Dennett’s Thesis isn’t Evidence for Very Interesting Claims

This post is part of a series discussing Daniel Dennett’s Breaking the Spell.

The main thrust of Daniel Dennett’s Breaking the Spell is that the history of religion is not incompatible with evolutionary theory.  That sounds a lot less exciting than an attack on religion, but it’s what the book is actually about.  Dennett’s book doesn’t mount up any direct evidence against the truth claims of religion, but it does make the argument that religion is something you might be reasonably likely to observe in a world where there was no god.  That means the mere existence of religion is not strong evidence for the existence of god.  Fine and dandy.

But that’s not really so big a claim.  This is the same basic argument that evolutionary scientists have over a seemingly maladaptive trait or altruism.  Is this phenomenon the kind of thing we really wouldn’t expect to see given what we know about evolution?  There are a variety of ways to try to explore these questions.  Sometimes we can play around with simpler analogues in species that reproduce and evolve swiftly and come up with general rules (like sex-signalling can turn into an our of control arms race).  Sometimes we can monkey around within a population and see what animals are using as signals (playing different parts of song to see which are mating calls and which are alerts, showing college students silhouettes with different waist to hip ratios).  And sometimes, we’re just using our imagination, to see if there’s a plausible path from a precursor creature to the behavior we know today.

I’m not knocking that method, mind you.  It can help us make predictions that we can test using the fossil or artifact record.  But Dennett does tend to treat it as a little more definitive than I would.  A plausible path from A to B may be more of a testament to human ingenuity than to the likelihood of a certain cultural practice emerging due to evolutionary cruft.

At a certain point in the book, Dennett admits that all the data and speculation he’s mounting up doesn’t have much to say about the truth claims (metaphysical, historical, or ethical) of any religion:

“Notice, too, that this [universality] leaves wide open the possibility that divination… is a mutualist meme because it’s true …After all, the reason why water is deemed essential to life in every human culture is that it is essential to life. For the moment, though, my point is just that divination, which appears just about everywhere in human culture… could be understood as a natural phenomenon, paying for itself in the biological coin of replication, whether or not it is actually a source of reliable information”

The obvious next question is how you would test whether divination is a cultural replicator because it’s a guide to truth or whether it’s remains popular for some other reason.  Dennett comes round to this question for the god proposition near the end of the book, and I’ll hit that section tomorrow.  But before we tackle those specific arguments, I think I’ve got a handy inversion to explain why the kind of evidence Dennett is gathering isn’t that relevant to an atheist vs theist argument.

Dennett asks repeatedly “Cui bono?”  Who benefits from a particular cultural practice?  How is it paying for itself, is the currency increased access to food? Mates? Resistance to disease?  But if a cultural practice benefiting people is a cue that it’s stated claims aren’t trustworthy, because they can be explained away, than we’re heading into really wacky epistemological territory.

“You can totally trust my philosophy, because living it is nearly impossible, really lowers the chance you’ll reproduce, and takes a major toll on your health.  It’s so evolutionarily disadvantageous it is  sustained only by our commitment to Truth, no matter the cost!”

I foresee a Stylite boom.  Reserve your pillars now.


About Leah Libresco

Leah Anthony Libresco graduated from Yale in 2011. She works as an Editorial Assistant at The American Conservative by day, and by night writes for Patheos about theology, philosophy, and math at She was received into the Catholic Church in November 2012."

  • Randy

    I would worry about distrusting philosophies that have obvious evolutionary advantages. If evolution explains why people find genocide revolting then we can just ignore that data? If we can “improve” society by doing it then reason should trump evolved instinct, should it not? We would have to get over our squeamishness but men have found ways to do that in the past.

    • Darren

      Yet, do we actually find genocide revolting? Do we find it revolting on the evolutionarily determined “rotting meat = Bad / childbearing hips = Good” operating system level? Or is our aversion to genocide a culturally based, application level aversion?

      If on the operating system level, then we would expect all humans, everywhere (baring the occasional defective) to be repulsed. This is not what we observe.

      If application level / cultural, then some cultures would be repulsed, and some not. This level is subject to meme model evolution, but not gene model.

      • Randy

        I don’t think so. Those who have committed genocide in Rwanda, Cambodia, Germany, etc. have talked about overcoming their revulsion. I have no direct experience but I take their word for it.

    • Luka Alexandrovich Nevskeyev

      Men have. It is exactly this line of thinking, in the eugenicists and Nietzsche and even Marx, which allowed the genocides of the 20th century in the name of racial purity, communism, and “choice”.

  • Adrian Ratnapala

    Dennett’s book doesn’t mount up any direct evidence against the truth claims of religion,

    Woohoo! Glad you recognized that. Also note, in this book Dennet doesn’t *try* to argue against the truth claims of religions, and goes out of his way to be polite to the religious. But because he is supposed to be one of the “Four Horsemen of New Atheism”, this book gets read through a filter that tries to turn it into “The God Delusion”.

    … but it does make the argument that religion is something you might be reasonably likely to observe in a world where there was no god. That means the mere existence of religion is not strong evidence for the existence of god. Fine and dandy.

    But that’s not really so big a claim.

    Yes and no. Once you take of the “God Delusion Filter” and stop trying to look far an argument against religion, you still find a HUGE topic: what is it that makes some of the most powerful social institutions work and survive? Why do such phenomena occure.

    So it is big. But perhaps it is not a “claim”, Dennet wants to say that the book is a kind of proposal for a new research field, and so it make only very vague, tentative claims for itself. That said, it the tentative claim is pretty big: Dennet things (tentatively) that religions are (vaguely) like critters that survive and change by a process similar to Darwinian evolution (except that the distinction between species and individual is blurred).

    • leahlibresco

      I don’t think it’s just the Horseman filter. Dennett spends a long while in the first 50 pages trying to jolly along possible religious readers and implies this book may be quite threatening to their faith, but they owe it to themselves to keep reading, no matter how upset or fearful it makes them.

      • Chris Hallquist

        Adrian is absolutely right, Horseman filter at work. My memory of the book, which seems confirmed by re-skimming the first chapter, is that Dennett’s main goal in the book is to advocate the scientific study of religion. He works to “jolly along” religious readers because such a project might turn up things that are threatening to religion, not because it’s Dennett’s goal to undermine religion.

        It’s also worth nothing this paragraph from Dennett’s review of The God Delusion:

        We agree about most matters, and have learned a lot from each other, but on one central issue we are not (yet) of one mind: Dawkins is quite sure that the world would be a better place if religion were hastened to extinction and I am still agnostic about that. I don’t know what could be put in religion’s place–or what would arise unbidden–so I am still eager to explore the prospect of reforming religion, a task that cries out for a better understanding of the phenomena, and hence a lot more research than has yet been attempted.

        • Smidoz

          I struggle to understand this scientific study of religion, what are we actually talking about? Studying what effects religion have on people, then deciding that those effects are caused by some natural phenomena (like chemical reactions in the brain) then saying, like Dawkins’ does, that we don’t need a god to explain them, therefore there isn’t one. Or, is the aim to scientifically prove there is a god? Which is the most common argument (veiled under “absence of evidence). Of course, proving there is a god by scientific means would result in whatever was proved being natural, not supernatural, in which case, it would not be a god. So in the greater scheme of scientific enquiry, atheism is untestable, so why should atheists expect theism to be testable?

      • deiseach

        Oh, noes! Bloke says that religion is a source of social cohesion, and that tribal group A exploited this for purposes such as going to war against tribal group B, by maintaining that their ju-ju was more powerful than the ju-ju of group B, so they would have the victory?

        I am so threatened, upset and fearful by this because, you know, I completely avoided any references in my reading of various apologetics or comparative religion texts to the 19th century Higher Criticism, the Literary-Historical Method, or the Preface to the first volume of Pope Benedict XVI’s trilogy “Jesus of Nazareth”.

        Why do writers persist in imagining that all religious believers are Fundamentalist American Protestants fixated on Biblical inerrancy, Biblical literalism, and with no exposure to the history of Christianity, never mind any other religions? Not even all American Fundamentalists are that rigid!

    • Darren

      Having read a small amount of Dennett (but not the book in question) I wondered if we would end up here.

      ”That said, it the tentative claim is pretty big: Dennet thinks (tentatively) that religions are (vaguely) like critters that survive and change by a process similar to Darwinian evolution (except that the distinction between species and individual is blurred).:

      Are we not back to memes as originally laid out in Dawkin’s “The Selfish Gene”? As I recall, Dennett made heavy use of the meme model in the later part of his “Darwin’s Dangerous Idea” and I would not be surprised one bit if I am confusing the two books (having read them around the same time).

      • leahlibresco

        Yup. Most of the book is on evolution of memes.

    • Randy

      Actually The God Delusion does not really try and prove atheism either. There is a minor argument in Chapter 3 but mostly Dawkins assumes atheism and just rants about how much better life is as an atheist and how terrible religion is. It is more an expression of anger then it is a rational argument.

  • deiseach

    I don’t quite get his point about divination; if you need a source of water, and you turn to a diviner, and that diviner doesn’t find you a source, it’s quite possible that your tribe or village will die. Unless the diviner strikes it lucky by the laws of chance so that they find water more times than they fail, how can divining pay for itself in “the biological coin of replication”, seeing as depending on diviners may leave you at greater risk of dying?

  • grok87

    We were sort of talking about something similar to this line of reasoning the other day in bible study. How perhaps the details about Jesus in the gospels most likely to be historically true are the “embarassing” ones. That he was crucified (“what a scandal!”), that he was baptized by John the Baptist (“wait, he was a follower himself at one point rather than a leader out of the gate???”), that he liked to party (“Look at him, a glutton and a drunk, a friend of tax collectors and sinners!”).

    It’s an interesting line of reasoning, but obviously one can show the absurdity of it by taking it to extremes (Much as Leah has done with her stylite reference). Truth is made up of MORE than the incongruous embarassing details which are out of sync with what one would have expected. The fact that some of the details of someone or something are consistent with expectations does not make them untrue.

    • Tom

      Or that he’s the “son of Mary”. That was a huge insult and the Gospel writers didn’t feel the need to protect him from it.

      • grok87

        Not really following your line of reasoning. I think the Jews of the era would have expected the Messiah to have been born of a woman, not sure why they would have looked on that as an insult…

        • Sam Urfer

          The phrase has implications of illegitimate fatherhood.

          • Ray

            It’s not particularly embarrassing when the “illegitimate father” is the One and only True God. Rather it legitimates Jesus’s claim to divinity (at least once you’ve managed to come up with an interpretation of Trinitarian claims that you can convince yourself isn’t self contradictory.)

            Basing divinity on divine parentage could be taken as being too Pagan maybe, but then whoever wrote the gospels had a number of other issues with Pharisaic Judaism, so I can’t imagine this would have caused all that much embarrassment — certainly not any more than the fact that the whole thing was written in Greek rather than Hebrew.

  • Ben L

    I think I have some explanation along a few fronts: one is that religion is so unlikely, has such a low prior, is contrary to occam’s razor, that showing that evolution is sufficient to explain it in its entirety *is* evidence it isn’t true. Secondly, there are a number of religious arguments that depend on God being logically necessary, or other lines of thinking. Again, if you show that God isn’t necessary for Religion, just the Idea You Might be Wrong, or an alternative explanation at all, is very threatening to the majority of believers. The cynic might say because they already know, beneath their rationalizations, they believe in something improbable. The same reason there are blasphemy laws. No one passes a law saying you can’t claim 2+2=5. If religion was as obviously true as some believers like to claim, there would be no need to pass laws against counter-arguments.

    • Anonymous

      You must be new here.

      How exactly did you calculate the prior?

      …there are a number of religious arguments that depend on God being logically necessary, or other lines of thinking. Again, if you show that God isn’t necessary for Religion…

      There is a difference between “logically necessary” and “necessary for Religion”. Even Friar Ockham studied Aristotle.

      • Ben L

        I’m not sure the comments are the place for an in depth debate on the prior probability of the supernatural. Let me instead clarify: yes, there is a difference between logically necessary and necessary for religion. Perhaps I should have clearly split them. But my main point was intended to be simple: If the natural world is sufficient to explain religion, it would take strong evidence to suppose something unnatural was actually causing it, especially if no line of cause and effect could be drawn between this unnatural thing and the world we see without passing through something that works just fine as a natural cause by itself. I simply wished to draw attention to how Dennet’s boring claims relate to a relatively standard atheist argument: if the non-supernatural explains everything, then supposing an additional supernatural beings is untenable. Therefore “mak[ing] the argument that religion is something you might be reasonably likely to observe in a world where there was no god,” is not a small claim, and is important. If you don’t believe the prior for God is low, then this line of thinking is pretty much useless to you, and I’m not willing to take the time to engage on that subject here. Lastly, I am not new.

        • Anonymous

          And now that you’ve rephrased your objection to make sense, it’s time to re-read Leah’s post. She’s taking what you admit is a relatively standard argument and responding, “Well, that actually doesn’t provide much evidence.” Re-stating the standard argument don’t really learn anyone anything.

          • Ben L

            And I don’t believe her, because the epistemological territory she goes off to isn’t implied by Dennet’s argument. You can both suspect someone’s motives when they have a lot to gain *and* when they seem crazy. I distrust both the tele-evangelist who tells me giving me all his money is the path to salvation and the unfortunate homeless man on the corner yelling about the end of the world. Note, these are extremes I don’t mean to accuse all Christians of being at all, but they are different reasons to be wrong, and you can’t reverse one and be right.

          • Anonymous

            What is this I don’t even…

            The only part of this response that I can even partially make sense of is the first statement. However, I’m still not sure what you mean by the epistemological territory that she “goes off to”. Are you saying that Dennett’s argument simply doesn’t speak to the argument about the existence of God? This could make sense, and you could claim that Leah just shouldn’t expect it to… but then you’d have to eschew the entire remainder of what you’ve said previously.

            Perhaps I’m just horribly missing your point… so I could use a clarification.

  • Aron Wall

    “You can totally trust my philosophy, because living it is nearly impossible, really lowers the chance you’ll reproduce, and takes a major toll on your health. It’s so evolutionarily disadvantageous it is sustained only by our commitment to Truth, no matter the cost!”

    Even if all of this antiworldliness happens to be true of some religion (and there have been ascetic religions with pretty dim views of procreation, for example), the skeptic could still say just that the religion has survival value because of its social ability to gain converts! Memes, not genes. And of course, any religion that exists, must by definition have been successful at gaining converts at some historical period. So it’s hard to refute this sort of claim. If the religion gains converts in a way not possible naturalistically (for example, by performing miracles in front of them) then perhaps it would be simpler to discuss the alleged supernatural interventions directly, rather than in terms of their indirect convert-obtaining effects.

  • Alex Godofsky

    re: your last paragraph

    Leah, if a lot of people firmly believed a thing that did, in fact, have those properties – that it was anti-selected, etc. – and held that commitment over many generations, etc., that WOULD be circumstantial evidence for its truth.

    • Aron Wall

      I’m not sure why you say it would be circumstatial evidence. Do you mean that people would be unmotivated to adopt a religion that was against their best actual interests, so they’re more likely to be believing it because it is true? Or do you mean that if the religion was in fact objectively bad for your survival, that its continuing to exist would be evidence that God was looking out for that group? Or is it something else?

      • Alex Godofsky

        Actually, I just retract the claim. It’s wrong, and causes confusion because the premise (“if the believe were held but anti-selected”) shouldn’t actually ever happen in the long run.

        The correct claim, the one I was aiming for but messed up on, is this:

        If belief X would be anti-selected if it were false, and we observe belief X, then that is evidence for X.

        If belief X would not be anti-selected even if it were false, then observing belief X is not evidence for X.

        Therefore, if we know that we observe belief X, and after that we discover that belief X would not be anti-selected even if it were false, that is evidence against X. It does this because (presumably) we had some prior that X is in the first bucket (would be anti-selected if false), and so we thought that observing belief X was evidence for X, but now we know that it isn’t.

        • Kristen inDallas

          Affirming a disjunct:
          If a phenomenon can be caused by A or B, affirming A does not invalidate B.

          Denying the antecedant:
          Not having evidence for is not the same as having evidence against.

          Sorry for being nitpicky – but all those if/then statements and Xs led me to the conclusion you’d be interested in logic arguments.

          • Ben L


            Maybe not the same, but you can’t convince me of a teapot in orbit without evidence.

          • Alex Godofsky

            I am using “evidence against” as shorthand for “this new information should make us update our priors in a way that reduces the weight we attach to the belief”.

            (This is also a response to LeRoi below).

          • LeRoi

            Thanks for clarifying your Bayesian shorthand, Alex – that makes sense.

        • LeRoi

          Very clear premises and conclusions. But I’m not sure that discovering the belief would be supported even if false counts as evidence against the belief, even if we used to think the belief would only be supported if it were true. If we think people would support the belief even if false, we’ve lost a piece of evidence for the belief – because now we don’t know what to think – but we haven’t gained any evidence against it. Or is this merely a semantic difference?

          Actually, what Kristen said – denying the antecedent.

    • Aron Wall

      As long as I’m here, I’d like to take the opportunity to shamelessly plug my new blog “Undivided Looking” to any of your readers who happen to be listening (just click on my name). I’m going to be attempting to popularize some fundamental physics ideas, aiming at geeky people who aren’t afraid of math, but don’t necessarily know much about physics, and are tired of trite metaphors as a substitute for understanding. I do gravity theory research for a living, so I actually know what I’m talking about.

      I’m also a Christian who will be blogging about “science and religion” issues in what I hope will be a new and engaging way. Your blog is frequented by a fairly reasonable and civil cross-section of the internet rationalist community, so I’m hoping that they might be willing to drop by and provide some friendly pushback.

      • leahlibresco

        I’ve had you in my RSS reader since about a week ago. :)

  • Elliott Scott

    David Bentley Hart has a great line in Atheist Delusions on this: “As for Dennett’s amazing discovery that the ‘natural desire for God’ is in fact a desire for God that is natural, it amounts to a revolution not of thought, only of syntax.”

  • keddaw

    Truth has an exceedingly strong evolutionary advantage.

    Given our psychology, so does something that creates an in-group/out-group separation.

    Religion comes under the second kind.

    • Anonymous

      Ipse dixit ftw!

    • Adam G.

      I can think of a number of scenarios where truth *doesn’t* have a survival advantage. For instance, people usually tend to overrate their attractiveness and the degree to which other people like them. Which, since you can’t procreate if you don’t copulate, and you can’t copulate if you’re too shy to ask, has survival advantages. In-group/out-group status has survival advantages? Well, then fictional (non-truthful) in-group/out-group distinctions–”imagined communities”–should have survival advantages.

      Basically, you’ve just made too big assertions and one big conclusion for which you provide no evidence and no argument. Presumably, since truth has a survival advantage, no one will mate with you today or you will be devoured by a lion.

    • Alan

      Given all the natural tendencies we have evolved that make it more difficult for us to discern the truth (think cognitive biases) I’m not sure why you would think truth has an evolutionary advantage let alone a strong one.

  • Irenist

    Atheist reductionist materialism seems to confer a survival advantage upon academics seeking tenure.

  • LeRoi

    I agree Dennett’s book is not as revolutionary or as bold as he might have wanted it to be. That he thought his proposal was new indicated meager knowledge of the subject. Dennett is actually a respected philosopher-of-mind, so he probably knows his stuff there, but he doesn’t seem to know up from down in philosophy of religion.
    After all, aren’t the history and philosophy of religion *all about* how to explain religion? Protestants and Catholics have been trying to demonstrate that the other side is of man and not of God for 500 years. Both have tried to do the same to every other religion – ever since the Church Fathers claimed the similarities between Christianity and the mystery religions came down to the devil mocking God. (yes, I know it’s much more nuanced than that – all truth is God’s truth and so on, but that’s not my point)

    One doesn’t need the pseudo-study of memes to see that cultural evolution is natural. Philosophers from the Greeks through the Renaissance took a very practical approach to politics and culture, which still illuminates when viewed through evolutionary eyes. And if one believes that religion is ultimately mostly about politics and culture, as Dennett does, then one is following in those philosophers’ footsteps.

    The conversation over the intersection of religion, politics, and culture has been going on for a long time. Dennett isn’t starting anything and has not added much. His book gets more attention than it deserves in intellectual circles that happen to be connected New Atheist controversies. His book will eventually be forgotten.

    Finally, Leah, I wish you’d take a gander at Richard Rorty’s “Contingency, Irony, and Solidarity,” or “Objectivity, Relativism, and Truth” – the first is my favorite but either is good. You may have already hit up neopragmatism in your school debating club, but I’d be interested to hear your thoughts. You think the best metaphysics comes from Christianity; Rorty thinks we don’t need metaphysics at all.

    • Irenist

      Second the Rorty suggestion! I’m a Thomist myself, but Rorty makes the case for the polar opposite position with thoughtfulness and charm. “Contingency, Irony, and Solidarity” is a great book, and its pragmatic nominalism and agenda of “philosophical deflationism” (i.e., epistemic humility) is one that every realist should grapple with.

  • Pingback: Reading Dennett’s Breaking the Spell [Index Post]