"With surveys showing a strong majority from conservative to liberal believing that religion is beneficial for society and for individuals, many Americans agree that their church-going nation is an exceptional, God blessed "shining city on the hill" that stands as an impressive example for an increasingly sceptical world. But in the other developed democracies religiosity continues to decline precipitously and avowed atheists often win high office, even as clergies warn about adverse societal consequences if a revival of creator belief does not occur."
- Gregory Paul

People spend a large amount of time and energy debating if faith is a good or bad thing for people in general to have. We have the arguments from belief and arguments from logic. All things being equal, well-formed logic beats belief. But both must bow before reality: logic trumps belief, but evidence trumps logic.

You can't objectively measure faith, but you can measure religion: you can count bums on seats on Sabbath days, and count how many people claim affiliation to religious groups. You can ask them to choose between evolution and a creator. And actually there is a lot of data about this around.

So a scientist called Gregory Paul asked in 2005: All other things being equal, are societies with more religion better off? By collating data from western democracies we can ask: what, if any is the correlation between levels of religiousness and social ills in these developed societies?

The results were published in The Journal of Religion and Society, in September 2005. And his verdict is that godly societies are worse off.

"In general, higher rates of belief in and worship of a creator correlate with higher rates of homicide, juvenile and adult mortality, STD infection rates, teen pregnancy, and abortion in the prosperous democracies. ... The United States is almost always the most dysfunctional of the developed democracies, sometimes spectacularly so, almost always scores poorly. "

"Higher rates of non-theism and acceptance of human evolution usually correlate with lower rates of dysfunction, and the least theistic nations are usually the least dysfunctional. ... In some cases the highly religious U.S. is an outlier in terms of societal dysfunction from less theistic but otherwise comparable secular developed democracies. In other cases, the correlations are strongly graded, sometimes outstandingly so."

"The United States is the only prosperous democracy where religion is really popular and we're the only nation among prosperous democracies to have really high murder rates."
- Gregory Paul


Of course the godly will try to pick holes in his arguments. Although The Times calls him "a social scientist", his primary qualification is as a paleontologist. He knows lots about theropod dinosaurs. He may know about evolution, and is clearly fed up with "intelligent design", but what does he know about social science and human population statistics? There have been many cases of scientists getting things spectacularly wrong outside of their area of expertise.

Correlation is not causation for one, so trying to suppress religion is not the answer. I don't think he's trying to claim that religion causes problems; merely asserting that the belief that "more God is better" is demonstrably false.

"The widely held fear that a Godless citizenry must experience societal disaster is therefore refuted."
- Gregory Paul

State efforts to actively stamp out religion have generally occurred in the failed communist eastern block, which were not correlated with positive effects on society. I have a suspicion that this happened not because the state wanted the best for the citizens, but because having a religion already got in the way of the loyalty and blind devotion to the state that was desired.

Evidence is as fallible as logic too - there seems to be good evidence, for instance, that the sun revolves around the earth. But on more detailed inspection, it's not so. In the social sciences it is particularly difficult to draw firm conclusions - one cannot conduct isolated experiments, and so must observe societies. Since they generally differ in so many ways at once, it is harder to pin down the effects of one variable.

Eighteen countries are included: Australia, Canada, Denmark, Great Britain, France, Germany, Holland, Ireland, Japan, Switzerland, Norway, Portugal, Austria, Spain, Italy, United States, Sweden and New Zealand. One of the criticism is that there is no objective criteria why these "developed and democratic" countries are included and others omitted. Where exactly you draw the line can make a big difference with such a small number of samples. As can which year you pick the data from, which is not even stated beyond "from the 1990s, most from the middle and latter half of the decade, or the early 2000s". Did he carefully pick the data to find the results, or just use what was to hand?

His statistical technique has been criticised as overly simplistic, with Scott Gilbreath calling the lack of regression analysis and multivariate analyses "inexcusable".

"This is not an attempt to present a definitive study that establishes cause and effect between religiosity, secularism, and societal health. Rather, the goal is to spark future research and debate on the issue."
- Gregory Paul

In other words: if you disagree, go find your own evidence. This is progress - inspecting reality is better than crafting rhetoric or telling us what you believe as an article of faith.

Criticisms that I feel are poorly founded

Correlation is not causation: True and well-known to all of us, including the author of the study. Yet if factor X correlates with factor Y, it suggest that it can be possible that one causes the other, or vice versa, and certainly casts a lot of doubt on people's claims that factor X causes the opposite of Y, which is precisely is commonly claimed for religion: that it makes societies and people better. The burden of proof has been shifted.

Religion can't be bad, some religious person or religious community do very good things: This aims to be a counterargument but is in fact not relevant. It could be the case that despite the actions of small numbers of devout, the general population is worse off with religion than without. It could be otherwise, but the evidence decides.

Religion can't correlate with factor X, most religion is against factor X: Unintended consequences are everywhere, which is why evidence is a good thing.

The correlation between religion and factor X is illusionary, in fact factor X correlates well with factor Y: Another argument that is not relevant. Any correlation between factor X and factor Y does not in itself remove the correlation between religion and factor X - all three can go together. Maybe factor X causes factor Y. Maybe religion causes both factor X and factor Y. Or factor X. causes factor Y and increases the odds of being religious. Or any other combination. The evidence does not say which, it only shows either a correlation, or no correlation.


Gregory S Paul: "Cross-National Correlations of Quantifiable Societal Health with Popular Religiosity and Secularism in the Prosperous Democracies: A First Look"

The Times of London article: Societies worse off 'when they have God on their side' - http://www.timesonline.co.uk/article/0,,2-1798944,00.html

transcript of an Australian radio broadcast interview: Study says belief in God may contribute to society's dysfunctions - http://www.abc.net.au/worldtoday/content/2005/s1470370.htm

Wikipedia biography of Gregory Paul: http://en.wikipedia.org/wiki/Gregory_Paul

Regis Nicoll criticies the findings: "Warning: Religion is Hazardous to Your Health" - http://www.crosswalk.com/news/weblogs/rnicoll/?adate=4/27/2006#1393761

Scott Gilbreath criticises: From our bulging How not to do statistics file: http://magicstatistics.blogspot.com/2005/09/from-our-bulging-how-not-to-do.html

George Gallup criticises: http://www.virtueonline.org/portal/modules/news/article.php?storyid=3094

There are two big problems with Paul's research inasmuch as it can be said to provide evidence for the effects of religiosity on a society: one, he doesn't measure the actual effects of religion, and two, he gives no evidence that the two sets of data he examines are at all connected.

This could be a case of a scientist playing too far away from his home pitch. Whereas in paleontology one normally looks at concrete finds or facts as isolated instances, in modern sociology it is almost universally dynamics over time that are considered the basic research unit. So to present two sets of statistics - numbers of religious people and number of, say, teen pregnancies - is by no means to prove that there is any causal link between them. Indeed, SF says as much in his own writeup. More importantly, however, it is entirely pointless to consider these number out of historical context, unless you can show that the levels of both church attendance and teen pregnancies has remained at a constant level in all post-industrial societies, which is patently absurd.

The US has not always been as religious as it is today. Although it did always have a higher rate of evangelism than many European countries, it was by no means a markedly more religious society than Italy, say, or Portugal. In the last 20 years a specific stream - you could call it a sect I suppose - of Christianity has steadily gained in popularity, contributing to a dramatic rise in church attendance numbers. At some point in the last 20 years, church attendance in Italy and the US would have been about the same (never mind for the moment the fact that the size and diversity of the US demographic landscape would have made a nonsense of that anyway); at this point, what was the ratio of other social ills per capita in these two countries?

It is equally important to examine the trending of the two phenomena alongside each other. What were the numbers like before the point at which the two countries started going in opposite directions (churchgoing up in the US, down in Italy)? Which of the two had the higher rate of teenage pregnancy in, say, the fifties, and how have they changed since? Was, for example, the rise in teen pregnancies in the US proportionate to the rise in church attendance? And was the opposite effect observed in Italy? We could get a bit of a shock if we find out that in fact religious affiliation started rising at the point when the numbers were the bleakest, because that would tend to indicate that religiosity is a reaction, or a corrective action, of a nation embattled by social ills, as opposed to being a cause at all.

Staying on the subject of teenage pregnancy, it is in fact on the wane in the US(1). The numbers are also down in the UK(2), as is church attendance(3); but the UK still has the highest teenage pregnancy rate in western Europe. What does that last fact mean? Does it mean anything at all? On the face of it it is significant that an increase in secularism has not wrought the same dramatic change in the numbers in the UK as it did in Scandinavian countries, for example - but surely in order to to examine that we have to set aside religion as the only factor and interest ourselves in other social dynamics and economic factors affecting the different countries.

Meanwhile Danish scientists have recently claimed that the decline in teenage pregnancies is more due to a decline in the fertility of young males(1) than any education or church based initiatives. If this is true to a statistically significant extent, then all the data we collect is effectively meaningless. Even if it isn't, we still haven't looked at contributing factors such as the availability of cheap and effective contraception, living standards and levels of poverty and education, the general permissiveness of social mores in non-sex related matters etc.

I've kept with teen pregnancy because it's an easy example to find statistics about. But it is not unique in its complexity. Briefly touching on violent crime, for example, we can see that despite the fact that Jesus would almost certainly be anti-gun, the rate of gun ownership in the religious US is very high, as is the rate of violent crime. In order to get any insight into possible connections between the two, we would need to properly understand the ratios of churchgoing folk owning guns, as well as whether they justify this ownership on religious grounds or on grounds of compatibility with generally more conservative opinions, which might in fact be at the basis of both behaviours.

I am not advancing any argument in favour of religion as a force for good, although it's not impossible to make one (the role of Christian movements in the abolition of slavery is well documented, though more recent examples are harder to come by). Certainly religiosity has not done the US any favours in terms of its foreign policy. But religion is so intertwined with the life and mores of a population that it would be difficult if not impossible to draw easy conclusions about its isolated impact on any trend across an entire society. We may as well say that as Japan is the most ethnically homogenous post industrialised society, and has the highest rate of education and technological market penetration, then ethnic diversity is bad for the education system.

(1) http://www.slate.com/id/2140985/?nav=fo

Log in or register to write something here or to contact authors.