Morality and religion: are they genetic adaptations?

February 14, 2010 • 7:58 am

In a new paper, “The origins of religion: evolved adaptation or byproduct?”,  Ilkka Pyysiäinen from the University of Helsinki and Marc Hauser from Harvard discuss the evolution of religion and morality (The paper was highlighted by Philip Ball in Nature News and by P.Z. at Pharyngula.)   The paper is divided roughly into two parts:

1. A discussion about whether religion is a direct adaptation, that is, whether there are specific genes that favor belief in religion, God, the afterlife, the supernatural, and so on—or whether religion is a “byproduct,” that is, that religious beliefs grow naturally out of other evolved features of the human mind.  Pyysiäinen and Hauser favor the latter.  They seem to agree with Pascal Boyer (see his Religion Explained), that faith is an outgrowth of the natural human tendency to attribute intentionality to objects, forces of nature, and the like.

I like the “byproduct” hypothesis, if for no other reason than it’s almost self-evidently true. Surely every human behavior is in some sense a byproduct of genes that evolved for other reasons.  And if religion, like music-making, jokes, and pornography, is an outgrowth of genes that have evolved for other reasons, then we need not make up adaptive stories favoring a “religion module” in the brain.  That imposes some restraint against the injudicious production of untestable stories.

That said, I don’t see decisive evidence one way or the other.  Pascal Boyer does make an excellent case for the “byproduct” hypothesis, but it is, after all, just an argument that sounds pretty good, without conclusive data.  I’m not sure exactly what data would support one hypothesis over the other, and in the end, if you can’t settle the issue the question becomes scientifically uninteresting.

2.  A discussion about whether morality grows out of religion, whether it’s culturally inherited, but not through religious teaching, or whether it’s innate.  Here I think we can approach an answer, at least in principle.  If morality tends to be similar among people of different faiths, despite their different moral teachings (yes, they all share the golden rule, but they differ in many other teachings), or if morality is similar in atheists and the faithful, or if children brought up without religion but exposed to different cultural teachings tend to converge on the same morality, then we have some evidence of an innate “moral grammar.”  Too, as philosophers have pointed out since the time of Plato, even the faithful admit of a morality antecedent to religion: if God told you to do something immoral, you wouldn’t automatically consider it moral.

Pyysiänen and Hauser’s work, described in their paper, supports the idea that morality does not derive from religion, but may in fact be innate.  This is based on the results of “morality tests” given to people of diverse religious beliefs and cultural backgrounds. (I urge you to take one of these tests, both to provide more data and to see exactly how they come to this conclusion. You can access the test here.) Regardless of background, faith, or upbringing, people tend to answer the questions in the same way.

Now there are two problems with taking this unanimity as evidence for an evolved morality.  First, people could simply be exposed to religious morality in their societies, and their internalizing of that morality could lead to consistent answers to Hauser’s questions.  Second, uniform morality could simply be a cultural adaptation without genetic roots: people have learned over time what sort of morality leads to a harmonious society, and we’re simply inculcated from birth with that secular morality.  Philip Ball raises this point in his Nature commentary:

It’s debatable, however, whether these moral tests are probing religion or culture as a moral-forming agency, because non-believers in a predominantly religious culture are likely to acquire the moral predispositions of the majority. Western culture, say, has long been shaped by Christian morality.

This is a reasonable objection, but I think it’s belied by the data.  I urge you to take the morality test, for if you do you’ll see that the kinds of moral dilemmas embodied therein have nothing to do with the kind of morality imparted by faith, nor do they really comport with the kind of moral teachings that we get from our parents, friends, and schools.  When you take this test, you might feel, as I did, that your answers are coming out of some intangible but innate wellspring of moral views. (That, of course, is not good evidence for a genetic adaptation!)

I wrote to Pyysiäinen and Hauser about Ball’s caveat, and Ilkka answered, speaking for the both of them (quote given with permission):

As far as we are dealing with people’s intuitive judgments, it is impossible to attribute these to learned religious views. People just do not respond in accordance with religious doctrines when they have no clue about how learned religious doctrines should be applied. I think Marc’s research shows that religious doctrines can only have an effect on especially salient topics such as abortion. When religious commitments differ and people yet produce quite similar judgments, this shows that religious commitments do not have causal power with regard to moral judgment. If religious people make moral judgments similar to those of nonreligious people, then there is no reason to suppose that religion is the driving force. This means that *explicit* religious commitment is not relevant. But, as you suggest, it might still be that religion affects explicitly nonreligious people’s judgments in an implicit way. However, this, then, means that explicit religious commitment is not the crucial factor.

The next decade, I think, will see an explosion of research about whether humans have an innate “moral grammar”.   To me, it’s the most interesting part of evolutionary psychology.

But the faitheists must weigh in as well. Over at Thoughts from Kansas, Josh Rosenau tries to make a silk purse out of what, for the faithful, is a pig’s ear:

In other words, morality is independent of religion or religiosity. Religion may be a means to pass down certain cultural norms about moral behavior, but there are plenty of other ways to do the same thing. As one theologian of my acquaintance put it, there are many paths to the top of the mountain.

Theists can take comfort in that notion, secure in the thought that their god(s) shaped the world so that everyone was led to moral behavior. Atheists can take this finding as further proof against the refrain of certain religious people that erosion of religious faith will result in erosion of morality. And the rest of us can take comfort in the notion that we’re behaving well, and the reasons why we behave well aren’t that important.

Well, if there are many paths to the top of the mountain, why are the faithful defaming, fighting, and even killing each other on the way?   Let’s face it: faitheists and liberal theologians can preach until they’re blue in the face that morality wasn’t dictated by God, but was—as Robert Wright suggests—simply a happy, inevitable and foreseen result of His creation.  But that’s not what religious scriptures say, nor what a huge number of the faithful really think. It always amuses me when accommodationists, especially the atheistic ones, tell religious people what they’re supposed to believe, or where they’re supposed to find comfort.


Pyysiäinen, I. and Hauser, M. 2010.  The origins of religion: evolved adaptation or by-product? Trends in Cognitive Science, online.

18 thoughts on “Morality and religion: are they genetic adaptations?

    1. Having just seen Avatar again (@ imax), it is obvious that morality in other species (e.g.: na’vi) comes from the bio-network but for humans comes from the need for Unobtainium!

  1. Pascal Boyer does make an excellent case for the “byproduct” hypothesis, but it is, after all, just an argument that sounds pretty good, without conclusive data. I’m not sure exactly what data would support one hypothesis over the other, and in the end, if you can’t settle the issue the question becomes scientifically uninteresting.

    I’m just a layman, and I have certainly not read Boyer’s text (yet). But I would think one would go about it something like this:

    Essentially you would like to know if religiosity depends on genes as causality (“religious module”) or correlation (“byproduct”). That means that you need to vary parameters. If relevant genes and religions depends the same on parameters it is causation. If not, it is correlation.

    If there was a one-one correspondence between genes and traits, all one had to do to falsify the hypothesis of “byproduct” would be to find one such “religion module” gene.

    Alas, that is not the case. Then you would have to vary parameters in other ways to compare genes and traits. That would involve looking over species or time.

    If there are genes that species share where traits important for religion can be traced, such as attributing intentionality, but at the same time attributing intentionality in say elephants (which can sense “self” in mirrors) doesn’t convey religious behavior in some sense (doesn’t seem so), supposedly they falsify the “module” hypothesis.

    If there are genes where traits important for religion can be traced, but at the same time weren’t swept to fixation during historical time (see for example John Hawks et al) when religion became a human trait, supposedly they falsify the “module” hypothesis as well.

    1. As a note, right or wrong, I believe attributing intentionality is a trait in other species. (Because they wouldn’t pass the mirror test otherwise, I think.) So I believe the “byproduct” hypothesis passes tests, while I don’t see any the “module” hypothesis has passed.

  2. Emergence, expediency, and reinforcement.

    Cooperation cannot exist without ethics and ethics cannot exist without moral sensitivity that develops and adjusts over time according to the needs of society.

    1. Cooperation cannot exist without ethics…

      Slime moulds show sophisticated co-operation.
      Do you contend therefore that slime moulds possess ‘ethics’?

  3. A few comments:

    First, Jerry, I’m intrigued that that you think evolutionary psychology might be a fruitful line of investigation. As I recall, in WEIT, you dissed much of it (following SJ Gould) as “just so” stories. Change of heart?

    Second, let’s assume, for the moment, we can get the “morality as a by-product of religion” hypothesis out of the way. This done, I still think it’s not helpful to pose the question as one of culture vs. innate faculties. This is just Galton’s old nature vs. nurture fallacy again. It will probably turn out that morality (like more prosaic human dispositions such as tanning) will turn out to be a product of both innate dispositions and cultural influences.

    Third, I agree that theists and non-theists respond similarly on trolley car type questions. But what if we ask different sorts of questions? Say, about abortion, women’s rights, suicide bombing, etc? Wouldn’t those questions reveal strong cultural and religious differences?

    Fourth, whatever contribution to our moral views our genes play will also be embedded with and indistinguishable from other putatively genetic dispositions, for example, the urge to take revenge, for men to dominate women, towards xenophobia, etc. So merely describing our genetic moral dispositions won’t tell us what dispositions we ought to have, since those dispositions were crafted in our ancestral environment unlike our own and, withal, were selected for their contribution to individual or inclusive fitness. not directly for their role in mantaining a harmonious society.

    So, while I find EP vastly interesting, it tells me much more about the sort of person I am (sometimes a depressing tale) and much less about the person I ought to be.

    1. In many religious traditions being in line with the wishes of god is the very definition of morality.
      For example, muslims will tell you that drinking alcohol is immoral, as is failure to fast during the Ramadan.
      Which makes me wonder what to make of unhesitating proclamations on the part of people who declare homosexuality or stem cell research “immoral”. Do they have at all a stronger standing than those who tell you if you share a beer with friends or eat during the day at a certain time of year you have committed an immoral act?

  4. I have always claimed to believe (for the sake of argument only) that there are those who must believe in absolute truth, absolute authority (especially gods, etc.), and become theists, creationists, climate-change deniers,anti-abortionists, political conservatives, conspiratorial thinkers, etc. And there are those (like myself) who find it very difficult to believe in anything much. We trust things (we believe) are real physical data — evolution, climate change, human rights, etc.
    This is, of course, a false dichotomy and hence should be on Fox news.

  5. “Western culture, say, has long been shaped by Christian morality.”

    Fortunately we’re getting away from that. The last Italian Inquisition ended over 100 years ago. I wonder though, what evidence do the authors have to even make such a statement? Where is this list of peculiarly jesus cult bits of morality which are not present in non-jesus-cult religions and yet so dominant in “western culture” (whatever the hell that means).

    Personally, I believe that anyone looking for a god gene or claiming that religion is a natural and inevitable outcome is simply wasting people’s time. It is trivial to formulate a more sensible hypothesis which does not require religion to be inevitable. As pointed out, this is not even a question of scientific interest. We may as well argue over the color of Russell’s Cosmic Teapot.

    1. Is religion is “inevitable”? If we follow the linguistic analogy, maybe it is. While human brain structure doesn’t specificly require Chinese or English, it does seem to be universally true that humans will use language — even in highly adaptive non-verbal ways (like tapping one’s fingers on a plastic keyboard to create a blog comment). One can imagine a post-Christian, post-Islam, etc world, where people in general live well-adapted to their inherent natures without having any *current* religion — but would it truly be religion-free, or, if we could see it from here, would we perceive their culture as simply having a different way of expressing these same religious attributes (err, byproducts)? Does giving money, say, to RSF simply alter the inherent mechanism for religious-style tithing and sacrifice to a different less-overt religion? Just asking.

      The teapot is blue, btw

  6. Morality was developed outside of religion, probably even before it, or at least in the very early stages of it. Morality is a system that regulates efficient life in society and the survival of the group. And it was developed, as anything else, by the ol’ good evolutionary method: trial and error.

    As any evolutionary process, this one never ends. And we’re yet to see if the morality was an error and what will happen next. So far, religion’s messing with the morality had some nasty consequences.

    Sure, religion helped spreading the system, but it’s not the only part of the culture that did (and does) the job.

  7. How can something evolved possess the desire to not acknowledge it’s origins? An evolved creature would not find in itself the desire to forget where it came from would it?

Leave a Reply