foundations to macro theories of cosmopolitanism and provides the needed empirical evidence to substantiate their claims. By illuminating what drives modern world citizens, this research contributes to a richer understanding of the bottom-up cosmopolitanism crucial for global governance.

In the sections that follow, I conceptualize cosmopolitan allegiance and explain why it makes sense to ground it in personal values. I then apply the SVT and derive testable hypotheses about how individual values affect world citizenship. The third section discusses my findings. I conclude by summarizing the results and laying out their broader implications.

Cosmopolitan allegiance and values

Cosmopolitan allegiance, as I employ it here, refers to an individual’s understanding of him-or herself as a citizen of the world (Tajfel, 1981; Tajfel and Turner, 1986). ⁴ It taps a person’s self-view anchored in a sense of membership in the world as a collectivity (McCall and Simmons, 1996; Owens et al., 2010; Stryker, 1980; Tajfel, 1981). World citizenship is an acquired attachment that individuals voluntarily adopt. It could not be assigned. What underpins this attachment is self-categorization. To explain why people see themselves as world citizens, accordingly, we need to examine the factors that lead to this self-categorization.

Conceptualizations of cosmopolitanism abound (Archibugi, 2003; Brown, 2009; Brown and Held, 2010a; Fine, 2003; Heater, 2002; Kleingeld, 1999; Lu, 2000; Scheffler, 1999; Vertovec and Cohen, 2002a; Waldron, 2010). As a result, there is a lot of theoretical discussion about what it means to be a world citizen. If we survey the literature with an analytical eye, however, we can identify four main ways to understand cosmopolitan motivation at the individual level. I categorize these as normative, economic, cultural, and liberal models. While each perspective is rich and not mutually exclusive with the others, stripped to its very core, each anchors world citizenship in a particular dominant motive. This taxonomy is, of course, a simplification, but it has analytical value. It approximates the four ideal-type images of world citizens prevalent in the literature (Cheah and Robbins, 1998; Delanty, 2006, 2012; Kleingeld and Brown, 2013; Nowicka and Rovisco, 2009; Vertovec and Cohen, 2002a). As such, this framework enables us to systematically study cosmopolitan motivation in practice.

Normative accounts associate world citizenship with the desire to protect and advance the well-being of others. World citizenship is marked by a moral commitment to the welfare of all human beings, regardless of their race, ethnicity, gender, or nationality (Brown and Held, 2010b; Nussbaum, 1996; Pogge, 1992). It represents a sense of obligation toward global compatriots. According to the universal moral logic of normative approaches, world citizens are other-regarding humanitarians.

Conversely, critics see world citizens as the new economic elites. Scholars have argued that the globalized world economy has created a class of world citizens who are privileged elites, typically, business executives, pursuing success and wealth. Instead of contributing to global prosperity and justice as some economic cosmopolitans from Adam Smith to Milton Friedman had hoped (Kleingeld, 2012), the skeptical model posits that globalization and the liberalization of markets have primarily served the interests of ‘cosmocrats’ (Micklethwait and Wooldridge, 2000; see also Furia, 2005). ⁵ Cosmocrats represent a global elite motivated by egoistic materialistic goals. They reap the benefits of free trade for self-aggrandizement (Micklethwait and Wooldridge, 2000). They have global experiences but no global moral consciousness or egalitarian aims. Unlike those striving for international financial justice, cosmocrats enjoy ‘the universalism of social and material self-interest that allows for the easy embrace of a consumerist cosmopolitanism’ (Cronin, 2006: 11). Thus, world citizenship looks very much like the ‘class consciousness of frequent travelers’ (Calhoun, 2002: 86).

The cultural perspective sees world citizenship as the expression of pluralism and tolerance (Delanty, 2006; Hannerz, 1990; Skrbis et al., 2004; Waldron, 2010). Ulf Hannerz’s (1990: 239) frequently evoked view summarizes the essence of the cultural perspective: ‘cosmopolitanism in a stricter sense includes a stance towards diversity itself … [it] is first of all an orientation, a willingness to engage with the Other.’ According to the cultural perspective, interlocking social networks (Castells, 1996; Cheah and Robbins, 1998), social mobility (Urry, 2000), hybrid cultures (Appiah, 2005), and allegiances (Benhabib, 2004) lie at the heart of transnational politics. Being a cosmopolitan therefore represents ‘two related abilities’: the ability to position oneself outside of one’s established territorial and cultural space and the ability to embrace new traditions (Held, quoted in Cronin, 2006: 11). From this point of view, world citizens are global multiculturalists.

Finally, the liberal perspective associates world citizenship with a search for independence (Bohman, 2007; Delanty, 2006; Nussbaum, 1996). Citizens of the world seek freedom from societal orthodoxies. Partly shaped by the assumption that individuals realize ‘the good life’ when they are free from interference (Bowden, 2003), this perspective particularly relates world citizenship to opposition to the nation-state and territorial politics. World citizens embrace liberty in a quest for new identities, experiences, and ways of life rather than conforming to imposed local attachments and obligations (Appiah, 1996; Delanty, 2006). They aspire to free agency. ⁶

Each of these approaches characterizes world citizenship as the manifestation of specific motives that map onto human values. To unpack these motives, I bring in values, translating these macro-assumptions about world citizenship into testable hypotheses.

Philosophers, anthropologists, sociologists, and psychologists have devoted a great deal of attention to values since the 20th century (Feldman, 2003; Rohan, 2000). Values constitute cognitive representations of context-independent goals (Rokeach, 1973; Schwartz, 1992). For Rokeach (1973: 5), a value ‘is an enduring belief that a specific mode of conduct or an end-state of existence is personally or socially preferable to an opposite or converse mode of conduct or end-state of existence.’ A value system reflects ‘an enduring organization’ of these enduring beliefs (Rokeach, 1973: 5). According to Schwartz (1992: 4): ‘values 1) are concepts or beliefs, 2) pertain to desirable end states or behaviors, 3) transcend specific situations, 4) guide selection or evaluation of behavior and events, and 5) are ordered by their relative importance.’

Drawing upon the insights of scholars who have explored the nexus between values and identities, in particular, the works of Roccas, Schwartz and Amit (2010), Feather (1994), and Hitlin (2003), I argue that individuals identify as world citizens when they see congruence between this self-understanding and their values. They do so because a value-congruent view of the self provides a channel for the realization of the trans-situational goals embedded in their values (Roccas et al., 2010). Self-view enables individuals to stay true to the principles guiding their lives. It also satisfies the need to cognitively represent and express one’s values (Roccas et al., 2002). Stated simply, a value-congruent self-understanding allows people to live their values.

A key mechanism underlying the human desire for congruence is self-verification (Swann et al., 2002). Individuals are motivated to verify their personhood and are thus attracted to different channels of validation, ranging from seeking self-consistent information (Robinson and Smith-Lovin, 1992) to choosing group memberships that validate their self-concept (Chen et al., 2004). A value-congruent attachment, therefore, evokes a sense of being true to oneself, and, as a result, enables individuals to verify their sense of who they are (Burke and Stets, 2009; Hitlin, 2007). Erickson (1995) refers to this phenomenon as ‘authenticity.’

The effect of values on identities has not gone unnoticed. Brewer and Roccas (2001: 225), for example, argue that the social self is shaped by the relative importance individuals place on individualism and collectivism values. Roccas et al. (2010) argue that values explain why some people are more likely to identify with the nation. Their results show that national attachment is positively associated with conservation values and negatively with openness-to-change values. Similarly, Hitlin (2007) demonstrates a relationship between values and volunteer identity, and Cleveland, Erdogan, Arikan et al. (2011) find a connection between values and cosmopolitan consumer traits. ⁷ In the next section, I turn to Schwartz’s theory to explain how different values might relate to cosmopolitan allegiance.

Applying the SVT

Schwartz and his colleagues have identified 10 values, each expressing a specific motivational goal (Schwartz, 1992, 1994, 2012; Schwartz and Bilsky, 1987; Schwartz and Boehnke, 2004). The SVT posits that personal values form a circular structure marked by two conflicts, as captured in Figure 1.

OPEN IN VIEWER

Figure 1.

An overview of the Schwartz theory of values.

Self-transcendence values comprised of universalism and benevolence represent sociotropic and humanistic concerns for others. These values rival self-enhancement values of achievement and power. Instead of concern for others, self-enhancement values emphasize promoting the interests of one’s self. The second conflict is between a preference for order and the status quo and a desire for change. Conformity, tradition, and security, namely, conservation values, stand in opposition to stimulation and self-direction, which make up openness-to-change values (Schwartz, 1992, 1994, 2012). Hedonism, meanwhile, has mixed characteristics that are associated with both self-enhancement and openness to change. Values adjacent to each other tap parallel motives, and are thus compatible with each other. Values located in opposite domains of the value circle tap divergent motives, and could conflict with one another. For example, tradition and conformity emphasize deference to rules and submission to dominant ways of life. These values are compatible with each other. However, they clash with self-direction and stimulation values centered on freedom, creativity, and change.

How, then, might personal values affect cosmopolitan allegiance? ⁸ Self-transcendence values emphasize going beyond self-interest. Universalism is characterized by other-regard and promoting the welfare of others. Benevolence is similar but bounded in scope. It represents care for the welfare of those closer to the self. Individuals who hold these values in high regard place emphasis on understanding, appreciation, tolerance, and care about the prosperity of other people. Commitment to these objectives, then, drives them to seek social justice, equality, peace, environmental conservation, friendship, forgiveness, and honesty. If the normative model of world citizenship is correct, universalism should move individuals to take on world citizenship. Since concern for the welfare of others is what defines the world citizen, those who score high on universalism should be highly attracted to world citizenship as an outlet to express and realize their values. Because benevolence emphasizes care for people in one’s closest network, I expect benevolence to exert a smaller effect on cosmopolitan allegiance than universalism.

Conversely, if the economic model of world citizenship is correct, self-enhancement rather than self-transcendence values will be congruent with world citizenship. People who value achievement and power care about personal success and competence, status, and prestige, and seek to dominate and control others. Hedonists are focused on pleasure and gratification. Self-enhancers take pride in being successful, pursue wealth and social recognition, and appreciate enjoying life. If world citizenship is the brotherhood of transnational capitalists motivated by self-regard, those who place emphasis on achievement, power, and hedonism will have reason to see themselves as citizens of the world. Thus, two hypotheses follow:

H1: The stronger self-transcendence values (universalism and benevolence), the more likely cosmopolitan allegiance.

H2: The stronger self-enhancement values (achievement, power, and hedonism), the more likely cosmopolitan allegiance.

Conservation values of tradition, conformity, and security stress individual restraint and conformity to convention, and mark a preference for order. Those who attribute importance to these values prioritize the status quo over change, defer to social rules, and censor their words and actions. They prefer impulse control, obedience, and discipline as well as care about security (security of the self, family, and the nation). Conformists honor the customs and norms that limit individual freedom for order. Traditionalists appreciate conventions and modesty. Those focused on security prioritize order. According to the liberal argument, the search for freedom and autonomy motivates world citizens. If so, conservation values should negatively relate to cosmopolitan allegiance. To the extent that world citizens are free agents, people who score high on conservation values will see a discrepancy between this role and their principles.

In contrast, openness-to-change values of self-direction and stimulation, should lead to cosmopolitan allegiance. Openness values feature intellectual and behavioral dynamism. Self-direction underscores independent thought and action; stimulation stresses diversity and novelty. Those who emphasize self-direction are curious, creative, and self-governing. Those who place importance on stimulation are attracted to novel experiences and discovery, and driven toward an exciting life. If the liberal account of world citizenship is correct, openness-to-change values should positively relate to cosmopolitan attachment. Individuals who prioritize openness will derive subjective utility from categorizing themselves as world citizens. In the same vein, if the cultural model of world citizenship is accurate, we should see a positive relationship between openness-to-change values and cosmopolitan allegiance. If being a citizen of the world involves a desire to engage people from different cultures, those who place emphasis on an exciting and diverse life, creativity, and change will be likely to accept world citizenship as self-descriptive. Therefore:

H3: The stronger conservation values (tradition, conformity, and security), the less likely cosmopolitan allegiance.

H4: The stronger openness-to-change values (stimulation and self-direction), the more likely cosmopolitan allegiance.

Data, method, and variables

I use data from the 2005–2008 wave of the WVS. ⁹ This wave included measures of the Schwartz values for the first time. My unit of empirical analysis is the survey respondent. I estimate three models. The first two are additive models that analyze the discrete effects of values on cosmopolitan allegiance. The third examines how a set of core values jointly affects world citizenship. Schwartz (2009) urges scholars not to use all 10 of the value variables in a single model. Proximate values are theoretically expected to be highly correlated with each other. Including strongly correlated values in a single model is not advisable because one cannot clearly differentiate the effect of one value on the outcome of interest from that of cognate values. ¹⁰ Accordingly, the first two models focus on how values on the opposite domains of the Schwarz value circle influence cosmopolitan allegiance. The first includes self-transcendence and self-enhancement values, and the second includes conservation and openness-to-change values. ¹¹ The third model investigates how universalism, benevolence, hedonism, achievement, stimulation, self-direction, and tradition values jointly shape cosmopolitanism.

Results

Table 1 presents the percentage of respondents who see themselves as world citizens and those who do not across countries and continents. Overall, about 78% of the participants in 57 countries see themselves as citizens of the world. If we focus only on those who ‘Strongly agreed’ with the statement ‘I am a citizen of the world,’ we see that about 30% of the respondents see themselves as world citizens. If we focus on those who ‘Agreed’ with the world citizenship statement, we observe that a little over 47% of the participants see themselves as world citizens. About 22% of the respondents do not identify as world citizens.

OPEN IN VIEWER

Table 1.

Percentage of respondents who see themselves as world citizens by countries and continents.

	Strongly agree	Agree	Disagree	Strongly disagree
Total	30.29	47.80	16.32	5.58
Continent/country
Europe	23.61	43.42	23.43	9.55
Andorra	25.65	61.52	11.32	1.50
Bulgaria	18.32	27.15	32.93	21.59
Cyprus	36.84	37.99	20.77	4.40
Finland	18.48	45.45	28.79	7.27
Germany	16.54	34.20	30.43	18.83
Italy	20.41	41.52	30.05	8.02
Moldova	27.25	37.31	29.12	6.32
Norway	31.29	46.63	15.45	6.63
Poland	22.06	51.50	21.20	5.25
Romania	16.57	37.32	29.67	16.44
Serbia	31.75	46.19	18.25	3.81
Slovenia	18.41	55.84	18.61	7.14
Spain	23.57	55.46	16.43	4.54
Sweden	18.32	66.70	14.07	0.91
Switzerland	33.31	46.68	15.88	4.13
Ukraine	27.61	32.83	25.76	13.80
North America	26.99	52.59	17.29	3.12
Canada	30.97	55.73	11.46	1.85
USA	19.92	47.00	27.68	5.40
Central/South America	30.61	53.24	13.53	2.62
Argentina	16.22	52.11	25.41	6.27
Brazil	26.94	51.65	19.18	2.23
Chile	27.31	45.69	20.15	6.85
Columbia	41.01	55.15	3.10	0.74
Mexico	35.40	55.21	7.31	2.07
Trinidad &Tobago	25.99	48.42	23.34	2.24
Uruguay	18.65	59.98	18.85	2.52
Australia	22.91	57.75	17.89	1.45
Australia	22.91	57.75	17.89	1.45
Asia	29.75	51.83	13.33	5.09
China	13.41	70.70	14.31	1.58
India	36.75	41.31	17.91	4.02
Indonesia	30.83	61.15	6.98	1.04
Iran	29.90	51.73	16.06	2.30
Japan	14.57	79.08	5.64	0.71
Jordan	37.29	35.18	14.69	12.84
Malaysia	41.00	49.08	9.08	0.83
South Korea	15.67	67.75	13.83	2.75
Thailand	32.96	63.19	3.72	0.13
Turkey	38.16	46.18	12.07	3.58
Vietnam	34.38	58.73	6.89	0.00
Africa	41.85	42.43	11.57	4.15
Burkina Faso	45.44	40.62	10.90	3.04
Egypt	28.61	28.91	23.68	18.79
Ethiopia	41.53	50.55	6.76	1.16
Ghana	46.30	39.76	12.68	1.27
Mali	61.80	32.77	4.55	0.88
Morocco	16.28	29.94	30.62	23.16
Rwanda	43.08	55.05	1.40	0.47
South Africa	42.25	44.29	10.17	3.28
Zambia	33.26	38.95	22.32	5.47

Note: Table entries represent the percentage of respondents who reported that they ‘Strongly agree,’ ‘Agree,’ ‘Disagree,’ or ‘Strongly disagree’ with the statement ‘I see myself as a citizen of the world.’ The total number of respondents who provided an answer was 65,269.

The finding that a large proportion of the respondents in the sample identify as world citizens is empirically intriguing. However, I take a cautious approach. When reporting the descriptive statistics in the following section, I only concentrate on the ‘Strongly agree’ answers. ¹⁴ Filtering out the ‘Agree’ answers under-represents the total proportion of world citizens. Because this strategy guards against misappropriating potentially half-hearted world citizens as cosmopolitans, it should increase our confidence in the results.

Cosmopolitan identification varies across continents and countries. About 42% of strong cosmopolitan identifiers live in Africa, 31% in Central and South America, 30% in Asia, 27% in North America, and about 24% and 23% in Europe and Australia, respectively. In Africa, Mali has by far the largest percentage, about 62%, of world citizens, whereas Morocco has the lowest, about 16%. In Ghana, Burkina Faso, Rwanda, South Africa, and Ethiopia, about 42% of the respondents identify as world citizens. Zambia and Egypt are on the lower end of the spectrum with 33% and 27%, respectively. In Asia, 41% of the Malaysian participants identify as world citizens. This is the highest on the continent.

Percentages in Turkey, Jordan, India, Vietnam, and Thailand range from 38% to 33%. About 30% of the Indonesians and Iranians in the sample see themselves as world citizens. In contrast, world citizenship is not widespread in South Korea, Japan, and China. China has by far the smallest proportion of world citizens. In the Australian mainland, about 23% of the respondents are strong cosmopolitan identifiers. In Central and South America, Argentina has the lowest and Columbia has the highest percentages of world citizens, at 16% and 41%, respectively. Uruguay also has a relatively small percentage of cosmopolitans, about 19%. While percentages in Trinidad and Tobago, Brazil, and Chile range from 26% to 27%, about 35% of the Mexicans in the sample see themselves as world citizens. In North America, Canada houses a substantially larger number of world citizens than the US. About 31% of the Canadian respondents strongly identify as world citizens, whereas only 20% of the US participants see themselves as cosmopolitans. In Europe, Cyprus has the highest percentage, at 37%, of cosmopolitan identifiers, while Germany and Romania have the lowest, at around 16.5%. Percentages in Switzerland, Serbia, Norway, Ukraine, and Moldova range from 33% to 27%. In Spain, Poland, and Italy, slightly over 20% of the respondents identify as world citizens. In Finland, Slovenia, Bulgaria, and Sweden, the proportion of cosmopolitans drops to 18%.

Which values drive world citizens across the globe? Table 2 presents the relative risk ratios obtained from three multinomial logistic regression estimations. The first two models analyze the discrete effect of values on the opposite sides of the Schwartz value circle on cosmopolitan allegiance. The third focuses on the joint effect of eight core values. I first report the results from the additive models followed by a discussion of the findings from the joint model.

OPEN IN VIEWER

Table 2.

Multinomial logistic regression results for the effect of values on cosmopolitan allegiance.

	Model 1			Model 2			Model 3
	Strongly agree	Agree	Disagree	Strongly agree	Agree	Disagree	Strongly agree	Agree	Disagree
Universalism	1.267****	1.106***	1.028				1.453****	1.247****	1.074*
	(0.050)	(0.042)	(0.041)				(0.055)	(0.045)	(0.041)
Benevolence	1.070*	1.047	1.041				1.231****	1.189****	1.094*
	(0.043)	(0.041)	(0.044)				(0.049)	(0.045)	(0.045)
Hedonism	1.053**	1.07***	1.028				1.115****	1.129****	1.039
	(0.027)	(0.027)	(0.027)				(0.33)	(0.032)	(0.032)
Achievement	1.079**	0.986	0.975				1.159***	1.047*	0.993
	(0.035)	(0.029)	(0.033)				(0.039)	(0.32)	(0.034)
Power	0.947*	0.966	1.032****
	(0.029)	(0.030)	(0.033)
Stimulation				1.182****	1.232****	1.141****	1.378****	1.376****	1.201****
				(0.034)	(0.034)	(0.034)	(0.040)	(0.038)	(0.036)
Self-direction				1.076**	1.049*	0.991	1.224****	1.155****	1.033
				(0.035)	(0.030)	(0.033)	(0.041)	(0.374)	(0.035)
Tradition				0.853****	0.875****	0.9177**	0.926**	0.930*	0.932*
				(0.030)	(0.030)	(0.033)	(0.036)	(0.035)	(0.038)
Conformity				0.939*	0.9337**	0.972
				(0.031)	(0.030)	(0.033)
Security				0.863****	0.910***	0.093*
				(0.030)	(0.031)	(0.034)
Trust	1.92****	1.959****	1.374****	1.79****	1.846****	1.32****	1.750****	1.792****	1.312****
	(0.152)	(0.150)	(0.113)	(0.141)	(0.142)	(0.110)	(0.138)	(0.137)	(0.108)
Religiosity	0.881*	0.757****	0.918	1.09	0.906	1.016	1.099	0.925	1.022
	(0.063)	(0.053)	(0.069)	(0.080)	(0.064)	(0.077)	(0.080)	(0.065)	(0.077)
Urban	1.027*	1.029*	1.035*	1.023	1.031**	1.03**	1.027*	1.030**	1.036**
	(0.016)	(0.016)	(0.017)	(0.016)	(0.016)	(0.017)	(0.167)	(0.016)	(0.017)
Left orientation	0.939****	0.944****	(0.977)*	0.941****	0.944****	0.976*	0.934****	0.940****	0.976*
	(0.012)	(0.012)	(0.013)	(0.012)	(0.012)	(0.013)	(0.012)	(0.012)	(0.013)
Education	1.018*	0.997	0.999****	1.017	0.994	0.998	1.009	0.990	0.997
	(0.011)	(0.010)	(0.011)	(0.11)	(0.010)	(0.011)	(0.001)	(0.010)	(0.011)
Income	1.100***	1.12****	1.058	1.074****	1.10****	1.049***	1.083*****	1.106****	1.051****
	(0.016)	(0.015)	(0.016)	(0.016)	(0.016)	(0.016)	(0.016)	(0.016)	(0.016)
Male	1.108*	1.010	1.026	0.982	0.898*	0.962	1.019	0.931	0.982
	(0.068)	(0.060)	(0.066)	(0.061)	(0.054)	(0.062)	(0.064)	(0.056)	(0.064)
Age	0.982****	0.984****	0.989****	0.990****	0.990****	0.993***	0.988	0.989****	0.992****
	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)	(0.002)
Europe	0.366***	0.570***	0.488****	0.384****	0.601***	0.492****	0.362****	0.562****	0.480***
	(0.060)	(0.093)	(0.082)	(0.193)	(0.098)	(0.082)	(0.605)	(0.092)	(0.080)
North America	0.925	1.623**	1.066	0.932	1.632**	1.046	0.833	1.470*	0.991
	(0.192)	(0.332)	(0.224)	(0.193)	(0.333)	(0.219)	(0.174)	(0.301)	(0.208)
Central/South America	1.354	2.232****	1.145	1.61**	2.64****	1.206	1.379*	2.270****	1.129
	(0.260)	(0.423)	(0.223)	(0.307)	(0.497)	(0.232)	(0.265)*	(0.431)	(0.219)
Australia	1.983	4.00****	2.472***	1.941*	3.81****	2.35**	1.745*	3.546****	2.26**
	(0.698)	(1.39)	(0.876)	(0.685)	(1.32)	(0.834)	(0.616)*	(1.235)	(0.802)
Asia	0.706*	0.8925	0.357****	0.835	1.049	0.380****	0.863	1.060	0.390****
	(0.124)	(0.155)	(0.065)	(0.146)	(0.182)	(0.068)	(0.153)	(0.185)	(0.071)
Africa	1.465*	1.451*	0.544***	1.192	1.213	0.503****	1.338*	1.326*	0.520****
	(0.281)	(0.276)	(0.107)	(0.232)	(0.235)	(0.101)	(0.262)	(0.255)	(0.104)
Wald chi2	1602.69****			1641.06****			1862.29****
Number of observations	24894			24894			24894
Log pseudo-likelihood	−28120.773			−28089.762			−27945.416

Notes: Table entries represent the relative risk ratios, with robust standard errors in parentheses. Values above 1 indicate a positive relationship and below 1 a negative relationship. ‘Strongly disagree’ is the reference category. Trust, religiosity, and continent variables are factor variables. Male is a dummy variable. Missing values were dummied out. *p ≤ 0.10; **p ≤ 0.05; ***p ≤ 0.01; ****p ≤ 0.001.

As predicted, self-transcendence values positively relate to cosmopolitan identity (Model 1). Universalism is the main motivator of cosmopolitan attachment. The firmer the commitment to the well-being of all human beings, the stronger the sense of world citizenship. Going from the minimum to the maximum of universalism, the predicted probability of strong cosmopolitan attachment more than doubles. Benevolence also positively associates with world citizenship. There is a 19% increase in the likelihood of strong cosmopolitan identification as we move from the minimum to the maximum of benevolence values. These two prime findings strongly support the normative model of world citizenship centered on concern for the well-being of human beings. As the importance placed on moral obligations rises, the attractiveness of world citizenship increases.

Yet what moves individuals to see themselves as world citizens is not merely other-regard. Self-interest-centered values also drive world citizens. Achievement and hedonism values are positively associated with world citizenship (Model 1). Moving from the lowest to the highest scores on achievement and hedonism, the probability that a respondent holds a strong commitment to world citizenship increases by 85% and 19%, respectively. These results indicate that values that reflect humanitarian and egoistic motivations are both compatible with cosmopolitan allegiance. Individuals who care about the welfare of others, as reflected in universalism and benevolence, are drawn to the role of world citizenship, but so are those who care about their personal interests and success, as evinced by achievement and hedonism values. These findings lend credence to both normative and economic understandings of world citizenship.

Contrary to expectations, however, power is negatively related to world citizenship (Model 1). People with a strong power orientation are less likely to see themselves as world citizens. This is an important finding. It suggests that world citizens have a distaste for controlling and dominating people, for seeking success at the expense of others. World citizens are interested in personal success, wealth, and pleasure but they self-enhance without distressing others. This suggests that world citizenship attracts people who do not conceive of self-interest in zero-sum terms.

As predicted, results show that conservation values negatively relate to cosmopolitan attachment (Model 2). Higher emphasis on tradition, conformity, and security inhibits cosmopolitan identity. For a unit increase in tradition, conformity, and security scores, the odds of strong attachment relative to no attachment decrease by about 15, 6, and 14 percentage points, respectively. For example, going from the minimum value to the maximum value of security, the predicted probability that a person answers ‘Disagree’ to the world citizenship question increases by about 36%. Similarly, the probability that a respondent answers ‘Agree’ decreases by about 11% when going from the minimum to the maximum of conformity. In the same vein, there is a 26% decrease in the predicted probability of cosmopolitan attachment when going from the lowest to the highest score on tradition values.

Stimulation and self-direction values are also associated positively with cosmopolitan identity. The odds of strong attachment as compared to no attachment increase by 18% for a unit increased in stimulation and by about 8% for an increase in self-direction scores. For instance, the probability that a respondent answers ‘Disagree’ declines by about 33% and answers ‘Agree’ rises by 36.7% when going from the minimum to the maximum value of stimulation. Similarly, the probability that someone chooses ‘Disagree’ declines by about 36.5% and the probability that someone chooses ‘Strongly agree’ increases by 37% when going from the lowest to the highest degree of self-direction.

These findings support the liberal and cultural models of world citizenship. As the importance placed on conformity, obedience, and order rise, the likelihood of cosmopolitan allegiance declines significantly. In contrast, the more a person values freedom, novelty, and openness, the more likely he or she is to identify as a world citizen. This implies that world citizens are both free agents and multiculturalists. They cherish independence and diversity alike.

The importance of values is clear, but trust also emerges as a critical predictor. The odds of strong cosmopolitan attachment relative to no attachment are 90% and 79% higher for individuals who trust others as opposed to cynics, according to Models 1 and 2, respectively. This is important, if not surprising. Without trust, any kind of tie to strangers seems unlikely. What is crucial to note is that even after we take into account the obvious effect of trust, values remain significant drivers of cosmopolitan attachment.

On average, demographic and political control variables have smaller effects on world citizenship. Urban residence, education, and income positively relate to cosmopolitan identity across both models. People who live in urban locations, are more educated, and have higher levels of income are driven toward world citizenship. This result is consistent with the findings in the existing literature. Individuals who are not religious, lean toward the Right, and are younger are also more likely to be cosmopolitan identifiers according to both models. Results for religiosity and age also echo the findings of earlier works (Norris, 2000; Norris and Inglehart, 2009). Thus far, findings on the role of ideology have been inconsistent. My analysis shows that Right political ideology is positively related to world citizenship (Pichler, 2012).

The effects of education and gender are unbalanced. In Model 1, the effect of education is inconsistent across levels of cosmopolitan attachment. In Model 2, the effect of education is not significant across any of the categories of the dependent variable. Norris and Inglehart (2009) found that men are more likely to see themselves as world citizens. Comparing strong attachment to no attachment, this is indeed the case according to Model 1. Yet there is a reversal if we look at Model 2. Comparing medium identification to no identification, women are more likely to see themselves as world citizens.

Results show that, in general, participants living in Africa, North America, Central and South America, and Australia are more likely to be cosmopolitan than those living in Europe and Asia. Continent effects across the additive models are largely consistent, and using country dummies does not significantly change the substantive results. When I split the data by continents and estimate separate models, the value-driven logic of cosmopolitan allegiance is still supported even though certain values become non-significant.

In Europe, universalism is the most important factor for explaining cosmopolitan allegiance. However, stimulation, self-direction, and opposition to tradition also exert large effects on world citizenship. This suggests that the motivational calculus of European world citizens is multidimensional, encompassing morality, diversity, and freedom. In North America, resistance to conformity is the most critical predictor of a cosmopolitan self-view. However, universalism also drives North Americans, indicating that freedom and morality both shape world citizenship in the region. In Asia, cosmopolitan attachment is largely shaped by self-transcendence values. In Central and South America, world citizens are primarily motivated by hedonism and achievement values. Mainland Australia seems to be an outlier. Stimulation values are associated with world citizenship but generalized trust exerts a larger effect on cosmopolitanism than values. In Africa, benevolence has the largest effect on cosmopolitan allegiance. The effects of openness and conservation values are in the hypothesized direction. The only exception to this pattern is security. In Africa, emphasis on personal and group security is consistent with world citizenship.

To summarize, results from Model 1 show that self-transcendence and self-enhancement values both motivate world citizens. Findings from Model 2 indicate that openness values contribute to world citizenship but conservation values hinder it. But how do alternative value orientations jointly relate to cosmopolitan allegiance? Model 3 addresses this question. It includes universalism, benevolence, hedonism, achievement, stimulation, self-direction, and tradition values. This combination of values makes it possible to analyze the motives hypothesized to inspire world citizens while evading the previously discussed problem of including strongly correlated values in a single estimation. ¹⁵

The far-right column of Table 2 presents the relative risk ratios obtained from the estimation of Model 3. Consistent with the previous findings, universalism, benevolence, hedonism, achievement, stimulation, and self-direction values lead to a sense of world citizenship but tradition values thwart this self-description. The risk that a respondent strongly identifies as a world citizen relative to the baseline category of no identification increases by a factor of 1.453 for a unit change in universalism and by a factor of 1.231 for a unit change in benevolence. These translate to over 100% and 50% increases when going from the minimum to the maximum of the universalism and benevolence scores, respectively.

Achievement also has a substantial positive influence on world citizenship. The predicted probability of strong identification rises to over 90% when moving from the minimum to the maximum of achievement. Hedonism exerts a smaller but significant effect. The odds of moderate attachment relative to no attachment increase by about 13% for a unit increase in hedonism. This corresponds to a 20% increase in the probability of world citizenship. Replicating earlier findings, stimulation and self-direction values also facilitate world citizenship. The predicted probability of strong cosmopolitan attachment rises by 38% and 75.5% when moving from the minimum to the maximum of stimulation and self-direction values, respectively. In contrast, the importance placed on tradition decreases the likelihood of world citizenship. For instance, the predicted probability of strong identification declines by about 5% when going from the minimum to the maximum of tradition.

Consonant with the earlier results, trust and urban residence positively relate to cosmopolitan attachment, while Left orientation and age relate negatively. Respondents who trust others, live in urban locations, and have higher incomes are more likely to see themselves as world citizens. Similarly, those who are on the Right side of the political spectrum and are younger are more likely to be drawn to world citizenship. The coefficients for religiosity, gender, and education are not significant in Model 3.

Findings for continent effects are mainly consistent with those obtained from the previous analysis. On balance, respondents from North America, Central and South America, Australia, and Africa are more likely to be cosmopolitan than those living in Europe and Asia. In Europe, universalism and achievement values have the largest effect on cosmopolitan allegiance compared to other values. Openness values of stimulation and self-direction are also critical motivators. However, hedonism has a negligible influence, and tradition is not statistically significant.

In North America, universalism, benevolence, achievement, hedonism, and stimulation each move world citizens. In accord with previous findings, in Central and South America, world citizens are primarily motivated by hedonism and achievement. In mainland Australia, stimulation values and trust are the key predictors of cosmopolitan allegiance. In Asia, universalism, hedonism, and stimulation values have large positive effects, and benevolence and self-direction have smaller positive effects on world citizenship. Interestingly, submission to tradition has the largest negative effect on world citizenship in Asia compared to other continents. The more an individual resists social orthodoxies, the more he or she is attracted to world citizenship. In Africa, values most conducive to cosmopolitan allegiance are benevolence and resistance to tradition. While regional differences should be further explored by future research, it is evident that values anchor the psychological underpinnings of world citizenship across cultures.

These findings show that cosmopolitan motivation in practice is complex. World citizens are driven by multiple motivations. Sociotropic and egoistic motives, as expressed by self-transcendence and self-enhancement values, and regard for diversity and independence, as expressed by openness values, jointly drive world citizens. Of course, I do not contend that every person with the relevant values will always identify as a world citizen. Rather, my analysis shows that in 57 countries across the globe, people with certain value priorities are more likely to see themselves as world citizens than others.

Conclusion

This article offers and tests a value explanation of the psychological underpinnings of world citizenship. I find that self-transcendence, self-enhancement (except power), and openness-to-change values, along with generalized trust, lead to the internalization of world citizenship, while conservation values obstruct this self-categorization.

Michael Walzer (1996: 126) famously noted that it is ‘odd’ for the world, the outermost circle of identity, to be a source of allegiance. The explanation of value-driven cosmopolitanism sheds light on this word puzzle. To my knowledge, my analysis is a first. Even though normative theorists have long associated cosmopolitan principles with certain values, no study has clarified and empirically tested the micro-motivational assumptions associated with these broad claims. The value analysis offered here contributes to the literature on cosmopolitanism by illuminating people’s world citizenship and providing the missing empirical evidence.

This study also helps establish an empirically grounded research agenda on cosmopolitanism. My analysis leads to a better understanding of who world citizens are and what they care about. I find that the normative, economic, cultural, and liberal models on world citizenship are all correct to different degrees. World citizenship is a multifaceted role compatible with varied outlooks on life. Self-transcendence values characterized by other-regard lead to the internalization of world citizenship, but so do self-enhancement values marked by self-regard. Openness-to-change values stressing change and independent thought and action also induce cosmopolitan identification. What hinders cosmopolitan attachment are conservation values that call for self-restraint and obedience to authority as well as power values that legitimate dominating and controlling others. These findings indicate that world citizenship in practice is a rich role congruent with multiple motivational goals. Belonging to the world at the mass level encompasses morality, liberty, and diversity, as well as self-interest.

My findings have important implications. Humanitarian and egoistic goals are sometimes thought to be antithetical to each other. Results cast doubt upon the conventional dichotomy of self- versus other-regard. At the mass level, sociotropic and selfish motives are congruent with world citizenship. World citizenship is not the fellowship of saints, and self-love does not exclude world citizenship. World citizenship appeals to self-transcenders and self-enhancers alike. Only self-enhancers who dominate and control others for personal achievement do not see themselves as cosmopolitans. This finding should encourage the realist critics of cosmopolitanism to revisit their pessimistic assumptions about individuals’ ability to be cosmopolitan and support cosmopolitan politics (Beardsworth, 2010; Deudney, 2007; Zolo, 1997). Similarly, cosmopolitan theory will benefit from exploring why humanitarian and selfish motives are both consistent with world citizenship.

Second, this study demonstrates that the universal morality of world citizenship is coupled with concern about liberty and diversity in the eyes of publics. In individuals’ minds, world citizenship extends beyond obligations toward global compatriots. It is also associated with independence, change, and diversity, as reflected in openness values. This finding provides critical empirical support for the claim that legitimate global governance must emphasize freedom and pluralism along with morality (Archibugi, 2008; Archibugi and Held, 1995; Brown, 2009; Franceschet, 2000; Held, 1997; Kaldor, 1999).

The results of this study also open up new avenues of research. I find that world citizens can be humanitarians, multiculturalists, freedom seekers, and egoists at the same time. This finding should inspire inquiries into types of cosmopolitans. For example, people prioritizing universalism and benevolence values may support more egalitarian forms of cosmopolitanism, while those valuing achievement and hedonism may place less emphasis on equality. Uncovering types of cosmopolitans will enrich normative theories of cosmopolitanism and add to studies on bottom-up cosmopolitanism.

My findings should also be of interest to scholars interested in how culture affects individuals’ commitment to world citizenship. For instance, Africa has the highest percentage of strong cosmopolitan identifiers, whereas Europe has the lowest. It could be that the effect of personal values on world citizenship is smaller when there is a salient regional community, such as Europe. Similarly, individuals in Africa may be more likely to see themselves as world citizens because many of the boundaries in the continent were arbitrarily drawn in the second half of the 20th century. Future studies of public cosmopolitanism should explore these questions.

The ideal of world citizenship is hardly new. It has existed for over 2000 years. The possibility of world citizenship, however, became particularly important in the post-Cold War era. This was because, as Thomas Pogge (1992: 48) explained, ‘[t]he human future suddenly seem[ed] open.’ Even though a number of cosmopolitan principles were enshrined in the post-Second World War international system, cosmopolitanism has gained new significance in international law and politics since the early 1990s (Archibugi and Held, 1995; Cheah, 2006; Held and McGrew, 2002; Linklater, 1998). Today, it is an important dimension of global governance. The establishment of the International Criminal Court (ICC), the adoption of the first resolution on the Responsibility to Protect (R2P), and the growing political will to democratize international institutions attest to this.

Managing complex interdependence and addressing problems of common concern starts with effective institutions and collaboration between state and non-state actors (Held and McGrew, 2002; Nye and Donahue, 2000). Yet it does not end there politically, and normatively, it should not end there. Global governance also requires some sense of world citizenship at the public level, simply because ‘diffuse support’ is necessary for the health of the international system (Easton, 1965). Whether a ‘truly’ communal bond is feasible on a global level is a separate question. However, world citizenship is likely to increase public support for international institutions and inspire individuals to act as members of the world community. While more work is needed to determine how much world citizens support cosmopolitan ideals, the findings reported here could inspire cautious optimism. If individuals motivated by self- and other-care, as reflected in self-transcendence and self-enhancement values, both identify as citizens of the world, cosmopolitan politics does not require the arrival of selfless saints. If multiculturalists moved by openness-to-change values see themselves as world citizens, ‘unity in diversity’ seems feasible. If conservationism is the main foil of cosmopolitanism, emphasis on individual rights and liberties in the discourse and practice of global governance could cultivate a sense of world citizenship.