Introduction

This chapter is a product of the research conducted in the Collaborative Research Center “Global Dynamics of Social Policy” at the University of Bremen. The center is funded by the Deutsche Forschungsgemeinschaft (DFG, German Research Foundation)—project number 374666841—SFB 1342.

In this chapter, we present an overview of the empirical results presented in this volume. We begin with a summary of our theoretical arguments and research design in the subsequent section of this chapter. We argue in favor of a combination of sociological neo-institutionalism with a concept of global cultural diversity. In our view, culture might be an important moderator of the global diffusion and institutionalization of both education systems and guiding principles of education. We briefly summarize the results of our analyses and juxtapose results and theoretical arguments. Since our results cover various aspects of global education and the theoretical arguments potentially allow a variety of different predictions, we also discuss whether the results are overall in line or in contradiction to these arguments.

In our view, the empirical results corroborate the theoretical assumptions of neo-institutionalism. At the same time, cultural diversity and cultural difference do indeed play a crucial role in moderating the diffusion process of Western educational institutions, ideational frames, and leitmotifs. In the concluding section of this chapter, we give an outlook for future research and argue that research on global education should take up the challenge and develop more sophisticated concepts of global culture, that are accessible to measurement and to relational methods of social network analysis, dimensional analysis, typology building, and classification.

Recapturing the Theoretical Approach and Research Design

Our theoretical concept follows sociological neo-institutionalism (Meyer and Rowan 1977; Meyer et al. 1997). In this perspective, institutions structure our taken-for-granted knowledge and our social order. During the transition into modernity, modern rational institutions became highly efficient bureaucracies and require appropriately educated and specialized administrative staff. Once established, these rational and bureaucratic institutions seem to take on a life of their own; rationalization became self-perpetuating and self-reinforcing. Citizens adapted to following the formalized rules, where the administrative staff operates in an “iron cage” (DiMaggio and Powell 1983) or “steel-hard casing” (Weber 1972) of depersonalization and dehumanization. A “caste” of managers, civil servants, and functionaries has monopolized political and economic power. Yet, this form of social order allows the organization of knowledge, science, and technology in a highly efficient way. The same is true for the economy and military. Following this, modern capitalism and bureaucracies tend to spread over the world. Most International Organizations (IOs) also show these institutional characteristics, so countries that aim to become acknowledged members of these IOs tend to adapt their own institutional structures—at least as a performative act of “myth and ceremony” (Meyer and Rowan 1977). These processes recently became obvious in terms of expansion of higher education and its effect on core characteristics of a society, for example, the rationalization of society and state, social and political mobilization as well as global integration by membership in IOs (Schofer et al. 2021).

There is considerable empirical evidence on the globalization of rational institutions, in particular with respect to education (Meyer et al. 1992). This evidence corroborates the basic idea of neo-institutionalism. Indeed, Western modern rational and bureaucratic organizations tend to spread around the world. Even the fact that adopting an institution occurs just at the “front stage” (Goffman 1959) in some countries, while their actual performance follows traditional routines and practices, is in line with the theory. It seems that these countries simply do have to play along with the game. They need to acquire legitimacy in the IOs they are a member in but also in their domestic population.

In our research, we started from sociological neo-institutionalism and enhanced the argument with a stronger focus on global cultural diversity. Max Weber already noticed the spread of Western rational culture and bureaucratic institutions around the globe, and he was interested in how different cultures related to rationalization (Weber 1972). These cultures developed along the lines of different religious views of the world and the afterworld. According to Weber, Protestant sects in Europe and North America and their specific theological approach to the problem of the theodicy initiated the cultural development toward rationalization, which is why Weber used the term “occidental rationalism”. The occidental culture took a very specific developmental path. Modern Western individualism, capitalism, and means–ends rationality was not necessarily appreciated in other cultural traditions. As studies in anthropology on trade and exchange (Mauss 1967; Fiske 1991, chp. 14) and recent interpretations of the history of economic thinking suggest, regarding “markets as morality-free zones” (Bowles 2017, 25) is rather inappropriate to human nature (Turner 2021, 110, 183). Yet, this is exactly what happens in today’s highly dynamic markets, particularly in financial markets and stock exchanges. Another example of the difference between culture as a legitimizing foundation of bureaucratic authority and both the legal state as well as societies based on lineages as the fundamental units of social organization is the response toward rectification of transgressions. Modern institutions operate sine ira et studio, without particular social obligations and emotions (Weber 1972, 129). In contrast, in societies based upon kinship there is a much lower tendency to accept a punishment executed by a nonmember of one’s kin group: “Punishment by a nonmember of a member’s misbehavior may itself be considered a transgression requiring rectification or inviting retaliation” (Bowles 2017, 141). Social order based on extended kinship groups was historically the norm rather than the exception, also in the northern countries, for example, Scotland and Ireland (Weiner 2013), and are comparatively important even today in some cultures (Haidt 2012; Inglehart 2018, 81–82). After the occidental rationalism became unleashed in the West, however, capitalism and institutionalized bureaucracies emerged in an astonishingly short historical period. Positive aspects of this development are enlightenment, progress in science and technology as well as civil rights and liberties at an unprecedented scale (Pinker 2018), at least for citizens in Western democracies. The flipside of the coin implies bureaucratically organized atrocities (Bauman 1989), mass exploitation, and environmental deterioration. Consequently, anthropologists recently described the highly individualistic European and North American culture as WEIRD (western, educated, industrialized, resourceful, democratic) (Schulz et al. 2019; Henrich 2020). Specific moral orientations come along with this WEIRD culture. Moral emotions are triggered by the evaluation of actions in six dimensions of what Haidt calls the “moral matrix”, and the relative weight of each dimension of this matrix considerably differs across cultures. The WEIRD “liberal moral matrix” puts substantial weight on the dimensions of care–harm, liberty–oppression, and fairness–cheating. In contrast, loyalty–betrayal, authority–subversion, and sanctity–degradation, which are quite important, for example, in India (Haidt 2012, 121–22), are not valued high, if not even rejected (Haidt 2012, 351).

Given these considerable average differences in cultural orientations between the WEIRD culture and other cultures (Henrich 2020, chap. 6), why should there be an undamped diffusion of Western institutions? If the legitimacy of such institutions requires specific cultural conditions, it will be likely that different cultures moderate the speed of the diffusion process, even if the process itself is hard to impede. This might hold especially in the institutional field of education because of its crucial role in the intergenerational reproduction of culture.

We developed our concept of ‘cultural spheres’ based on existing cultural typologies that suggest a wide-ranging set of characteristics for a country’s culture. In our standardized analyses, our methodological approach was to recode cultural characteristics into binary variables and to generate a two-mode network of cultural proximity. In so doing, we extracted fuzzy-set clusters of global cultures from the data. Moreover, analyzing documents on educational issues and programs published by IOs, we combined standardized and nonstandardized methods of text analysis. In addition to the macro-quantitative approach, which included most countries in the world, our qualitative research design is based on a 2 × 2 table. One dimension was ‘derivative vs. dedicated’ IOs; the second dimension was whether these organizations were open to any state or restricted by specific criteria, such as geographical position or religion (Windzio and Martens 2021).

We analyzed whether the valued network of cultural proximity had an influence on the diffusion of compulsory education, whether there are different trajectories of adjusting the length of compulsory education and whether globalization affects the performance of secondary education systems. Moreover, we investigated regional clusters in the OECD discourse on education, which differ according to similarity in economic and cultural conditions. Using qualitative methods, we checked similarities and differences in ideational framing of the education purpose in four selected IOs, what kind of ideational framing the SEAMEO and the ICESCO apply, both of which are not universalistic organizations, and whether we find different frames when we analyze these documents using standardized methods of text analysis.

Summarizing Empirical Results in the Light of the Theory

The chapters collected in our volume focus on various issues but are integrated by our theoretical framework outlined in Chap. 1. Each of the nine empirical chapters presents empirical results related to our theoretical framework.

Chapter 2 investigates diffusion of compulsory education from 1789 to 2010 and shows that there are strong and consistent effects of similarity in the cultural spheres network, whereas there is no effect of ties in the network of colonial legacies. Moreover, the cultural clusters ‘WEIRD’ and ‘Catholic, Spanish-speaking’, but also ‘Eastern Europe’ show much higher adoption rates than ‘dominantly Muslim’ countries, or countries that are ‘not dominantly Muslim African/East Asian’. Since GDP per capita and levels of democratization have been controlled in this analysis, results indicate that cultural similarity does matter in the diffusion process. The five cultural spheres we extracted from the data by using two-mode network analysis do indeed show considerably different inclinations to adopt compulsory education. State organized compulsory education originated in Europe and Western countries, then quickly spread among ‘Catholic, Spanish-speaking’, and ‘Eastern European’ countries. The cultural spheres of ‘dominantly Muslim’ and ‘not dominantly Muslim African/East Asian’ countries lagged behind in this process. And this is what we expected according to our theory. Transferring important tasks to the state and public institutions and thereby exposing your children to a more or less standardized curriculum and giving them access to literacy and mass communication might accelerate the process of individualization and thereby could support the ‘Westernization’ of culture and society. Moreover, the role of the state in the Islamic tradition could be different from the Western tradition. Certainly, imposing shari’a law on the society or regarding the state as a caliphate where the leader’s government should work on the realization of a divine law and order is close to the ideology of Islamism rather than Islam (Tibi 1994, 2012). However, it “ is likely that many mainstream Muslims would theoretically welcome a renewal of the caliphate, but feel that it is not for them to bring this about” (Silverstein 2010, 75). Silverstein’s speculation can prove false, but one reason why states in Muslim countries were comparatively weak in history, although at the same time often authoritarian, could be that interpretations of Qur’an and the Sunna by Muslim scholars did indeed suggest a kind of divine order (Tibi 1994). There are plenty of Surahs in the Qur’an related to educational goals and practices (Antes 1991, 70–71). If religious education in Qur’anic schools provided basic literacy, the religious tradition was perhaps a good alternative and to some degree a substitute for state-regulated compulsory education.

There are also huge intra-cultural differences in the sequences on the durations of compulsory schooling, as presented in Chap. 3. Interestingly, the cluster ‘Catholic, Spanish-speaking’ has comparatively long durations, but it is due to an overall trend to introduce compulsory preprimary education. In contrast, WEIRD countries tend to hold back from implementing such policies, even though these policies are propagated by the UNESCO. Clustering according to the sequences of length of compulsory education reveals that cluster 2 ‘stagnant short’ (N = 28) consists of mainly Arabic and African countries, but there is no clear distinction with respect to cultural spheres when the clustering is based on the sequences. However, there is a correspondence between cluster membership and gender rights: in the cluster ‘Long and extending’, where the duration of compulsory education is comparatively long, but nevertheless still extending, the mean value of the gender rights scale are highest, whereas it is lowest in the cluster ‘Late and stagnant’. Hence, there might be a correspondence between culture and duration of compulsory education, since gender rights or the tendency toward their absence, is an important aspect of the so-called honor cultures, which include segmentary lineages or patriarchal clans (Nisbett and Cohen 1996; Basáñez 2016; Henrich 2020, 283).

A further question was whether globalization has an impact on the performance of secondary education systems. The influence of IOs on domestic education policies has been clearly shown in previous studies (Martens et al. 2014), but it is not yet clear whether the actual performance is affected by globalization. In our study, we used global trade, global migration, and global student mobility as indicators of globalization. We measured each of these dimensions as networks and thereby captured recent ideas about globalization in a rather direct manner. In Chap. 4, we applied a model for network evolution that is able to separate the effects of selection into a particular network tie from the influence these ties have on actor attributes, in our case, on the performance of a country’s secondary education system. Nonetheless, our empirical results did not illustrate a significant selection of network ties according to similarity in PISA scores or rankings, nor any effect from social influence. This is an interesting result when considering the background of the responsiveness of some countries to the results of the PISA study (Martens et al. 2014). Rankings published by the OECD had considerable effects and triggered intensive debates in some countries. Surely, the top-down influence by ‘naming and shaming’ and by benchmarking the performance of education systems is a different mechanism than horizontal interdependencies via networks of trade, migration, and student mobility. Nevertheless, this is what we would expect according to the theories of globalization: the stronger the interconnectedness of countries, the stronger are the lateral influences in the world system. If this argument is still correct, then our results indicate that it depends on the respective dimension of the network in which countries are tied to each other. At least we can conclude that the three network dimensions investigated in our study do not influence educational performance, despite the policy changes triggered by global actors, such as the OECD. Perhaps our results also indicate that the implementation of policy reforms is at times rather a performative act (Steiner-Khamsi 2012) and ‘myth and ceremony’ (Meyer and Rowan 1977).

Chapter 5 focused on the four major IOs, the World Bank, OECD, UNESCO, ILO, and their ideational framing of education. The World Bank now acknowledges the limits of a liberal market model but its traditional economic paradigm is still present. The OECD follows a twofold education leitmotif, namely, the individual benefit and a better quality of life due to education, but also better employability and higher wages. However, the states are supposed to also benefit from the overall increase in education due to economic prosperity. The UNESCO does not neglect the economic consequences of education, but addresses the improvement of people’s living conditions, a better inclusion of marginalized groups, reducing poverty, and improving social justice and social cohesion. The leitmotifs of the ILO became more diverse over time as well. Originally, the ILO managed labor markets and employability, but now also covers equal opportunities and social cohesion, even though the improvement for individual workers’ conditions is still considered highly important. The most important result of this comparative analysis is the tendency toward integrativeideas. Whereas these organizations formerly dealt with different issues—World Bank and OECD with economic growth; UNESCO and ILO with quality of employment, social issues, quality of life, and social cohesion—these issues seem now to converge toward more integrative concepts. In our view, if there is some commonsense about the importance of particular issues, in this case, social justice and the improvement of life, they become institutionalized expectations in the social environment of the organizations so that it becomes almost impossible to ignore these issues. Proponents of new institutionalism may not be surprised by this result.

Chapter 6 investigates the references in OECD publications. References to European countries seem to be more prevalent compared with references to non-European countries. Certainly, this has to do with the fact that most OECD countries are European. However, references occur quite often within regional clusters, even though European countries, mainly Finland and Germany, receive most references. So, the “Westernization of education is just as much a Finlandization, a Europeanization, or even a ‘localization’ of education policy” (Chap. 6 in this volume). It has been shown in previous research that Finland, as the best performer in the PISA 2000, did set the mark for many countries, whose representatives tried to learn about Finland’s education system (Martens et al. 2014). Once a particular Finnish strategy has been identified as effective for success and entered public discourse, for instance, the inclusion of anybody (Herrmann 2020), it is hardly possible not to consider its adoption.

Chapters 7 and 8 analyze the educational leitmotifs of the SEAMEO and the ICESCO as regional or cultural specific IOs. The Islamic IO on education, ICESCO, and its partner organizations ALESCO and ABEGS, are aware of the fact that high-quality education is of crucial importance in international relations and for global economic and political competition. Nevertheless, the ICESCO puts particular emphasis on the preservation of the cultural and spiritual roots of its member countries, almost all of which are dominantly Muslim, and these issues rank much higher than in other regional organizations. Islamic educational IOs perform the balancing act of acknowledging their embeddedness into global markets, on the one hand, and considering “Westernized” education an impediment to the spread and proliferation of Islamic cultural roots, on the other hand. The influence of the Islamic tradition and philosophy is obvious in many documents. These IOs have their own distinct ideas on education and follow their own educational leitmotifs, without neglecting, however, the requirements of inclusion into global markets and the world system. Accordingly, these results are in line with the idea of ‘cultural spheres’ that moderate the diffusion of Western institutions, rationality and culture, even though there is no serious alternative to the adoption of institutionalized education in general and in the long run. In a similar way, the SEAMEO emphasizes traditional norms and cultural values of the particular regional context of their member states and is at the same time interested in economic and developmental benefits of global markets and the integration into the world system. The two studies in Chaps. 7 and 8 reveal the tension between the worldwide institutionalization of Western culture and rationality, which dominates global markets and many important and powerful IOs, on the one hand, and the countries’ own particular local or religious cultural orientations, on the other hand. To some degree, our results even indicate resistance of some cultures in the world against the expansion of the WEIRD cultural model.

Our mixed-methods approach to the analysis of ideational framings and leitmotifs of education enabled us to enhance the political view and to contrast the results of the qualitative studies with results from standardized text analysis. Analyzing more than 1,600 documents published by six IOs, namely, the ILO, World Bank, OECD, UNESCO, ICESCO, and SEAMEO, Chap. 9 shows that these various IOs focus on different topics in the field of education. IOs restricted in membership tend to focus on their particular topics, which is Islam for the ICESCO and issues concerning the Southeast Asian world region for SEAMEO. In contrast, the OECD focuses more on universal issues related to education, higher education, and the economy. Moreover, also the heterogeneity of topics differs enormously between these IOs. Why the most prevalent three topics of the respective organization cover between a fourth and a half of their overall topics, the SEAMEO devotes a major part of its communication (78.6%) just to one topic, namely, “development, SE Asia”. The same holds for the ICESCO, but to a lesser degree: the ICESCO concentrates 54.7% of its overall communication covered by our 70 topics on “Islam, culture and the state”. These results fit quite well to the qualitative document analyses in Chaps. 7 and 8. We also found that the UNESCO communicates around 7% of its overall communication (covered by our 70 topics) to the issue “Islam, culture and the state”. At first sight, this seems to be surprising since the UNESCO is a universalistic IO, dedicated to education and has no restrictions in membership. Maybe, this result reflects an attempt of the UNESCO to cover particular interests of its member states and in so doing, the UNESCO has to acknowledge global cultural diversity. Indeed, culture is one of the major fields in which this IO is highly active.

Overall, our results are in line with our theoretical assumptions. First, institutionalized education is considered highly important by IOs who focus their activities on cultural or regional contexts. Education is essential for countries to compete on global markets and to participate in global trade. The more the postindustrial modes of production and consumption are globalized, the more important education becomes on a global scale. Surely, globalization entails competition between states in the field of education (Wallerstein 2004). If competition was the major stimulus of why countries institutionalize education, however, it would be difficult to explain why countries sometimes adopt these institutions just at the ‘front stage’ (Goffman 1959), as a ‘myth and ceremony’ (Meyer and Rowan 1977), or a ‘performative act’ (Steiner-Khamsi 2012). Moreover, why would the regional or cultural specific IOs put such a strong emphasis on the preservation of their regional or religious traditions? Instead, neo-institutionalism argues that once a practice is institutionalized as taken-for-granted, it becomes increasingly difficult for countries to refuse to adopt it, even though they adopt it at the ‘front stage’ in order to gain legitimacy. In this regard, neo-institutionalism is certainly correct at the global scale, where there is a considerable pull toward adopting educational institutions.

But, if the global spread of educational institutions is also fostered by universalistic and dedicated IOs, why then do regional and culturally specific IOs exist at all? IOs, such as the SEAMEO and the ICESCO, make particularistic claims and develop particularistic programs. In their view, their missions seem to not be covered by universalistic IOs, e.g. by the World Bank or the UNESCO. Their particularism indicates the existence of forces working against Western-dominated isomorphism in the field of education. Since education is strongly linked to the intergenerational reproduction of culture, it is not surprising that forces working against global diffusion and isomorphism reside in what we have called ‘cultural spheres’. This conclusion fits well to recent studies highlighting the increasing importance of identity—ethnic, religious, cultural or political—during the last decades (Tibi 2012; Fukuyama 2018). Empirical findings pointing to the ‘performative act’ of adoption (Steiner-Khamsi 2012) are of particular interest because they integrate both perspectives: on the one hand, countries must somehow become active and address institutional innovation propagated by influential IOs. Otherwise, they do not acquire legitimacy, neither at the level of the global state system organized within the IOs nor in their domestic population and electorate. On the other hand, they do not fully acquire these institutions and keep operating in their traditional way on the ‘back stage’. This argument highlights that the two theoretical perspectives outlined in the introduction of this volume—isomorphism and cultural spheres—do not contradict but rather complement each other.

We are well aware that a systematic inclusion of the concept of culture in international comparative education studies is far from being trivial (Anderson-Levitt 2012). It is indeed like “nailing the pudding to the wall”, as the German political scientist Max Kaase argued with reference to the concept of political culture (Kaase 1983). However, disciplines such as cultural psychology, anthropology (Henrich et al. 2010), and cultural comparative sociology (Inglehart 2018; Norris and Inglehart 2011) have recently improved their theoretical concepts and empirical measurements of culture (Mohr et al. 2019). Analytically, therefore, we should not hesitate to systematically include culture in our studies—we should simply accept the challenge and account more systematically for one of the core concepts in sociology and anthropology.

In addition, we are also aware that integrating culture in the social science studies can cause quite a stir from a normative and political-activist perspective. It is important to note that analyzing cultural differences does not necessarily imply a negative valuation of other cultures. This is an important lesson to be learned from cultural psychology (Shweder 2003). Certainly, the editors of this volume are born and socialized in post-1968 Germany, so part of our identity is shaped by the WEIRD culture. Both of us do highly appreciate civil liberties, personal freedom, equality, egalitarian gender roles, democracy, social security, and a well-functioning legal state—we are even in the comfortable situation of being able to take all this for granted. If persons adhere to liberal-democratic values, they would like to see these values realized for other people in the world, if they desire these values as well. So do we.

But being shaped by a particular culture and adhering to particular values should not come along with imposing this culture and values to others. As Haidt argues, if actors observe moral transgressions they respond with emotions generated by neuronal processes. Which kind of transgression triggers such a response is to a considerable part ‘hard-wired’. All mentally healthy people are empathic and generate negative emotions when they see people suffering (Haidt 2012, 72–75). Another part of our responses toward moral transgression, however, results from cultural factors and cognitive judgments (Shweder 2003, 36–37). Maybe, it will never be possible “to draw a sharp line between, on the one hand, emotions whose manifestation is organized by the generative templates in the human brain and, on the other, emotions whose arousal by neurological and body systems is regulated by socially constructed rules” (Turner 2000, 131), but this distinction is nevertheless important. For many people from secular WEIRD cultures, the moral dimension ‘sanctity-degradation’ is not important, whereas not respecting sanctity is regarded as a serious transgression by many people, for example, in India (Haidt 2012, 121–122, 173). Who are we to say that the emotional response of moral disgust is inappropriate when observing such a transgression? Not respecting or even denying Indians’ moral emotions would be Eurocentric. Cultural moral psychology has shown that to some degree the moral matrix in other cultures differs from ours. It is definitely a virtue, not only for scientists, to recognize this difference and to put our own moral matrix into perspective. In a global view, we are rather WEIRD than the norm. This does not mean relativism if we adhere to liberal-democratic and egalitarian values and argue in favor of them. Yet, given the historical evolution of the WEIRD culture, it is not surprising that other cultures differ from us. And these differences are an interesting subject to scientific inquiry.

Outlook for Future Research

In this volume, we analyzed different “Global Pathways to Education” and focused on cultural spheres, networks, and international organizations as explanatory variables. We started from a combination of theoretical insights that originate from sociological neo-institutionalism and the comparative analysis of global cultures. We assume that our approach is innovative and could contribute to the further development of neo-institutionalism.

Moreover, the research methods applied in this volume are innovative. Our methods range from qualitative document analysis, macro-quantitative longitudinal analysis, social network analysis, sequence analysis, and standardized text analysis. Future research could further elaborate on our theory-driven multi-methods perspective. Thereby, the set of IOs analyzed in this volume could be enhanced as well. Currently, we have identified 30 active IOs in the field of education. Of course, such a considerable extension of the number of IOs is a challenge, particularly for qualitative analyses. But also, the standardized methods of text analysis require an accountable selection and collection of documents and their preparation (adaption of formats, deletion of stop words, word stemming) comes with a considerable workload. Nevertheless, the classification of IOs in the dimensions of dedicated versus derivative and open versus restricted seems to be fruitful. Thus, it is worth to put even more effort into the analysis of regional or culturally specific IOs compared with universalistic IOs, such as the World Bank, ILO, or the UNESCO.

Obviously, the cultural clusters resulting from our two-mode social network analysis depend on the respective cultural characteristics we included in the analysis. Also, extending or changing this set of characteristics would change the results. Therefore, we should think about collecting more data on global cultures. Moreover, future research should systematically account for the consequences of education policy in different cultural spheres and their populations. Individual level data, as provided, for example, by the World Value Survey, can be used to refine our typology of cultures (Basáñez 2016; Inglehart 2018). Moreover, this data also includes information on respondents’ education, labor market integration, but also on, for example, gender role orientations and the inclusion of females into education. This information should be used in future studies in order to assess the actual, micro-level consequences of education policies in different cultural spheres.

Finally, in addition to the collection of more data on the introduction of particular policies, the network diffusion analysis should be systematically extended to continuous variables. These variables could be, for example, the share of female students in primary, secondary, and higher education, or the share of migrants and minorities relative to the overall population, by defining thresholds and the discrete event of adoption by crossing the respective threshold (Meyer et al. 1992).