David Smilde's new book addresses the classic question of who converts to Evangelism and why, by asking whether people can actively decide to believe, and if so, why they sometimes (often?) decide against believing (or, at least, converting). Smilde's focus is Caracas, Venezuela, and his intensive sample of about fifty informants is male (which is itself an innovative step in a field dominated by studies of women). Smilde concentrates on the individual level rather than on the larger ‘social level outcomes’ of the Evangelical movement. At this level, he finds that Evangelical ideas allow individuals to feel they are ‘deciding to believe’. He also finds that this feeling depends on the ‘microcontexts’ of social networks: when surrounded by a family with a hostile orientation to Evangelicalism, for example, the individual is frequently prevented from converting. Such constraints, however, are also experienced as the effect of the individual's own will. Thus, Smilde concludes, behavioural outcomes in the religious field result from complex combinations of free will and cultural/social constraint.

This is a valuable, important book. It is an excellent example of the qualitative turn in sociology, in which traditional tools of survey and statistics are fleshed out and extended by rich life historical material, descriptions of specific cases and episodes, and where long stretches of interview transcript take up much of the text. An early chapter on the everyday violence of poor neighbourhoods in Caracas – above all, the pressure on men to take vengeance – deploys high-quality ethnography to show how conversion offers an escape route from the vicious cycle of male vendettas. The final substantive chapter, which focuses on two illustrative life histories, clearly reveals long-term individual social and psychological processes of affiliation (and disaffiliation) with Evangelical churches.

The book is especially strong in taking theology seriously, such as when it discusses the various ways converts explain the ‘failure’ of their new identities to solve particular life problems. Indeed, one of the most interesting parts of the book is its finding that Evangelical ideas can impede as well as facilitate effective evangelisation, by making evangelisers more, rather than less, judgmental of repeat and stubborn backsliders. The point is important and suggests a useful line of future research. It may be that as long as resistance to evangelisation can be interpreted as the action of the devil, the evangeliser may remain motivated to push ahead; but as evidence accumulates that the resistance is the result of pride and wilfulness, the evangeliser may give up.

There were a few ways the text could have been improved. First, while it is quite useful to point out that family enforcement of ‘conformity’ can impede conversion, Smilde never makes it entirely clear why the same process cannot also operate to facilitate conversion, as the convert-to-be tries to conform to the example of influential family members. Second, I doubted some of Smilde's claims about the prevalence of particular beliefs. While, for example, some of his informants may have subscribed to the Manichean view that God is not omnipotent in face of the devil (p. 138), it would have helped to compare the frequency of such an idea with what is certainly the more orthodox view, based on the Book of Job, of an omnipotent God using the devil instrumentally. I was also surprised that Smilde did not balance his treatment of some informants' view that their assent to God was a kind of ‘surrender’ to a force majeure with the tradition within Evangelicalism of appealing to the cognitive understanding of the Gospel, that ‘accepting Jesus’ is the expression of a mature understanding of truth.

The concluding chapter is also less than entirely convincing. It depends in part on Smilde's effort to distinguish his own approach from what he calls ‘conventionalism’ (culture is an arbitrary set of symbols disconnected from social reality), and ‘determinism’ (culture is determined mechanically by social reality). The problem is that it is not clear which scholars adhere to one or other of these two extreme views, thus making it hard to evaluate the claim that this book charts a genuinely new course. The idea that religious beliefs can be rationally and deliberately held does not strike me as particularly new. Certainly there is a venerable tradition in anthropology of ‘cognitivists’ – from Taylor and Frazer to Malinowski – who have pointed out the rationality of belief in this or that god, spirit, charm or spell.

In addition, though I understand Smilde decided not to focus on the broader social and political implications of his findings, he might have at least raised some issues for further analysis. In this connection, for example, it would have been useful to think about the implications of Evangelical ideas about the role of God and ‘business’. Over half of Smilde's sample thought that accepting the Lord should have as an aim the improvement of one's ‘business’ (presumably meaning one's waged job as well). The spiritual dimensions of business – how someone must be able to provide for family, have discretionary income to give to the poor, be successful to serve as witness to others – all remain unanalysed.

But these are minor quibbles. Overall, Reason to Believe raises the level of discussion of Protestant conversion in Latin America – and elsewhere – to a new level of sophistication. Its splendid combination of qualitative and quantitative methodologies will make it a model for anyone conducting research on these issues for many years to come.