The Secular Transition: A bibliographic essay

David Voas

9 June 2026

pdf version

Introduction

  All scientific theories have a lineage, and it can be helpful to trace the development of ideas from their roots and across various branches. This essay is one attempt to show how the concept of secularisation emerged and evolved over many decades. It aims, in particular, to provide a history of the origins and elaboration of the secular transition model. Because science is often a story of competing explanations, there will be signposts to alternative theories of religious change and to different routes within the secularisation paradigm. The focus, though, is on the secular transition, with increasing detail as we reach the present day.

The term ‘secularisation’ is used to mean different things. Descriptively, it sometimes refers to the separation of various institutions and spheres of life (law, medicine, education, and so on) from religious control, a process also labelled ‘differentiation’. Occasionally it is taken to mean that religion becomes a private matter and is removed from the public sphere. (See Casanova 1994.) Increasingly, however, the term is applied to declining levels of religious identity, practice and belief among individuals.

To complicate matters further, though, some scholars maintain that secularisation (even used descriptively) implies that people are becoming secular rationalists rather than merely becoming less religious. “Secularization happens. But so do a lot of other processes, including ones that perpetuate and strengthen the sacred, spiritual, enchanted, magical, supernatural, occult, ecstatic, and divine in modern life—even if outside the structures of traditional religion.” (Smith 2025: 369) On this definition, secularisation means not simply that religious involvement is weakening, but that that people are adopting a naturalistic worldview.

Often, however, ‘secularisation’ is used not descriptively as a synonym for religious decline but rather as a label for the main family of theories about what has caused that decline. The idea uniting this theoretical approach is that “modernisation creates problems for religion,” in the words of Steve Bruce (2002). Some consequences of modernisation are uncontroversial; for example, virtually all social scientists agree that key sectors have largely separated from religion. Many, though not all, agree that religion has increasingly become a private matter in highly developed societies. What continues to be contested is the idea that individuals are losing interest in religion.

Early works

  In The Positive Philosophy (1830-1842), Auguste Comte suggests that knowledge and society develop through three stages, the first being dominated by religion and the last by science. That theory of social evolution is an unequivocal expression of the secularisation thesis. But Comte subsequently suggested that people need a religion without God, and variants of this idea occur in later work. Likewise, Émile Durkheim’s key contributions to secularisation theory are found in The Division of Labour in Society (1893), where he discusses the shrinking domain of religion and the rise of individualism. Although The Elementary Forms of Religious Life (1912) is a foundational text for the sociology of religion, its conclusion that religion is socially necessary could be seen as an early instance of the ‘change, not decline’ thesis.

  Max Weber is more clearly the forefather of modern theories of secularisation, even if he did not use the term except in relation to legal systems. In his 1917 lecture “Science as a Vocation”, he refers to disenchantment and the implications of rationalized perspectives for religious belief. These ideas are much expanded in his posthumously published work Economy and Society (1921-22). He writes about the replacement of magic by more rational and bureaucratic forms of action, with religious belief pushed out of the public sphere, the move away from revelation in favour of rational-legal authority, the fact that different spheres of life (including the economy, politics and science) become autonomous, and the way that religion itself becomes more orientated to this world rather than the next.

  Through his translations of Weber, Talcott Parsons played an important role in promoting secularisation theory in the English-speaking academy. At the same time, however, he shifted its focus from disenchantment to differentiation. And while Parsons accepted that modernisation causes religion to decline in social significance, he supposed that at an individual level we would see change rather than decline in identity, practice and belief. (See in particular the last book published in his lifetime, Action Theory and the Human Condition, 1978.)

The contemporary debate

  Two books published almost simultaneously mark the beginning of the contemporary debate over secularisation. In Religion in Secular Society (1966) [and see the ‘50th anniversary’ edition from 2016 with introductions and appendices by Steve Bruce], Bryan Wilson argued that religion has lost its social significance, whatever beliefs and practices individuals might maintain in private. Religious institutions no longer shape education, politics, law, or public morality in the way they once did. In The Sacred Canopy (1967), Peter Berger argued that modernisation led to a ‘pluralisation’ of worldviews, undermining the taken-for-granted status of any single religious tradition. When multiple competing truth claims exist side by side, the plausibility structures supporting religious belief weaken, leading to a “crisis of credibility” for religion.

  Although the secularisation thesis is often regarded as having been especially influential in this period, there was never a consensus. Indeed, prominent opposing views appeared at the same time, notably David Martin (1965) “Towards eliminating the concept of secularization”, in the Penguin Survey of the Social Sciences, edited by Julius Gould, and Thomas Luckmann (1967) The Invisible Religion: The Problem of Religion in Modern Society, in which he argued that religiosity would persist in idiosyncratic forms. Likewise, Robert Bellah (1964) “Religious evolution” in the American Sociological Review implies support for the change, not decline position, as does his famous 1967 essay “Civil religion in America” in Daedalus. A 1967 article in the Journal for the Scientific Study of Religion expresses exasperation at the ambiguity of the term ‘secularisation’, with abandonment offered as an option (Larry Shiner, “The concept of secularization in empirical research”).

  Other dissenting voices soon appeared, with both Mary Douglas (in Natural Symbols: Explorations in Cosmology, 1970) and Andrew Greeley (in Unsecular Man: The Persistence of Religion, 1972) arguing that religion is an inescapable feature of the human condition. In short, the secularisation thesis was always contentious.

  That said, it seems fair to say that the debate did not become intense until the last decade and a half of the 20th century (Stolz 2020). During the interim, David Martin decided that secularisation was worth analysing, first in his 1969 article “Notes for a general theory of secularisation” in the European Journal of Sociology, and subsequently in A General Theory of Secularization (1978). (Much later he developed the ideas further in On Secularization: Towards a Revised General Theory, 2005.) These works in fact advance not a general theory but the view that secularisation depends on the cultural and historical context.

  Other noteworthy contributions from this period include Bryan Wilson (1976) Contemporary Transformations of Religion and Karel Dobbelaere (1981) “Secularization: A multi-dimensional concept” (published by Current Sociology), later revised as Secularization: An Analysis at Three Levels (2002). The three levels are societal, organizational, and individual. Distinguished scholars continued to object, however, for example Daniel Bell (1977) “The return of the sacred? The argument on the future of religion” in the British Journal of Sociology.

Contention

  The controversy escalated in the mid-1980s. In his 1987 article “Toward desacralizing secularization theory” in Social Forces, Jeffrey Hadden argued that secularisation theory was more ideology than empirical social science. Two years earlier, in 1985, Rodney Stark and William Bainbridge set out a different theory of religious change. In The Future of Religion: Secularization, Revival and Cult Formation, they asserted that religion operates in a perpetual cycle of decline, revival, and innovation. Humans are naturally religious, and religious vitality is a function of market forces. Rational choice theory had arrived in the sociology of religion.

  A substantial body of work followed, including Rodney Stark and Laurence Iannaccone (1994) “A supply-side reinterpretation of the ‘secularization’ of Europe” in the Journal for the Scientific Study of Religion, which sought to explain religious decline as the result of supply-side factors. Notoriously, Rodney Stark (1999) declared secularisation theory to be dead and buried in the article “Secularization, R.I.P.” in the journal Sociology of Religion. The full application of the market model to religion is found in Rodney Stark and Roger Finke (2000) Acts of Faith: Explaining the Human Side of Religion.

  Criticisms of secularisation theory came from other quarters as well. A particularly well-known article from 1993 is “Work in progress toward a new paradigm for the sociological study of religion in the United States” by Stephen Warner in the American Journal of Sociology. Like Stark et al., he emphasized the role of religious choice in producing growth or (where it is absent) decline. He is more concerned with the social and historical particularities of the American context, though, than with devising a universal theory of religious behaviour.

  Grace Davie (1994) Religion in Britain Since 1945: Believing Without Belonging popularised the idea that religious decline might be limited to institutional participation. José Casanova (1994) Public Religions in the Modern World argued that religion has not disappeared from public life. He used case studies from Spain, Poland, Brazil, and the US to argue that religions, far from becoming confined to the private realm, are actually deprivatizing, actively engaging in the public sphere.

  The appeal of the secularisation thesis was undoubtedly compromised in the 1980s and 90s by the growth of Islamic movements in the Middle and Near East and the spread of evangelical and Pentecostal Christianity in Latin America and elsewhere. Peter Berger changed his mind about the link between modernisation and religious decline, commenting that the world “is as furiously religious as it ever was” in his introduction to The Desecularization of the World: Resurgent Religion and World Politics (1999). Any form of religious revival was taken to contradict the expectation of secularisation, even when it was observed in societies that were far from modern. Various books and articles highlighted particular cases, e.g. David Martin (1990) Tongues of Fire: The Explosion of Protestantism in Latin America.

  It was especially the enduring religiosity of the United States, though, that seemed to contradict the idea that modernisation produces religious decline. Mark Chaves (1994) tackled the issue in the journal Social Forces by framing “Secularization as declining religious authority”. In retrospect we can see that declining affiliation and churchgoing was underway in the US (and not just in mainline Protestant denominations), but it was not until considerably later that it became impossible to ignore.

Empirical research

  Investigation of religious change expanded rapidly during the 1990s, and the work was increasingly empirical and statistical. Although quantitative research on religion can be traced back at least to the Middletown studies in the 1920s, and The Religious Factor (1961) by Gerhard Lenski was a particular landmark, most of the work mentioned above is strikingly theoretical. The first chapter of Wilson (1966) is a close examination of numerical measures of religious involvement in England, but the remainder of the book is largely devoted to developing a theoretical understanding of secularisation. Berger (1967) is even less concerned with empirical analysis.

  By 2001, however, Mark Chaves and Philip Gorski could write a review article on “Religious pluralism and religious participation” for the Annual Review of Sociology that assessed 193 tests of the hypotheses. The abstract states that

For more than a decade, sociologists of religion have been debating the answer to a basic question: What is the relationship between religious pluralism and religious vitality? The old wisdom was that the relationship was negative, that pluralism undermines vitality. This view has been challenged by advocates of a supply-side model of religious vitality. They argue that the relationship is positive—that pluralism increases vitality—and this empirical claim has become foundational to the larger project of applying economic theory to religion. We review the relevant evidence and reach a straightforward conclusion: The empirical evidence does not support the claim that religious pluralism is positively associated with religious participation in any general sense.

  In the following year (2002), David Voas, Daniel Olson and Alasdair Crockett went even further in “Religious pluralism and participation: Why previous research is wrong” in the American Sociological Review. They found that “Almost all of these findings (both positive and negative) should be abandoned. The associations reported do not reflect the effects of pluralism but a previously overlooked mathematical relationship between measures of religious participation and the index of pluralism.” Nearly two decades later, Olson was able to circumvent this problem using sophisticated statistical methods, finding that religious diversity leads to declining religious involvement, as predicted by secularisation theories: see Olson, Jung, Marshall and Voas (2020) “Sacred canopies or religious markets? The effect of county-level religious diversity on later changes in religious involvement” in the Journal for the Scientific Study of Religion.

  To return to the point, the controversies over rival theories of religious change gave a boost to increasingly rigorous empirical research. Some of the arguments since have tended to oppose ‘scientific’ perspectives with more impressionistic cultural studies.

Classical secularisation theory

  The secularisation thesis is not a single theory; it is the proposition at the heart of a paradigm (Kuhn 1962) or scientific research programme (Lakatos 1970, Voas 2020). It has been developed in four distinct directions. These avenues of research often overlap and are far from being mutually exclusive, but it is useful to distinguish them. The theories (with leading proponents) are:

• Classical (Bruce)

• Existential security (Norris & Inglehart)

• Gender-fertility (Brown, Jenkins, Inglehart)

• Secular transition (Voas, Stolz)

  Steve Bruce was the leading proponent of the secularisation thesis during the 1990s and subsequently. In 1992 he edited a collection entitled Religion and Modernization: Sociologists and Historians Debate the Secularization Thesis; in 1999 he produced Choice and Religion: A Critique of Rational Choice Theory. His summation the theory and evidence in God is Dead: Secularization in the West (2002) was the key reference for a decade, until it was superseded by Secularization: In Defence of an Unfashionable Theory (2011). Interestingly, the one-line statement cited here (“The basic proposition is that modernization creates problems for religion”) appears in the 2002 book but was not repeated in 2011.

  Bruce is the direct descendent of the classical writers on the topic, from Max Weber through Bryan Wilson, Karel Dobbelaere and the early Peter Berger. He defends the standard view: modernisation creates multiple problems for religion, and the multiplicity of causes is such that secularisation is over-determined, at least in Western societies. Religious decline is long and gradual; it is irreversible but not inevitable, because it depends on the right conditions being in place. In this respect Bruce goes much further in the direction of generality than David Martin, who focuses on cultural and historical contingency, but not as far as David Voas and Jörg Stolz, who argue for the universality of the link between modernisation and secularisation.

  Most of the work on secularisation could be seen as supportive of Bruce’s classical version of the theory. The alternative theories outlined below differ largely in emphasis, for example in giving more importance to specific causes, or suggesting that the theory applies more or less broadly, or that decline has occurred rapidly rather than slowly. A major empirical study supporting the theory is Religion and Modernity: An International Comparison, by Detlef Pollack and Gergely Rosta (2017); another work looking at evidence from around the world is Beyond Doubt: The Secularization of Society by Isabella Kasselstrand, Phil Zuckerman and Ryan Cragun (2023). The Plight of Western Religion: The Eclipse of the Other-worldly by Paul Gifford (2019) is a more discursive treatment that emphasizes the erosion of supernaturalism in modern societies.

  Jörg Stolz has made major contributions to the secular transition model (see below), but much of his theoretical and empirical work is also supportive of classical secularisation theory. In the 2009 article “Explaining religiosity: Towards a unified theoretical model” in the British Journal of Sociology, Stolz shows that many factors, including some drawn from rational choice perspectives, can be combined in a comprehensive explanatory framework.

  Stolz develops that framework more fully in (Un)Believing in Modern Society: Religion, Spirituality, and Religious-Secular Competition (co-authored by Könemann, Schneuwly Purdie, Englberger, and Krüggeler, published in German, French and English in 2014, 2015 and 2016 respectively). Although presented as a new theoretical perspective, the work is perhaps best viewed as an enhanced version of secularisation theory. Stolz et al. highlight the way that attractive alternatives to religious belief and practice appear in the modern world, with individuals in the “me-society” being free to choose secular options. Empirically, they show that qualitative retrospective data point to the same religious decline as quantitative data. See also Jörg Stolz and Pascal Tanner (2017) “Elements of a theory of religious-secular competition” in Política & Sociedade. Stolz’s presidential address to the International Society for the Sociology of Religion, published in 2020 in Social Compass as “Secularization theories in the 21st century: Ideas, evidence, problems”, provides an overview of progress and challenges in research on this topic.

Existential security theory

  Meanwhile, Pippa Norris and Ronald Inglehart used World Values Survey data from dozens of countries around the world in their book Sacred and Secular (2004, second edition 2011). They pointed to the crucial role of ‘existential security’. Societies where people feel vulnerable (physically, financially or socially) tend to be religious, while those with safe environments, stable institutions, sound economies and welfare systems are sufficiently secure that religion seems redundant.

  This version of secularisation theory has received much attention, in part because Inglehart was one of the foremost modernisation theorists of recent decades, and in part because the explanation is consistent with the strong association between prosperity and secularity at a national level. It is much less clear, however, that the proposed mechanism operates as suggested at the individual level. Tim Immerzeel and Frank van Tubergen (2011) report corroborating evidence in “Religion as reassurance? Testing the insecurity theory in 26 European countries” in European Sociological Review. By contrast, Franz Höllinger and Johanna Muckenhuber (2019) find little support at the individual level in “Religiousness and existential insecurity: A cross-national comparative analysis on the macro- and micro-level” in International Sociology, and Francesco Molteni (2025) is similarly ambivalent in “Rising security and religious decline: Refining and extending insecurity theory” in the journal Sociology of Religion. Molteni’s 2021 book A Need for Religion: Insecurity and Religiosity in the Contemporary World makes it apparent that existential (in)security may not feature prominently among the mechanisms producing religiosity or secularity.

Gender, sexuality and fertility theory

  The theory connecting low fertility and changing gender relations with religious decline derives from the work of Callum Brown, in particular his books The Death of Christian Britain: Understanding Secularisation 1800-2000 (2001, second edition 2009), Religion and the Demographic Revolution: Women and Secularisation in Canada, Ireland, UK and USA since the 1960s (2012), and The Battle for Christian Britain: Sex, Humanists and Secularisation, 1945-1980 (2019).

  One can be critical of Brown’s view that secularisation was the product of a cultural revolution in the 1960s while still regarding the gender-and-fertility theme as being among the most innovative and important ideas in recent research on this topic. In Religion and the Demographic Revolution, he claims that the 1960s saw three of the greatest social changes in the West: secularisation, falls in fertility and marriage, and a revolution in women’s lives. “The hypothesis of this book is that the incidence and spread of these three trends from the 1960s were intimately and causatively interconnected” (Brown, 2012, p. 1). Brown focuses on gender as a key factor in religious decline, and he presents a richly discursive treatment of cultural history to develop the connection. Attempting to summarize, he identifies two main factors:

First in significance in my mind is the impulse of women’s desires – for sexual liberation, control of their own fertility, and freedom to construct their self in clothes and appearance according to declines in moral constraint … Second comes the ideological impulse of second-wave feminism. … the intellectualized idea of women’s subordination and need for liberation tended to make religious faith a victim of second-wave feminism (Brown, 2012, pp. 263-4).

  The next step empirically was taken by another historian, Philip Jenkins, who assembled data from every part of the world in Fertility and Faith: The Demographic Revolution and the Transformation of World Religions (2020). “Rarely remarked even by expert observers, a direct relationship exists between the fertility rates of a community – the number of children that a typical woman bears during her lifetime – and that society’s degree of religious fervor and commitment” (Jenkins, 2020, p. 3). As population growth has been falling almost everywhere outside sub-Saharan Africa, the implication is that organized religion is in trouble around the world. The jacket blurb states that “the religious character of many non-European areas is highly likely to move in the direction of sweeping secularization.”

  Theoretically, however, Jenkins takes a step back; he is reluctant to commit himself on the causal mechanisms. He simply says that “when such correlations occur frequently and examples accumulate, it is probable that we are dealing with a real relationship between the two elements” (Jenkins, 2020, p. 12). Does reproduction suffer when religious commitment declines, or is there something about low fertility that leads to loss of faith? Or is there an underlying cause for both? Jenkins does concede, though, that “Changes in sexual morality above all were powerful drivers in the movement away from religion … The harder the churches fought to resist those changes, the more they undermined their own prestige and power” (Jenkins, 2020, pp. 61-2).

  The same idea was developed theoretically by Ronald Inglehart in Religion’s Sudden Decline: What’s Causing it, and What Comes Next? (2021), the last book published before his death. Inglehart is much more specific than Jenkins about the supposed mechanisms and causal connections. He suggested that the connection between fertility and faith has changed. “Throughout most of history, religious institutions were able to impose pro-fertility norms. But the causal relationship is reciprocal and the dominant direction can be reversed: if pro-fertility norms come to be seen as outmoded and repressive, their rejection also brings rejection of religion” (Inglehart, 2021, p. 11). His argument is that people in late modern societies are committed to autonomy, self-expression, and individual control over their bodies, sexuality and intimate relationships. Religions typically seek to regulate reproductive behaviour, promote fertility, and defend traditional norms around sex, marriage and gender roles. Given the resources to pursue their own choices, people will leave religion behind. A number of questions remain unanswered, such as what triggers the reversal in the direction of causality, how much of the change occurs in adults as opposed to during socialisation, and what the mechanisms are at the individual level. We might suspect that value change rooted in the prosperity, complexity, diversity, and freedom of modern society is the cause of both the ‘second demographic transition’ and secularisation.

Secular transition theory

  In “The continuing secular transition”, a paper originally presented at a conference in 2000 and ultimately published in 2007 in The Role of Religion in Modern Societies (edited by Detlef Pollack and Daniel Olson with a copyright date of 2008), Voas argued that what he called the secular transition is inseparable from other transitions that constitute modernisation, e.g. the industrial revolution, the decline in mortality, or equalisation in the status of women. A transition is a permanent large-scale change; it is not cyclical or recurring. Specifically, Voas points to parallels between the fertility transition—the global decline in birth rates—and what might be called the secular transition, the move away from religion. He predicted that secularisation would spread across the world in essentially the same order as the fertility transition.

  The theory that modernisation leads to a secular transition is an extension of the classical secularisation thesis marked by forthright support for several key elements:

• The scope is not limited to the Christian West: it applies universally.

• Secularisation often comes very late in the process of modernisation, but it is inevitable.

• Decline is usually irreversible, with circumstances that could produce temporary revivals needing to be defined.

• Socialisation in childhood is crucial for the transmission of religion, and religious identity, practice and belief are relatively settled in the mid-20s; subsequent lifecycle and period effects are modest or absent.

• Generational replacement is thus the main driver of religious decline, which in consequence tends to be gradual but self-reinforcing: less religious parents raise less religious children.

• Alternative spirituality and other forms of religious diversification are outcomes of the early stages of secularisation and staging posts on the way to more complete secularity, not stable and persistent replacements for organised religion.

• A country’s history and culture may influence when the secular transition starts, but once underway there is a common pattern of religious decline, a rise and then fall in ‘fuzzy fidelity’, and finally a situation where most of the population is secular.

• The transition from nearly everyone in the society being religious to most people being unreligious extends over more than two centuries.

• The complexity and simultaneity of social transformations produced by modernisation ensure that religious decline is over-determined, and we should not expect to identify a small number of necessary and sufficient causes.

  Voas pursued some of these hypotheses in subsequent work. He offered a model to explain the downwards trajectory of baptisms in England in the 2003 article “Intermarriage and the demography of secularisation” in the British Journal of Sociology. With Alasdair Crockett, he pointed out that the transmission of religious identity, practice and belief is so weak that “in Britain institutional religion now has a half-life of one generation” in “Religion in Britain: Neither believing nor belonging” in the journal of Sociology in 2005. The following year (2006), they identified cohort replacement as the proximate cause of secularisation in “Generations of decline: Religious change in twentieth-century Britain” in the Journal for the Scientific Study of Religion.

  In “The rise and fall of fuzzy fidelity in Europe” (2009) in the European Sociological Review, Voas wrote that

Many people are neither regular churchgoers nor self-consciously non-religious. The term ‘fuzzy fidelity’ describes this casual loyalty to tradition. Religion usually plays only a minor role in the lives of such people. Religious change in European countries follows a common trajectory whereby fuzzy fidelity rises and then falls over a very extended period.

  The model described in the article was found to apply to the United States as well; see Simon Brauer (2018) “The surprising predictable decline of religion in the United States” in the Journal for the Scientific Study of Religion. And indeed, Voas and Chaves answered “no” to the question “Is the United States a counterexample to the secularization thesis?” in the American Journal of Sociology in 2016, not only because “American religiosity has been declining for decades,” but importantly because “this decline has been produced by the generational patterns underlying religious decline elsewhere in the West: each successive cohort is less religious than the preceding one.”

  The article “Modeling fuzzy fidelity: Using microsimulation to explore age, period, and cohort effects in secularization” in the Journal of Religion and Demography (by Puga-Gonzalez, Voas, Kiszkiel, Bacon, Wildman, Talmont-Kaminski, and Shults, 2022) reports on computational findings that support the primary role of “cohort replacement, based on weakened transmission of religiosity as a function of the social environment.”

  In the 2025 article “Invisible secularity: American theism beyond belief” in the journal Social Forces, Voas writes that “The apparent persistence of belief in God conceals erosion in its substance, strength, salience, and stability. ... Contrary to claims that apparent secularization masks enduring (if unorthodox) invisible religion, implicit religion, diffused religion, lived religion, and so on, the evidence shows that continuing religious involvement disguises waning religiosity.”

  Ryan Cragun and Ronald Lawson analysed “The secular transition: The worldwide growth of Mormons, Jehovah’s Witnesses, and Seventh-day Adventists” (2010) in the journal Sociology of Religion; they suggest (echoing Steve Bruce) that modernisation can initially provide a propitious environment for religious growth, and particularly for strict sects.

  Sarah Wilkins-Laflamme addresses the key issue in her 2021 article “A tale of decline or change? Working toward a complementary understanding of secular transition and individual spiritualization theories” in the Journal for the Scientific Study of Religion. International Social Survey Programme (ISSP) data from 46 countries around the world show that there has been both change and decline, but decline is the dominant trend and “this decline also leaves the door open for more individualized spiritualities to spread.”

  In parallel with this work by Voas and others, Jörg Stolz and colleagues have done important research on testing and refining the secular transition model. In two articles dealing with the case of Germany, Jörg Stolz, Detlef Pollack, Nan Dirk De Graaf and (for the second) Jean-Philippe Antonietti explain how and why East Germany does not conform to the usual pattern; see “Can the state accelerate the secular transition? Secularization in East and West Germany as a natural experiment” (2020) in the European Sociological Review, and “Losing my religion as a natural experiment: How state pressure and taxes led to church disaffiliations between 1940 and 2010 in Germany” (2021) in the Journal for the Scientific Study of Religion.

  Religiosity in France follows the expected trajectory, as shown by Stolz, Ferruccio Biolcati-Rinaldi and Francesco Molteni in their 2021 article, “Is France exceptionally irreligious? A comparative test of the cohort replacement theory” in L’Année sociologique. Interestingly, Voas (2008) notes that France was in the vanguard of the fertility transition, and “One could conjecture that secularization will spread across the world in essentially the same order.” The findings of Stolz et al. suggest that “France began on its secularization path earlier or from a lower level than other countries.”

  Georgia experienced a religious revival in the post-Soviet period that challenges the notion that secularisation is irreversible. Stolz, Alexi Gugushvili, Francesco Molteni and Jean-Philippe Antonietti address this issue in their 2023 article “A counterexample to secularization theory? Assessing the Georgian religious revival” in the British Journal of Sociology. Stolz and Voas tackle the general problem of “Explaining religious revival in the context of long-term secularization” (2023) in the journal Religions.

  In a remarkable article published in 2024, “Can we explain the generation gap in churchgoing?” in the Journal for the Scientific Study of Religion, Stolz (with Oliver Lipps, David Voas and Jean-Philippe Antonietti) reports findings that could change the way sociologists think about secularisation. “Parental religiosity [is] the most important predictor of child religiosity, but no attributes of the parents or the immediate context seem able to explain the fact that children are consistently less religious than their parents.” Stolz et al. suggest that “modernization operates indirectly on the religious transmission process, through socially dominant values, ideologies, and worldviews, rather than directly through attributes of the parents and their immediate context.” This account is consistent with our current understanding of the fertility transition, in which “it seems that the diffusion of ideas and norms, through personal networks, the workplace, or mass media changes the probability of having fewer children.”

  And in 2025, Stolz, along with Nan Dirk de Graaf, Conrad Hackett and Jean-Philippe Antonietti, produced what is probably the most important contribution to secular transition theory since Voas (2009). In “The three stages of religious decline around the world” in Nature Communications, they propose that “that secular transition happens in three steps: first, public ritual participation declines; second, the importance of religion to individuals declines; and third, people shed religious affiliation. We test this model using datasets from surveys in 111 countries.” The results are the best evidence so far that the secular transition is universal and follows a common pattern.

The opposition

  Although religious decline has become increasingly apparent in the United States, Latin America, East Asia and elsewhere, the title Secularisation: In Defence of an Unpopular Theory (Bruce 2011) remains apt. For many sociologists of religion and probably most scholars in religious studies, the mantra is ‘change, not decline’. An interesting variant is ‘decline, not secularisation’: a frank acknowledgement that organised religion has lost ground, combined with a rejection of the secularisation thesis.

  The most prominent example is the book A Secular Age (2007) by the philosopher Charles Taylor. In 1500, religious belief was taken for granted; by 2000, it was merely an option. People now operate in an ‘immanent frame’ of everyday naturalism. But according to Taylor it is secularity without secularisation, which he labels a ‘subtraction story’. What we see instead is an explosion (or ‘nova’) of spiritual options. People still want what religion provides, but the ‘conditions of belief’ have changed, so now some effort is needed. Taylor seems disinclined to accept that people are increasingly adjusting to life without supernaturalism.

  The historian Hugh McLeod pays more attention to ordinary people. His 2000 book Secularisation in Western Europe 1848-1914 emphasizes national particularities, and to that extent he takes a sceptical view of the link between modernisation and secularisation. The Religious Crisis of the 1960’s (2007) is influenced by Callum Brown’s view that change came suddenly in that decade, but McLeod tells a multi-causal story about upheavals in both the Christian churches and the general culture.

  One of the most surprising recent contributions is from the sociologist Christian Smith, a longstanding critic of the secularisation thesis. The title Why Religion Went Obsolete: The Demise of Traditional Faith in America (2025) points to the seriousness of religious decline in the United States. Remarkably, he claims that religion is not merely doing poorly at the moment in America but is actually obsolete (or perhaps more accurately obsolescent). Like Callum Brown, he argues that this collapse occurred suddenly (around the millennium in the US, rather than in the 1960s), and he also appeals to local factors and historical contingency. These scholars are thus able to accept the reality of waning religious identity and participation while continuing to contest the idea that their national cases are instances of a general phenomenon.

Dimensions of disagreement

  As is evident from the literature review above – abbreviated as it is – there is a striking lack of agreement about the nature and causes of declining religious identity, practice and belief. At least four dimensions of disagreement are apparent.

First, some scholars assert that decline is limited to ‘church’ religion, which is being replaced by other forms of religious practice or spirituality. They argue that what we observe is change, not decline. People may move away from organised religion, but most continue to be religious or spiritual in different ways. The supposed change takes many forms: a shift towards new religious movements, or alternative spirituality, or private forms of ‘lived religion’, and so on. The impulse to see transformation rather than subtraction is not new: it has existed since the earliest debates on this topic.

The second dimension of disagreement concerns whether there are common causes operating across different societies. The leading paradigm is the secularisation research programme. The core idea is that modernisation creates problems for religion, and as a result religious practice, identity and belief tend to be lower in highly developed countries than in more traditional societies. The alternative view is that religious change and decline are the consequence of particular social, political, historical and cultural circumstances applying to the specific society in question. It is thus possible to accept the reality of decline while denying that it is a consequence of modernisation. A popular line of research between roughly 1985 and 2010 suggested that where religious involvement is low, it is because the supply of (not the demand for) religious goods and services is weak.

Thirdly, there is a tension between simple and complex explanations. Some scholars have tried to identify one key factor as the driver of the process of change or decline, while others take a multi-causal view.

Fourthly, there is disagreement about the speed of the process. The standard position was that secularisation develops over centuries. More recently, a number of authors have argued that religious decline has been sudden, at least in some countries.

Although these dimensions have been presented as dichotomies, the boundaries are not always sharp. It is debatable, for example, where David Martin and Christian Smith stand on the question of whether the main story is decline or simply change. If we treat these authors as theorists of religious decline and then focus on the last three dimensions, the classification below emerges:

Theories of religious decline: A taxonomy

It may be helpful to justify the location of each theory within this taxonomy.

Culturalist theories

The theories here labelled ‘Culturalist’ are based on an acknowledgement that organised religion has lost ground, combined with a rejection of the classical secularisation thesis. Although his 1978 book is entitled A General Theory of Secularization, David Martin does not accept the idea that modernisation inevitably brings about the decline of religion or that there is one general path from religion to secularity. He argues that even within Western Europe, church-state relations were highly varied, and this variation influenced the strength or weakness of religion in modern times.

Martin saw the differentiation of social spheres, which has played out in varied ways, as a key force in shaping the fate of religion. He produced a typology of how different relationships between church, state, and society produce different trajectories of religious change. Where the church stood in relation to the ancien régime, whether or not it became identified with national resistance or class domination, what kind of revolution (or reform) a society underwent, all served to determine how religion fared as those societies developed. Thus secularisation depends on the specificities of the context.

The process is gradual and even potentially reversible, with the success of the Pentecostal movement showing that religion can thrive with modernisation rather than being eroded by it. Even societies with broadly similar cultural and institutional roots can follow markedly different religious trajectories depending on their underlying political foundations.

In short, Martin emphasises historical and cultural particularity rather than universality. Change extends over a long period. Although he sees the relationship between church and state as an important determinant of the course of secularisation, he recognises many influences on religion and secularity.

The ‘culturist’ label might also be applied to other scholars who explain religious decline in local rather than global terms, with Christian Smith being a prominent case in point. For many years the US was held up as a decisive counter-example to the secularisation thesis. Now that religious decline is obvious, many of the explanations for American secularity appeal to national political and cultural conditions. Christian Smith is thus able to accept the reality of waning religious identity and participation in the US while continuing to contest the idea that it is simply one instance of a Western or even global phenomenon. (The thesis explaining disaffiliation as a liberal reaction to the religious right is an example of the same tendency; see Hout and Fischer 2002, 2014.)

In Why Religion Went Obsolete, Smith’s focus is entirely on the United States, and he pins responsibility on a panoply of factors. The main part of the book is headed “Perfect Storms Converging.” Over five chapters, there are 43 sections describing those storms.

The local contingency is implicit rather than explicit – Smith never says that his explanation only applies to the United States – but most of the factors on the list are specific to the 1990s and 2000s in the US. Smith provides a very particular story about American culture around the turn of the millennium to explain why religion became obsolete.

Nobody in the preceding decades foresaw the big picture that was developing or what the consequences might be for religion. As late as the twilight of the 1980s, it seemed that traditional religion was ‘back.’ That was an illusion, however. Once that decade had passed, many trending forces of the 1960s and 1970s gained new momentum. All it took then was an unpredictable set of further events to push religion into terminal decline (Smith 2025: 127).

Waning religious involvement is thus held to be a recent occurrence. He writes: “All of the evidence about traditional American religion’s recent troubles points to two crucial decades of decisive change: the 1990s and 2000s. ... I can be even more precise. The pivotal year marking the launch of the new zeitgeist that remade religion was 1991” (Smith 2025: 12-13).

The distinctions in this taxonomy are not always sharp. In fairness to Smith, he writes that “The sociocultural conditions and forces that drove American traditional religion into obsolescence were a long time in the making” (Smith 2025: 127). Nevertheless, it seems fair to say that he views religious decline in America as sudden, multi-causal, and locally contingent.

Gender-fertility theory

In The Death of Christian Britain, Brown argues that religious decline in Britain was sudden, mono-causal, and locally contingent. From the blurb to the second edition: “Callum Brown challenges the generally held view that secularization was a long and gradual process dating from the industrial revolution. Instead, he argues that it has been a catastrophic and abrupt cultural revolution starting in the 1960s.”

In Religion’s Sudden Decline, Ronald Inglehart (2021) moved away from the gradualist, multi-causal view that links secularisation to human development, or indeed the focus on existential security in Sacred and Secular (Norris & Inglehart 2004). The blurb refers to “this dramatic worldwide shift”, noting that “virtually all major religions encouraged women to stay home and produce as many children as possible; and they sternly discouraged divorce, abortion, homosexuality, contraception, and any other form of sexual behavior not linked with reproduction. These norms … [are] no longer necessary for societal survival. ... The resulting tension, together with the fact that rising existential security has made people less dependent on religion, opened the way for an exodus from religion.”

Both scholars, then, are firmly behind the notion of sudden change. Both see something in the gender / sexuality / fertility nexus as the primary driver of secularisation; Brown places more emphasis on gender equality while Inglehart emphasises the shift to self-determination in sex and reproduction, but the basic idea is the same. The difference between them lies in their treatment of cultural contingency. As a historian, Brown wants to study the specific details of each case and is reluctant to give any role to broad and amorphous concepts of ‘modernisation’. As a social scientist and modernisation theorist, Inglehart is much more inclined to see the drama in global rather than local terms.

Existential security theory

The scope of Norris and Ingelhart’s theory is clearly global. Rising existential security is a gradual process. And although they might not maintain that security is the only cause of secularisation, the central role given to this factor justifies treating their theory as monocausal.

Classical theory

The traditional view is that secularisation is a gradual process with deep historical roots. See, for example, Figure 2.1 in Steve Bruce (2011). The classical theory of secularisation describes the process as gradual, multi-causal, and potentially global, though Bruce focuses on the Western world and is reluctant to make claims about what will happen beyond it.

Secular transition theory

The process of the secular transition extends over at least a couple of hundred years, as illustrated in Voas (2009). The theory is firmly in the gradualist category.

It is also general rather than particular to any region. Stolz et al. (2025) look at every country in the world for which we have suitable data and provide evidence that the secular transition is universal and follows a common pattern. They write that “It is well established that religion declines in Western countries. The secular transition model provides a prominent theoretical explanation for this phenomenon (Voas 2008, 2009). … We have presented a revised model of secular transition according to which cohorts drop dimensions of religiosity in a specific order over time … Summing up, our results suggest a worldwide secular transition has been underway for many decades. Religion seems to be declining in three stages worldwide.”

Summary assessments

Survey evidence casts doubt on the idea that ‘the millennial zeitgeist’ is to blame for the demise of traditional American religion. Religious involvement in the United States has been waning for several decades; it is not even apparent that the pace of religious decline quickened during the 1990s and 2000s. The obvious change since 1991 has been the rise of the nones, which (at least according to Stolz et al. 2025) is an end stage of religious obsolescence.

The gender-fertility theory is an important innovation, but much more investigation is needed. Causality does not have to run in just one direction. Inglehart suggests that it used to go from religion to fertility, and now it has been reversed. Jenkins writes that “A shift to lower fertility encourages declining religiosity, which in turn would discourage religious enthusiasm, and so on, in a kind of feedback loop” (p. 14). As suggested above, we might suspect that value change rooted in the prosperity, complexity, diversity, and freedom of modern society is the cause of both the ‘second demographic transition’ and secularisation. Monocausal explanations seem too simple to account for religious decline.

If existential security is indeed the key mechanism driving secularisation, we should be able to see it in operation at the individual level. The evidence suggests that we do not. Applied to whole societies, it simply amounts to reiterating that safe environments, stable institutions, healthy economies and good welfare systems are associated with religious decline, and that sounds very much like the classical theories about modernisation undermining religion.

The theory of secularisation remains the pre-eminent account of religious change in the modern world. It is broad enough to encompass multiple explanatory factors, and to that extent is best seen as a research programme or paradigm rather than a single theory. It has gradually been extended and refined over many years. The theory that a secular transition is underway is a promising development of the core ideas; it builds on the work of Wilson, Berger, Dobbelaere, Bruce and others through additional hypotheses and empirical testing.

The evidence is accumulating that a secular transition is really taking place, following a common pattern around the world, and that multiple features of modernisation erode religious identity, practice and belief, typically over a long period.

Acknowledgements

  I am grateful to all of the scholars who have contributed to research on this topic. My thanks to Jörg Stolz for his comments on an earlier version of this essay, to Ferruccio Biolcati Rinaldi for his comments on a seminar presentation, and to Steve Bruce for his support over a long period.

How to cite this essay

Voas, David (2026) The secular transition: A bibliographic essay. www.seculartransition.org/bibliographic. Version 2, 9 June 2026.