Sean Carroll debunks the “fine-tuning” argument for God

December 31, 2015 • 2:00 pm

I don’t know how many readers watched Official Website Physicist™ Sean Carroll debate theologian William Lane Craig on whether cosmology gives evidence for God, but nonbeliever Carroll clearly won (I’m not unbiased, of course!). Here Carroll takes apart the argument that the so-called “fine tuning” of the physical constants of the Universe constitutes evidence for God (the “FTA”). Since, as Carroll notes, this is the most “sophisticated” argument for God in theologians’ poorly stocked arsenal, it’s incumbent on us to understand why it’s wrong. I deal with this in Faith Versus Fact, but here’s a nine-minute primer. You might want to watch it before you get drunk tonight.

Carroll is a fluid and eloquent speaker, anticipating and then answering his audience’s objections before they’re even uttered.

Carroll’s written summary of the debate, including the fine-tuning argument, can be found in his post at Preposterious Universe. Here are his five arguments (in his words) why the FTA doesn’t prove theism:

  1. We don’t really know that the universe is tuned specifically for life, since we don’t know the conditions under which life is possible.
  2. Fine-tuning for life would only potentially be relevant if we already accepted naturalism; God could create life under arbitrary physical conditions.
  3. Apparent fine-tunings may be explained by dynamical mechanisms or improved notions of probability.
  4. The multiverse is a perfectly viable naturalistic explanation.
  5. If God had finely-tuned the universe for life, it would look very different indeed. [Carroll considers this his most important point. Here he goes into not only the cosmos, but the nature of human culture which, Carroll avers, comports much better with naturalism than with theism.]

He goes on in his post to explain his own intentions and to dissect Craig’s responses.

I should add that Sean is giving the prestigious Gifford Lectures in October of next year in Glasgow. That lectureship, originally endowed to promote the study of “natural theology” (the observation of nature as evidence for God), has had some prestigious honorees. They include, for example, William James, whose talks became The Varieties of Religious Experience. Other lecturers included Paul Tillich, Hannah Arendt, Arthur Eddington, Reinhold Neibuhr, Carl Sagan, J.B.S. Haldane, and Steven Pinker.  You can see that there are some lecturers, like Haldane, Sagan, and Pinker, who weren’t espousing natural theology at all, but talking about straight naturalism. Kudos on the organizers for starting to include nonbelievers, who, after all, probably have something more substantive to say.

When I asked Sean if he was going to turn the lectures into a book, which I believe is expected, he said he already had. And indeed, the book is already on Amazon, scheduled for publication by Dutton in May. Click on the screenshot to go to the listing:


I expect that this book will be very good, summarizing Carroll’s views on the implications of particle physics and cosmology for philosophy and our own self-image. Here’s the advance summary:

In short chapters filled with intriguing historical anecdotes, personal asides, and rigorous exposition, readers learn the difference between how the world works at the quantum level, the cosmic level, and the human level–and then how each connects to the other.  Carroll’s presentation of the principles that have guided the scientific revolution from Darwin and Einstein to the origins of life, consciousness, and the universe is dazzlingly unique.

Carroll shows how an avalanche of discoveries in the past few hundred years has changed our world and what really matters to us. Our lives are dwarfed like never before by the immensity of space and time, but they are redeemed by our capacity to comprehend it and give it meaning.

The Big Picture is an unprecedented scientific worldview, a tour de force that will sit on shelves alongside the works of Stephen Hawking, Carl Sagan, Daniel Dennett, and E. O. Wilson for years to come.

I’ll be reading it for sure.


131 thoughts on “Sean Carroll debunks the “fine-tuning” argument for God

  1. Hi Jerry– Hopefully you will enjoy the physics and philosophy parts of the new book, there’s some stuff in there that you don’t see in other places. And hopefully you won’t mind the discussion of evolution, although it focuses on abstract principles rather than dealing with the messy reality of evolution here on Earth. You won’t like my compatibilist take on free will, but maybe I’ll convert you!

    1. Oooo I look forward to the compatibility bit since I keep trying for someone to convince me of that argument.

      1. The only way someone could convince me of this argument is with the relevant evidence, as in the end the question of free will is a scientific question. So far, all the available evidence leads me to conclude that there’s no such thing as free will – even our emotions are not free (they are the product of our brains, genes, hormone levels, upbringing, etc). What’s more our brain changes as it develops throughout life, a child’s brain thinks in a different way than a teenager’s brain, and a teenager’s brain differs from an adult’s brain. Where is free will here?

      2. Never mind a convincing argument; I’d be satisfied at this point with a coherent definition of the term that everybody can agree upon.

        We all experience the phenomenon of imagining different potential futures, freely flipping in our minds from the expected results of one option available to us to another. The outcomes we imagine resulting from the choices we perceive open to us heavily influence the actions we actually take.

        All that feels like we’re freely hopping from one alternate universe to another, willing all sorts of realities to manifest with reckless an unconstrained abandon. But the actual reality is that the entire process is perfectly causal and deterministic, and none of it under any conscious control at all. The trickles of water may make fascinating and incredibly intricate patterns as they cascade down the hillside, but there’s nothing guiding the water itself, nothing creating the patterns save the Universe’s Schrödinger equation.

        People who argue for free will inevitably point to that mental decision-making process as the phenomenon in question. And the phenomenon is undoubtedly very real and constitutes an ever-present and priceless component of human experience. But to label it as either free or willful or freely willed or anything like that…to do so is to mistrake the illusion for reality and, ultimately, to miss the point entirely of the fact that we are completely and inescapably an integral part of this Universe. We ourselves are nothing special, but the Universe itself is amazing, and any and all specialness we possess is but a reflection and small apportioning of the wonderfulness of the Universe.


        1. Ben,

          What a wonderful statement on the question of what is free will, and the illusion of experiencing it. Well done!

        2. And as I usually would argue, there are dubious and arguable assumptions in Ben’s use of words like “Feels like” and “Illusion.”

          The reason it “feels” like we have a choice isn’t born some strange “illusion.” It comes from the fact we are often thinking reasonable, true things when we are engaged in decision making. My son recently had to decide between
          two courses of action: 1. Accept a golf scholarship and go directly from high school to university or 2. Take a year off first. What kind of process was he going through during this process? Well, obviously he didn’t t think he could do both – he could only do one or the other.
          So, like we all do everyday, he considered the likely outcomes IF he took option 1 vs option 2 – e.g. If he took a year off to work on his golf game he’d go into the scholarship and university team a better golfer. But IF he
          didn’t take a year off, he’d be out of university sooner (true) and also he may have a smoother move into the academics of university, while he’s in the groove of school and have retained what he learned better than taking a year off.

          Not once did I ever hear my son appeal to his magical contra-causal powers when he explained his thinking on the subject because, of course, that didn’t play at all, nor could it contribute at all, to solving the real world problems
          he had to solve.

          There is nothing “illusory” about entertaining possible futures in the if/then mode in which we normally decide our actions. And we think (if we are thinking clearly) perfectly plausible and often true chains of reasoning. THAT’s why we don’t feel in the grips of an illusion when
          making decisions – because for the most part, no important part of our decision making relies on this purported illusion.

          (Why repeat that point having said it before? Because surely there are always new folks here lurking who may not have heard that viewpoint before. Since Ben had his say, I just had mine, I’ll leave it at that).


          1. The decision-making process your son went through is very real and very important. Indeed, it may well be the most precious thing there is about being human, for it defines who we are as individuals.

            But why sully and burden that process with the theologically-loaded term, “free will”? What is “free” about it? How does it explain the existence of evil? Which god gave it to us so we might freely and willfully choose this heaven or that hell after being exposed to which holy writ?

            Your son thought long and hard about his future and made the best decision he was capable of making. Is that not enough? What do you gain from the faery dust of “free will” that you must sprinkle it on top?

            Misa and I are constantly saying to each other that we have no choice in this or that — especially whether or not we are to love each other. And it’s true. How could I not love her? It’s impossible. You might as well tell the Sun to leave the sky or ask a baby not to cry.

            And so it is with even the tiniest, most inconsequential of decisions. Did I have any choice of whether or not to put some pineapple in the yoghurt I ate for breakfast this morning? Sure, I can construct imaginary worlds in which I didn’t…but, in the real world, given the circumstances at the time, it was inevitable. All those other hypotheticals under which conditions I didn’t put pineapple in the yoghurt…they’re all imaginary, not real. In the real world, I put pineapple in the yoghurt, and I had no choice but to do so. Tomorrow I might or might not put pineapple in the yoghurt — and, indeed, I might not even have any yoghurt at all. We shall see. But, whatever tomorrow may bring, I shall have no choice in the matter.

            And, of course. I do my best to bring about the best possible future I can. But, even there, I have no choice. How could I not do my best? Sure, my best is often inadequate, sometimes less than what, in hindsight, I think I perhaps could have managed. But, whatever the deficit…if that stumbling block were not in the way, I would have done better (and maybe I will next time). But the stumbling block was there, and I stumbled, still doing the best I was able to despite stumbling.

            All the world’s a stage, and all the men and women merely players. We have our exits and entrances, and in our times play many parts. But we play our parts and read our scripts, discovering from within as the play unfolds what our true natures reveal themselves to be. What role do you wish you would play? Think carefully now and rehearse your lines, for you will not get a second chance once the spotlight shines upon you.



            1. ‘The decision-making process your son went through is very real and very important. Indeed, it may well be the most precious thing there is about being human, for it defines who we are as individuals.

              ‘But why sully and burden that process with the theologically-loaded term, “free will”?’

              But Vaal nowhere mentioned free will. Perhaps you should read, quite carefully, what he actually said, as opposed to what you would like to think he said, and refrain from imposing your unfounded assumptions…

              1. It’s a continuation of a very long-standing discussion. Vaal is an advocate for Compatibilism, the proposition that Free Will is a real phenomenon that is compatible with the natural laws as we already understand them. He distinguishes his position from those who advocate for contra-casual Free Will.

                Most of the disagreement comes from the understanding of the word, “free.” As you can read in Vaal’s post, he sees freedom in the decision-making process. In stark contrast, though I see vastly more complexity and sophistication in the decision-making process than there is in a thermostat’s operation, I see no freedom in either. We have an illusion of freedom, yes, but it’s the freedom of the puppet to love its strings.



              2. Ben…

                Are you using a new computer these days? You comments are getting munged. Some examples:

                “…of the word, “free.” As you…”
                “…term, “free will”?…”

              3. My guess is somehow his comments are getting encoded in Windows-1252 and the site is interpreting them as UTF-8. I’d imagine the replies are not being typed directly into the comment box on the site for this to be happening. An email response with different encoding perhaps?

              4. Close. I’ve recently upgraded to OS X El Capitan. It seems to be doing the “smart” quotes thing…no clue what sort of encoding it’s using nor why it wouldn’t use UTF-8. (Using the mail interface with the built-in

                I might try to fix it at some point…too busy with other things right this moment, though….



              5. Yours look smart, at least in the email I got. I’m sure mine won’t have changed…haven’t even pretended to look at it yet….



              6. Hmm… I’ve been running El Capitan since it was released without that problem. I’m got “Use smart quotes and dashes” checked in my keyboard preferences. But here in Safari they aren’t being inserted.

        3. Ben, everything you say there shows that you’re still totally misunderstanding what compatibilism is all about.

          Compatibilists are not seeking something that is free of deterministic constraints. It is a concept about freedom from *social* constraints.

          1. Thus the importance of definitions. And the definitions should be laid out before the arguing begins.

              1. do you mean like….

                the unique ability of A ROBOT to exercise control over their conduct in the manner necessary for moral responsibility

              2. To quote the same article:

                “it would seem unreasonable to say of a person that she deserves blame and punishment for her conduct if it turned out that she was not at any point in time in control of it.”

                So the answer is NO for robots or animals; control over their own conduct is too limited to deserve to be punished.

                I’m not aware of any judicial system that contradicts this.

              3. Peepuk –

                How do we have any more control over our own “conduct” than a robot has over theirs? Do I or a robot need the ability to “reprogram” ourselves to a sufficient degree to be considered morally responsible? Yet, our ability or inability to “reprogram” ourselves is a function of our existing program. A program that we had no part in designing and are therefore not responsible for.

                I didn’t select my genes or parents or upbringing or environment any more than a robot selected its mechatronics or programming. I am a more advanced machine than the robots we are currently producing but only in degree.

                At what point does a complex machine become “morally responsible” despite it never taking part in it’s initial design, programming or environment?

              4. Probably we agree on all points. I don’t think free will is real, but we all seem to have this strong illusion.

                Previous definition of free will is only provided because people seem not to now what we mean by that concept, resulting in endless discussions.

      3. I like to think of the issue modeled as a billiard table with balls whizzing this way and that. Free will is like saying the 4 ball is trying to decide whether to hit the 9 ball at 4 degree glancing blow leading to a bumper impact, or a 5 degree carom into the corner pocket.

      4. FWIW, I think it is a matter of compatibility between determinism and moral responsibility, rather than between determinism and free will. We have moral responsibility because we can foresee the effects of our actions on others. I know that if I hit someone with a crowbar I will hurt or even kill them, so if I do it I have a moral responsibility for that action. That is why the law recognizes diminished capacity for those who cannot understand the effects of their actions. What else does morality mean to an atheist other than a due consideration for the welfare of other sentient beings? I refuse to relinquish exclusive possession of moral responsibility to the religious.

        1. “We have moral responsibility because we can foresee the effects of our actions on others.”

          Do you think that a highly aggressive lowlife with poor impulse control (and drug addiction to boot) who beats someone to death during a robbery has the same capability of foreseeing his actions as you, and therefore has the same moral responsibility as you?

        2. The only way that “free will” and morality ever seem to actually fit together is that “free will” is used as the excuse to shift the blame from the gods to us humans.

          If you witnessed a priest raping a child in the name of a god, you wouldn’t hesitate to call 9-1-1, at the very least. Yet the same god whom the priest invokes in order to cow the child into obedience and silence…that god does nothing whatsoever, not even call 9-1-1, despite allegedly possessing superhuman ability and compassion…because the god values the priest’s freedom to force his will upon a child more than the god values the most obvious and essential human norms of morality. Similarly, though the god must certainly know of the mid-ocean earthquake that will soon send a tsunami crashing over hundreds of thousands of people, the god would rather they be free to die in the ignorance he wills than that they should have the choice of whether to flee or drown.

          Outside of such religious contexts, the incoherency of “free will” is worse than useless. Even if you formulate some redefinition of the term that applies to a real-world phenomenon, it’s only likely to be used to place blame and therefore justify retribution. Much better to focus on the future, looking to the past only for help in determining how to make the most of what is yet to come. In the context of criminal justice, that means rehabilitation (and quarantine until rehabilitation can be reasonably trusted to be effective). Even less sexy, it means a solid social infrastructure, including education and a safety net and mental health facilities, so you don’t even get crime in the first place. What use “free will,” of any flavor, in such deliberations?



          1. I miss the mobs with burning torches marching toward the castle chanting:

            “Rehab no! He must die”.
            “Rehab no! He must die”.
            “Rehab no! He must die”.
            “Rehab no! He must die”…

      5. Starting the new year – oh no, it’s yesterday’s post – by wanting to be flogged with a dead horse. Not a good way to start the year.
        Sorry, I was just seeing the Dogbert cartoon, and now there is an image in my mind which I can’t shift.

  2. I think that there’s an even more basic problem with the fine-tuning argument: it depends on the starting assumption that humans are special. Without that it gets nowhere. And yet, what is the conclusion of the argument? It’s that humans are special!

    Afterall, fine-tuning for what? To answer that question you have to assume that humans matter in some way, and that a universe without humans would be “wrong”.

    1. I don’t think there is circularity. The “special” quality that leads to the probabilistic fine tuning problem defines “life” as “anything self-aware that can ask questions about its own universe”. The question is, how unlikely is *that*?

      This leads to Carroll’s #1. Life that looks like humans is unlikely, but that’s not the requirement. We have no idea what variety of self-aware questioning beings might arise in universes with different physical laws. So we just don’t know how unlikely “life” really is.

      Whether Carroll’s #1 is valid is open to argument. There do not seem to be many parameter values that lead to anything remotely stable. You’d think that stable matter of some kind is required for life to evolve, at least.

      1. The “special” quality that leads to the probabilistic fine tuning problem defines “life” as “anything self-aware that can ask questions about its own universe”.

        Ok, but why are you asking about “anything self-aware that can ask questions” (as oppose to amorphous nondescript stuff that doesn’t ask questions) in the first place?

        The very question requires us to regard humans as special; but that is the whole issue under consideration. If humans are not special, then the universe is not fine tuned.

        1. Just to be clear, I’m talking here about the real scientific fine tuning problem here (nothing to do with using it to argue for God).

          It’s a problem that arises because the best models that we have for “how to build a universe” have a lot of free parameters, and seem to allow a vast number of possible universes, most of which would be incompatible with life.

          To be clear, it’s not a “paradox” – it’s just a problem that admits several possible answers, and we don’t know the correct one.

          I put up a post below to explain better.

          1. I’ve never liked this argument either. Scientists who talk about fine tuning due to the free parameters make a variety of mistakes, but the most egregious one in my opinion is that they simplify things to much. They will talk about how if you vary X by some small amount Y, the universe would be inhospitable for life. They are forgetting that the free parameters can all vary at the same time. If you have parameter X, and add a small amount Y to it, the universe is inhospitable, but then if you take parameter Z, and change it by a small amount W, suddenly everything falls back into place again.

            The point being, no one has solved the insane multi-dimensional problem of how large the “habitable zone” is in the hyperspace formed by all the parameters being allowed to vary.

            1. When physicists talk about the fine tuning “problem”, they mean “something interesting that we don’t yet have a satisfactory explanation for, and that seems to require one.”

              You, and several others on here, seem to be saying “But there IS probably an answer that we haven’t found yet, so there is no problem”.

              I’m left scratching my head – something we don’t yet know the answer to is called – wait for it – “a problem”!

              There’s no implication that the fine tuning problem in physics is some kind of irreconcilable paradox.

      2. Yes, Ralph, I agree. Although most religious people do bring humans into the fine-tuning argument, more rational people can still wonder why the universe is capable of producing intelligent life of any kind. The multiverse does seem to answer the question, but it is not clear to me whether this is “cheating”…I think at least some people believe in the multiverse just because it is a solution to this problem.

        But I maintain that the fine-tuning problem does not exist for theists, only for naturalists. Theists already believe in disembodied souls and a god, so they think intelligent “life” can exist without any physical constraints whatsoever. Under their belief system, any universe at all can support (or perhaps a better term would be “coexist with”) intelligent life. So they have a different problem. They have to explain why any physical universe exists at all.

    2. Good one!

      My own objection slips in between that one (why constraints) and Sean’s #1 (unknown constraints) as I see the religious finetuning argument as mistaking prior probability with posterior likelihood (mistaken constraints).

      E.g. it is very unlikely that a stone will tumble the way it does during an avalanche. Yet tumble it did if an avalanche happened, and calling that ‘finetuning’ without quantifying finetuning is a no go, we could have anthropic selection for one. [Which analysis pretty much relies on Sean’s observation, I now see. I guess the religious perversion of probability theory is what gets me.]

      1. I think that’s unfair to the fine tuning argument. The outcome under consideration is a very natural category with a broad range. Intelligent life. Or even life of any kind.

    3. “it depends on the starting assumption that humans are special.”

      Agreed, which is why the response that I make to the fine-tuning argument is that if any of the parameters changed, there would be something else and the universe doesn’t care.

      1. The physics suggests this doesn’t work though. The fine tuning problem is that if the any of the parameters changed slightly, there would probably be nothing.

        See here, for example.

        We can agree that none of this implies “therefore God”. But among scientific cosmologists, there IS a widely acknowledged fine tuning problem, with many views on the likely correct naturalistic explanation.

        1. “physics suggests this doesn’t work though.”

          No, I don’t think it does and I don’t think that all physicists agree that there is a fine-tuning problem. I’ve essentially restated the Anthropic Principle and some physicists hold this view.

          I do agree that physicists are interested in why the constants have the values that they do, but some seem to acknowledge that the answer might be “just because”.

          1. Ok, let me try to be clearer.

            Here Sean is responding specifically to the “fine tuning implies God” argument, and doing so eloquently.

            But whilst agreeing with Sean 100%, that does not mean that there is no genuine “fine tuning problem” within the physics community, about which there are differing opinions about the correct naturalistic explanation.

            I’m not sure that many physicists think the answer is “just because”, but in any event it’s a real question within physics, it’s not just about an argument with dumb theologians.

            1. “’m not sure that many physicists think the answer is “just because””

              Of course not, because you can’t do science that way. But if you keep drilling down, you’ll eventually get to a “just because”. Assuming we discover why the constants have the values that they do, we can still ask the reasons for those reasons, and it’s turtles all the way down.

              It also seems to me that it’s a failure of imagination to think that the range of possible universes is merely limited to having different values for the constants.

              1. “It also seems to me that it’s a failure of imagination to think that the range of possible universes is merely limited to having different values for the constants.”


              2. “Assuming we discover why the constants have the values that they do…”

                But that’s just it – we have been trying really hard for decades to discover this. And, so far, failed completely. The lack of a successful explanation so far is precisely what me mean by “the fine tuning problem”.

              3. That isn’t my understanding of the fine-tuning “problem”. It’s not merely why they have the values that they do, but why they have the values needed to produce life.

                Based on that distinction, one can be interested in an explanation for the values of the constants without considering it a problem.

              4. I think most physicists realize that the laws are also in play. But it is very unclear how that impacts the fine tuning argument.

              5. “I think most physicists realize that the laws are also in play…unclear how that impacts the fine tuning argument. ”

                Because then you can’t really argue that most possible universe give rise to nothing. This strikes me as unknowable.

    4. I would argue that Carroll is right in his opening statement where he says “I’m not sure there even is a fine tuning argument.”

      The traditional analogy is to a lottery. The multiverse answer is equivalent to saying there were a lot of tickets bought, so the odds of somebody winning is high. But here’s my question which I think undermines the whole argument and which goes back to Carroll’s first comment: do you know the odds of winning? Show me the data on what values the Plank Constant can attain. What is the range of values? And in that range, what is the granularity? Show me the same for all the others.

      You can’t, because nobody knows those things. Nobody knows what values the constants could attain. Nobody knows the range or how many discrete values lie within the range. In terms of the lottery analogy, we have no idea what the odds of winning are. Which means, as Carroll said, we have no idea whether there really is a fine tuning problem or not.

      The other relevant point that Carroll makes about this is that theism is undefined and unlikely to contribute data. Its possible someone, at some future time, will figure out the possible range of values for the Plank constant. Would you care to guess whether that person is a physicist or a theologian? It will be physicists themselves who look for and evaluate whether fine tuning is even an issue; theologians will sit on the sidelines, claiming it is with no actual data to back them up, until somebody else does the hard work for them.

      1. The traditional analogy is to a lottery. The multiverse answer is equivalent to saying there were a lot of tickets bought, so the odds of somebody winning is high.

        Assume a lottery where the winning payout is a predetermined amount, independent of how many tickets are sold. Every ticket has the same chance of winning, and there may be 0, 1, or many winners. (A real-life example of such a lottery is the “numbers racket”, where the payout is around 600 to 1 if you correctly predict 3 effectively random digits.)

        The analogy with the (weak) anthropic principle is that winning = being in a universe that supports intelligent life, and the number of tickets = the number of universes.

        If you ask a winner of such a lottery, they won’t know how many tickets were sold.

    5. Hmm, could I raise an argument that the universe is fine-tuned for rocks, and that humans are a minor “skin irritant” (in Terry Pratchett’s phrase) which exists only to provide socks.
      Yeah, I think I could make that fly. I’ll just need to “fine tune” my pig-cannon.

  3. Excellent. And i will be putting the book on my already too long list of must-read books.

    I enjoyed his list of rebuttals. They were knocking it out of the park, to put it mildly. But as yet I did not understand his 3rd one on the list which starts at ~ 2 minutes in. The issue there is the claim that if the early expansion of the universe was at a slightly different rate
    (1 part in 10^60 different), then no life as we know it. But, he says, if one considers a correct derivation of General Relativity, then the probability that the expansion rate will be just what it was becomes certain. I did not get that part. Does he mean that the expansion rates will be different in different places in a multiverse, and somewhere it will be the expansion rate of our early universe?

    1. I don’t know if it is related, but Sean has published a paper that effectively comes to that conclusion, cosmological space is likely flat.

      “We use the invariant measure on solutions to Einstein’s equation to quantify
      the problems of cosmological fine-tuning. The most natural interpretation of the measure
      is the flatness problem does not exist; almost all Robertson-Walker cosmologies
      are spatially flat.”

      [ ]

      I guess that if you see inflation, that constraint guarantees that inflation will happen despite that it needs to be finetuned to start – which I think Sean has argued numerous times.

      Now how much that measure is reasonable (it _is_ invariant) is arguable at a guess.


      1. OK, I stumbled on the probabilities there. A better way to describe it is that if inflation is the only way to make long lasting universes it will happen. (I fear I am relying on anthropic reasoning now, so I’ll stop there. “Before you get too drunk” was a beer ago – I am trying to handle the special day without pain killers after a traffic accident, and just found out that beer helps.)

        1. Terribly sorry about the car accident. May you recover fully and quickly. (notice, I removed the “dear Lord” from the beginning of that prayer). If the beer loses its effect, I’m helping myself to some Port with fudge. That might be more effective.

    2. #3 is best stated as:
      There may be better models for how to build universes that don’t have free parameters. It may be that the conditions in our universe are, in fact, the only possible conditions, and we just haven’t found the correct model yet.

  4. Sean’s arguments are always so clear and articulate, they make me almost feel sorry for his opponents. But, after a few minutes reflection I don’t mind that they must suffer for their sins.
    In a debate with Lawrence Krause, WLC was asked what new evidence would change his mind about Christian and accept the other side’s view.
    “Nothing”, he mumbled.
    In other words WLC and his unfortunate type are completely committed and cannot even appreciate the joy of learning and discovery. It makes for really awful arguments when you can’t even apply what rationality you have to a serious question about our human existence.

    1. When I see the rigid belief systems of others there presents the opportunity to look at my own.
      Many atheists, myself included, call themselves humanists yet the society, indeed global ‘morality’ of such thoughts are not replicated in natural systems.
      Nihilism is the truth in the broader moral/ethical debate, yet most atheists will grasp the lack of abstract ‘moral’ certainty but will say, “Yes thats right but don’t call me a Nihilist”.
      Less socially isolating for the individual?

      1. Scientific minded atheists reject nihilism because of biology: “Moral nihilists assert that morality does not inherently exist, and that any established moral values are abstractly contrived.”

        A lot of species have evolved inherent morals that are not abstractly contrived but observably useful, so nihilism is unfactual.

        1. But some, no matter how scientific, might be afflicted by depression that makes them perpetually cynical, fatalistic, and functional nihilists.

          It is reason alone that lifts us toward the realization that nihilism is unfactual.

          1. The ‘selfishall gene’ allows for gradations of morality, ie I against my brother, my brother and I against our cousin, my cousin and my brother and I against the neighbours etc.
            What reason allows is the very special ‘group’ behaviour that the selfish theory denies.

        2. I don’t think that’s true.

          The two competitors of moral-nihilism are moral-universalism and moral-relativism. It’s easy to see that these are not compatible with value-free-science.

          Biology or physics cannot tell us what’s good or wrong, so some form of moral-skepticism is the only game in town if you want to be compatible with value-free-science.

          The few moral nihilist I know all seem to be consequentialists; they have no other means to justify their actions.

      2. “‘morality’ of such thoughts are not replicated in natural systems.”

        But, if you think of human life as a natural system you don’t have so much of a conflict. At least I don’t. Natural morality is easy to account for, especially since the advent of secular Western civilizations which operate mainly independent of authoritarian creeds.

  5. To me, the fine-tuning question is a variant of the “why is there something rather than nothing” question.

    Answer: Because there is something and we are here. PERIOD.

    1. I can’t remember who it was, but someone else in debate with WLC – I think a chemist, with something of the manner of Betrand Russell about him – said (something like), “You might not like it, but that’s the way it is.”


  6. Sean Carroll may be an atheist, but he clearly dabbles in the dark arts of black magic, too. How else to explain that he never seems to age? He’s cosmology’s perpetual teenager, the Dick Clark of particle physics.

    Might there be a portrait of him reflecting the wear-and-tear of the years kept under lock-and-key somewhere in the bowels of Cal-Tech’s physics department?

      1. There’s also Tom Cruise. I’ve always attributed his forever-young appearance to the “sea org” Scientologists coming by his manse once of month to place him in suspended animation and submerge him in a tank of doctor fish, which eat away his top epidermal layer, leaving him as smooth and blushing as a baby’s bottom.

        1. There is quit a bit that can be done with a cleaver scalpel to de-age an aging star. Not to mention the various software post-production techniques that can carve years away making a worn face fresh and new.

    1. Now that you mention it, you never saw Sean Carroll and Dick Clark in the same place at the same time. Why is that?

        1. With Carroll, I would guess its an unpublished theoretical physics paper that elegantly describes in a single equation how the entire universe functions. Every year, the equation (and thus the universe) gets less and less elegant, while Carroll stays the same. 🙂

    2. Have you recently been reading a Charlie Stross book? Because I met that exact simile while unwinding form the drive at Hili-awful o’clock this morning.

  7. I’d like to point out that, setting aside the “fine tuning therefore Jeebus” nonsense, there really is a valid fine tuning problem within scientific cosmology, for which we don’t know the correct naturalistic explanation.

    The only workable models that we have yet devised for our universe (string theory et al) seem to have a very large number of free parameters. In other words, there appear to be many possible physical laws that universes can have, and no apparent way to choose our specific universe without making arbitrary choices for these parameter values. If most of the possible universes are not compatible with life, then we have some explaining to do.

    Possible resolutions are:

    (1) Multiverse + anthropic principle. Huge variety of universes (possibly infinitely many) DO all exist. Anthropic principle explains why we see ours.

    (2) There is a better model for how universes are built, that does not have all these free parameters, in other words, the conditions in our universe “fall out” of the model quite naturally as the only valid solution. This is Sean Carroll’s #3.

    (3) Life is not as unlikely as it appears. This is Sean Carroll’s #1. Might be combined with a weaker form of (2) above, where a better model might just constrain things somewhat better, and allow a narrower range of possible universes.

    1. (1) Multiverse + anthropic principle. Huge variety of universes (possibly infinitely many) DO all exist. Anthropic principle explains why we see ours.

      (1.5) Anthropic principle + nothing else. If there is only one universe, and if it were not suitable for intelligent life, then the issue of why it is not suitable would not arise. So the mere existence of the issue presupposes its resolution.

      This is related to the difference between prior and posterior probabilities. The prior probability that a lone universe with randomly selected parameters is suitable for intelligent life might be very very small, but the posterior probability (given that this is an issue at all) is 1. From a prior probability point of view, it could be that we just got very very lucky, and we wouldn’t notice if we hadn’t.

      Some folks have used Bayes’ theorem to try to show me that this reasoning is unsound, but I didn’t get it. Anyone else want to give it a go? If this reasoning is unsound then I really would like to understand why.

      1. No, that’s not quite right. The problem arises because of the models we have so far. See it as a problem with the prediction of the current models, if you like. String theory seems to predict ~10^100 possible kinds of universe, most of them incompatible with life (probably, although see Sean #1).

        So to avoid a probability problem, you need all the universes to actually exist (Multiverse), then the Anthropic Principle is enough.

        The alternative is finding a better model that just doesn’t allow all the possible universes, doesn’t have all the free parameters, and says that any possible universe looks like our universe.

        1. <blockquoteSo to avoid a probability problem, you need all the universes to actually exist (Multiverse)….

          My (attempted) point was that there isn’t any prior probability problem to be avoided. No matter how vastly improbable an intelligent-life-suitable universe is, the anthropic principle ensures that either (1) we got vastly improbably lucky and have such a universe, or (2) nobody cares.

          1. The problem as I see it is that “we just got lucky” isn’t an explanation. Any conceivable state of affairs can be “explained” the same way: “That’s just the way it is.”

            To count as a good explanation, you have to explain why things are this way and not some other way. Anthropic arguments can’t do that unless all the other ways actually exist.

            1. I don’t think there is enough information to provide an “explanation”. We can only outline the possibilities. The way things are could be lucky, meaning it is improbable, or there are some constraints we don’t understand which limit how much things can differ from what we experience, in which case, it was not lucky but likely that we exist.

        2. This is a different problem though. I think Barbara is right that even if there were a very low prior probability that things ended up this way, that’s not an argument for God. They did end up this way. A somewhat stronger argument for God would be to say that there’s a zero probability things naturally happened the way they did, but I think the real problem is assigning prior probabilities to a scenario where we have no comparable scenarios to figure out what the prior probability should be.

          A rare event no long has a low probability after it has already occurred. Theism holds no advantage by simply declaring that the fine tuning has a low chance of occurring. The only thing they can derive from this is that we don’t have a good explanation for it yet; they don’t get to fill in specific versions of their preferred deity and call it a day.

      2. Bayesian probabilities.

        This is a huge simplification, but it should give the idea.

        All the models that we have come up with so far (such as string theory) seem to look like this:

        10^100 types of universe.
        1 type compatible with life.
        Random universe generator, all types equally likely.

        Bayes under this model:

        Priors for two hypotheses:
        H1: Generate One Universe, p=0.5
        H2: Generate Multiverse, p=0.5

        Data: At least one universe exists, with life

        Posterior probabilities (approx):
        H1: p=0
        H2: p=1


        Hence the argument – we need either a Multiverse, or a different model where life is less unlikely.

        1. Priors for two hypotheses:
          H1: Generate One Universe, p=0.5
          H2: Generate Multiverse, p=0.5

          To put it charitably, that is sheer speculation. Bayesean inference using only one observation datum (below) requires credible prior probabilities; otherwise one is severely prejudging the outcome.
          (If there are very many observations then the prior probabilities don’t matter much, but in such cases one is effectively using a maximum-liklihood approach rather than a Bayesean one.)

          [Datum]: At least one universe exists, with [intelligent] life
          [Conditional] probabilities (approx):
          [P(datum given H1)]=0
          [P(datum given H2)]=1

          By the anthropic principle, the probability of the datum is 1, regardless of the number of universes:
          if we’re doing these calculations then for sure there is intelligent(ish) life in some universe, no matter how many or how few.

          So P(datum given H1)=1. The datum provides no discriminitory power between H1 and H2.

          1. I’m scarcely an expert on Bayesian reasoning, but it seems to me all you’ve shown is that P(datum given datum) = 1.

          2. To be honest, I shy away from expressing it in Bayesian terms, because I don’t think it’s particularly helpful. I think Greg Kusnick’s comment above about finding a “good” explanations is a much better way to state it.

            But in any event, I don’t think your criticism of my crude Bayesian analysis is correct:

            If the stated model is assumed, then the single data point is highly informative, because a universe with life is so vanishingly unlikely under H1 and the stated model. Technically, the posterior probabilities are robust to the prior. No prior is going to shift those posterior probabilities unless you think that the prior probability of the Multiverse Hypothesis is less than 1 in 10^100.

            So, really, the assumption to challenge there is the model, not the prior.

            That’s just the issue for physics and cosmology: with all the work that has gone into string theory, we can’t find a model that doesn’t look like this, that doesn’t have the free parameters. We don’t have a model that naturally constrains the physical constants. That’s what physicists mean by the fine tuning problem.

    2. I would argue that there isn’t. Unless you have empirical evidence for what values the physical constants of the universe could attain. Show me the paper that describes, with evidence, what values the Plank constant can attain in universes. Show me in a paper where it describes – based on evidence – the set of discrete (or even analogy/continuous) values the constant can attain within that rannge.

      Do you have that? Until you do, you really have no idea how probable or improbable this universe is. You have a winning lottery ticket, but you don’t know the odds of winning. That’s not a sufficient amount of knowledge to say your odds of winning are so low that they pose a deep philosophical problem.

      1. Yes, agreed, but that’s option (2) in my comment. If we find a better model without free parameters that constrains the physical constants, then the fine tuning problem goes away. That’s exactly what the fine tuning problem means – we have not yet found such a model.

        1. You’re still making a mistake. Our theories do not say these parameters are free, they don’t say anything about their freedom or lack of it at all. The equation E=mc^2 doesn’t tell you that the constant C could have other values in other universes. It doesn’t’ tell you it can’t. It doesn’t tell you anything about c at all, except for a statement of what it’s observed to be. And the same thing is true for all our other equations and fundamental constants.

          You’re mistaking “no mathematical dependence on axioms of the theory” with “can be multiple values.” You cannot conclude the latter from the former; that is a logical error.

          1. I recommend the Andrei Linde article at the Edge link that I posted just above, since I’m obviously not expressing this well. He explains the motivation for the multiverse quite well. Snippet:

            “…since we know that the best of the theories developed so far allow about 10^500 different universes, anybody who argues that the universe must have same properties everywhere would have to prove that only one of these 10^500 universes is possible….Can we return back to the old picture of a single universe? Possibly, but in order to do it, three conditions should be met: One should invent a better cosmological theory, one should invent a better theory of fundamental interactions, and one should propose an alternative explanation for the miraculous coincidences.”

    3. Here’s Sean himself, talking physics:

      “..[O]ur observed universe is highly non-generic, and in the past it was even more non-generic, or “finely tuned.”….
      We find that inflation is very unlikely, in the sense that a negligibly small fraction of possible universes experience a period of inflation. On the other hand, our universe is unlikely, by exactly the same criterion. So the observable universe didn’t “just happen”; it is either picked out by some general principle, perhaps something to do with the wave function of the universe, or it’s generated dynamically by some process within a larger multiverse. And inflation might end up playing a crucial role in the story. We don’t know yet, but it’s important to lay out the options to help us find our way.”

    4. When we talk about “fine tuning” on websites such as this or in debates with WLC, we mean fine tuning as an argument for god.

      Of course we should work to discover why the universe is the way it is, but when we say “there’s no fine tuning problem” it means there’s nothing about the simple fact that the universe is the way it is that necessitates a frantic explanation like god. It’s just as Torbjörn writes above: the rock took the path it did, and the universe looks the way it looks.

      1. Well, of course I realize that.

        But the problem is, a lot of commenters on here seem to have misconstrued Sean’s excellent rhetoric when he’s debating WLC. Sean isn’t saying that there’s nothing interesting about the apparent fine tuning that we see, he’s not saying physicists are not working on it, he’s not saying “nothing to see here”. He’s saying “God did it” is not a good explanation. But it IS one of the biggest and most interesting open problems in modern cosmology.

        Perhaps it would be clearer if it were called “the apparent fine tuning problem”, since one of the proposed solutions is that there isn’t really fine tuning, i.e. Sean’s point #1. But physicists do just call it “the fine tuning problem”.

        Anyway, it seems that several commenters on here seem to think there’s really nothing to explain at all. That’s just not true – physicists think there is. See Sean’s comments that I posted above, when he’s talking about physics, rather than debating a theologian.

        1. Yes, I think that’s right. Many people here seem to be thinking only of ways to cut off the god inference. But almost any conceivable set of laws and constraints would not lead to intelligent life. So it seems to me there is an interesting problem, whose answer is probably the multiverse.

  8. I’d like to see theists take an a priori position on the question of which is more consistent with theism, a universe in which our planet is the only locus of life (or intelligent life) or a universe in which life (or intelligent life) is rife (relatively speaking). I’ve a feeling that, left to their own devices, theists could and would argue that one either way.

  9. Brilliant as usual, Carroll is eloquent, quick, and shattering.

    Cosmology gets hit too, not just evolution. As a physicists, I have to echo Weinberg’s thoughts: evolution is done and finished, there are no theistic holes, there never were since Darwin. Cosmology has some interesting anthropomorphic metaphysical kernels still smoldering, but Carroll and Krauss can reasonably smash them all.

  10. The most simple, eloquent and convincing argument against fine tuning I’ve ever heard is this: The solar system might harbour life other than on earth but it is, for the most part, uninhabitable.

    Ignoring the empty space between the planets, not much of each planet is hospitable to life anyway. Certainly, organisms can live at extremes but to say that the universe is fine tuned so that some as yet undiscovered bacteria can eke out a meagre existence on Mars is almost laughable.

    And this is to say nothing of the expanse of the galaxy, the Local Group, Laniakea, and the rest of the universe. Most of it is just barren space and what little specks of dust are to be found in between, they are hardly welcoming to life.

    99.999999999999999999999999999999999% of the universe is an uninhabitable mess.

    As Hitchens might be heard to exclaim, “THAT’S SOME FINE TUNING!!!”

  11. I suppose we can seriously consider that the universe was fine-tuned to make a lot of Hydrogen. The basis for that view is that the early universe probably steered like a juggernaut to make lots of H. This seems a more credible fine-tuning-argument-for-something argument than trying to spin that it was fine tuned for life.

    1. The Universe, as I understand it, is even more fine-tuned to make black holes than it is to make hydrogen…and, at long enough timescales even the black holes evaporate.

      What the Universe is really and truly fine-tuned for…is to turn everything into nothing.

      Then again, as Lawrence Krauss has observed, that exact type of nothing is itself quite unstable (given enough time), and will spontaneously create everything from it.

      So any talk of the Universe being fine-tuned for this or that is just so much meaningless babble. The Universe is what it is, and it’s up to us to make of it what we will.



  12. Awesome but he kind of contradicts himself when he says theism shouldn’t expect anything because theism is ill-defined, but then he rattles off a list of things we should expect from theism vs. things we should expect from naturalism.

  13. I read about this book on Sean’s blog and I can’t wait to read it.
    He never fails to blow my mind. One of “our side’s” most valuable
    players IMO.

  14. God as a First Cause.
    However long and erudite are the arguments, none can fully explain the “cause” that produced the “effect” of life. In other words “the cause” is inexplicable. So it could be given the word “God” to refer to it and no atheist can prove otherwise to such a naming. But this name “God” has been given to describe only something that is inexplicable: it does not entitle the theist to expand its original meaning of “inexplicable” to become such as a concrete belief in the detailed, personal God as explained and expressed in the precise terms of the words of the Bible, Quran, etc.
    The atheists’ true target is this unjustified leap from “God” (as meaning “the inexplicable”) magically changing into the omniscient “God” of an Established Religion.

    However here’s a thought, perhaps atheism is only acceptable to persons who live a fulfilling and comfortable existence, -for whom work, for example, is more akin to a hobby. Those whose lives are not so enjoyably rewarding, even acutely uncomfortable, then for them maybe some persuasive fiction like a loving Father with promise of a possible and enjoyable afterlife is a desirable, even “useful”, idea and a not so unreasonable (compensatory) way of life.

  15. I’ve never been bamboozled by fine tuning as it’s only interesting when viewed in the rear view mirror. Why does a protein fold that way? Because it must based on the characteristics of the materials that comprise it. There’s no magic here. Where is the ‘fine tuning’ in the distance of the sun from the earth? If the distance is much shorter or longer then life as we know it doesn’t exist. How is this fine tuning? Is it fine tuning that some individuals in a population are better suited for successful reproduction? No. Some are faster. Some slower. The slow gives way to the fast. Or fast to slow. No magic.

  16. Is Carroll not missing Douglas Adams’ puddle argument? That the laws of this universe produced this one life form that can think about its origins, but that with the physical constants tuned differently, other self-conscious life forms might have developed. That we are the necessary result of the settings we find, not that the constants were set so that we could exist.

    1. To be honest, I’ve never found the puddle argument very persuasive. Yes, we are the right shape for our hole, but it does not follow that any hole of arbitrary shape can contain a puddle. (Holes in vertical surfaces typically can’t.)

      Even if we define “hole” in such a way that holding water is a necessary property of holes (in this narrow technical sense), we’re still left with the question of why holes of that particular kind should exist in the first place. Adams doesn’t come close to answering that; he just assumes it as a given.

      1. Of course, you can overstretch (and overthink) any analogy. Addams speaks of one puddle in one hole. Just like we are one self-conscious life-form in one universe. His point is, I believe, that we think the universe was made for us, because we would not exist with different constants, missing the points that a) we are the way we are because they are tuned as they are, and b) other life-forms could have evolved with a different fine-tuning (who might think the universe was made for them).

      2. I don’t think Adams was even trying to answer this question. His puddle analogy always struck me as addressing Creationist arguments that it is amazing the Earth suits us so well, as if it were we, not the Earth, that appeared first.

        1. Fair enough, but if Adams isn’t addressing cosmological fine-tuning, then Carroll can’t be faulted for ignoring him.

Leave a Reply