The rise of individual agency in conceptions of society: Textbooks worldwide,1950

Abstract

Spanish

French

A broadly recognized sociological insight is that rising levels of individualism increasingly characterize a growing number of countries. This article examines the extent to which schooling is altered by, and transmits, this core cultural shift. It analyzes 476 secondary school social science textbooks from 78 countries from 1950 to 2011 to see whether they increasingly portray society as made up of agentic individual actors of all sorts (e.g., children, women, minorities). It is found that emphases on older social institutions remain stable, but there are striking worldwide increases in emphases on people, especially ones empowered with rights. This global peopling of social science instruction, especially strong in the recent neoliberal decades, characterizes every type of country and textbook analyzed, and occurs over and above other features of the sample.

Keywords

Agency empowerment individualism rights textbooks

Historically, the social sciences tended to see societies in structural and cultural terms, with the participating people and groups embedded in fixed roles as a matter of tradition, custom, and culture. Especially with the Enlightenment, and what is called the ‘invention of society,’ social life became re-envisioned as the product of purposive (often rational) people, groups, and states (Toulmin, 1990). In the last half of the twentieth century, these liberal lines of thought became dominant cultural ideologies, increasingly locating purposive action in the individual, and human agency became central to understanding societal processes (Meyer and Jepperson, 2000). Agency was increasingly seen, especially with the rise of neoliberalism, as ultimately located in individuals. These changes were anticipated and partly captured by the idea that the honor of the family was superseded by the dignity of the individual as organizing cultural principle (Berger et al., 1973). They were also developed in writings on ‘the religion of the individual’ (Durkheim, 1969: 28) and the sacral character of the person (Goffman, 1959).

The rise of individual empowerment was not simply an intellectual evolution: it underlay the restructuring of contemporary cultures, especially after the Second World War. Individuals around the world, now massively schooled (Baker, 2014), are seen to choose and act. They often and increasingly take this perspective themselves: in the modern drama, they are the core actors. Social groups become purposive organizations, infused with actorhood ultimately rooted in individual rather than corporate authority (Bromley and Meyer, 2015). Individuals become endowed with a growing range of rights, stemming not just from their national states but also universally, from greatly expanded notions of their inherent entitlement to justice and equality (Skrentny, 2009; Stacy, 2009; Therborn, 2000). Individual rights and capacities come to be envisioned as the legitimate frame of reference for development (Sen, 2001) as well as security (UNDP, 1994). In short, individual agency is a core cultural ideological theme in today’s world. This human agency perspective is captured in the economist and Nobel prize-winner Amartya Sen’s view that governments should view people not as patients but as active agents of change.

The contrast, here, is with accounts of society emphasizing the dominance of institutionalized structures: collectives with histories, authority, and legitimacy of their own. Through much of the post-Enlightenment period, the dominant actors were national states and their political and military elites. Common persons were subsumed under these larger social structures rather than envisioned as playing active roles in shaping society. These structures tended to be reified – rooted in race, religion, ethnicity, land, and language – and institutionalized in polities, economies, and cultures. The central institution, which educational systems were in part constructed to support, was the national state (Ramirez and Boli, 1987), and the stories of human history tended to focus on it.

The rise of liberal and neoliberal models reshaped these older social structures. Discussions of neoliberalism tend to emphasize its economic dimensions: the retreat of the state, privatization, and the deregulation of economies (Fourcade-Gourinchas and Babb, 2002). But its distinctive cultural feature is an emphasis on the individual as ultimate locus of authority, leading to a great many transformations, which encompass, but are not limited to, the economy. For example, political orders are to reflect and serve the people: democracy came to be preferred, even in unlikely polities (Hafner-Burton and Tsutsui, 2005). Economic structures are seen as the product of the choices of consumers, producers, and innovators. Family arrangements are to reflect the choices and interests of their participants in matters such as individual control over marriage, divorce, and numbers of children (Thornton et al., 2015). Even religious and related cosmologies are legitimately matters of individual opinion and choice (e.g., Wuthnow, 1998). Related, science expanded to new domains, reflecting expanded perceptions of human capability for understanding the natural and social world (Drori et al., 2003).

We address this core point: Attributed agency – the rights-bearing entitlement, empowerment, rationality, and capacity of individual persons – becomes a central account of society and social change. We study secondary school social science textbooks across the world over the period since the Second World War, to see how much their depictions of society come to celebrate people seen as agentic individuals. Education is a central mechanism by which models of society are formed and transmitted, and textbooks are important lenses through which such models are formed and communicated (Schissler, 1989). Textbooks represent institutionalized understandings – often backed by the authority of national states and other elites – about the nature of society. The occasional ‘textbook wars’ that arise indicate the importance attached to them as authoritative statements. Our analyses draw on a dataset of 476 history, civics, and social studies textbooks from 78 countries for the time period since the Second World War. Using factor analysis, descriptive statistics, and multilevel modeling we find a dramatic increase in textbook emphases on human agency (across subjects and world regions), contrasting with stagnation in accounts of social structure. The rise in agentic depictions of society is particularly striking in the most recent neoliberal period (1995–2011).

We contribute to existing research in two main ways. First, the empirical findings present insight into changes in the intended socialization of young people around the world: our data cover the entire period since the Second World War. Globally, the intended curricula may increasingly send to students and others in society the message that all individuals ought to be empowered and agentic rather than passive and powerless to act. Second, our analysis speaks to the structure-versus-agency dispute in the social sciences (e.g., Giddens 1979; Sewell, 1992). Our findings imply that the debate itself can be thought of as a product of change in how society is culturally envisioned.

Background

It is widely understood that contemporary societies celebrate the agentic individual, and the term ‘actor’ came into common usage in the social sciences to describe their growing depiction as a fundamental building block of society (Hwang and Colyvas, 2011, 2013). Existing research highlights these worldwide changes in social organization and ideology. Previous studies find international treaties emphasizing human rights – and in particular empowerment and agency – in great and growing numbers (Elliott, 2007). The treaties that appear to be most consequential focus on violations of bodily integrity (Cole and Ramirez, 2013). Relatedly, educational systems focused on individual educational advancement explode around the world (Meyer et al., 1992b; Schofer and Meyer, 2005). Prior textbook studies illustrate a worldwide shift toward human rights and student-centrism (Bromley et al., 2011; Meyer et al., 2010). Even the family experiences such a shift as countries across the globe revise their criminal laws to protect persons rather than collective entities (Frank et al., 2010). In organizational research, the phenomenon of actorhood captures the reconstruction of entities as organizational actors bearing rights and responsibilities for purposive action (Brunsson and Sahlin-Andersson, 2000; Drori et al., 2009). In our view, organizational actorhood is undergirded by the rise of individualism: expanded and rationalized agency attributed to individuals allows for their assembly into purposive organizations (Bromley and Meyer, 2015).

A broad mix of forces produced these new models of national and international society. The first half of the twentieth century demonstrated the inadequacy of a world of conflicting nationalist states: It had generated two massive wars, a Great Depression, and horrific violations of the most basic human rights. International interdependence was inescapably increasing on military, political, economic, and social fronts, but could not be managed with anything like a global state-like political order. With the collapse of most communisms at the end of the 1980s, neoliberal triumphalism accelerated. Celebrations of the individual expanded, even over the formerly charismatic national states (cf. Ruggie, 1982, 1998). These lost much of their old legitimated primordiality rooted in dynasties, races, religions, and histories (Strange, 1996), and were partly reconstructed by various ‘new public management’ ideologies (Hood, 1991). Values that privilege individual agency and choice underpin these changes, with older structures reassembled as organizations. Theories of the assembly of individuals into organizations blossomed (e.g., Coleman, 1990). Overall, individuals went through a great social promotion with the expansion of rights and schooling.

Argument

We suppose that broad changes in ideology and social organization that attribute increasing levels of agency to individuals are depicted in general knowledge systems. We can find them in survey research, which now spans the globe (Kamens, 2012). And they appear in national and global policies – for instance in the numerous supra-national human rights instruments (e.g., Wotipka and Tsutsui, 2008). In this article, we look for them in secondary school social science textbooks. As secondary education has become widespread in practically all countries, this system is a central place to look for institutionalized depictions. Educational curricula follow global patterns, and textbooks are central carriers of the knowledge patterns involved (e.g., Meyer et al., 1992a). To our knowledge, no cross-national, longitudinal textbook studies have examined explicitly the rise of individual agency. Existing studies focus more narrowly on related issues like human rights, student empowerment, minority rights, or global citizenship (Bromley et al., 2011; Buckner and Russell, 2013; Meyer et al., 2010; Terra and Bromley, 2012).

We examine history, civics, and social studies books from around the world for the time period since the Second World War. History was heavily dominant in the early decades, and remains central, though it is partially replaced in the curriculum by the more social-scientized topics of civics and social studies in a process itself relevant to our arguments (Benavot and Amadio, 2005; Wong, 1991; for universities see Frank and Gabler, 2006). Unlike the traditional topics of civics and history, respectively focused on citizenship and national/global histories, social studies is a newer subject that integrates material from these and related subjects in a more socially oriented approach (Wong, 1991).

We examine the textbooks around a central hypothesis. We expect to find increases in textbook foci on the agency of individuals – and groups seen as made up of individuals, rather than having a corporate character. Specific measures are discussed below, but obvious examples include topics of rights and active participation. For example, in a section entitled ‘YOU – The Most Important Human Rights Defender,’ a Canadian civics textbook asserts, ‘The majority of defenders of human rights are ordinary people. They see that something is unjust and try to right it … the dedication of people to causes that they believe in is the most important catalyst for change in our school, neighborhood, country, and world’ (Watt et al., 2000: 138). Along these lines, a South African textbook (Dilley et al., 2008) educates students on the rights of children (see Figure 1).

Figure 1.

Emphases on agency and empowerment in South African textbook (Dilley et al., 2008).

We contrast our focus on agency with foci on collective social structure and culture such as emphases on religion, language, the state, or the military. For example, a social studies textbook for Bahamian secondary schools includes a chapter on important religions in the Bahamas (see Figure 2) plus chapters on Bahamian culture and government (Sealey and Curry, 2006). Similarly, a book on Pakistan Studies for Classes 9–10 includes chapters entitled ‘Establishment of Pakistan,’ ‘Natural Resources,’ and ‘Pakistani Culture’ (Punjab Textbook Board, 2002). These chapters teach students about the national social structures in which their lives are seen as embedded.

Figure 2.

Emphasis on social structure in Bahamian textbook (Sealey and Curry, 2006).

In analyzing the data, we test a number of subsidiary hypotheses. (1) Individual agency might be especially emphasized in the ‘neoliberal’ period since the 1990s. In this post-communist era, the rights and powers of individuals, and of organizations seen as agentic extensions of individuals, have been emphasized worldwide (Elliott, 2007, 2014; Ruggie, 1998). (2) Agency might especially be emphasized in developed Western country textbooks. In some views, the whole recent period has been depicted as reflecting Western (or American) triumphalism (Thornton et al., 2015). (3) Agency might be given weakest emphasis in history books, and be dramatized more in the civics and social studies curricula. History is often depicted as focusing on collectives – especially the state or nation conceived as unitary – and only slowly coming to emphasize the lives of ordinary people in society (FitzGerald, 1979; Loewen, 1995).

Data

Historical data on the content of schooling on a widely cross-national basis are difficult to obtain. Enrollment data are more commonly available, but these tell us little about changes in the substantive foci of education. For more recent periods curricular standards are available, but these appear only in the last few decades for many countries. In contrast, textbooks offer a way to compare the content of education across countries quite far back in time. Although textbooks provide a way to develop standardized and reliable measures to examine educational content cross-nationally and longitudinally, a drawback is they cannot be sampled completely at random. For many countries it is implausible to find all the textbooks used in a given period; there are rarely even lists of what textbooks were used in the early decades of our study. Our approach has been to try to maximize our coverage of countries, periods, and books; and to look at trends among various sub-samples of countries and regions to consider how the dataset might shape our results. The consistent nature of our findings is reassuring.

We searched for textbooks in a variety of ways. By far the most important was the extensive collection (178,000 volumes from 173 countries) in the library of the Georg Eckert Institute in Braunschweig, Germany. This library collects social science books from everywhere in the world. The collection is extensive, but there are fewer books from developing countries and in earlier periods. We selected 345 history, civics, and social studies textbooks for grades 6–12 from the Georg Eckert Institute. We then supplemented this collection with books from other libraries (including at universities) and by calling on colleagues around the world to help us find books. We ended up with 476 textbooks from 78 countries. Of these, 287 were history books, while 128 focused on civics, and 104 on social studies. We classified the books by their predominant character.¹ In selecting our books, the rationale was to have as balanced a distribution as possible across subjects, decades, and regions, though we were often constrained by availability. History books were the most common (especially in the earlier decades) and more books were available in the most recent decades (the 1990s and beyond) and from Western nations. For some countries, not many books were available in library collections and so we coded all the books we could obtain within our grade and subject specifications. For countries well-represented in existing collections, we carefully chose textbooks to maximize variation in terms of content and publication date and to avoid over-representation of a single country in our dataset. We excluded curricular materials such as teachers’ guidebooks and student exercise books. Appendix A shows how the books are distributed across countries and time periods.

The books were coded with a standardized scheme, developed over several years and with much pre-testing (available from the authors: items relevant to the present analysis are shown in Appendix B). It took about an hour to code a typical book. Further time was taken to train coders who were fluent in languages we do not command. Reliability checks suggested that our efforts to get standardized coding were successful – inter-coder reliabilities on items typically fell in the range of 80–95%.

Measures

Structure

As a background contrast to our focus on individuals, we measure the presence of interpretations of society as social structure. We include 13 dichotomous items capturing the presence of discussions of the nation-state and other core social orders. Our measures include mentions of the military, independence, independence movements, national symbols, and territory; we also include depictions of democratization and colonization/decolonization. Beyond the nation-state, the books were coded for any discussions of the economy, the knowledge system, and culture. The central idea is that these items depict a national (or supra-national) society made up of social structures providing social order and change.

To examine the coherence of these items, we conducted a factor analysis on the tetrachoric correlation matrix among the 13 items. This revealed two factors with Eigenvalues above the rule-of-thumb cut-off of 1, with one notably higher at 3.5 (vs. 1.6). This factor accounted for 32% of the variance among the items, while the factor with the next highest Eigenvalue only explained 15%. Thus, we re-estimated the factor analysis limiting the number of factors to one. All factor loadings on the resultant factor were above 0.4, with the majority falling in the 0.5 to 0.8 range. A Kuder–Richardson coefficient of 0.79 indicated internal consistency among the items, clearly above a 0.7 standard. We build an overall measure simply adding up the number of indicators coded as present in a book, out of the total of 13.

Agency

To measure emphases on individual agency we coded the books for 15 dichotomous indicators capturing discussions of individual rights in various substantive areas, general individual rights, and student-centered pedagogical approaches. Our coding included 10 items that indicate various types of rights including human and civic rights, the rights of children, women, and ethnic, racial, or religious minorities, and substantive rights to health, education, language, religion, and culture. As notions of equality and empowerment arise, in part captured by ideas of rights, discussions of particular groups rise in tandem; we code for any discussions of women, children and ethnic, racial, or religious minorities. Rights tend to inhere in members of these groups as individuals, contrasting for example to rights associated with indigenous peoples or workers seen as acting collectively (Elliott, 2007, 2011).² Importantly, in the most recent time period, not only are women, children, and minorities more likely to be discussed, but when discussed they are also more likely to be discussed as having rights (Nakagawa and Wotipka, 2016). Lastly, we include pedagogical indicators that treat students as active participants in learning rather than passive recipients of pre-existing facts; we measure whether books include role-playing activities (requiring the student to view roles as flexible), and whether books suggest students should get involved in the world (see Bromley et al. [2011] for a more elaborate analysis of curricular ‘student-centrism’).

To support this assembly of items, we conducted a factor analysis on their tetrachoric correlation matrix. This pointed to two factors with Eigenvalues above 1, but with one much higher at 5.3 (vs. 1.6). This factor captured 48% of the variance among the items; we therefore re-ran the factor analysis restricting the number of factors to one. Most factor loadings fell between 0.55 and 0.95. Only two fell below 0.4: role-play (0.28) and minorities (0.34). A Kuder–Richardson coefficient of 0.82 indicated satisfactory internal consistency between the items. In our factor analysis, the two dichotomous indicators measuring discussion of women and of cultural rights drop out because of collinearity. In view of this, we opted for a composite measure of agency that simply added up the number of indicators coded as present in a book. This enabled us to include discussions of women and of cultural rights in our index.

Beyond the core indicators of agency described above, we also coded indicators we saw as closely related. We report results on these indicators below.

Results

Descriptive analyses

Our core hypothesis has a descriptive character – it is about changes in textbook foci on topics over time. Further below, we report multivariate analyses intended to further test this core hypothesis in addition to our subsidiary hypotheses.

Background: Foci on social structure

In Table 1 we report the proportions of textbooks that cover our 13 social structure topics in each of three time periods: 1950–1974, 1975–1994, and 1995–2011 (chosen to show any special effects of the recent neoliberal period). We also report in Table 1 the means of our simple summary measure (‘structure index’). We present the data as baseline information, with which our findings of change in emphases on agency can be compared.

Table 1.

Structure: Changes in indicators and index in textbooks over time.

	1950–1974^a	1975–1994^b	1995–2011
	(n = 123)	(n = 156)	(n = 197)
Proportion of textbooks with any discussion of:^c
democratization	0.69	0.71	0.70
colonization and decolonization	0.63	0.69	0.60
the economy	0.81	0.85+ †	0.91
natural resources and production (e.g., mining)	0.58	0.59	0.59
science (e.g., physics, chemistry, biology)	0.37	0.39– *	0.29
inventions (e.g., light bulb, printing press)	0.43	0.44– †	0.35
language	0.28+ *	0.37	0.39
religion	0.63	0.66	0.63
the national military	0.61	0.58	0.53
national independence	0.69	0.64	0.64
national symbols (e.g., flag, anthem)	0.46	0.43	0.49
national territory	0.63	0.62	0.62
the national independence movement	0.54	0.56	0.50
Mean of structure index	7.36	7.52	7.23

Significance indicates the results of a t-test comparing the difference between proportions in periods 1 and 2.

Significance indicates the results of a t-test comparing the difference between proportions in periods 2 and 3.

For all but one item, ‘any discussion’ was defined as the presence of at least a paragraph in the book on the topic. For ‘national independence movement,’ the question asked whether it was discussed as a social movement.

p < .05, †p < .1 (one-tailed tests).

The results show remarkable consistency over time. Nine of the 13 items show modest positive changes between the first and the second periods, while four show negative ones. Importantly, only one of the changes is statistically significant, and the overall index shows an insignificant change. Between the second and third periods, five of the items show positive differences, while eight show negative ones. Only three of the differences are even modestly significant – one positive and two negative. The overall index shows a negative, but insignificant, drift.

Changes are of limited interest. Attention to the national military goes down a little, reflecting a general change in textbook discussions of war (see Lachmann and Mitchell, 2014). Attention to the economy drifts upward. But the clear overall conclusion from the data is that textbook discussions of various elements of social structure do not much change in frequency over time.

Foci on agency

In Table 2a we report the proportions of textbooks that discuss our 15 topics on individual agency in each of the three time periods. We also report results for an overall measure aggregating the items into an index.

Table 2a.

Agency: Changes in indicators and index in textbooks over time.

	1950–1974^a	1975–1994^b	1995–2011
	(n = 123)	(n = 156)	(n = 197)
Proportion of textbooks with:
any discussion of:^c
human rights	0.30+ **	0.45+ †	0.52
children	0.31+ †	0.39+ †	0.48
women	0.41	0.38+ ***	0.58
racial, ethnic, or religious minorities	0.48	0.55	0.60
the rights of:
children	0.09	0.13+ †	0.21
women	0.17	0.17+ ***	0.39
racial, ethnic, or religious minorities	0.09+ *	0.17+ *	0.24
citizen rights	0.59	0.63	0.66
health rights	0.05	0.06+ *	0.14
educational rights	0.13	0.17+ †	0.23
language rights	0.05	0.05+ †	0.10
religious rights	0.15	0.12+ *	0.20
cultural rights	0.07	0.06+ †	0.10
student role-playing activities	0.13+ *	0.21+ ***	0.43
suggestions for student involvement (e.g., volunteer)	0.26	0.26+ **	0.40
Mean of agency index	3.26+ *	3.80+ ***	5.29

Significance indicates the results of a t-test comparing the difference between proportions in periods 1 and 2.

Significance indicates the results of a t-test comparing the difference between proportions in periods 2 and 3.

For the various group variables and the citizen rights item, ‘any discussion’ was defined as the presence of at least a paragraph in the book on the topic. For all other items, we simply measured whether the book mentions the item at all. For the group rights, issue rights, and group discrimination items, this was conditional on the book mentioning the group or issue in at least a paragraph.

***

p < .001, **p < .01, *p < .05, †p < .1 (one-tailed tests).

The results are striking. There is a general upward shift between the first two periods. Eleven items show increases, and only four decline. Only four of the changes – all upward ones – are statistically significant, and the overall index is only marginally so. Between the second and third periods, however, increases are dramatic. All 15 items move upward – 13 significantly so. And the overall index shows a very substantial, and significant, increase.

In Table 2b, we show some related results: changes on indicators we thought relevant to the agency issue, but which were not included in the overall index for a variety of reasons.³ In some cases – environmental rights and gay/lesbian discussions – changes were in the direction of increased agency, but mentions in the books were too infrequent. In other cases, factor loadings were low in initial factor analyses (this was an issue for discussions of immigrants/refugees, open-ended questions, and the assumption that students should develop their own opinions). In the case of the five items referring to the extent to which various kinds of people experience discrimination,⁴ we thought some interpretations might see these as not really attributing agency to the groups involved. In our view, certainly supported by some initial factor analyses, perceptions of a history of disadvantaged people as reflecting discrimination in fact attribute rights to the individuals involved, rather than simply seeing them as suffering from disadvantage in history (see Bonacker [2015] and Kastner [2015] on the expanding rights of the victimized).⁵ In any event, the comparisons in Table 2b add weight to the earlier ones. Almost all the coefficients are positive, and most are significant, especially in the comparisons of the second and third periods. These findings are striking, and certainly support our main argument.⁶ While pictures of society as social structure remain relatively constant through the period since the Second World War, pictures of society as defined or infused by human agency rise dramatically.⁷ The stability of structures is somewhat surprising as our informal expectation was that we would find an overall decline in emphasis on social structure over the postwar period. Its continued presence, combined with the significant association between book length and emphases on agency, suggests that the rise of agentic depictions of society does not simply replace the older structural vision. Instead, there is a process of sedimentation where newer images of people and society are layered on top of older ones.⁸

Table 2b.

Agency: Changes in related indicators in textbooks over time.

	1950–1974^a	1975–1994^b	1995–2011
	(n = 123)	(n = 156)	(n = 197)
Proportion of textbooks with:
open-ended questions	0.63+ *	0.74+ ***	0.88
an assumption that students should form own opinions	0.14+ ***	0.34+ **	0.50
any discussion of:^c
immigrants or refugees	0.33+ ***	0.54	0.52
gays or lesbians	0.00+ *	0.02+ **	0.08
environmental rights	0.00+ †	0.01+ ***	0.09
discrimination against:
children	0.09	0.08+ **	0.19
women	0.13	0.13+ ***	0.38
immigrants or refugees	0.08+ **	0.21+ *	0.31
racial, ethnic, or religious minorities	0.19	0.22+ **	0.36
gays or lesbians	0.01	0.02+ *	0.06

Significance indicates the results of a t-test comparing the difference between proportions in periods 1 and 2.

Significance indicates the results of a t-test comparing the difference between proportions in periods 2 and 3.

***

p < .001, **p < .01, *p < .05, †p < .1 (one-tailed tests).

Subject and regional comparisons

In Table 3, we carry our descriptive analyses further. First, we compare history, social studies, and civics books on our two indices over time, to see if all sorts of books independently show the same broad patterns of change. Some lines of thought suggest that changes might especially characterize the field of history, which has come under rather sustained attack from ‘progressive’ points of view (e.g., FitzGerald, 1979; Loewen, 1995; Ward, 2007).

Table 3.

Changes in structure and agency indices in textbooks over time by subject and by region.

	Structure index			Agency index
	1950–1974^a	1975–1994^b	1995–2011	1950–1974^a	1975–1994^b	1995–2011
By subject:^c
Social studies	7.67	6.30	7.21	3.80	4.15+ **	6.16
Civics	4.97+ †	6.18	5.28	3.88+ **	5.79	6.17
History	8.28	8.38	8.47	2.83	3.17+ **	4.15
By region:^d
Asia	8.05	7.13	7.43	3.10	4.07+ †	5.70
Eastern Europe	7.25	8.00	8.16	2.75	3.10	4.58
Latin America & Caribbean	7.29	7.84	8.05	2.88	3.31+ *	5.70
Middle East & North Africa	8.60	8.00	7.95	2.20	2.38+ †	3.71
Sub-Saharan Africa	7.59	7.46	7.20	2.65	3.37+ **	5.73
West^e	6.84	7.21– †	6.02	3.85+ *	4.98	5.62

Significance indicates the results of a t-test comparing the difference between proportions in periods 1 and 2.

Significance indicates the results of a t-test comparing the difference between proportions in periods 2 and 3.

The number of books for each subject by time period are: social studies (15, 27, 62), civics (34, 34, 60), history (78, 108, 101). Subject categories can overlap.

The number of books for each region by time period are: Asia (20, 15, 40), Eastern Europe (4, 10, 31), Latin America & Caribbean (17, 32, 20), Middle East & North Africa (10, 16, 21), Sub-Saharan Africa (17, 35, 30), West (55, 48, 55).

The ‘West’ category includes North America and Western & Central Europe.

***

p < .001, **p < .01, *p < .05, †p < .1 (one-tailed tests).

Second, we compare world regions in the same spirit, to see if the basic changes describe all the regions. A common argument has it that the dramatic contemporary emphasis on human agency is a phenomenon principally of the Western or developed world.

Thus in the top half of Table 3, we show scores on the structural and agency indices across time separately for history, civics, and social studies textbooks. In the bottom half of the table, we show the same data for textbooks classified by country regional location.

Despite the fact that many of the cells in the table are based on small numbers of textbooks, the data show very consistent effects across both region and textbook subject area. All three subject area categories show substantial (and often significant) increases on the agency index. All six regional categories show similarly substantial increases. And in the same way, none of the relevant comparisons on the structural measure show overall increases or decreases worthy of note. Regions and subject areas differ in their foci, as we discuss below: the point here is that all the regions and subject areas show the same general patterns of change over time.

We can conclude that the overall patterns we describe – little change on the interpretation of society as structural, and strong increases in depictions of society as involving individual agency – characterize all the world’s regions, and all of the subject areas.⁹

Multivariate analyses

In order to further examine the observation of a dramatic increase in textbook depictions of society as reflecting individual agency, we conduct additional analyses of a multivariate character, focusing on our agency index. Because we have data at two levels – textbook and country – we use multilevel linear modeling.¹⁰ Hausman tests failed to reject the null hypothesis that the coefficients estimated by random- vs. fixed-effects models were systematically different. We thus estimate two-level random-intercept models, though we also conducted fixed-effects models to check the robustness of our results – noting few meaningful differences, as reported below.¹¹

Our main goal is to see whether our ‘time period’ effects hold up with multivariate controls, so every analysis included dummy variables for our two later time periods (1975–1994 and 1995–2011) with the first period, 1950–1974, as the reference category. We include two controls suggested by previous work: a continuous variable for the number of pages in the textbook (longer books could obviously include more topics), and a dummy variable measuring whether the book is intended for senior, rather than junior, secondary school.

We also included dummy variables for the subject focus of the textbooks – noting whether the book had something of a civics or a social studies focus (with purely history books as the reference category). These controls are of substantive interest for our study, since work described in the previous section suggests that a shift from ‘history’ toward social science characterizes contemporary curricular change, and that this shift is related to our core hypotheses. Our study cannot usefully describe the shift, as we did not sample books with this issue in mind (though we noted increases over time in civics and social studies books). But we can see if history books give less attention to individual agency. And we can see if any such association moderates our core over-time comparisons.

At the country level, in the hierarchical models, we include several standard variables. Gross domestic product (GDP) per capita might show a positive ‘modernization’ association with depictions of agency, and democracy would be even more likely to show such an association. We also characterized countries on their memberships in international non-governmental organizations (INGOs) – and thus connections to a broadly liberal global educational culture (Boli and Thomas, 1999). More important, we included a dummy variable for Western countries (i.e., Europe and the Anglo-American cases). A substantial literature suggests that these countries provide continuing pressures for liberal cultural emphases on human agency (Thornton et al., 2015).¹²

Analyses are presented in Table 4. The main findings relevant to our argument are clear. Textbook emphases on individual human agency show increases between the first two periods, and strong increases between the second and third periods. These associations hold up across all the models.

Table 4.

Multilevel models predicting agency in textbooks: coefficients and (standard errors).

	Model 1	Model 2	Model 3
Textbook-level
Time periods (ref. 1950–1974)
1975–1994	0.893*	0.880*	0.821*
	(0.372)	(0.359)	(0.382)
1995–2011	2.424***	2.124***	1.753***
	(0.364)	(0.357)	(0.423)
Number of pages	0.003*	0.004**	0.003*
	(0.001)	(0.001)	(0.001)
High school textbook	–0.186	0.038	0.010
	(0.333)	(0.327)	(0.326)
Civics textbook		1.749***	1.680***
		(0.321)	(0.323)
Social studies textbook		1.217***	1.246***
		(0.349)	(0.348)
Country-level
West			0.109
			(0.551)
Level of democracy			0.086**
			(0.030)
GDP/ capita (log)			–0.124
			(0.229)
INGO memberships (residualized, × 0.1)			0.004
			(0.004)
Constant	2.415***	1.328**	0.717
	(0.469)	(0.483)	(1.531)
Variance of the constant (level 2 random effect)	1.542	1.258	0.958
	(0.494)	(0.426)	(0.363)

Note: n at the textbook-level is 476. n at the country-level is 78.

***

p < .001, **p < .01, *p < .05.

Beyond this, book length is positively associated with our dependent variable, plausibly indicating that new emphases on agency are layered on top of existing material; older emphases are not fully replaced. Whether the book is senior or junior secondary school in focus does not seem to matter. But notably, textbooks with civics or social studies foci score much higher on the agency measure than purely history textbooks (the reference category). The finding of the special character of history books is of substantive interest, given the relative decline in the importance of history (in contrast to social science) in curricula in recent decades. Presumably, modern students are to make history more than study it.

The regional comparisons are of modest interest. In supplementary analyses not reported here, we also included regional dummies for Asia, Eastern Europe, Latin America and the Caribbean, Middle East and North Africa, and Sub-Saharan Africa (with West as the reference category). All regions were lower than the West on agency, as we might expect. But these differences were largely statistically insignificant; the only significant difference from the West was for textbooks from the Middle East and North Africa. Democracy produces the expected increase in emphasis on agency, but not a strong one. National log GDP/capita and residualized INGO memberships show weak associations. In further supplementary analyses, we estimated models including measures for official development assistance and international trade, for secondary and tertiary enrollment, for countries’ age (years since independence), and countries’ colonial history. Apart from colonial history, none of these showed significant associations with our outcome. A variable measuring whether a country had been colonized by Britain showed a marginally significant positive association with our outcome, without substantively changing the results reported here. Most importantly, our main finding – the strong effect of the current neoliberal period on depictions of human agency – holds up through these analyses.

In summary, our findings show that while textbook emphases on social structure remain stable, foci on human agency have gone up sharply. The trend is especially striking in the most recent neoliberal era (1995–2011) and evident across all regions and subject areas. The strong neoliberal effect is validated in multivariate analyses, which also point to stronger emphases on agency in civics and social studies textbooks (compared to history) and, weakly so, in more democratic nations. A unit entitled ‘Democracy and Human Rights’ in a 2008 Kenyan social studies textbook illustrates this celebration of individual agency amid persisting social structures (Ondieki et al., 2008, 247–248). The text describes the ‘Responsibilities of an individual in a democracy’ and goes on to elaborate the agentic powers and rights of individuals within the social structure of democracy. Students are taught that individuals must not only participate in democratic processes, but also speak out against corruption, mismanagement of public funds, and destruction of the environment. An activity has the students discuss cases of corruption and how they would prevent them. This example dramatizes that while social structures – like democracy – persist in textbooks, they today are inhabited by active and empowered individuals as opposed to obedient citizens.

Discussion and conclusion

In the knowledge system of the medieval world, human agency was little celebrated – but human social structure was not either. Order arose from the supra-empirical worlds of theology and law (e.g., Clark, 2006; Frank and Meyer, 2007). Subsequent transformations, epitomized by the Enlightenment, brought social structure into focus, and created the social sciences. But these changes also brought into centrality the potentials for the management of society, and the purposive creation of the social world. Partly this involved the expanded conception of the human individual, dramatized in liberal economic, political, social, and religious ideologies. And partly it involved a conception of empowered social structure, carried by the ideologies celebrating the national state and its institutions. Both these themes had strong roots in the Western religious tradition, which sacralized both the collective (as in the Church) and the individual (as in the soul). Both elements gained greater and greater articulation over the whole post-Enlightenment period, with the rise of knowledge structures focused on the individual (in psychology, economics, and [especially American] sociology), and the corresponding rise in the articulation of the nation-state collective in law, politics, and history.

The Second World War, and the whole pattern of events in the first half of the twentieth century, dramatically shifted the balance. Interpretations of the ontology of a legitimate social order shifted toward notions of the rights and powers – the agency – of human individuals, and the organizational and political structures they (rather than the gods of history, tradition, religion, or race) choose to create (Thomas et al., 1987). In the most recent decades, with the rise of neoliberal constructions, the celebration of the agentic individual as the center of society further intensified.

Existing work has dramatized how this ‘cult of the individual’ reshapes contemporary cultures, reorganizing economies, polities, and organizations around empowered and purposive human actors. Our contribution here has been to demonstrate its greatly expanded centrality in textbook cultural accounts worldwide. Analyses of a global dataset of textbooks in the social sciences since 1950 confirmed striking growth in textbook foci on human agency worldwide. As hypothesized, this infusion of textbooks with agentic depictions of society is particularly strong in the neoliberal period since 1995. The findings also support our prediction that books in social studies and civics contain higher emphases on agency than those in history. National characteristics do not seem to matter much, though we find some support for our expectation that democratic political development leads to stronger foci on agency. Contrasting with the rise in narratives around individual actors, we had modestly speculated that textbook accounts of social structure may be on the decline. Our results, however, show stability in structural depictions of society.

As a whole, these findings have important implications for the knowledge systems of modern societies. Education not only reflects the cultural triumph of individualism, it also transmits it to future generations and societies at large. Increasingly, it appears that students are taught to take the fate of society into their own hands rather than passively accept determination of their lives by elites and collectivities. The special emphasis on this worldview in social studies and civics is particularly significant, given the increasing importance of these subjects in national curricula around the world. Future work could examine whether this growing focus on agency in the cultural artifacts of schooling extends to classroom practice or the perceptions of teachers and students more broadly.

While our emphasis has been on this dramatic shift toward agency, the unexpected finding of the persistence of social structure is interesting in itself. The Kenyan example discussed above suggests that structure and agency may be reconciled in textbooks by populating structures with agentic individuals. But we can envision alternative possibilities. For instance, agency and structure could be depicted as existing alongside each other without integration, or perhaps in contestation, with individualism undercutting notions of responsibility and citizen participation as a member of the collective (see e.g., Giroux [2005] for a critical view). More detailed analyses of the relationship between structural and agentic accounts of society – in schooling and elsewhere – could thus be a fruitful area for future studies.

Our findings also illustrate that the expanding attribution of agency to individuals is best envisioned as a historical process and continuum, rather than an inherent and fixed state. Within this broad trend of increasing attributed agency, it seems plausible that the phenomenon is multidimensional and will take on varied meanings over time and between countries. For example, envisioning one’s self or others as having the capability to make purposeful choices is one key dimension of agency, but another is translating choices into action (Bromley, 2016; Samman and Santos, 2009). Individuals can make choices, but outside a system of equality that allows choices to be transformed into desired actions and outcomes they are less agentic. Similarly, individuals can be given equality, but without the ability to make choices they are less agentic. A useful issue for future research is to examine systematic variations within the broad frame of increasing attributed agency documented here (e.g., perhaps emphases on equality come before emphases on choice in Scandinavian countries).

In the expanded liberal and neoliberal schemes, it has seemed important to school students with conceptions of social life as created by purposive and responsible individual ‘actors,’ and as modifiable by them. Students, and people in society generally, should learn that ‘you can make a difference.’ History is not driven only or mainly by elites, or by social forces outside clear control. ‘Man,’ as it were, ‘Makes Himself’ (Childe, 1936).

Footnotes

Acknowledgements

Participants in Stanford’s Comparative Workshop provided useful comments. Most of the textbooks analyzed come from the Georg Eckert Institute in Braunschweig, Germany, whose staff were extraordinarily helpful. We also thank the Institute for Research in the Social Sciences at Stanford University.

Funding

This work was supported by the Spencer Foundation (200600003) and the National Research Foundation (Korea).

Notes

Author biographies

Julia Lerch is a doctoral candidate in Education at Stanford University. Her research focuses on globalization and comparative education; current projects are on worldwide curricula shifts during the twentieth century as well as the expansion and professionalization of global educational governance.

Patricia Bromley is an Assistant Professor of Education and (by courtesy) Sociology at Stanford University. Her research focuses on comparative education and non-profit organizations; recent publications appear in Nonprofit and Voluntary Sector Quarterly, Administration and Society, and American Sociological Review.

Francisco O Ramirez is Professor of Education and (by courtesy) Sociology at Stanford University. His research interests focus on the globalization of (a) human rights, (b) rationalized university organization, and (c) the political incorporation of women. Recent publications may be found in American Sociological Review, Higher Education, and Comparative Education.

John W Meyer is Professor of Sociology, Emeritus, at Stanford University. He has contributed to organizational theory, comparative education, and the sociology of education, developing institutional theory. Recent projects are on the organizational impact of globalization (Bromley and Meyer, Hyper-Organization, Oxford, 2015), and world curricula in mass and higher education.

References

Baker

(2014) The Schooled Society: Education and the Transformation of Global Culture. Stanford, CA: Stanford University Press.

Benavot

with Amadio

(2005) A Global Study of Intended Instructional Time and Official School Curricula, 1980–2000. UNESCO background report. Geneva: International Bureau of Education.

Berger

Kellner

(1973) The Homeless Mind: Modernization and Consciousness. New York: Random House.

Boli

Thomas

(1999) Constructing World Culture: International Nongovernmental Organizations since 1875. Stanford, CA: Stanford University Press.

Bonacker

(2015) The inclusion of victims and the globalization of criminal law. In: Holzer

Kastner

Werron

(eds) From Globalization to World Society. New York: Routledge.

Bromley

(2016) Empowered individualism in world culture: Agency and equality in Canadian textbooks, 1871–2006. European Journal of Cultural and Political Sociology 3(2–3): 177–200.

Bromley

Meyer

(2015) Hyper-Organization: The Global Expansion of Organization. Oxford: Oxford University Press.

Bromley

Meyer

Ramirez

(2011) Student-centeredness in social science textbooks: Cross-national analyses, 1970–2005. Social Forces 90(2): 547–570.

Brunsson

Sahlin-Andersson

(2000) Constructing organizations: the example of public sector reform. Organization Studies 21(4): 721–746.

10.

Buckner

Russell

(2013) Portraying the global: Cross-national trends in textbooks’ portrayal of globalization and global citizenship. International Studies Quarterly 57(4): 738–750.

11.

Childe

(1936) Man Makes Himself. London: Watts and Company.

12.

Clark

(2006) Academic Charisma and the Origins of the Research University. Chicago: University of Chicago Press.

13.

Cole

Ramirez

(2013) Conditional decoupling: Assessing the impact of national human rights institutions. American Sociological Review 78(4): 702–725.

14.

Coleman

(1990) Foundations of Social Theory. Cambridge, MA: Harvard University Press.

15.

Dilley

Clitheroe

Bottaro

. B (2008) Life Orientation, Learner’s Book, Grade 12. Cape Town: Oxford University Press.

16.

Drori

Meyer

Hwang

(2009) Global organization: Rationalization and actorhood as dominant scripts. Research in the Sociology of Organizations 27: 17–43.

17.

Drori

Meyer

Ramirez

Schofer

(2003) Science in the Modern World Polity. Stanford, CA: Stanford University Press.

18.

Durkheim

(1969) Individualism and the intellectuals. Trans. S Lukes. Political Studies 17: 14–30.

19.

Elliott

(2007) Human rights and the triumph of the individual in world culture. Cultural Sociology 1(3): 353–363.

20.

Elliott

(2011) The institutional expansion of human rights, 1863–2003: A comprehensive dataset of international instruments. Journal of Peace Research 48(4): 537–546.

21.

Elliott

(2014) The institutionalization of human rights and its discontents: A world cultural perspective. Cultural Sociology 8(4): 407–425.

22.

Feenstra

Inklaar

Timmer

(2013) Penn World Tables. Groningen: University of Groningen.

23.

FitzGerald

(1979) America Revised: History Schoolbooks in the Twentieth Century. Boston: Little, Brown.

24.

Fourcade-Gourinchas

Babb

(2002) The rebirth of the liberal creed: Paths to neoliberalism in four countries. American Journal of Sociology 108(3): 533–579.

25.

Frank

Gabler

(2006) Reconstructing the University: Worldwide Shifts in Academia in the 20th Century. Stanford, CA: Stanford University Press.

26.

Frank

Meyer

(2007) University expansion and the knowledge society. Theory and Society 36(4): 287–311.

27.

Frank

Camp

Boutcher

(2010) Worldwide trends in the criminal regulation of sex, 1945 to 2005. American Sociological Review 75(6): 867–893.

28.

Giddens

(1979) Central Problems in Social Theory: Action, Structure, and Contradiction in Social Analysis. Berkeley: University of California Press.

29.

Giroux

(2005) The terror of neoliberalism: Rethinking the significance of cultural politics. College Literature 32(1): 1–19.

30.

Goffman

(1959) The Presentation of Self in Everyday Life. New York: Doubleday.

31.

Hafner-Burton

Tsutsui

(2005) Human rights in a globalizing world: The paradox of empty promises. American Journal of Sociology 110(5): 1373–1411.

32.

Hood

(1991) A public management for all seasons? Public Administration 69(1): 3–19.

33.

Hwang

Colyvas

(2011) Problematizing actors and institutions in institutional work. Journal of Management Inquiry 20(1): 62–66.

34.

Hwang

Colyvas

(2013) Actors, actors! actors? The proliferation of the actor and its consequences. In: European Group for Organization Studies Annual Meetings 4, Montreal, Canada, July.

35.

Jorgenson

(2006) Global warming and the neglected greenhouse gas: A cross-national study of the social causes of methane emissions intensity, 1995. Social Forces 84(3): 1779–1798.

36.

Jorgenson

Clark

(2009) The economy, military, and ecologically unequal exchange relationships in comparative perspective: A panel study of the ecological footprints of nations, 1975–2000. Social Problems 56(4): 621–646.

37.

Kamens

(2012) Beyond the Nation-State: The Reconstruction of Nationhood and Citizenship. Bingley: Emerald.

38.

Kastner

(2015) The globalization of transitional justice: On the diffusion of norms, standards, and institutions of post-conflict justice in world society. In: Holzer

Kastner

Werron

(eds) From Globalization to World Society. New York: Routledge.

39.

Lachmann

Mitchell

(2014) The changing face of war in textbooks: Depictions of World War II and Vietnam, 1970–2009. Sociology of Education 87(3): 188–203.

40.

Loewen

(1995) Lies My Teacher Told Me: Everything Your American History Textbook Got Wrong. New York: Free Press.

41.

Marshall

Gurr

Jaggers

(2013) Polity IV Project. Vienna, VA: Center for Systemic Peace.

42.

Meyer

Jepperson

(2000) The ‘actors’ of modern society: The cultural construction of social agency. Sociological Theory 18(1): 100–120.

43.

Meyer

Bromley

Ramirez

(2010) Human rights in social science textbooks: Cross-national analyses, 1975–2006. Sociology of Education 83(2): 3111–3134.

44.

Meyer

Kamens

Benavot

. (1992a) School Knowledge for the Masses: World Models and National Curricula in the Twentieth Century. London: Falmer.

45.

Meyer

Ramirez

Soysal

(1992b) World expansion of mass education, 1870–1970. Sociology of Education 65(2): 128–149.

46.

Nakagawa

Wotipka

(2016) The worldwide incorporation of women and women’s rights discourse in social science textbooks, 1970– 2008. Comparative Education Review 60(3): 501–529.

47.

Ondieki

Mbugua

Muraya

(2008) Comprehensive Social Studies. Nairobi: Longhorn.

48.

Punjab Textbook Board (2002) Pakistan Studies 9–10. Lahore: Sofi Sons.

49.

Ramirez

Boli

(1987) The political construction of mass schooling: European origins and worldwide institutionalization. Sociology of Education 60(1): 2–17.

50.

Raudenbush

Bryk

(2002) Hierarchical Linear Models: Applications and Data Analysis Methods, 2nd edn. Thousand Oaks, CA: SAGE.

51.

Ruggie

(1982) International regimes, transactions, and change: Embedded liberalism in the post-war economic order. International Organization 36(2): 379–415.

52.

Ruggie

(1998) Globalization and the embedded liberalism compromise: The end of an era? In: Streeck

(ed.) Internationale Wirtschaft, Nationale Demokratie. Frankfurt: Campus Verlag.

53.

Samman

Santos

(2009) Agency and empowerment: A review of concepts, indicators and empirical evidence. Prepared for the 2009 Human Development Report in Latin America and the Caribbean. Oxford: Oxford Poverty and Human Development Initiative.

54.

Schissler

(1989) Limitations and priorities for international social studies textbook research. The International Journal of Social Education 4(3): 81–89.

55.

Schofer

Meyer

(2005) The world-wide expansion of higher education in the twentieth century. American Sociological Review 70(6): 898–920.

56.

Sealey

Curry

(2006) Social Studies for Bahamian Secondary Schools. Nassau, Bahamas: Macmillan.

57.

Sen

(2001) Development as Freedom. Oxford: Oxford University Press.

58.

Sewell

Jr (1992) A theory of structure: Duality, agency, and transformation. American Journal of Sociology 98(1): 1–29.

59.

Skrentny

(2009) The Minority Rights Revolution. Cambridge, MA: Harvard University Press.

60.

Snijders

TAB

Bosker

(1999) Multilevel Analysis: An Introduction to Basic and Advanced Multilevel Modeling. Thousand Oaks, CA: SAGE.

61.

Stacy

(2009). Human Rights for the 21st Century: Sovereignty, Civil Society, Culture. Stanford, CA: Stanford University Press.

62.

Strange

(1996) The Retreat of the State. Cambridge: Cambridge University Press.

63.

Terra

Bromley

(2012) The globalization of multicultural education in social science textbooks: Cross-national analyses, 1950–2010. Multicultural Perspectives 14(3): 136–143.

64.

Therborn

(2000) Globalizations, dimensions, historical waves, regional effects, normative governance. International Sociology 15(2): 151–179.

65.

Thomas

Meyer

Ramirez

Boli

(1987) Institutional Structure: Constituting State, Society, and the Individual. Beverly Hills, CA: SAGE.

66.

Thornton

Dorius

Swindle

(2015) Developmental idealism: The cultural foundations of world development programs. Sociology of Development 1(2): 69–112.

67.

Toulmin

(1990) Cosmopolis: The Hidden Agenda of Modernity. New York: Free Press.

68.

UNDP (United Nations Development Program) (1994) Human Development Report 1994. New York: UNDP.

69.

Union of International Associations (various years) Yearbook of International Associations. Munich: Saur.

70.

Ward

(2007) History in the Making. New York: The New Press.

71.

Watt

Sinfield

Hawkes

(2000) Civics Today. Scarborough, ONT: Nelson Thomson Learning.

72.

Wong

(1991) The evolution of social science instruction, 1900–86. Sociology of Education 64(1): 33–47.

73.

World Bank (2013) World Bank Development Indicators. Washington, DC: World Bank.

74.

Wotipka

Tsutsui

(2008) Global human rights and state sovereignty: State ratification of international human rights treaties, 1965–20011. Sociological Forum 23(4): 724–754.

75.

Wuthnow

(1998) After Heaven: Spirituality in America Since the 1950s. Berkeley: University of California Press.

The rise of individual agency in conceptions of society: Textbooks worldwide,1950–2011

Abstract

Keywords

Background

Argument

Data

Measures

Structure

Agency

Results

Descriptive analyses

Background: Foci on social structure

Foci on agency

Subject and regional comparisons

Multivariate analyses

Discussion and conclusion

Footnotes

Acknowledgements

Funding

Notes

Author biographies

References