Most secular people I know would probably be alarmed by the idea of a future in which the human race is highly religious. And they would also probably find it highly unlikely. Indeed, I have occasionally heard a sentiment expressed by secular people that goes something like this: the beliefs of most religions are so preposterous that humanity will obviously become less religious as our knowledge and understanding expands.
When I encounter this sentiment, my response to the person expressing it is always the same: How many secular people do you know with big families? Inevitably, the answer is always the same:
I ask this question because the logic of the matter is inescapable: the human race of the future will be composed of the descendants of whoever breeds the most.
That religious people have more children than non-religious people cannot be argued. Worldwide, the difference is staggering. According to a report by the Pew Research Center, people who are unaffiliated with any religion have 1.7 children per woman, well below the replacement fertility rate of 2.1 children per woman. In comparison, Christians and Muslims—the two largest religions, at a combined 54% of the world population—have 2.7 and 3.1 children per woman, respectively.
What can be argued, though, is whether religious people have more children because they are religious, or if there is some other factor which causes people to both be more religious and have more children.
Some other factor like, say, poverty.
Intuitively, poverty as a driver of both fertility and religiosity makes a lot of sense. Worldwide, there is a clear link between how developed a country is and it's fertility rate. And there is also a clear link between how developed a country is and how religious it's people are. Indeed, the largest driver of currently high Christian and Muslim fertility levels is the fact that they are the predominant religions in sub-Saharan Africa, which is one of the least developed regions in the world.
A common assumption, then, is that the rest of the world will also become less religious and enter a similar demographic transition—to much reduced fertility—as it becomes more developed. And this seems like a reasonable assumption; ultimately we're all just human. It is interesting, though, to consider what happens to a population once the process of demographic transition has kicked in.
Because it turns out that the link between religiosity and fertility doesn't go away completely once a society goes through this demographic transition. Religion may be much less important to people's daily lives. And fertility may be much lower—well below replacement levels—but religious people still have more children (p. 16-17). Even when you control for things like wealth and education (p. 119-120).
But that's not all.
It also turns out that religiosity is a heritable trait (p. 49-55). It's only moderately heritable, but the implication is profound nonetheless (it'll just take longer to be realized than if religiosity were strongly heritable). And the implication is this: if religious people have more children, and religiosity is heritable, then, over time, the genes for religiosity should come to dominate the gene pool.
It doesn't matter that an increasing percentage of people in the developed world are unaffiliated with any religion, because they aren't having enough children to replace themselves. Secular genes will live on—since religious and secular people will continue to intermarry and have children together—but they will become increasingly less common as below-replacement fertility continues to siphon them out of the gene pool.
There's something else interesting going on at the same time. The same demographic study which demonstrated that religious women in Europe and North America have more children than their secular counterparts also found something else: the degree of religiosity correlates with the number of offspring a religious woman has (p.16-17). That is, the more religious a woman is, the more children she will have.
Things get especially interesting out at the very end of the religiosity spectrum, among the most intensely religious groups, some of whom have fertility rates greater than 6 children per woman. So far there are only a few religious groups in the developed world with fertility rates this high—the Amish, Mennonites, Hutterites, and Haredi Jews—and they represent a minuscule share of the population, but that can change incredibly fast with a fertility rate more than 3.5x higher than secular people.
For instance, the Amish population in the U.S. and Canada has grown from only about 5,000 individuals in 1920 to over 240,000 in 2010. That's almost 50 times larger in just 90 years. As a point of comparison, in the same time period the U.S. population only grew about 3 times larger and the Canadian population only grew about 4 times larger. Or, to put it another way, the Amish have been growing 12 times faster than the population around them for almost 100 years. And those were 100 years during which the surrounding society saw the advent of automobiles, plane flight, radio, television, computers, the Internet, and mobile phones—essentially everything that makes up modern life.
In Israel, Haredi Jews, who separate themselves from secular society, much like the Amish, are quickly becoming a major portion of the population. As recently as 1990, they comprised only 5% of the Jewish population. Now, they comprise almost 10%. And that number should continue to rise, since they currently make up 29% of the Jewish population under 20 years of age.
So on one hand, we have the below-replacement fertility of secular people siphoning secular genes out of the gene pool. And on the other hand, we have the super-high fertility of ultra-religious sects pumping religious genes into the gene pool. If it continues to be the case that religiosity and fertility are linked—however it is that they are linked—the long term consequences are inescapable: the human race will eventually become more religious (even though it may become less religious for a time while the developing world undergoes demographic transition).
While secular people have been all up in arms about religious fundamentalists not believing in evolution, the religious fundamentalists have been busy, well, winning at evolution.
Religion, fertility and genes: a dual inheritance model, by Robert Rowthorn, is a paper which models what happens to a population's gene pool if there is a high-fertility religious minority (such as the Amish) outbreeding a low-fertility secular majority.
The inevitable rise of Amish machines, by Razib Khan, is an article which examines ways in which the fertility differential between religious and secular people may not hold over the long term.
Genetic and Environmental Influences on the Traditional Moral Values Triad, by Laura B. Koenig & Thomas J. Bouchard Jr., is a chapter from a larger book on the science of religion that examines the heritability of religiosity (as well as other conservative values).
Religion, Religiousness and Fertility in the U.S. and in Europe by Tomas Frejka & Charles Westhoff, is a paper which establishes the fertility differential between religious and secular people in Europe and the United States.
Thanks to Monika Radon-Jones and Guru Khalsa for reading drafts of this essay and providing essential feedback.