[University home]

Institute for Social Change

A Transatlantic Comparison of Religion’s Role in Society


Robert D. Putnam (Harvard)

David Voas (Manchester)

David E. Campbell (Notre Dame)

Project co-ordinator:

Thomas Sander (Harvard)


Siobhan McAndrew (Manchester)

Rodney Ling (Manchester)

The Harvard-Manchester project on the Transatlantic Comparison of Religion’s Role in Society aims to accomplish three important goals: 1) to test the robustness of Putnam et al.’s earlier work measuring religious beliefs and their social impact in the US in the related but distinct national ‘laboratories’ of Britain, Ireland and Northern Ireland; 2) to explore similarities and differences in religion’s role in contemporary society in four national settings in which the role of religion has been historically very different; and 3) to construct first class new datasets in the UK and Ireland to help quantitative non-religionists (like economists or political scientists) to explore the importance of religion and to encourage the rigorous quantitative study of religion in Western Europe. 

Objective I:  Replicating research on the changing role of religion in America

With the financial support of the Templeton Foundation, the Harvard research team conducted an extensive – arguably the largest ever – survey examining the links between religion and social capital  in the United States (Faith Matters 2006, or FM 2006). FM 2006 yielded high quality data that enabled a follow-up, or panel, survey: FM 2007. We are building on the foundation laid by these two surveys in the United States by conducting parallel studies in Great Britain, Northern Ireland, and Ireland.

Research on the links between religion and social capital across multiple nations is a natural extension of the existing social capital literature. Indeed, a concern for religion was present at the creation of social capital as a theoretical concept. In his pioneering work on social capital, Coleman argued that the high academic performance of urban Catholic school students stemmed from the interlocking social connections within the neighborhoods surrounding the typical parish school, facilitated by the ethnic and religious homogeneity of these communities.  Putnam then exported the concept of social capital to Italy as an explanation for the differential performance of regional governments.  Again, religion was shown to matter. But in highlighting the significance of religion’s impact on social capital, Putnam also demonstrated that the links between them defy simplistic generalization.  While Coleman found high levels of social capital within urban Catholic parishes, in Southern Italy Putnam found that the dominance of the rigidly hierarchical Italian Catholic Church sapped the region of social capital.

Upon re-importing social capital back to the US, Putnam found a strong empirical connection between religion and social capital. “As a rough rule of thumb, our evidence shows, nearly half of all associational memberships in America are church related, half of all personal philanthropy is religious in character, and half of all volunteering occurs in a religious context.”  His conclusions echo Alexis de Tocqueville, who commented on the high levels of religious involvement among Americans in the early 1800s. Indeed, in famously describing how “Americans of all ages, all stations in life, and all types of dispositions are forever forming associations”, Tocqueville placed religious and moral associations at the top of the list.  Yet notwithstanding the general conclusion – from both past and present – that high levels of religiosity and social capital walk hand-in-hand, puzzles concerning the distribution and operation of these variables remain.

The findings from the Faith Matters surveys in the US need to be replicated in other countries.  To take a single illustration, we know from the US study that religious people are better neighbors and better citizens – voting more, volunteering more, and so on.  Our understanding of the import of this generalization would be greatly strengthened if we knew whether or not this feature of religion is unique to the US, or is also true in very different social contexts, such as those exemplified by Britain, Ireland, and Northern Ireland.  In the US, it is not the theology or simply social networks that explains extraordinarily high level of secular engagement and secular philanthropy, but rather membership in a ‘moral community’.  Is such a pathway similar or different in the UK or Ireland? 

Objective II: Learning from transatlantic cross-national comparison

The logic of comparison is fundamental to virtually all science, especially social science.  Cross-national comparison has long been used by scholars and observers as the basis for developing and confirming insights into social, political, and religious dynamics.  Take two relevant examples. Tocqueville’s legendary analysis of Democracy in America was grounded in comparison between France and America.  Because of this binocular vision, Tocqueville could see and illuminate essential features of each country that had been invisible to observers who had focused only on one or the other case.  Similarly, Weber’s classic analysis of the links between Protestantism and capitalism relied on comparisons of economic development in various national settings, some predominantly Protestant and others not.

The proposed project will focus on four cases that all share certain fundamental features:  All are historically Christian, all are representative democracies, all are socioeconomically advanced, with post-industrial economies, high levels of education, and modern states and political parties. 

On the other hand, the role of religion in these polities and societies illuminates some profound contrasts.  At the most descriptive level, religious observance in the US has been historically been relatively high and roughly constant (roughly 40% reported weekly church attendance) over the last half century.  While religiosity was once very high in Great Britain (consider the religious wars of the 17th century), for most of the 20th century it has represented the classic case of long-term secularization case, and currently rates of reported weekly church attendance have fallen well below 10%.   Ireland until quite recently had extremely high rates of religiosity (reported weekly church attendance through 1980 was above 90%), but over the last quarter century has seen very rapid secularization.  Church attendance has been cut virtually in half in barely two decades and continues to plummet. Our fourth case, Northern Ireland, has been (at least until very recently) one of the archetypal cases in the modern world of religiously-based violence, a society in which religious beliefs and affiliations have dominated much of personal, social, and political behavior.  

Against that backdrop the proposed comparison would be fruitful for many reasons.  First, these cases present different starting points in terms of the founding of the modern state and the role that religion played in that foundation.  More recently, each of these places has experienced considerable social, economic and political change – all of which affect the role of religion and offer excellent data upon which to examine further changes in society that seem destined or likely to occur. 

Objective III:  Fostering and facilitating research by other scholars and experts

Our survey will produce a rich source of data for social scientists from multiple nations and multiple disciplines: economics, political science, sociology, and others.  We expect that these datasets will be rapidly exploited by many other scholars, often on themes very different from those outlined here.  And just as the Harvard team’s work on social capital in the US has spawned an entirely new field of research over the last decade, we have some hope that the surveys proposed here may help revitalize the scientific study of religion in Europe.

In Britain, the study formed part of the 2008 British Social Attitudes (BSA) survey administered by National Centre for Social Research (NatCen), Britain’s pre-eminent survey on social attitudes.  Given the social salience of Muslim identity in Britain, we are also carrying out an over-sample of Muslims in the UK.  This parallel study, and those in Northern Ireland and the Republic of Ireland, are being conducted as a stand-alone surveys by Ipsos-MORI.

The British survey runs in conjunction with the International Social Survey Programme (ISSP) 2008 religion module.  David Voas (Manchester) received a grant from NORFACE (the European consortium of funding councils) both to add additional questions (beyond the core 60 ISSP module questions) and to increase the sample size in Great Britain, Northern Ireland, Ireland, Denmark and the Netherlands.  This conjunction of the ISSP, NORFACE and Faith Matters modules makes the 2008 BSA survey the largest and best survey of religion in Britain ever conducted.