May Atheism succeed demographically?

5 May 2013 by Michael Blume, posted in Evolutionary Studies

In the wake of the fruitful explosion of Evolutionary Studies of Religion, Evolutionary Studies of Atheism are making inroads, too. I was glad to be able to blog about Dominic Johnson's respective hypotheses last summer and recently about a ground-breaking paper by Ara Norenzayan and Will M. Gervais.

Although we are beginning to understand the emergence and expansion of non-religious worldviews better than ever, a major question remains: Are human populations lacking any beliefs in superempirical agents doomed to demographic extinction?

It's not only that religious people tend (on average) to have far "more" children than their non-religious peers - it's also about that ominous fact that we do not know about a single non-religious group or population that managed to retain at least the reproductive replacement level for just a century. There have been numerous attempts since Greek and Indian antiquity, not to speak about those late Western cultures and countries - but up to now, there has not been a single demographic success on the communal level.

Reasons for not having children

A new recent survey about the reasons of many Germans not to have kids pointed out three main arguments:

1. 60% of the respondents claimed that they wanted to be "free and independent".

2. 58% pointed out that raising children would be "too expensive".

3. 51% argued that "the professional career is more important than having a family".

Religious vs. Non-Religious Arguments

To me, it seems to be rather obvious that religious traditions are able to counter such arguments, referring to their respective superempirical agents (such as ancestors or gods): Parts of the individual freedom should be submitted to their will, money and professional self-fulfillment should not be held higher than the needs of communities and families. Even if you chose not to have children (for example as a religious Celibate), you should contribute to the survival and cooperation of the religious group! And the superempirical agents are watching and judging your behaviors!

In contrast, I can't see any convincing arguments acceptable to an educated non-believer. Philosophically, Society or Evolution are no absolute values and the simple answer that there might be "too many humans" around would be able to counter and silence any remaining moralistic claims.

Therefore, the empirical irony remains: The more Atheism is flourishing numerically, the more Religion(s) are winning out evolutionarily.

If there's a God, he seems to sport a certain sense of humour...

27 Responses to “May Atheism succeed demographically?”

  1. Robin Reply | Permalink

    Is there any high quality data on conversion of secular to religious and vice versa and any suggestion of any changes in these parameters with time? One could speculate in the past if you were born into a religious family you would be more likely to remain religious, is there any evidence that this has changed, at least in the western world?

  2. Adam DeCoste Reply | Permalink

    I can't tell which is more plausible: that the author has unwittingly recapitulated Social Darwinism or that he's done so knowingly (perhaps as part of a subtle joke?)..

  3. Michael Blume Reply | Permalink


    Yes, although the field of study is rather new, there are some datasets available starting to explore the matter. For example, you might want to read: VID Working Paper 04/08: Vegard Skirbekk, Anne Goujon, and Eric Kaufmann, "Secularism or Catholicism? The Religious Composition of the United States to 2043", Vienna Institute of Demography, with relevant empirical data on religious demography & religious-secular switches.

    @Adam DeCoste

    Both of your assumptions are highly implausible. I understand that empirical facts can be bothersome for religious and atheist worlviews alike; but I will never give in to any attempts to suppress them. You might want to check out the integrity of Susan Blackmore on that:

  4. JayMan Reply | Permalink

    Yes. I have a whole slew of findings that I've discussed on my blog which detail this trend. In the U.S. at least, not only can you correlate fertility disparities with religiosity, but with political orientation (favoring conservatives) and broad personality traits (favoring those less "open to experience" on the Big Five model). In short, a certain mindset is being selected against, that is of the modern liberal secular type. See:

    Liberalism, HBD, Population, and Solutions for the Future | JayMan's Blog

    The Liberal/Conservative Baby Gap: Time Depth | JayMan's Blog

    Expectations and reality: a window into the liberal-conservative baby gap | JayMan's Blog

    Another reminder… | JayMan's Blog

  5. Michael Blume Reply | Permalink


    Thanks for your links and interesting data therein (although I would disagree with some of your political viewpoints, especially concerning "eugenics", which I believe is just another word for more-or-less concealed racism. I believe in freedom and would protest against states or scientists trying to "plan" evolutionary processes among human beings.). Nevertheless, I think you might enjoy my Web-Resources on Religion & Reproduction I collected here:

    Looking forward to reading more about your empirical findings!

  6. Michael Sullivan Reply | Permalink

    I think the children of those religiosos are much more seduceable by modern lifestyles and supporting technologies than in prior generations. No matter how slick your faith-based spin doctor, not only the implications of these technologies but the empirical scientific fact of their existence strain the basic mythology of pre-industrial agrarian belief structures. Add the instantaneous communication of the internet with video games, MTV, spring break and youthful rebelliousness and their kids are leaping on the coming secular bandwagon.

  7. Michael Blume Reply | Permalink

    @Michael Sullivan

    I tend to agree: There are some indicators that new media are driving secularising processes. Partially on the same route, they seem to quicken the fertility drop. Various liberal religious traditions are starting to adapt their "pre-industrial agrarian belief structures" (check for example the new CBS-report about the Universalist Unitarians in the US), whereas strict religious groups seem to raise their fences (for example the Old Order Amish). It seems that religious history will be far from boring in this upcoming century!

  8. prochoice Reply | Permalink

    Your source is a tobacco firm think tank well known for rightwinger views.
    This blaming of women who want to decide for themselves is a very old rhetoric in Germanspeaking countries - Hitler and Goebbels USED it because the Vaticanlobby had paved the way.
    Would you dare to translate German statistics?
    In the Weiman Republic, when people first could have no religion, the Jews lost members - until Hitler defined who was Jewish with the Nuremberg laws.
    This was commonly blamed on their education.
    In the 1960ies the Lutherans began to accept the campaign for schools for the lower classes - and began to loose members in the 100,000s per year.
    When it became clear that the two churchtax religions were no longer the halves, but only 2 smaller thirds of population, the government gave up some delay tricks in statistics, but not the church tax.
    So PLEASE: How die we (my people, the non-church-members) become the biggest group in Germany?
    Not by the birthrate.

  9. prochoice Reply | Permalink

    Sorry correction of Terms:
    Judaism lost members, from 2% something to 1.6%, in German this distinction between religion and persons of Jewish extraction does not exist.
    The wish to get out of religion in the Weimar Republic was blamed on education - for many Jewishborn people, a small number of Lutherans of the better classes, and very few male Catholics who were supposed to go to seminary, but left religion.
    End of 5-year churchtax statistics-delay: 2000 or 2001.

  10. ohwilleke Reply | Permalink

    "we do not know about a single non-religious group or population that managed to retain at least the reproductive replacement level for just a century."

    Non-religious worldviews have been vanishingly rare until the late 19th century, and weren't widespread in whole communities until the Russian Revolution, less than a century ago. This was mostly because science (e.g. Darwin, Maxwell) was not developed enough to provide a viable alternative until then intellectually. People who would be atheists now, would have been deists, transcendentalists, Unitarians, or nominal pagans who were more devoted to philosophies like Epicurianism. Development and education generally, which coincide with non-religion, are probably the stronger drivers of reduced lifetime fertility.

  11. Michael Blume Reply | Permalink


    I have an hard truth to tell you: All human beings - religious and non-religious alike - are going to die, The difference is in their reproduction.

    I am a strong advocate of sexual liberty. Everyone should be free to chose, as they are in Germany nowadays. The reproductive gap is one of the results of this freedom.

    Concerning your references to Judaism: In Germany, USA and Israel, orthodox Jews tend to have far more children than their secular peers.


    You named it already: Non-religious movements never managed to be successful, although they are known since antiquity. And religious groups such as the Mormons (founded 1830) managed to retain higher fertility levels while becoming wealthy and educated. Non-religious movements failed in every instance. Of course, this could change in the future - but as to now, we do not know about a single, promising case.

  12. Dominik Reply | Permalink

    Dear Michael,

    I am not a social scientist but in my humble opinion, your statistical methods are inherently flawed and strongly distorted. Most of all because you're choosing your data in a highly selective manner.
    Let me give you an example:
    There are about 850000 catholic nuns and monks and about 400000 catholic priests (quickly googled, non reliable www research). Their production rate is known to be, let's say close to ( ;) ) zero and their number is declining. Ergo strong religious beliefs are an evolutionary disadvantge.
    If you would evaluate a statistical relevant, randomly selected group of people and ask them for their religious beliefs and the amount of their children, education level, social status and other known to be reproduction relevant information and use the correct statistical tools for evaluation (median, standard deviation etc. etc.), I'd bet one of my kidneys that the effects are completely inside of normal statistical fluctuation.
    Making assumptions of an evolutionary impact based on this data is non-sense and it would be even if the data would be correct (see above), regarding the time frames you're talking about. Could you sum up the evolutionary impact of smoking? Of driving cars? Of using mobile phones?

    This is a textbook example for "cum hoc ergo propter hoc".


    • BarryG Reply | Permalink

      Up voted for knowing about selection bias. Please breed and raise your kids the same way.

  13. Michael Blume Reply | Permalink

    Dear Dominik,

    it is always stunning to experience people trying to defend their worldviews by distorting and belittling scientific positions.

    The text presented here is a blogpost presentation, not the entirety of my scientific work including a doctor and various teaching positions. I linked to my homepage where you can find tons of data including various censusses, studies from other scientists (with the same results overall) and case studies (such as, for example, the Amish) for free. And although place was short, I even included a sentence about religious Celibates contributing to the overall survival and reproduction performances of their respective groups - nothing really mysterious, it's widely known as the "helpers at the nest"-strategy among animals.

    You are perfectly free either to enjoy serious science or to use the very tricks applied by people trying to "disprove" evolution by attacking straw men. It's your choice,

  14. Zachary Stansfield Reply | Permalink

    I've always assumed that atheism (or strict forms of religious belief) generally require a sort of elite position in the world (e.g. sufficient wealth to ponder such issues without great loss of self-esteem). Given that there is a strong negative correlation between wealth and offspring production (e.g. when comparing across societies), shouldn't this largely explain the effect? Namely, that the people who are capable of controlling and reducing their number of offspring are also those most able to question religious sentiments? This statement of course assumes some potential lag time, as factors like cultural heritage might promote continued religiosity and/or high levels of offspring production even after wealth has been acquired, such as following immigration to a wealthy society.

    (Apologies if I missed this point somewhere else in the comments.)

  15. Michael Blume Reply | Permalink

    @Zachary Stansfield

    Yes, Atheism is flourishing among those raised in rather secure lifes, which often (but not always) combines with wealth and education.

    At this point, religion and family turn from a more-or-less fixed necessity to a matter of choice(s). And these choices turn out to be related. Thus, the reproductive gap among the higher educated religious and non-religious turns out to be bigger than among those less educated. Modern education looks like a demographic & evolutionary bottleneck, with only the (very) religious squeezing through with more than two children.

    Here's a graph included with data from an Australian census including birth rates of religious & non-religious of various education (p. 162):

  16. Doug B Reply | Permalink

    An interesting blog post, but perhaps a little off the point. It might be good to remember that atheism is still a "-theism", which is to say a religion. The particular sect of atheism you seemed to be concerned about (secular liberalism) is one that pretty much espouses a generally Protestant ethical system, so perhaps it might be more accurate to describe it as an unusual Protestant sect (at least in Europe and North America).

    "Atheists are more sure that there is no God than the Pope is that there is one."

    It is true that you can model a religion as any population model, with P(t+1) = P(t) + births (people born into the faith or converted) - deaths (people who pass away or leave the faith), and any population model that doesn't reach replacement (deaths > births) will end up with P(t) = 0 for all t > time of extinction. (See Shakers). A guess the question is what are the coefficients for births, conversion, deaths and disillusionment and how do you estimate them?

    The lack of actual births mean the religion has to make up for that with conversion, and be relatively agressive about it if it is to survive. Something to think about.

  17. Len Reply | Permalink

    I understand agnosticism but not atheism. There is a universe with logical rules that we call "laws of nature." An atheist can shout down and shut up his religious critics but he can never answer why there is anything at all rather than nothing. A community of such atheists might live out their lives. But their children and grandchildren will grow impatient with their parent's dogma and begin to wonder about those things all over again. Atheism has not rational grounding to sustain itself over time. I don't think I'm the first person to observe this.

  18. Nameless Reply | Permalink

    One should not discount the power of the basic law of natural selection. Even if we don't see an atheist subgroup with fertility above replacement, one will probably turn up sooner or later, and, if it has a reproductive advantage (e.g. because atheists are more intelligent) and if natural selection somehow gets turned back on in the Western society (it's been pretty much nonexistent since we eliminated all major causes of child mortality in the early 20th century), that group will eventually dominate the society.

    There is an interesting tangentially related fact that I came across recently. You all know about the current drive towards acceptance of gays and same-sex marriage. There is much wringing of hands (in some circles) about how that's going to dig an even deeper demographic grave for the Western society.
    It is less known (but still known) that homosexuality was perfectly acceptable in the Ancient Rome. What most people don't know is that, for most of the history of the Empire, Romans did not have the concept of exclusive same-sex marriage. They had the concept of marriage, or "matrimony" (from the word "mother"), which, by definition, was procreative and therefore opposite-sex, and men could enter into a wide variety of extramarital relationships with other men.

    Roman example shows us that, in a society that accepts gays in general, there may be one subset - one where gays choose to lead complicated social lives involving traditional marriages and openly acknowledged extramarital partners - that has selective advantage over others, and eventually it dominates the entire culture.

    Likewise, while atheism in general seems destined for demographic death, there is a slice of atheists with 2 or 3 children, and their atheist children will carry some of the social values of their parents and will probably have more children then atheists of today, etc. etc. The process is underway.

  19. Happy Atheist Unitarian Reply | Permalink

    Very funny, if unserious, article. All we really know in the U.S. is that according to the Census atheists are the fastest-growing demographic. The so-called "long-term" is sheer speculation. Anyway, the author willingly overlooks that in the most technologically advanced societies, as well as huge societies which are getting there, like China, atheism is well established ad likely to remain so for the foreseeable future. Only 2%-6% of all western Europeans report they are "religious" (whether they officially believe in god or not is just a technicality, in my opinion, since they tend not do a blessed thing about it).

    So far from going extinct due to a specious misreading of history (which was all written by statist believers), atheism (especially in the guise of its socially accepted brother, agnosticism) seems to have the brightest future, in my humble opinion, thank god!

  20. David Marshall Reply | Permalink

    An interesting perspective that emphasizes one of many variables. Chinese young people tend to buy into atheism in college, according to limited research I've done so far, but then seem to lose their non-religion in large numbers later on. Children of atheists also tend to go astray at high rates. The growth in secularism in Europe and America may continue, or reverse, as it has in the past.

    What atheist societies have reproduced? Maybe the neo-Confucian literati of post-Song China. (Not always strictly atheist, but close enough.) All in all, though, I think the man is onto something. But one can only guess what the future will bring.

    • BarryG Reply | Permalink

      The future will bring robots, genetic engineering and post humanism along with massive war and refugee crisis from human induced climate change but no messiah, no second coming, no God to save us but ourselves.

      • David Marshall Reply | Permalink

        The future will bring lots of predictions about the future that don't come to pass, and lots of stuff no one expected.

  21. BarryG Reply | Permalink

    They breed but they bleed.

    That is, they have more children, most of whom get sick of the religion, water it down, leave it.

  22. Jack Decker Reply | Permalink

    And then there has been the mass extinction of religion in the countries of USSR and China as well as other communist countries. The absolute decline of religion in the UK and ALL other industrialized countries. Skeptics might not be reproducing as fast as religious people but we're winning over FAR more people after birth that more than makes up for the lower reproductive rate.

  23. Michael Blume Reply | Permalink

    @Doug B.

    If we define religiosity as beliefs in superempirical agents, atheism does not qualify as a religion. And it doesn't show the community- and family-building potentials we found among many religious movements.


    I agree. I don't know of any human worldview who managed to have a thoroughly rational foundation. We are evolved animals after all, not machines.

  24. Michael Blume Reply | Permalink


    Thanks for your secular prophecy! And the nameless prophet proclaims: "Even if we don't see an atheist subgroup with fertility above replacement, one will probably turn up sooner or later" - Just allow me to ask humbly: How do you know?

    "Likewise, while atheism in general seems destined for demographic death, there is a slice of atheists with 2 or 3 children, and their atheist children will carry some of the social values of their parents and will probably have more children then atheists of today." - As to now, intergenerational traditions are relying on religious beliefs to survive and to spread. I would love to see a convincing example of such an atheistic substitute, but I don't know one yet. And as a good scientist, I want to see data before I start believing! ;-)

    @Happy Atheist Unitarian

    Actually, we are exploring the very factors leading to increasing levels of secularization:

    But then, the former Soviet Union and China (both brought up by @Jack Decker) are perfect (if sad) examples of demographically declinig societies with quickly (re-)growing religious minorities...

Leave a Reply

× 7 = forty nine