On October 4 2019, editors at The Conversation became aware the Current Biology paper mentioned in this article had been retracted. The author of this article was not involved in the research and his argument goes further than just discussing the now retracted study.
It is certainly true that humans don’t need religion to be moral. But whether religion generally reduces or impedes kindness is less obvious, as the link between religion and morality is a notoriously difficult one to untangle. Although many researchers have tried to provide answers on empirical grounds, these answers often seem to fall predictably along partisan lines.
A new study of more than 1,100 children in six different countries has found that children growing up in non-religious households are better at sharing than those from religious homes. The researchers behind the study, published in Current Biology, conclude that teaching children about morality in some sort of secular setting will therefore increase human kindness.
The experiment
This new research – which investigated the attitudes of children from the US, Canada, Jordan, Turkey, South Africa, and China – was based on a game in which the children were asked to share stickers with an anonymous person from the same school.
Parents in religious households generally reported that their children had more empathy and sensitivity for justice than did non-religious parents involved in the study. However, the children’s actual attitudes and behaviours seem to contradict their parents’ expectations. Non-religious children were in fact more generous than their devout counterparts: they were more willing to give stickers away to their peers. The experiment also showed that non-religious children were also less likely to judge interpersonal harm – for example, pushing and bumping – as “mean”, and perpetrators of such harm as deserving of punishment.
It is in many ways an impressive study: over a thousand children aged five to 12 across six countries. Unfortunately, it only compared children from households that identified as Christian, Muslim or non-religious. The study suggested that Christians and Muslims differed on how punitive they were: Muslims were more punitive than both Christians and the non-religious, who were quite similar. But it would have been nice to know where, for example, Chinese Buddhists or Taoists, fitted in.
Contradictory findings
So why can we not (yet) confidently proclaim that religion makes the world a morally worse place? Surely a study comparing religious and non-religious people on their moral behaviour is all that we need. As it turns out, a lot of people have tried to do just that, and the evidence on the matter is ambiguous at best.
On one hand, we know that religious people often report that they do charitable giving and volunteering; on the other, a recent study looking at participants’ behaviour at five random points during the day for three days found no relationship between religiosity and everyday moral behaviour.
This disparity has led some social scientists to conclude that religious people tend to exaggerate their charitable activities – implying a connection between religion and dishonesty. That said, reminders of religious concepts (known as “priming”) have been shown to decrease cheating behaviour, though not all kinds of religious believers are less likely to cheat than non-religious peers.
The notion that there are different kinds of religious people is an important one. There are so many different religions and approaches to religion that may well matter in answering questions about religion’s relationship with morality.
A recent study of 67 countries found that different religious beliefs lead to different crime rates: belief in hell was associated with lower crime rates, whereas belief in heaven was associated with higher crime rates. Similarly, research has even found that subliminally priming people with the words “religion” and “God” before making a moral judgement in an experiment can have different effects on generosity – while the word “religion” can increase generosity and trust towards members within a group, “God” can increase such behaviours toward members outside a group. To make matters messier still, religious priming has also been shown to both increase and decrease prejudice against racial and sexual minorities.
This is just some of the recent research in this area, yet we can already see there is no single answer to the question of exactly how religion affects morality. This is in part because “religion” is a very broad concept. But so is “morality”: we have already covered at least generosity, honesty, trust and prejudice in this article.
So, how should we interpret the new study? The authors may be right to say that their findings challenge the notion that religion is vital for moral development, but this is hardly a new finding – let’s just hope that this is the last nail in that dusty coffin. But on the matter of whether secularisation is good for “human kindness”, well, that depends on what you mean by “human kindness”. We probably all agree that sharing stickers is a morally good thing, but the judgements of interpersonal harm seem more ambivalent.
Is it really kinder to let perpetrators of interpersonal harm go unpunished (or punished lightly)? In fact, even the matter of sharing stickers is open to interpretation. Perhaps religious children are just more likely to share with specific people, such as members of their own group (or people in need or people who otherwise deserve it). Seen in this light, their behaviour could be described as loyal or fair, or some other trait that most of us would consider virtuous. These questions about the relative goodness of different values are all contentious – we shouldn’t pretend they are not.
None of this is to say that research on religion and morality is fruitless. On the contrary. This new study challenges some strongly held stereotypes about religion and morality, and that can only be a good thing. It has also detected some interesting differences between two religions that are worth exploring even further.
Going forward, what is needed is an even more nuanced approach. The question is neither “is religion necessary for morality?”, nor “is religion evil?”. Rather, it is – the admittedly cumbersome – “what aspects of religion are related to which aspects of morality, under what conditions?”. Judging by the current state of the evidence, it looks like psychologists and other social scientists still have some work to do.