A recent BBC article claims that “Religion may become extinct in nine nations, study says.” (1) It references a study which actually suggests that religion might be going away in 85 regions for which they have data. (2) Obviously, this is a profoundly important finding if it is valid. The study is somewhat mathematically dense. Most of our readers will not want to wade through that complexity.
There was only one explicit prediction made in the study, that “nearly 70% of the Netherlands will be non-affiliated by mid century.” The study starts with a very simple model that assumes that everyone is either affiliated with religion or not and that a single number reflects the average experience of everyone in terms of the average religious affiliation that he or she has with friends and neighbors. The formula includes a single number which represents the expected relative utility of being not affiliated with religion.
That utility number is provided for four regions. the autonomous Aland islands region of Finland (0.63), Schwyz Canton in Switzerland (0.70), Vienna Province in Austria (0.58), the Netherlands (0.56). The utility of being affiliated with religion is one minus these numbers. The religious affiliation utility for the Netherlands would 1-0.56 or 0.44. The average utility number for religious non-affiliation seems to be 0.65. The study includes a chart with all of the data sets plotted around a line assuming 0.65. The obvious implication is that religion has a relative utility of close to half that, only 0.35.
Many of our readers will agree that going to church and having a moral idiot tell you how to live your life has a significant downside. Moreover the church would want you to contribute a substantial proportion of your income to support “God's work.” There seems to be an increase in the number of people who know that prayer does not work to change outcomes. Elisabeth Edwards before she died told people that she did not believe in a God who intervened in daily life. Francis Collins asserts that the purpose of prayer is not to get things from God but to align our thinking with God. In both cases a major assumed reward of religion in the past is not there.
There are several levels of more sophisticated modeling included in the study. They develop a more complex model with many differences in religious affiliation and social connections with others. There was another numerical simulation that look at what would happen if the affiliated and the non-affiliated populations had low levels of interaction. In both cases the simulations resulted in the same results, a dramatic decline in religious affiliation. The fact is that the assumption of a massively greater utility in not being affiliated with religion will overwhelm any assumed model of local conditions and social interactions. The only difference that the study reports is a difference in how long it takes for the massive turnover toward non-affiliation with religion to take hold.
Obviously there are areas where there is great utility in religious affiliation. There are places in America where an admission that you are an atheist will result in no one being willing to talk with you. If you are renting you can expect to be told in a very short period of time that you have to move out. In Pakistan anyone who had been affiliated with Islam can expect that nearly everyone in his or her community will want a convert to humanism to be put to death. A recent poll found that 30 percent of Muslims in England felt the same way. In all these cases the utility of religious affiliation will be substantially higher than 0.35 assumed for the 85 regions in this study. In Pakistan the utility value of religion would have to be greater than 0.99. Death is a massive disincentive to move toward religious non-affiliation.
This study does present a promising picture of a move away from organized religion. However, the failure to model the more complex utility functions which obviously exist in support of religion means this study cannot be trusted. The primitivism of religion will remain alive and well in all areas where severe sanctions can be imposed on non-believers.
The study's authors are in two cases members of the Department of Engineering Sciences and Applied Mathematics of Northwestern University. The other author is associated with the Research Corporation for Science Advancement and Department of Physics, University of Arizona. These are mathematically sophisticated scientists who are not trained in social sciences. They are working outside their fields. My guess is that they are excessively enamored with their equations and failed to see that the obvious problems with their study.
The bottom line rests on the machinery of government. If free speech is allowed and secularists can associate with others without severe sanctions religion will die out. However, the other side is fighting hard to get legal privilege for religion. The fight has not been won and we should not let this study convince us that we don't have to do anything to have religion go away. The do not make that case.
1 Religion may become extinct in nine nations... http://www.bbc.co.uk/news/science-environment-12811197
2 A mathematical model of social group competition with application to the growth of religious non-affiliation.