At my other weblog I looked at some of the data on the international data on religion. There are two positions in regards to religious trends which always crop up.
* That in the medium-to-long term religion, in particular supernatural religion, will disappear.
* That in the short-to-medium term we are in the midst of a “religious revival.”
The first position has been held more or less by some intellectuals since the Enlightenment. The second position is something that I’m familiar with contemporaneously. The reality is that the world is not going through a revival in religiosity if by religiosity you mean active participation in religious activities and supernatural beliefs. Rather, if anything there is less of an emphasis on religious belief among younger people than the older.
Rather, there are two separate specific trends which lead people to conclude there is a religious revival, and one more general one. In the case of the last churches and religious actions are salient and catch our attention, so we’ll always be more aware of religious practice and proselytization than religious apathy. In the first two cases you have seen a revival in religious identity on a cultural level across the post and quasi-Communist world. Russia is now robustly identified as an Orthodox society, while Cambodia has given nominal recognition to the role of Theravada Buddhism in Khmer society. Even in China there is a trend toward rehabilitating Confucius, while the Central Asian Muslim nations have rediscovered their Islamic and Turkic identities. This does not mean that these societies are particularly religious on a world wide scale. Granted, it does seem that in many of these nations both religious affiliation and belief have gone up, but only from a very low threshold imposed by state atheism. Secondarily, societies which are very secular in religious belief and practice may still reassert their customary religious identity in the face of pluralism; this is occurring in much of Europe with the rise of Muslim minorities in traditionally Christian societies.
Finally, I’d like to add some data to the discussion. The correlation between religiosity and lack of wealth is well known. I looked in the WVS 2005 and created an index of religiosity by weighting the % who assert that religion is very important, rather important, not very important and finally not important at all. In society where everyone thinks religion is very important the index would be 3, and for not important it would be 0. For the independent variable I just took GDP per capita purchasing power parity. I generated three plots.
1) Religiosity index vs. GDP PPP per capita
2) Log(Religiosity index) vs. GDP PPP per capita
3) Log(Religiosity index) vs. Log(GDP PPP per capita)
The correlation assuming linearity is -0.68. The plots tend to “compress” at the high end of religiosity and low end of GDP PPP, so that’s my rationale for log transformation. The trend is important, but the deviations show you civilizational clusters too.
Religiosity vs. GDP
Log(Religiosity) vs. GDP
Log(Religiosity) vs. Log(GDP)
(note: the curves of best fit are LOESS)
Razib Khan’s degrees are in biochemistry and biology. He has blogged about genetics since 2002, previously worked in software development, is an Unz Foundation Junior Fellow and lives in the western US. He loves habaneros.
December 6th, 2009 at 9:00 pm
There is also a Pew Study data available from a few years ago (as part of Pew Global Attitudes Project). In fact, here is an Atlantic Monthly article that used the data to plot GDP vs religiosity (overall the shape looks the same as your plot, but I think the measure of religiosity in particular cases is slightly different). This may provide additional data points. As is the case in your plot, US sticks out. Alan Wolfe attributes it to the free religious marketplace in the US.
December 6th, 2009 at 9:22 pm
i have that plot on my other blog i think (from pew). their correlation is higher. they used fewer nations i think, and also only put in the % who said religion was “very important” as the dependent variable. this removes information as very polarized societies will seem more religious than they are (e.g., spain).
December 6th, 2009 at 10:22 pm
The post-war religiosity has to be seen in the following light: the pre-war West was under the sway of communism and naziism: “Scientific” philosophies that led to gut-wrenching suffering.
A turn back to the tried-and-true religions of the past was inevitable, and has seemingly run its course after the Neocon-Islamic wars showed that religion of any kind is the problem.
Another thing to watch is the self-destruction of the Catholic Church in the West under Ratzinger.
December 7th, 2009 at 12:34 am
You could add another dimension to the graphs by putting the names of the countries in different colours, according to what the dominant religion of the country is.
December 7th, 2009 at 6:40 am
It is not just GDP per capita that matters, but also how the wealth is distributed and what social safety networks are available. When people feel secure, they seem less interested in turning to religion, which explains the far reaching secularization of the Scandinavian countries, such as Norway and Sweden. It should be noted that these countries were once very religion-dominated, with official state churches like in England. Religion did not suffer from repression, it just faded away.
December 9th, 2009 at 1:47 pm
Religiosity is measured here in relative terms to respective societies – in a way: proportions of these societies. Hence, the curves ought also to be based on POPULATION NUMBERS, not just dots, corresponding to countries. And the shape would come out completely different.