One of the most common ideas about religion is that it’s somehow intrinsically connected with morality. Showcasing just how widespread this assumption is, a recent study found that people around the world tend to implicitly (or not-so-implicitly) believe that, compared with religious believers, nonbelievers are less moral and more prone to criminality. One of the study’s co-authors, anthropologist Dimitris Xygalatas, summed up the study’s results last week in the web magazine The Conversation. There, he argued that in fact religion isn’t really all that connected with people’s actual morality. But is it really that simple?
Xygalatas argues that, although most people believe that religion is important for morality, the facts are that our religiousness (or lack thereof) has surprisingly little correlation with our actual behavior. He also suggests that belief in punishing, moralizing gods may have helped large societies to evolve, but implies that these “big god” religions are now a sort of vestige, like the human tailbone.
The connection between religion and morality is a lot more complicated than most people intuitively believe, and it’s good that we’re studying it more intensively. But there are several problems with Xygalatas’s think piece that muddy the waters, making it more difficult to clearly assess how and whether religion is or isn’t related to morality. Let’s examine them here.
1. The rhetoric conflates prescriptive with descriptive ideas about morality
There are two ways to talk about morality: prescriptively or descriptively. For example, “human sacrifice is wrong!” is a prescriptive statement. But “the Aztecs believed that human sacrifice was moral, and in fact necessary to support their society” is a descriptive or neutral statement of fact.
Descriptive morality is, by definition, relativistic and value-neutral. It doesn’t make any claims about what’s truly moral in any absolute sense. And of course, cultures and religions are very different in what they believe about right and wrong. Xygalatas highlights this relativism, writing that we
might assume that religious commitment is a sign of virtue, or even that morality cannot exist without religion.…Both of these assumptions, however, are problematic.…For instance, in the 19th century, Mormons considered polygamy a moral imperative, while Catholics saw it as a mortal sin.
The problem here isn’t that Mormons and Catholics had different ideas about the morals of marriage. It’s that Xygalatas uses this difference to counter the claim that religion is intrinsically connected with morality. This rhetorical strategy depends on a subtle slip from a descriptive into a prescriptive way of thinking. Why? Because the fact that there are vast differences in moral norms between different religious groups actually underscores that religions try to influence their adherents’ moral beliefs. And it only undermines the claim that religion is intrinsically connected with morality if you’re comparing things against an absolute standard for moral behavior. If you argue that this relativism means that religion actually doesn’t actually make people moral, then by necessity you’ve got a core assumption about absolute morality that you’re comparing religious moralities against.
So what is the prescriptive morality that Xygalatas is drawing on? Well, he cautions that religions can cause divisions and hatred between groups, citing Martin Luther’s antisemitism as another reason to discount the connection between religion and morality. Yet as despicable as Luther’s views on Judaism were, we’re making a prescriptive claim when we argue that religion’s role in fostering between-group divisions means that it doesn’t have an intrinsic connection with morality. And, importantly, the prescriptive standard we’re drawing on is the one that dominates in late liberal modernity: the post-Enlightenment ethic that people of all groups should be treated equally, that individual liberty and fairness are paramount, and that loyalty to groups, religion, or traditions is generally suspect. In other words…
2. The essay defines morality as liberal morality
Many of the studies that Xygalatas cites have to do with cheating, fairness, and reciprocity. For example, one study found that capuchin monkeys refused offers of reward for tasks if they saw other monkeys receiving better rewards for completing the same task. According to Xygalatas, this finding implies that “even our primate cousins…have innate moral predispositions.”
However, such a conception of “moral predispositions” assumes that rejection of unfairness and enforcement of equitable treatment are the core of morality. But the moral psychologist Jonathan Haidt would disagree. Haidt and his team have found that human psychology has at least five basic moral foundations, or intuitions. Political liberals use only two: maximizing fairness and minimizing interpersonal harm. Conservatives, on the other hand, also consider being loyal to one’s group, respecting authority, and maintaining standards of purity and sanctity to be moral imperatives. Liberalism downplays these three “binding” foundations of loyalty, authority, and sanctity, while conservatives equally value all five dimensions.*
So when Xygalatas implies that religious parochiality, or ingroup exclusivism, disqualifies religion from playing a central role in morality per se, he could easily be interpreted as asserting that liberal morality is the “correct” (prescriptive) morality. After all, religions typically make people more conservative, valuing local standards and bounded traditions more than universal ones. For a liberal, this effect is problematic.
But Haidt and his team call the conservative moral foundations “binding” for a reason: conservative morality produces all-important social cohesion. In a recent interview on the NPR program On Being, Haidt argued that,
if you let liberals run everything, they tend to burn up social capital, but conservatives tend to focus more on building up social structures that actually do allow us to flourish in some ways. You do need order. You do need some restrictions. You do need some boundaries.
How does this association between conservative moral values and religion play out in American society? Well, for one thing, religious adherents tend to belong to more groups and organizations. They’re also more likely to vote than the non-religious. In other words, in the United States, being religious is positively correlated with a number of civic behaviors that benefit the “group” at large. These civic behaviors don’t necessarily translate into more openness to outgroups. In fact, they probably have the opposite effect. But you can’t have a democratic society without an engaged, participatory citizenry. By holding more “conservative” morals, religious adherents help support the institutions that we all depend on.
However, while acknowledging that “a sincere belief in a punishing supernatural watcher was the best…public signal of compliance with social norms” in long-ago, archaic societies, Xygalatas seems to claim that religion no longer has an effect on civic behaviors today:
religious individuals claim to be more altruistic, compassionate, honest, civic and charitable than nonreligious ones.…But when we look at actual behavior, these differences are nowhere to be found.
This is a pointed, empirically verifiable statement, but the data don’t actually support it. This leads me to the critique that…
3. The essay gets some crucial facts wrong about religion’s effects on behaviors
Xygalatas claims that religious adherents don’t actually contribute more to charity than the non-religious, and seems to argue that there’s no measurable difference in civic commitment between believers and non-believers. But a substantial body of evidence contradicts both of these claims.
In the United States, social scientists and nonprofit experts have long accepted that religiously active Americans donate more to charities than the nonreligious. A recent study found that 65% of religiously affiliated Americans report giving to charity, whereas only 56% of the nonaffiliated make charitable donations. The original study itself (PDF available here, signup required) goes into more detail, showing that nearly three-quarters of American charitable donations go to religiously affiliated organizations or congregations, and that more than half of charitable benefactors cite religious commitment as a reason for giving.
This difference extends even to secular charities. For example, 58% who say that religion is “very important” in their lives donate to secular charities, whereas only 42% of those who say religion is “not very or not at all important” give. Similarly, 60% of frequent religious service attendees give to secular causes, compared with only 49% of infrequent attendees. These differences are both statistically significant, even when demographic and socioeconomic factors are controlled.
But Xygalatas argues that religious believers only say they’re more generous. So wouldn’t all those differences disappear if we could look at actual tax receipts and see who’s really giving? Unlikely. A recent study reported in The Chronicle of Philanthropy used tax data, not self-reports, to discover that regions of the U.S. that verifiably gave the highest percentage of income to charity were also the most religious. Utah led the pack by far, donating more than $65 of every $1000 of income. According to Pew, Utah is the 11th most religious state in the country, with 61% of adults saying they pray daily.
In addition, four of the five most generous large cities in the country were in the Sun Belt South, where religiosity is higher than elsewhere. The Washington Post argues that “red states” are more financially generous precisely because they’re more religious: “Of the states that gave the most to charity in 2012, the top 17 all voted for Mitt Romney that year.” If you look at the two maps shown in that Post article, you’ll see that the color-coded maps of the most religious and the most charitable regions of the country are virtually identical. And remember, this data on charitable giving is from tax records. It’s not just self-reports.
What about other kinds of altruistic or civic behaviors? We’ve already seen that voting turnout tends to be higher among religious adherents. Other studies show that religious believers are, in fact, also more likely to volunteer. One study in the prestigious American Sociological Review reported multilevel analyses showing that, across 53 different countries,
frequent churchgoers are more active in volunteer work…Furthermore, religious volunteering has a strong spillover effect, implying that religious citizens also volunteer more for secular organizations.
So, in general, religiosity tends to make people more likely to vote, to participate in civic organizations, to volunteer, and to donate money to charities, both religious and secular. These facts are hard to square with Xygalatas’ argument that supposed differences between believers and nonbelievers are “nowhere to be found” when it comes to real-world civic or prosocial behavior.
But they don’t lead to the conclusion that religious people are “more moral” than the nonreligious, either. Why? Because making qualitative judgments about right and wrong implies a prescriptive understanding of morality.
Yes, religious people belong to more groups, volunteer more, and give more to charity. But the majority of their charitable donations go to their own congregations or denominations, as well as to religiously affiliated charities. Additionally, Robert Putnam and David Campbell, authors of American Grace, found that religious congregations in America tend to be ethnically homogeneous.
So, just like Jonathan Haidt points out, religion makes people more conservative, or “groupish.” Conservatives and the religious tend to care more about their own groups, and they’re more willing to sacrifice for those ingroups than liberals are. But they’re not cosmopolitan universalists, and they can be closed off to outsiders. Religious “morality” comes with both upsides and downsides.
However, without people to join organizations, maintain institutions, and sacrifice for the in-group, there wouldn’t be any ingroups. This prospect might sound wonderful to a secular liberal, until the moment she realized that her in-group was the reason she had roads to drive on, institutions to educate and care for her, and food to eat. Humans can’t actually live without in-groups, no matter how inspiring Enlightenment ideals of universalism might be. This is why we shouldn’t discount religious forms of morality, which brings me to my final point:
4. The essay seems to express an implicitly teleological (directional, purpose-oriented) view of human cultural evolution
Citing the work of psychologist Ara Norenzayan, Xygalatas argues that “big god” religions helped enable the rise of complex cultures like those of South Asia (India) and the Fertile Crescent – the predecessors of our own global, industrial civilization. These religions’ doctrines included gods who could punish transgressors in the afterlife, thus encouraging adherents to cooperate and abide by social norms:
How were people to know whom to trust? Religion provided an answer by introducing beliefs about all-knowing, all-powerful gods who punish moral transgressions. As human societies grew larger, so did the occurrence of such beliefs. And in the absence of efficient secular institutions, the fear of God was crucial for establishing and maintaining social order.
But the essay leads us to think that today, in the 21st century, things are different, concluding that now “we have other ways of policing morality, but this evolutionary heritage is still with us.” In other words, while early civilizations in Babylon or the Indus Valley needed punishing gods to keep people in line, now we have stable governments with rule of law, rendering religion at least partially obsolete. Yet we still irrationally retain an intuitive expectation that religious belief will predict adherence to social norms.
This interpretation is similar to certain assumptions, common among prewar anthropologists, that human culture would progressively evolve away from religion and toward science and reason. One example is the early British anthropologist E.B. Tylor’s concept of “survivals,” or cultural institutions that once served a purpose but now only persist because of inertia. Religion was the prime example of a survival for Tylor, and Xygalatas’s interpretation of our religious intuitions bears some similarities to this characterization.
Yet it’s hard to support the claim that our intuitions about religion and morality are only “survivals.” We know that religious adherents do, in fact, behave differently than the nonreligious in a variety of domains: they join organizations and groups more, they drink less and do fewer drugs, and they tend to be more conscientious than the nonreligious. These findings are pretty robust, and together they point to a solid connection between religiousness and adherence to social conventions and norms – whatever those happen to be.
This doesn’t mean religious people are more “moral” in any absolute sense, because we’re thinking about morality descriptively, not prescriptively. Remember, Aztec religion convinced people of the moral righteousness of human sacrifice, and the Inquisition enjoyed the hearty support of many Catholic faithful. Religion doesn’t make people more liberal, in the sense of respecting human rights or treating people fairly. It makes people more compliant to social norms.
But in a descriptive framework, this means that, yes, religion absolutely does have an intrinsic connection with morality. And that connection is probably not just a vestige of the past, because, in any given society or subculture, religion plays a major role in choosing what counts as morality, and then in incentivizing, cajoling, threatening, and socializing people – usually at least partially successfully – to abide by those contingent standards.
To sum up, morality doesn’t just boil down to treating people fairly or being nice. For the vast majority of people around the world and throughout history, the domain of morality has also included local cultural norms such as “treat elders with respect,” “send your relatives Christmas cards,” or “carry yearly sacrifices to the temple.” From a WEIRD (Western, Educated, Industrialized, Rich, and Democratic) perspective, these injunctions seem like mere conventions, not actual moral precepts. But that’s the point. We’re operating with a thinned-out understanding of morality.
Xygalatas is a stellar scholar who knows the anthropological literature backward and forward. None of what I’ve said here will be new to him. But when talking about a subject as sensitive (and, in today’s context, polarizing) as religion, it’s important to get both the facts and the interpretations as right as possible. And the facts are that religion does have a pretty deep-set connection to morality, particularly the kind of morality that “binds” communities together rather than liberating individuals. We might think we’ve outgrown the need for that sort of morality, but the institutions we depend on, from universities to neighborhood trusts to stable (???) governments, depend substantially on people who follow social norms, believe in duty, and are loyal to organizations and groups. Hopefully, we’ll never have to find out just how much we depend on them. But if we do, we’ll realize that “religious” morality doesn’t just belong in the past after all.
* You can take a version of the Moral Foundations Survey, as well as a bunch of surveys on religion, morals, spirituality, and other things, here.
EDIT: I accidentally published this post before it was ready, so I’ve made some minor edits (added links, etc.) since it went live.