The Secular Sacred Canopy: A Paradigm Revision in the Study of Secularization

Abstract

This article revisits and revises the secularization paradigm in light of late-stage secularization. Whereas secularization theory, influenced by Peter Berger, holds that structural differentiation, rationalization, pluralism, and individualization erode religious authority, we argue that these same processes facilitated the emergence of a secular sacred canopy. Drawing from recent developments in cultural sociology, we demonstrate how scientific empiricism, naturalism, and a romantic liberal imaginary collectively function as a secular equivalent to a religious cosmology. By theorizing secularization in cultural sociological terms, we are able to explain the rise and resilience of cradle nonreligion, the shift from “religion” to “spirituality,” the ascendance of right-wing “religious” populisms, and the paradoxical nature of the heretical imperative in highly secular societies (and contexts). We conclude by outlining the methodological, theoretical, and explanatory implications of our paradigm revision.

Modernization causes problems for religion. If the secularization paradigm can be reduced to a thesis, this is it (e.g., Bruce 2011; Kasselstrand, Zuckerman, and Cragun 2023; Wilson [1966] 2016). Despite challenges from market theories (Stark and Bainbridge 1980; Warner 1993) and from approaches emphasizing the rise of “vicarious” or “lived” religion in modernity (Ammerman 2014; Davie 1994), scholarship shows that religion (and Christianity in particular) has dramatically declined across measures of belief, belonging, and behavior in Western liberal democracies, including in the United States. If we understand secularization to be the empirical process of religious decline, then the empirical evidence supporting it is robust (Kasselstrand et al. 2023; Stolz 2020; Voas and Chaves 2016).

However, the secularization paradigm is “both a theory and a process” (Kasselstrand et al. 2023:21). The theory has had many champions over the years, but few have contributed to its elaboration as much as the late Peter Berger. Indeed, despite the fact that Berger later recanted his earlier work on secularization (see Berger 1999), many of his ideas remain cornerstones of the secularization paradigm (Dobbelaere 2011). For instance, secularization theorists agree with Berger (1967) that structural differentiation tears apart and ultimately destroys the “sacred canopy” that provided support to the religious legitimations and plausibility structures that organized life in premodern societies. Likewise, they follow Berger in emphasizing the modern process of rationalization and its corrosive effects on supernatural beliefs. Furthermore, they accept Berger’s (1979) argument that the pluralism typical of modern society weakens religious conviction and certainty by forcing individuals to reflect on and choose their beliefs amid an array of alternative religious and nonreligious options. In other words, Berger’s (1979:30) claim that this “heretical imperative” is “a root phenomenon of modernity” that functions to destabilize religiosity continues to receive widespread approval. Finally, secularization theorists agree with Berger (1967:133) that owing to individualization, religion in modernity increasingly becomes a matter of subjective “choice” or “preference” and weakens considerably as a result. In fact, according to secularization theorists, these classical arguments of Berger’s—which foreground structural differentiation, rationalization, pluralism, and individualization (conceived as core features of modernization)—are among the best explanations we have for why religion has declined so consistently (and dramatically) in Europe and North America (see Kasselstrand et al. 2023; Voas and Chaves 2016; Zuckerman, Galen, and Pasquale 2016). ¹

Given Berger’s outsized influence, we think it worthwhile to revisit his ideas in light of contemporary social conditions—conditions quite unlike those in which he first articulated his theory. When Berger published The Sacred Canopy in 1967, the West was certainly in the midst of secularizing—albeit to varying degrees across regional and national contexts—but organized religion (Christianity, specifically) still remained hegemonic, especially in the United States. Berger’s arguments were elaborated against a public culture of normative religiosity, which although weakening, was nevertheless powerfully present (Putnam and Campbell 2010). By contrast, in many Western countries today, organized religion is a shell of its former self. For example, belief in God has lost its majority status in a number of Western countries (falling below a 50 percent population share), including in Great Britain, the Netherlands, Norway, and Sweden; and it is below the 60 percent threshold in many more nations, including Australia, Canada, Denmark, Finland, France, and New Zealand. ² Bruce and Voas (2023:1) observe that in most twenty-first-century liberal democracies, religion has “declined steeply in power, popularity, and plausibility.” Recent studies make clear that with each successive generation, religion’s relevance has declined so much that in some contexts, the cultural “default” among the young is no longer religion but “nonreligion” (Brown 2017; Wilkins-Laflamme 2023; Woodhead 2017). This unprecedented change in social conditions—which we characterize as a shift from “early to late-stage secularization”—raises critical questions about how we ought to understand the nature of secularization as a theoretical paradigm and not just as an empirical process.

One interpretation of late-stage secularization holds that the once-religious sacred canopy typical of premodernity has been supplanted by a highly differentiated, instrumentally rational (utilitarian), and pluralistic social order that lacks any overarching cultural legitimations, moral imperatives, or plausibility structures. We can call this the “Weberian interpretation” because it draws inspiration from a conventional reading of Weber (1946), in which modernity is considered a “disenchanted” “iron cage.” On this view, modern secular societies are ostensibly bereft of any integrative symbolic universes or shared normative meaning systems.

However, an alternative interpretation holds that in contexts of late-stage secularization, the once-religious sacred canopy has been supplanted by a now-secular “sacred canopy,” replete with its own cultural legitimations, moral imperatives, and plausibility structures. We can call this the “Durkheimian interpretation” because it was Durkheim ([1912] 1995) who famously argued that the “sacred,” although historically originating in religion, nevertheless lives on in secular modernity (Alexander 2003; Drori et al. 2003; Seidman 1985). Because humans are fundamentally cultural animals and because human societies cannot endure without some kind of moral-cum-symbolic integration, the notion that secular modernity is without a shared symbolic universe or overarching moral meaning system remains sociologically dubious (Jepperson and Meyer 2020).

In this article, we seek to elaborate and substantiate this Durkheimian interpretation.

We conceive of this project as a paradigm revision in the study of secularization and in our conception of modernity more broadly (Kuhn [1962] 2012). Thus, in line with earlier secularization theory, we espouse a substantive definition of religion—as comprising beliefs, actions, and traditions based on the existence of supernatural entities or powers (see Bruce 2011). Unlike the religious economies approach, which reflects a distinct paradigm (Warner 1993), we conceive of our proposal as an attempt to revise—albeit significantly—the secularization paradigm so as to integrate insights afforded by the “cultural turn” in the social sciences (Lizardo 2017; Mohr and Rawlings 2010; Simko and Olick 2021; Swidler 2001).

Postcolonial scholars and critical theorists have criticized the secularization paradigm for presenting secularization as a uniform and teleological process rather than a historically and culturally contingent project (Asad 2003; Josephson-Storm 2019). Others have argued that the paradigm ignores the particularities of national and regional contexts while also disregarding the role of symbolic entrepreneurs and social movements in catalyzing religious change (Smith 2003). And scholars such as Taylor (2007) and Joas (2021) have criticized the secularization paradigm on the grounds that it fails to sufficiently theorize secularity. We agree with and affirm each of these critiques, yet in our view, the issues they raise stem from a more foundational problem with the current secularization paradigm—its “thin” conception of culture. That is, despite its virtues, the secularization paradigm currently suffers owing to its failure to theorize modern society as “a cultural system, with its institutions and functions as cultural products and with its individual and associational elements as culturally constructed” (Meyer 2020b:29). Like other “realist” theories, ³ secularization theory tends to take for granted the cultural “myths” of modern Western societies rather than treating them as phenomena to be explained (Meyer 2020b; Thomas and Meyer 1984). So although we agree with secularization theorists that structural differentiation, rationalization, pluralism, and individualization cause problems for religion, we conceptualize each of these processes in much more cultural sociological terms.

For instance, in our view, structural differentiation has not led to the dissolution of a shared symbolic order but has in fact taken place alongside the gradual rise and consolidation of a secular sacred canopy that is global in reach (Drori et al. 2003; Meyer 2010). Furthermore, although rationalization remains a core feature of modernity, rationalization is itself a cultural script, which, far from always supporting instrumental rationality and technological efficiency, often undermines them (DiMaggio and Powell 1983; Drori, Meyer, and Hwang 2006; Meyer, Boli, and Thomas 2009). Finally, although pluralism and individualization have indeed contributed to the decline of religion, this is not because the mere experience of diversity deterministically leads to a more individualistic and thus secular consciousness but, rather, because of the significant cultural authority and ontological standing granted to “actorhood” in secular societies and the moral imperatives attached to the modern ideal of agency (Meyer 2010; Meyer and Jepperson 2000).

Lizardo (2017) argues that cultural sociologists must distinguish analytically between personal culture and public culture because they operate differently and are often not empirically correlated. Personal culture refers to culture that is internalized by individuals and comes in two forms—declarative (e.g., beliefs, values, worldviews) and nondeclarative (e.g., skills, dispositions, schemas)—each of which are acquired through distinct processes of enculturation. Public culture, by contrast, refers to externalized (extrapersonal) forms of culture (e.g., models, codes, symbols, scripts) that are institutionalized (Zucker 1977) and imbued with authority and legitimacy (Lamont and Lareau 1988). Lizardo’s insistence on separating these levels of culture stems from the empirical fact that structural mismatches—or “decoupling” (Bromley and Powell 2012)—between these levels are commonplace. For instance, personal declarative culture can and often does conflict with personal nondeclarative culture (Khan and Jerolmack 2013), and both can sit in tension with and even oppose public culture (Patterson 2014; Swidler 2001). Thus, Lizardo (2017:110) concludes, “the theoretical action lies precisely at the intersection of declarative and nondeclarative [personal] culture and the link of both of these with institutionalized public culture.” These cultural sociological insights matter for the study of secularization, we contend, because they compel sociologists of religion to expand our focus on individual-level indicators (e.g., the three Bs of belief, belonging, and behavior)—narrowly concerned with personal culture—to include systematic analysis of institutionalized public culture. Stated succinctly, we aim to shift the analytic focus from “what people believe” to what the moral-cum-symbolic order authorizes and legitimates.

In the first section of this article, we revisit Berger’s classic statements on secularization. We make the case that at the core of his theory lies a profound ambiguity. This ambiguity revolves around whether secularization entails the rise of instrumental rationality, epistemic and moral relativism, and the ever-present threat of meaninglessness or “disenchantment” (à la Weber) or spells the emergence and consolidation of a secular “sacred canopy” that is culturally, morally, and epistemically coherent and integrative (à la Durkheim). We show that Berger vacillates between these two incommensurable interpretations, leading to an unresolved tension in his sociological thought (Hjelm 2024). We then show that the very same tension haunts contemporary secularization theory.

In the second section, we suggest this theoretical tension can be resolved by viewing the Weberian and Durkheimian accounts of modernity as capturing, in ideal-typical terms, distinct stages of the process of secularization. The Weberian account of modernity is most germane in times and contexts of early to mid-stage secularization, when the religious sacred canopy of old has yet to be wholly supplanted. In such contexts, a growing “cultural mismatch” (Smith 2025:62) exists between the emergent secular public culture and the residual religious personal culture (both declarative and nondeclarative; Williams 1977). Owing to this lack of cultural and social structural cohesiveness, few taken-for-granted meanings exist, and thus, Weberian diagnoses of “disenchantment” will resonate widely. By contrast, in contexts of late-stage secularization, when nonreligion is prevalent at the levels of both public and personal culture (at least in specific regions or generational populations), we suggest the Durkheimian account becomes more apposite because such contexts institutionalize a “secular sacred canopy” replete with its own cultural models, legitimations, and plausibility structures.

We then elaborate the theoretical basis of a Durkheimian account of secular modernity by reviewing recent sociological developments, inaugurated by the cultural turn (Alexander, Jacobs, and Smith 2010; Meyer and Jepperson 2020). We contend that secularization theorists would benefit greatly from integrating the insights of cultural sociology. Following that, we synthesize an array of theoretical and empirical scholarship to present the contours of what we call the “secular sacred canopy.” This secular cosmology is composed of three broad cultural models: (1) an empiricist epistemology, which holds that science is the authoritative means of identifying true knowledge; (2) a naturalist ontology, which posits that the natural world exhausts reality; and (3) a romantic liberal imaginary, which conceives of the individual actor as both ontologically prior and “sacred” and prescribes the good life as one of authentic self-realization. We then show how this imaginary is institutionalized at the level of public culture across twenty-first-century liberal democracies. Thus, far from resembling instrumentally rational (utilitarian) and relativistic social orders, highly secular societies (and regional and generational contexts) more accurately resemble comprehensive symbolic universes that powerfully socialize individuals into secularity.

Finally, we demonstrate the theoretical and empirical payoffs of our account. We argue that this account not only improves the secularization paradigm’s ability to explain religious decline by bringing to light its cultural and institutional dimensions but also enables us to explain or shed light on a host of other significant religion-related developments, such as the rise and resilience of cradle nonreligion, the turn away from “religion” to “spirituality,” the ascendance of right-wing “religious” populisms, and the paradoxical nature of the heretical imperative in highly secular societies (and contexts).

Weberian And Durkheimian Strands In Peter Berger’s Account Of Secularization

At the core of Berger’s account of secularization lies a deep ambiguity. Berger contends that structural differentiation pluralizes institutions and legitimations, leading individuals to experience both “as relatively unstable and unreliable” (Berger, Berger, and Kellner 1973). Consequently,

The modern individual . . . lives in a world of choice, in sharp contrast with the world of fate inhabited by traditional man [or woman]. . . . The modern individual must stop and pause where premodern men [and women] could act in unreflective spontaneity. Quite simply, the modern individual must engage in more deliberate thinking—not because he [or she] is more intelligent, not because he [or she] is on some sort of higher level of consciousness, but because his [or her] social situation forces him [or her] to do this. (Berger 1979:19–20)

With the fall of the sacred canopy and the rise of modern pluralism, the religious and nonreligious alike are forced to become “heretics.” And this, Berger concludes, causes problems for religion.

This interpretation of modernity seems cogent except it raises a critical question about secularization: If modernity pluralizes, weakens, and destabilizes authorities, legitimations, and plausibility structures in general, forcing all modern individuals to reflect on and choose their own beliefs, then why would religious belief systems be the only ones threatened? Or put another way: If structural differentiation and pluralism lead to a social condition where all legitimations lose their all-encompassing and taken-for-granted character, then why would it not equally undermine nonreligious legitimations? Consider, for instance, that this account simply assumes that when confronted with a lack of plausible legitimations, moderns will opt for an instrumentally rational outlook—that is, “weigh[ing] between the alternatives and choos[ing] the option that is more efficient, more effective, and/or more reliable” (Kasselstrand et al. 2023:29)—rather than supposing that such an outlook is itself culturally scripted and thus relies on its own models, legitimations, and plausibility structures (Meyer 1983).

Interestingly, much of Berger’s work seems to indeed suggest the religious and the nonreligious are equally affected by the relativizing effects of modernity (Berger 1967:156; Berger et al. 1973:78; Berger and Luckmann 1995:38). In The Heretical Imperative, Berger (1979:30–31) observes that “heresy [the conscious choice to depart from orthodoxy], once the occupation of marginal and eccentric types, has become a much more general condition; indeed, heresy has become universalized.” Here, one finds a conventionally Weberian image of the modern individual: as inhabiting a differentiated and rationalized social order—conceived, in “realist” terms, as dominated by instrumental rationality and technological efficiency (Meyer et al. 2009)—that no longer provides moral direction, cultural integration, or protection from anomic terror. Indeed, this is the “iron cage” of Weberian nightmares. Something closely approximating this view can be found in contemporary secularization theory (and social theory more broadly). For instance, Kasselstrand et al. (2023:33) explicitly endorse Berger’s view that “religious pluralism” is a “fundamental mechanism of secularization” that undermines “plausibility structures, causing sacred canopies to collapse.” And Bruce (2011:2) defines rationalization as “the displacement, in matters of behavior, of religious rules and principles by demands that accord with strictly technical criteria.” In these accounts, modernity is framed in conventionally Weberian terms—as a differentiated and rationalized order where individuals no longer inhabit an overarching nomos.

To be sure, this Weberian account of modernity is not without merit. Yet the problem it poses for secularization theory is that it fails to explain why, if modernity imposes on the religious and nonreligious alike the same epistemic and moral challenges, the outcome should be wholesale secularization. Indeed, why should such an epistemically disorienting social condition favor secularity over religiosity? Thus, in contemporary secularization theory, there is a paradoxical insistence that modern pluralism is intensely corrosive of religious authority, legitimations, and traditions while somehow not similarly undermining secular authority, legitimations, and traditions—and it is not made clear why this should be.

In Durkheim, we can find an alternative explanation for the rise and resilience of secularity. Consider, for instance, how Berger’s (1967:107) The Sacred Canopy both describes modernity as bereft of strong plausibility structures and emphasizes “the rise of science as an autonomous, thoroughly secular perspective on the world.” Or consider Berger’s (1967:125) claim that “the peculiar Christian theodicy of suffering lost its plausibility and thereby the way was opened for a variety of secularized soteriologies.” Such statements suggest that when Berger (1967:127) writes “secularization has resulted in a widespread collapse of the plausibility of traditional religious definitions of reality,” the unspoken social implication is that such religious definitions of reality have been supplanted by strongly socially supported secular ones.

Furthermore, if we return to the building blocks of Berger’s sociology of knowledge, which begins from the assumption that “[m]an’s [sic] self-production is always, and of necessity, a social enterprise” (Berger and Luckmann 1966:51), then the claim that secular modernity is without a nomos seems wildly implausible. Indeed, such a claim seems to implicitly assume that whereas becoming religious requires socialization, becoming secular does not—a view that is antithetical to Berger and Luckman’s (1966) classic treatise on how humans collectively construct the social world.

In truth, secularization theorists have begun to come around to this view that modernity is by no means a “disenchanted” or anomic wasteland, where instrumentally rational (utilitarian) individuals live in constant existential dread. For instance, in a review of secularization theories, Stolz (2020) highlights the importance of secular competition in spurring religious decline (see also Stolz, Lipps, and Voas 2024). Similarly, in their work on religious nones, Zuckerman et al. (2016) underscore the importance of secular socialization (see also Blankholm 2022; Catto et al. 2023; Frost 2019). These studies make clear that although secularization undoubtedly entails the decline of religion, this is clearly not all it entails.

Both Berger’s classical work on secularization and contemporary secularization theory thus contain two conflicting strands. On one hand, they stress with Weber (albeit to different degrees) the existentially disorienting character of “disenchanted” modernity in the wake of secularization—which leaves the modern individual confronting a pluralistic and relativistic order. On the other, they follow Durkheim in depicting a world where the religious believer is forced to struggle against a dominant secular frame.

From Early To Late-Stage Secularization: Toward A Durkheimian Account Of Secularization

How might we reconcile these conflicting interpretations? We argue that we should read them as capturing distinct ideal-typical stages of secularization. In the early to mid-stages of secularization, something approximating the Weberian image of “disenchanted” modernity applies. At this earlier and more transitional stage, neither the residual premodern religious legitimations of old nor the new emergent secular legitimations have yet to achieve taken-for-granted status, owing to a divergence between the secular(izing) public culture and the residual religious personal culture of the majority population (Williams 1977).

Again, by public culture, we refer to extrapersonal dimensions of culture (models, codes, symbols, and scripts) that are enforced and reproduced through macro- and meso-level institutions and organizations (e.g., law, education, economy, healthcare, media), whereas personal culture refers to forms of culture that are embodied and individualized (e.g., beliefs, values, attitudes, and dispositions; see Lizardo 2017). We see the process of “secular transition” (Voas 2008) as initially taking place at the level of public culture—via select carrier groups and elites embedded in specific institutional spheres (Martin 1978; Smith 2003)—and only subsequently affecting personal culture across the majority population. ⁴ In early to mid-stage secularization, being personally nonreligious is a reality for some individuals, but it is not yet widespread. Under such conditions, feelings of moral relativism and epistemic uncertainty—and what Durkheim ([1897] 1951) referred to as “anomie”—are much more likely (Abrutyn and Mueller 2018). ⁵

With the shift from mid- to late-stage secularization, this context changes. At this advanced stage of the secularization process, the once prevailing religious default—defined by a coupling, albeit never total, of religious public culture and religious personal culture—has been replaced with a secular default—again, defined as a relative coupling of secular public culture and secular personal culture—that ensures religion will have only a marginal presence or influence in society (Kasselstrand et al. 2023:120).

The upshot of this interpretation is that it usefully accounts for the fact that scholarship on secularization has historically been developed within a context of early to mid-stage secularization. That is, Berger developed his account of secularization at a time when Christianity was much more publicly and personally influential than it is today in both North America and Europe, even if the process of secularization had been underway for some time. In a similar vein, the early scholarship on the nonreligious makes clear the extent to which nonreligious identities were mostly forged against a dominant religious consensus (LeDrew 2013; Smith 2011; Zuckerman et al. 2016). Given these background conditions, it would have been strange for secularization theorists to foreground the existence of a regnant secular nomos, given that it did not yet exist.

Following the seminal works of Martin (1978, 2017), who demonstrated how distinct socio-historical factors across different regional contexts can delay or accelerate the secularization process and imbue it with specific characteristics, we see different countries, regions, and generational cohorts currently finding themselves at different stages of the secularization process (Voas 2008). For instance, we would argue that the United States is currently at a mid-stage point in the secularization process: no longer in the early stages, with its now higher rates of nonreligion among Americans born in the 1990s and later, but not yet having reached late-stage secularization (see Stolz et al. 2025). Consequently, the Weberian image of “disenchantment” and epistemic uncertainty is relatively apposite for understanding the U.S. context and its heightened polarization as this country continues to emerge out of a Christian sacred canopy and religious and secular legitimations jostle for power and public presence (Campbell, Layman, and Green 2020). This said, some regions in the United States, including the Pacific Northwest and the Northeast, with their much higher rates of nonreligion (Block 2017; Wilkins-Laflamme 2014), could be seen as approaching later-stage secularization. The Scandinavian countries, Great Britain, France, and the Netherlands all exemplify late-stage secularization.

In summary, this framework both usefully explains the tensions inherent in Berger’s account of secularization and makes clear the appropriate direction for future scholarship on secularization. We do not wish to dismiss the insights contained in the Weberian account of “disenchanted” modernity, but we nevertheless contend that as regards contexts of late-stage secularization, we must turn to Durkheim.

Theorizing the Secular Sacred Canopy

What do we mean by a “secular sacred canopy”? Simply put, in highly secular societies (or highly secularized generational or regional contexts), a relatively cohesive and integrated symbolic order exists that collectively naturalizes a secular nomos. A secular sacred canopy may not reference a supernatural deity, but it is nevertheless “sacred” in the Durkheimian sense (Lynch 2012). Importantly, we conceive of the secular sacred canopy as operating at a highly abstract, even implicit, level (see Simko and Olick 2021), taking the form of institutionalized cultural models that regulate and structure public culture in secular society (Jepperson and Meyer 2020). Thus, in elaborating this secular sacred canopy, we do not deny the existence of diversity at the level of personal culture (i.e., beliefs, values, or dispositions). Rather, we see this canopy as institutionalized at the level of legitimate public culture such that it shapes what “goes without saying” in most modern institutional domains (Ridgeway and Correll 2006). In other words, rather than naturalizing a specific belief system or worldview (cf. Taves 2020), we conceive of the secular sacred canopy as functioning more generally as a secular cosmology that grants plausibility to certain types of legitimations over others (Drori et al. 2003).

We are not the first to propose that something like this exists. Perhaps the best-known version is Taylor’s (2007) account of the “immanent frame,” which emphasizes the distinctive “conditions of belief” characteristic of Western liberal democracies today. Thinking along similar lines, Casanova (2009) and Asad (2003) make the case that far from being a neutral playing field for diverse belief systems, modernity should be understood as a moral-cum-political project that is no less symbolically imposing than was premodernity. Still, in our view, it is Durkheim and those who have carried forward his theoretical tradition who provide us with the theoretical resources necessary to complete this task. For, as we have discussed, although convention has it that Weber pessimistically diagnosed the modern world as bereft of an overarching meaning system, Durkheim took a quite different view. This alternative account of modernity was premised on Durkheim’s ([1912] 1995) conviction that social life is fundamentally “religious” in character. For Durkheim, the sacred/profane binary, historically central to religion, is in fact constitutive of human thought. What the Durkheimian tradition allows us to see, which the conventional Weberian account obscures, is the extent to which differentiated and rationalized modernity rests on profoundly cultural foundations (Meyer 2009). Thus, sociologists working in this tradition have made the case that it is a mistake to assume that secular modernity is without myths and sacred forms of its own (Alexander 2003; Lynch 2012; Meyer and Rowan 1977).

For instance, John Meyer and his fellow neo-institutional theorists argue that far from being sites of narrow instrumental rationality and technical efficiency, modern organizations depend profoundly on shared symbols, models, and cosmologies (Frank, Meyer, and Miyahara 1995; Meyer and Rowan 1977). These scholars have documented the existence of highly institutionalized and standardized cultural scripts across modern scientific organizations (Drori et al. 2003), environmental organizations (Meyer et al. 1997), educational institutions (Boli, Ramirez, and Meyer 1985), state institutions (Thomas and Meyer 1984), and nonprofit organizations (Bromley and Meyer 2014). This work shows how such scripts chiefly serve the purpose of legitimation rather than increasing technological efficiency or instrumental rationality as such (Meyer and Rowan 1977). Thus, according to Jepperson (2020:76), realist accounts of rationalization (like those typical of secularization theory) fail to see that “principles of rational organizing are themselves socially constructed and reconstructed.” For this reason, such theories “often underplay the highly constructed, scripted, and legitimated character of modern actorhood” (Meyer and Jepperson 2000:101). Accordingly, whereas secularization theory holds that secularization (the process) is the result of parents granting their children autonomy and independence (Kasselstrand et al. 2023; Thiessen and Wilkins-Laflamme 2020), neo-institutional theory foregrounds the extent to which ideals of “autonomy” and “independence” are culturally scripted and institutionally mandated in modern life.

In a more hermeneutic vein, Alexander and his fellow cultural sociologists have demonstrated, across myriad empirical case studies, the profoundly symbolic character of modern life (Alexander et al. 2010). For instance, Alexander (1990) documents how the seemingly uber-rational discourse surrounding the computer is in fact profoundly moral and nonrational—presupposing a vision of moral progress and human emancipation that motivates and justifies ceaseless technological innovation. Moreover, he has demonstrated the centrality of moral binaries in structuring and animating even the most rational of discourses—be they concerned with democracy (Alexander 2006), historical injustice (Alexander 2012), or social science itself (Alexander 1995).

In summary, the Durkheimian tradition and cultural sociology more broadly furnish us with the theoretical resources necessary to theorize late-stage secularization and thus elaborate the idea of a secular sacred canopy. In the next section, we draw from an array of recent empirical studies to flesh out the nature of the secular sacred canopy that exists in highly secular societies (and contexts).

The Secular Sacred Canopy: A Substantive Elaboration

The secular “sacred canopy” should be understood as a broad cultural cosmology, comprising a configuration of institutionalized cultural models that regulate and structure public culture in secular society. We analytically distinguish three pillars: epistemological, ontological, and moral.

The epistemological model naturalizes science as the main mode of knowing and of determining what is true in Western societies (Drori et al. 2003; Toulmin 1990). Crucially, this does not mean that everyone inhabiting the secular sacred canopy follows the scientific method in their everyday lives (Weber 1946:139). Rather, it means that science holds a public legitimacy and cultural authority that alternative modes of knowing do not. Thus, under conditions of late-stage secularization, scientific authority is conferred a high epistemic status (Wuthnow 2012) such that it “goes without saying” (Ridgeway and Correll 2006). As a result, in highly secular contexts, claims to truth must be shown to be science-based or to directly challenge the limitations of current science (e.g., by saying science just has not caught up yet; Drori and Meyer 2006). Scientific empiricism is naturalized across a wide array of institutional domains. For example, in educational curriculums (Bromley 2010), economic development discourse (Schofer, Ramirez, and Meyer 2000), environmentalism (Bromley, Meyer, and Ramirez 2011), and in medicine and healthcare (Frank et al. 1995). Indeed, from policymaking to parenting, counseling to consumption, management to media, scientific “evidence-based” reasoning reigns supreme (Marques et al. 2025). Drori et al. (2003:25) observe, “All the sectors of rationalized society—the economy, political system, family and socialization, medicine, and others—rest on a cultural base that is substantially scientific in character. . . . The scientization of society, or the penetration of science into everyday life, is extreme.”

In conceiving of science as a core pillar of the secular sacred canopy, we are implicitly challenging the “realist” view that science has achieved its dominance simply by virtue of its instrumental efficacy or functionality. Not only is it the case that “much of what science is about falls well outside the scope of the utilitarian” (Drori et al. 2003:297), but we must not forget that instrumental rationality rests on distinct cultural foundations (Drori et al. 2006). Thus, we conceive of science less as a rational instrument than “a penetrating meaning system” (Drori et al. 2003:12) that generates “a modern, rational interpretation of world order and offering this logic as a secular interpretive grid for natural and social life” (Drori et al. 2003:23).

The second model of the secular sacred canopy is a naturalist ontology, which holds that reality is exhausted by nature (thereby denying the supernatural; Papineau 2023). In secular modernity, it is widely taken for granted in modern institutions that the natural world—including human cognitive, moral, and social phenomena—operates according to predictable and universalistic forces and laws (Weber 1946). Of the ontology naturalized in Western culture, Meyer and Jepperson (2000:103) report,

Nature is represented by elaborate lawfully defined entities set in imagined lawful relation to each other; the laws involved are held to be universally binding rather than culture-specific. More and more domains are incorporated over time in this cognitive system: psyches, elements of human society, the ecosystem, the forces of history, of the universe and its current physical operation, the evolution of life, species, and human language. Animist and spiritual forces are marginalized or purged, relative to other cultural ontologies.

The empiricist epistemology and the naturalist ontology of secular modernity logically imply one another; effects have natural causes that can be identified through the scientific method (Drori et al. 2003). Here, we wish to emphasize the cultural work these components of the secular sacred canopy accomplish: It is because of the institutionalized authority of science and ontological naturalism that means-end rationality, standardization, and technological efficiency (i.e., rationalization) are taken for granted by many in highly secular contexts (Drori and Meyer 2006).

Weber (1946) famously described this process as one of “intellectualization,” which he saw as pivotal to the process of “disenchantment.” Yet Weber failed to sufficiently appreciate the extent to which science and naturalism function as parts of an interlocking cosmology that provides cognitive frames or general models “within which human social actors exist and maintain their universal status, or legitimacy, as agentic entities” (Drori et al. 2003:36).

Before we move on, we wish to stress once more that the institutionalization of science and naturalism at the level of public culture does not mean all moderns refrain from holding nonscientific or spiritual beliefs in their private lives. As noted earlier, the “decoupling” of personal culture (both declarative and nondeclarative) from public models and scripts is a consistent feature of social life (Bromley and Powell 2012; Lizardo 2017). It may well be (indeed, we suggest it is the case) that many moderns privately resist or disregard the public authority conferred on science and naturalism. However, although decoupling of personal culture from public culture is common, this does not mean institutionalized models and scripts are impotent (Bromley and Powell 2012). As Meyer (1977:76) explains, “The effects of [modern] myths inhere, not in the fact that individuals believe them, but in the fact that they ‘know’ everyone else does, and thus that ‘for all practical purposes’ the myths are true.” Correll et al. (2017:301) refer to this as “third-order inference”; that is, institutionalized public culture exerts influence on individuals not because it “is widely endorsed as a personal conviction, but because it is what everyone knows that everyone knows.” In line with this analysis, we contend that the cultural models of scientific empiricism and naturalism function as legitimate public culture (Lamont and Lareau 1988) such that although many moderns may privately espouse religious or nonscientific beliefs and practices, the power and presence of these beliefs and practices is profoundly curtailed owing to their lack of authority and legitimacy in most modern institutional domains.

The third model of the secular sacred canopy takes the form of a romantic liberal imaginary. We return to Durkheim’s ([1912] 1995) insight that modernity gives rise to a new sacred form—the “cult of the individual.” This form of individualism, argued Durkheim, sacralizes the individual person (what neo-institutionalist theorists call the “actor”; see Meyer and Jepperson 2000), investing the abstract human person with ontological and moral supremacy. What Durkheim refers to as the cult of the individual is, in effect, the Enlightenment tradition of liberalism—which sacralizes not merely the individual person but also with it, the cultural ideal of autonomy (Gray 1995).

The institutionalization of autonomy as a cultural script remains a central feature of the secular sacred canopy (Alexander 2006; Meyer and Jepperson 2000), yet it is not alone. The 1960s counterculture had a profound impact on modern Western societies. What Parsons ([1974] 1984) referred to as the “expressive revolution” brought the expressive individualism that animated the counterculture from the cultural margins into the mainstream (see Bellah et al. 1985). Post-1960s liberal democracies presuppose a human actor with inner depths who harbors deep within them a true, unique self (Taylor 1992). As a result, a sacred ideal in modern societies is authenticity. Inspired by the dual traditions of liberalism (Gray 1995) and romanticism (Abrams 1973), these societies invest the individual person with great rights and powers while prescribing the imperative of self-realization—that is, the good life entails becoming the self one truly is (Rogers 1961). We thus refer to the late modern imaginary as “romantic liberal” because it naturalizes the cultural traditions of romanticism—with its sacralization of authenticity and liberalism—with its moral focus on individual personhood and sacralization of autonomy (Rosenblum 1987; Watts 2022).

Of course, just as legitimations require plausibility structures to survive, so too does a cultural imaginary require social support. We thus now adumbrate the remarkable extent to which a romantic liberal imaginary is naturalized across post-1960s Western societies. Consider, first, the fact that an almost unanimous consensus exists across social theoretical traditions that modernity raises up the abstract individual or actor to a god-like status (Alexander 2003; Beck and Beck-Gernsheim 2002; Joas 2013). According to Frank and Meyer (2002:93), “Individual personhood is the standardized master identity of the modern system, celebrated worldwide in documents such as the United Nations Declaration of Human Rights and in global institutions such as democracy, the free market, and the psychological science.” Moreover, the form of individualism that reigns supreme in the post-1960s era is expressive in nature—that is, a form of individualism that sacralizes both (liberal) autonomy and (romantic) authenticity (Bellah et al. 1985; Parsons [1974] 1984).

Of the legal culture born of the 1960s, Friedman (1990:35) writes, “The new variety of individualism stresses self-expression, that is, cultivating the inner human being, expanding the self, developing the special qualities and uniqueness of each person.” One finds the same expressive concern for authenticity in modern political culture, with its emphasis on recognition (Lamont 2023). With the shift from industrial to postindustrial society (Bell 1973) and the rise of consumer culture, expressive individualism has become increasingly integral to the economic sphere as well (Florida 2002). Indeed, not only is the romantic ideal of authenticity central to consumer culture, but we have also begun to see a “corporate turn to personal authenticity” (Fleming 2009:ix) whereby employees are encouraged to treat work as a means to self-realization (Boltanski and Chiapello 2005).

Why does a romantic liberal imaginary favor a secular, as opposed to a religious, outlook? By stressing the moral supremacy of the abstract individual and sacralizing the ideals of autonomy and authenticity, this imaginary fundamentally weakens the standing (and thus plausibility) of traditional, external religious authority, legitimations, and plausibility structures (Smith 2025). Because individuals are considered supreme, moral authority must lie within individual selves (rather than, for instance, a sacred text or a religious leader). However, this effect is not chiefly the result of a lack of an overarching symbolic order; rather, it is a consequence of the nomos secular modernity enshrines.

In turn, although we agree with secularization theorists that structural differentiation, pluralism, and individualization have contributed to the decline of religion, this is not, in our view, because modern pluralism undermines “plausibility structures, causing sacred canopies to collapse” (Kasselstrand et al. 2023:33). On the contrary, in secular modernity, a romantic liberal imaginary remains a key source of moral-cum-symbolic cohesion, providing moderns in the twenty-first century with a shared model with which to interpret reality and normatively orient their lives (Watts 2022). So although racial, gender, and sexual pluralism may flourish in some contexts in the twenty-first century (cf. Willaime 2006), this follows from the institutionalization of expressive individualism. The romantic liberal imaginary both enables new kinds of pluralism and constrains others: It is morally mandated that individuals pursue self-realization and refrain from inhibiting the self-realization of others.

In closing, we wish to reiterate once more that the existence of a sacred canopy (religious or secular) does not entail a complete lack of dissent or diversity at the micro-level. Although this cultural configuration exerts pressure in contexts of late-stage secularization, decoupling of personal culture from public culture is routine. Nevertheless, it remains the case that all sacred canopies—be they religious or secular—institutionalize a symbolic-cum-moral order that individuals cannot avoid reckoning with.

Applying the Revised Secularization Paradigm

What are the theoretical and empirical payoffs of the paradigm revision we propose? We now elaborate how a more cultural sociological account of secularization both better explains the nature of late-stage secularization and illuminates a number of other significant religion-related empirical developments.

The rise and resilience of cradle nonreligion

The secularization paradigm does not struggle to explain the dramatic decline of organized religion in the modern world. However, it does have trouble accounting for the remarkable resilience of nonreligion. Studies show that most members of younger generations who were raised with no religion (cradle nonreligious) remain nonreligious as adults despite the fact that nonreligious parents often do not consciously strive to raise nonreligious children (Manning 2013; Thiessen and Wilkins-Laflamme 2020). It seems highly unlikely that were modernity equally destructive of religious and secular authorities, legitimations, and plausibility structures, the former would have fared so poorly relative to the latter. Indeed, the consolidation of a secular sacred canopy seems the most plausible explanation for why secularity has become normalized among younger generations and in many regions across the West.

Our macro-level account is supported by a host of meso-level studies. For instance, studies of modern educational curriculums identify the institutionalization of scientific empiricism, naturalism, and romantic liberal ideals of autonomy and authenticity (Manning 2013; Strhan, Lee, and Shillitoe 2024; Thiessen and Wilkins-Laflamme 2020; Wilkins-Laflamme 2023). Likewise, Strhan and Shillitoe (2019) draw attention to “banal” forms of secular socialization—that is, the many ways ordinary interactions and experiences in modern life subtly affirm secularity. These studies show that under conditions of late-stage secularization, institutional sites from the family to education to associational life more broadly increasingly socialize children to become expressive individualists who take for granted the authority of science and naturalism (see Catto et al. 2023).

Some secularization theorists acknowledge the existence of “secular socialization,” but like other realist accounts, they tend to think in terms of secular values (e.g., autonomy and independence) and secular belief systems (e.g., atheism or agnosticism) that naturally emerge as a result of a decline in religious socialization. What is more, secularization theorists conceive of secular socialization as in fact producing more autonomous and free-thinking individuals, who are disposed to reject religious orthodoxy, tradition, and conformism (Zuckerman et al. 2016). Our account, by contrast, reframes “secular socialization” as a process of rejecting one set of cultural models and scripts (religious ones) in favor of others (secular ones)—that is, a romantic liberal tradition that presupposes the moral primacy of individual actors and the cultural imperatives of autonomy and authenticity. Similarly, secularization theorists take for granted the authority of science and naturalism in the modern world. By contrast, our account treats these as phenomena to be explained, conceiving of them less as technical instruments in the service of instrumental rationality than as core pillars of a widely institutionalized cultural configuration.

The spiritual turn

Organized religion (especially traditional Christianity) has suffered as a result of secularization, yet more individualistic forms of “spirituality” have thrived (Houtman and Aupers 2007; Roof 1999; Schnabel et al. 2025; Wuthnow 1998). Although churchgoing, belief in orthodoxy, and religious affiliation have declined, practices such as meditation and prayer, “spiritual” groups and virtual networks, and belief in a higher power or a cosmic energy and life force have grown among significant segments of Western populations (Kucinskas and Stewart 2022; Steensland, Kucinskas, and Sun 2022). This “spiritual turn” (Houtman and Aupers 2007; Watts 2022) engendered a debate from the 1990s onward that pits the secularization paradigm (foregrounding religious decline) against the “spiritualization” or “individualization” paradigm (tracking religious change, not decline; Davie 1994; Heelas and Woodhead 2005; Luckmann 1967). Another upshot of our paradigm revision is that it enables us to reconcile these paradigms in a way that does justice to both of their core insights.

Secularization theorists are correct that the turn away from “religion” to “spirituality” among some moderns has not reversed the overall decline of religion (Bruce 2017). This is most evident in contexts of late-stage secularization, such as in Scandinavia and much of Western Europe (Stolz et al. 2025). However, it is also true of mid-stage secularized America (Chaves 2017). What is taking place is religious change (spiritualization) within a larger trend of religious decline (secularization; Tromp, Pless, and Houtman 2022; Wilkins-Laflamme 2021).

However, proponents of the spiritualization paradigm are also correct that the secularization paradigm, as it stands, fails to appreciate the extent to which contemporary spirituality, rather than reflecting a privatized religiosity that lacks institutionalization and plausibility structures, in fact reflects a long-standing religious tradition that shares elective affinities with the secular sacred canopy. Watts and Houtman (2024) argue that “spirituality” signals a mystical religious formation that presupposes an experiential epistemology (truth is ascertained through spiritual experiences), an immanent ontology (a notion of the divine as an omnipresent and immanent life spirit, energy, or life force that permeates the natural world), and a soteriology of self-realization (the ultimate purpose of life is to realize one’s true self; see also Bender 2010; Fuller 2017; Hannegraaff 1996; Heelas 1996; Lynch 2007; Schmidt 2012). To be sure, contemporary or “holistic” spirituality sits in tension with the secular sacred canopy, yet it also remains the case, as Woodhead (1999:120) remarks, that it is a religious form “superbly adapted to life in late modernity.”

How so? Consider, first, that the mystical tradition of spirituality emerges, in part, out of an attempt to “reconcile religion with contemporary scientific knowledge, and in particular in attempts to ground spirituality in a contemporary scientific cosmology” (Lynch 2007:29). Thus, historian of the New Age—an important carrier movement of contemporary spirituality—Hannegraaff (1996:62) remarks that the tradition of spirituality holds a “high regard for modern science” because science is interpreted “in such a way as to legitimate a spiritual worldview.” Indeed, this explains why many spiritual practitioners interpret their experiential epistemology as not only compatible with scientific empiricism but also expanding it (Hammer 2004).

Furthermore, spirituality’s immanent ontology—which conceives of “the divine as both the imagination behind the cosmos and a life-sustaining energy of the cosmos” (Lynch 2007:45)—is compatible with various forms of “religious” and “open-ended” naturalism. As Asprem (2014:78) explains, “These are positions that hold divine, spiritual, or magical agents to be at work within nature (immanence), and hold, furthermore, that these traces of spiritual activity in the world can be uncovered, understood, and even manipulated by human agents through rational and empirical means.” Asprem (2014:10) concludes that these forms of “natural supernaturalism” (Abrams 1973) have been the epistemological starting point for

a whole range of new natural theologies that have since become features of twentieth-century history of religion. These natural theologies undergird the post-war “New Age science” discourse . . . and they permeate theological conceptions of “occulture” and “alternative spirituality” that so frequently call for a synthesis of cutting-edge science with “religion,” “esotericism,” or “spirituality.”

Finally, it is not difficult to see how spirituality’s soteriology of self-realization is consonant with a romantic liberal imaginary given that both spirituality and romantic liberalism sacralize the ideals of autonomy and authenticity (Watts 2022). ⁶ In other words, spirituality embodies an “individualized” (Schnabel et al. 2025) religiosity that receives legitimation from the tradition of expressive individualism and modern scripts of actorhood.

In summary, the spiritual turn does not contradict the secularization paradigm (Bruce 2017) but is in fact the by-product of the mystical tradition of spirituality’s deep elective affinities with the secular sacred canopy. ⁷ In turn, we agree with Campbell (1978:153; see also Fuller 2017) that secularization and spiritualization have been coterminous processes ⁸ :

This form of religion [contemporary spirituality] has thus prospered at the expense of the others because it is better adapted to survive in a scientific, secular culture. The characteristics which give mystical religion its adaptive advantage in this sense are its monism, relativism, tolerance, syncretism, and above all, its individualism.

The rise of right-wing “religious” populisms

One of the most noteworthy political developments of recent years has been the rise of right-wing “religious” populisms in a number of Western democracies. Studies of the United States, for instance, have documented the impact of “White Christian Nationalism” (Gorski and Perry 2022; Whitehead and Perry 2020), and studies of the European context report the growth of civilizational rhetoric tied to a conception of “Christian culture” (Brubaker 2017). A leading interpretation of right-populism sees it as extending the “Culture Wars” that pit religious traditionalists against secular modernists (Hunter 1991). On this account, right-wing populism is understood as a religiously motivated backlash against the modern secular order (i.e., the secular sacred canopy; Gorski and Perry 2022; Norris and Inglehart 2019). This may be the case in contexts of early to mid-stage secularization, as older religious legitimations vie with newer secular ones for public presence and power. However, this interpretation fails to appreciate the considerable extent to which the political right has secularized (and liberalized) over the past half-century (Baldassari and Park 2020; Hout 2021).

Furthermore, accounts that treat right-wing populism as a resurgence of traditional religion fail to distinguish between religious traditionalism and secular authoritarianism (Daenekindt, de Koster, and van der Waal 2017). It is telling that identification with “White Christian Nationalism” is not, in fact, correlated with Christian belief or practice in the United States (Whitehead and Perry 2020). According to Cremer (2023:7), right-wing populists should be interpreted as championing a “secularised ‘Christianism’, an ethno-cultural identifier of the nation and a symbol of whiteness.” In other words, right-wing populism is fueled by a “new secular identitarian right” and less by a traditionalist religious right (see also Smith 2025).

Of course, none of this contradicts the standard account of secularization, which would predict that with the advance of secularization, religious symbols may be invoked by secular elites for the purpose of “cultural defense” (Kasselstrand et al. 2023). But our account shows that right-wing populism emerges from within the secular sacred canopy even as it contests one of its foundational pillars.

The rise of these movements reflects an emerging cleavage in modern societies, less between the religious right and the secular left than between liberal cosmopolitans and postliberal (or anti-liberal) communitarians (see also Bonikowski 2017; Cremer 2023; De Wilde et al. 2019). Translating this into our own terms, we see this as a divide between those who embrace the expressive individualism naturalized in the secular sacred canopy and those who seek to revive more communal forms of identity and belonging—what we call “expressive collectivisms” (Broćić and Watts forthcoming). ⁹ On this account, right-wing populist movements embrace the romantic value of authenticity but like nationalist movements before them, collectivize it while simultaneously asserting the moral primacy of the nation (over the individual; Lindholm 2013). ¹⁰

In summary, the ascendance of right-wing “religious” populisms is, in our view, a consequence of secularization; the political invocation of Christian symbolism by right-wing elites is evidence of the waning authority of traditional religious authority (Cremer 2023; Smith 2025). The ontological and moral standing granted to the abstract individual under the regnant secular sacred canopy has fueled a postliberal (or anti-liberal) backlash among right-wing communitarians who seek less to contest the legitimacy of science or naturalism ¹¹ —the other pillars of the canopy—than to revive the standing of the (White, male) nation (Jepperson and Meyer 2020; Lerch, Bromley, and Meyer 2022).

The paradoxical nature of the heretical imperative under late-stage secularization

Recall that for Berger (1979), modernity forces all of us to become “heretics” because no authoritative tradition or orthodoxy exists. This, we argued, is a reasonable depiction of early to mid-stage secularization. However, if we are correct that a secular sacred canopy has established itself under late-stage secularization, then something like a secular orthodoxy exists in these contexts. And if this is the case, then it follows that it is only the religious who are forced to become “heretics,” not the nonreligious. This is paradoxical. Yet empirical research bears it out.

In interviews with younger religious and nonreligious North Americans, we found it is the religious who are the most reflexive as regards their religious identity—that is, those who feel the need to reflect on and justify their choices both to themselves and to others in a wider secular context. By contrast, nonreligious or secular individuals who live in highly secular contexts largely take for granted their secularity and do not feel the need to question or reflect on it much (Wilkins-Laflamme 2023). Additionally, it is young adults who were raised in religious households who are more likely to disaffiliate from their childhood religion in present-day contexts, to choose to leave their original parental religion for nonreligion (an estimated 24 percent in Canada and 23 percent in the United States). In contrast, among individuals raised by nonreligious parents, only 11 percent in Canada and 9 percent in the United States convert to a religion in adult life (see Wilkins-Laflamme 2023:135–45). Of course, this is not to suggest being secular or nonreligious exempts one from having to reflect at all. It simply means that with respect to (non)religious identity, it is the religious and not the secular who are forced to become “heretics” because it is the religious whose beliefs and identities depart from those legitimated and authorized by the secular sacred canopy. Hence, religious reflexivity has become a late modern imperative (Martí 2015).

In fact, religious reflexivity is not merely an imperative for the social structural reasons that Berger identified: It is equally a cultural imperative, tied to the moral primacy of the individual actor and the romantic liberal ideals of autonomy and authenticity (Meyer and Jepperson 2000). The paradoxes thus intensify: According to social theorists Giddens (1991) and Beck and Beck-Gernsheim (2002), to be modern is to be “reflexive.” Crucially, this is both a descriptive and a normative claim. Descriptively, it is incorrect because it is the religious who are forced to reflect in contexts of later-stage secularization in a way the secular are not. However, as a normative claim, it remains an important facet of the secular sacred canopy. The roots of this ideal lie in the Enlightenment. According to Enlightenment liberals such as Kant ([1784] 2006), to be an autonomous individual is to question taken-for-granted authority, to think for oneself, and thus to be reflexive. The Enlightenment project emerged at a time when the Christian sacred canopy of premodernity retained considerable power. Thus, the liberal conception of autonomy was formulated in opposition to an authoritative religious orthodoxy—hence, for Kant, religiosity remained a barrier to individual autonomy (Bristow 2023). But what becomes of this ideal when the normative background shifts from religious to secular? Does the Enlightenment ideal of the autonomous individual (not to mention the romantic ideal of authenticity) presuppose an orthodoxy against which to contend? And if so, what follows if modern orthodoxy shifts from being religious to being secular?

Discussion And Conclusions

The secularization paradigm’s core proposition—that modernization causes problems for religion—is arguably “beyond doubt” (Kasselstrand et al. 2023). And yet, dissatisfaction with the paradigm is both widespread and warranted (Gorski and Altinordu 2008; Joas 2021; Smith 2025; Taylor 2007). We seek to revise this paradigm to better account not only for the early stages of the process of secularization but also for its later stages. To do so, we integrated recent developments in cultural sociology that take seriously Berger and Luckmann’s (1966) contention that humans are cultural animals who inhabit symbolic orders by investigating “religious” and “secular” societies as, although distinct, equally cultural systems (Geertz 1973). The crux of our thesis is that in contexts of late-stage secularization, modernity should be understood as institutionalizing a secular sacred canopy where the broad models of scientific empiricism, naturalism, and a romantic liberal imaginary shape and structure public culture. In turn, secularization must be reframed as a process of cultural transformation, not merely religious decline.

What are the implications of this paradigm revision? Methodologically, our account compels secularization theorists to expand our analytic focus beyond micro-level measures of belief, belonging, and behavior to include systematic analyses of public culture. Specifically, we must take seriously the structuring and regulating role of cultural models, codes, symbols, and scripts institutionalized across domains such as education, medicine, professional science, media, law, nongovernmental organizations, and the state bureaucracy. Heeding Lizardo’s (2017) counsel, we must be attentive to the dynamic and variable relation between public and personal culture.

This means secularization theorists must become much more familiar with recent developments in cultural sociology and its conceptual toolkit (Edgell 2012). Thus, in addition to adopting the distinction between personal and public culture (Lizardo 2017), we also used the concept of the sacred (Durkheim [1912] 1995); institutional models and scripts (Meyer 2010); residual, emergent, and dominant cultural formations (Williams 1977); decoupling (Bromley and Powell 2012); cultural mismatch (Smith 2025); and third-order inference (Correll et al. 2017). No doubt, these concepts only scratch the surface of the field, and so we strongly encourage a closer rapprochement between the sociology of religion and cultural sociology.

One might object that our paradigm revision sheds no additional light on the causes of secularization, merely reframing existing processes. We disagree. Although a cultural sociological approach does indeed reframe the “standard” mechanisms of structural differentiation, rationalization, pluralism, and individualization, it also brings to light at least two cultural mechanisms that secularization theorists have overlooked. The first is institutional isomorphism (DiMaggio and Powell 1983). Secularization has advanced, in part, owing to the global diffusion and mimetic adoption of its cultural models, fueled less by demands for technical efficiency than by a desire for legitimacy. This raises the question of global variation (cf. Wohlrab-Sahr and Burchardt 2012). “World polity” scholarship suggests the West has exported its own secular sacred canopy to other parts of the world through processes of isomorphism, globalization, and ongoing colonial power dynamics (Meyer 2007). However, given that the secular sacred canopy we sketch here emerged in opposition to (while also inheriting much from) an earlier Christian sacred canopy, in line with the scholarship on “multiple modernities” (Eisenstadt 2000), we think the existence of multiple (secular and religious) sacred canopies across the globe remains likely (see Smolkin 2018; van Mulukom et al. 2023; Weir 2023). ¹²

The second mechanism is processes of sacralization and symbolic pollution (Watts and Reimer 2024). The mainstreaming of expressive individualism in the 1960s was critical to the consolidation of a romantic liberal imaginary—and with it, both secularization and spiritualization. Equally, the sacralization of the autonomous and authentic individual was instrumental to the symbolic pollution of organized religion (especially traditional Christianity). Indeed, it is because of the institutionalization of expressive individualism that organized religion is experienced by many as an “iron cage” (Schnabel et al. 2025).

Finally, a critical implication of our paradigm revision is that secularization is neither inevitable nor wholly stable because it relies on the active reproduction of a cultural configuration—a secular sacred canopy—whose symbolic equilibrium is dynamic and susceptible to challenge and mutation. Consider that each of the three pillars we identified, despite their institutionalization, are also subject to contestation. For instance, although contemporary spirituality holds elective affinities with the secular sacred canopy, it also critiques “conventional” science in favor of a “spiritualized science” (Hammer 2004:210) predicated on a “religious naturalism” (Asprem 2014). In turn, were this more “spiritualized” conception of science to be institutionalized in, say, education, professional science, or state policy, it could lay the groundwork for a mutation, if not a reversal, in the process of secularization. Likewise, were the dominant romantic liberal imaginary to be displaced by the emergent postliberal imaginary, this would entail a major transformation in the symbolic-cum-moral order. As noted, we do not believe this would entail a revival of a residual culture of traditional religion; it could simply entail a shift from a secular sacred canopy grounded in the sacrality of the individual to one anchored in the sacrality of the nation. Yet if the autonomous and authentic individual ceased to serve as the sacred center of moral life, it is difficult to predict what this would entail for secularity.

Secularization is indeed unfolding in many modern societies, but it does so as a dynamic, historically situated, and culturally mediated process rather than as a uniform or determined trajectory. Our paradigm revision thus helps clarify both the persistence and the fragility of secularization—and in so doing, opens new theoretical and empirical directions for the study of religion and modernity more broadly.