Jul 20

Fine Tuning and a Beginningless Past

One of the most popular arguments for theism is the Fine Tuning Argument (FTA). The FTA is generally formulated something like this:

  1. The “constants” of the universe either arrived by chance or design
  2. The chance is overwhelmingly small
  3. Therefore, they must have arrived by design

There are a lot of things we could discuss about the FTA. It poses a particularly intriguing problem due to all the unknowns involved. However, I want to focus on a single aspect. If it is correct that the past is beginningless, then chance actually poses no problem.

The biggest hurdle to people accepting the possibility of a beginningless past, at least in my experience, are a strand of arguments regarding actual infinites. The thing about actual infinites is they grate against our intuitions as they are presented in the aforementioned argument. They are, in a sense, a completed infinite or an infinite set of things. This tends to bring about questions like, “How can an infinite ever be complete?” It’s a good and difficult question, but I hope to offer a different perspective on these arguments that may give you something to at least think about, even if you aren’t convinced.

So how might we respond to this difficulty? To start, the proponent of the FTA will often not have a problem with a potential infinite. They will likely grant the possibility of a potentially infinite future – namely, Heaven. I question whether this is really so different than asserting a beginningless past. One states that for any moment chosen in the future, there will be a later moment. The other says the same, but for the past. This symmetry can be seen in the following figure:

It does not seem clear at all why we should accept one, but not the other. Confronted with this, the proponents will often turn to another sort of argument that seemingly applies to past events.

So, what are these arguments? I’ll give a few varieties and you should see their basic form:

You cannot create an infinite set through successive addition (i.e., 1, 2, 3, 4 … ∞).

If you knock down the first of an infinite set of dominoes, you will never knock down the entire set.

If you start filling an infinite hole with an infinite amount of dirt and an infinite amount of time to shovel, you will never fill the hole.

The gist of the arguments is that we would never reach “now” given an infinite past. I initially found these arguments very persuasive. But after seeing them in a different light, I came to realize they don’t actually address the idea of a beginningless past head on enough to be convincing. You’ll see that these arguments all rely on starting somewhere. You begin adding, you begin the dominoes, and you begin filling the hole. But who has asserted that we are beginning anything? The dominoes have always been falling. If the concern is that we  can never reach the present, you can say, “Ok, choose any moment in the past and it will be countably far away from the present.” Their argument, though, doesn’t rely on counting from a past moment to this one – it relies on counting from the first past moment. But that is precisely what the beginningless past theory says we will not find.

I propose that the popular arguments against a beginningless past merely seem to have force because we aren’t framing the issue correctly. And, if a beginningless past is an option, what do we make of the “chance” problem presented in the FTA?

Similar Posts:


2 pings

Skip to comment form

  1. Ryan

    Is the implication of your argument that, given infinite time, our universe would necessarily come to exist because the probability of its existence would reach (or approach?) 100%? If so, I disagree because the possible universes are themselves infinite. But this does still mean that chance isn’t an issue, since we can no longer speak of the probability of our universe occurring. We can’t divide one kind of infinity by another. I’m sorry if I’ve misunderstood; infinity is a frustrating subject.

    I’m just not convinced by any theistic argument attempting to confront how we came to be, whether it’s this or the cosmological argument or something else. They’re really a waste of time because they contain so many questionable premises. Do we have a beginningless past or not? Is the concept of a Creator outside space, time, and cause and effect any less absurd than the concept of endless time with no Prime Cause? Does the answer even matter if we learn nothing important about the supposed Creator from any of this? We don’t have to take a position; it is perfectly acceptable to admit that we don’t know. But to admit that is to admit, if one is honest with himself, that atheism must trump theism in this case because it is theism that makes the positive claim.

  2. Mike


    I don’t think I would say it would necessarily come to exist (although, I can see that argument being made). Rather, I would say that the small chance is no longer a problem. Infinity is frustrating and I’m no expert myself, but I think I can make that particular point without discussing infinity. Let me try framing it a bit differently and see if my point comes out better.

    First, I think we need to recognize the unstated assumption that there is some natural universe creation mechanism. This exists also in (1) of the FTA, though, so it doesn’t seem like a big deal. Then, we figure that through this mechanism, the chance that certain constants would be what they are is very small, let’s make up a number and say the odds are 1 in 10^50 to make a universe like ours. Now, let’s define the time it would take for the mechanism to create a randomly delivered universe and call it “m” moments (m could be a second, a billion years, whatever it takes). So, for every m that passes, the mechanism creates some universe. Probably many of these do not support life and don’t make it very far.

    So, if we can say that the mechanism has existed for more than 10^50m, then the chance shouldn’t bother us. In the case of no beginning, then we can definitely say the mechanism has existed much longer than 10^50m. If only 1m or 1,000m or 1,000,000m had passed, then the chance might be reason for concern. But for the beginningless past, it’s no problem.

    If you roll a die three times in a row, the chances of rolling six all three times is approx. 0.5%. You might be surprised if you sat down once and rolled the die with a goal of three sixes in a row and got it on the first try. However, what if you did a million trials and it happened? Then, it wouldn’t be a big deal. That is essentially my point. It’s not that it’s necessary, but that it is no longer unlikely.

    I agree that it’s problematic for theists to try and draw these sorts of conclusions from the unkown, and I also agree that it’s perfectly fine to say we don’t know. I say it all the time. But these arguments, especially fine tuning and infinity arguments, present a lot of intuitive appeal, so I’m hoping to offer reasons why maybe that intuition is mistaken to avoid people giving the argument more credit than it’s due.

  3. andrew gray

    I would (and do) argue that existence is inevitable. The fact that we exist proves it.

    if you think of time as the measure of motion, or the measure of the increase of entropy, then before there was stuff, time had no meaning. All of your rolls of the dice happen at once. So once it is possible for a universe to exist, the probability of it existing is 1.

    Second,once you move far enough away from a universe such that no part of it can ever interact with you, and since time is relative, you’re pretty much back where you started. There’s no time where you are now, so as soon as a universe becomes possible where you are, it must exist.

    Repeat ad infinitum. I’m not convinced, though, that there is an infinite number of possible universes. If there is not, then inevitably a universe identical to this one in every possible way — down to the last quantum event — must exist somewhere else in space or time (not to mention all the universes that differ on the basis of one or more quantum events that went a different way).

    The question remains whether or not it makes sense to consider two identical universes as actually being different things. It makes more sense to me that there exists exactly one copy of every possible universe.

  4. Michael

    Here’s something to point out the absurdity of the artificial distinction between potential and absolute infinities:

    The universe began last Tuesday at approximately 10am and is extending in both the forward and backward directions from there, according to rules of physics which are impacted by the direction of flow. There’s nothing mathematically distinct between this model and the model with an infinitely long ago past, but this one is only a potential infinity, not an absolute one.

    Additionally, the set { … , -2, -1, 0, 1, 2, … } is the same as the set { 0, 1, -1, 2, -2, … }, despite one having a clearly defined ‘first’ element. Which brings me to the key point: the difference between potential and absolute infinities isn’t a difference in the structure of the claim, it’s a difference in how we feel about the words involved.

  5. Mike

    Andrew, is it inevitable or just really ridiculously highly probable that we will exist? I don’t mean us in particular because we obviously do exist, but I mean a universe just like ours.

    I’m interested in the things you say about time too. I’ve got some reading planned on how physics now shapes our understanding of time, but haven’t gotten to it yet.

  6. Ryan


    “Repeat ad infinitum. I’m not convinced, though, that there is an infinite number of possible universes.”

    A universe with even just our particular laws of physics could probably have infinite variations. In this universe, I am typing a response to you; in another, everything might have been identical up until this point, but I choose to not respond to you. Now, if you think that the nature of the Prime Cause determines all consequent events and that we have no free will, perhaps this isn’t convincing. But you could still imagine infinite universes arising from infinite variations of the Prime Cause.

    Thus I think the existence of our universe is neither inevitable nor highly probable because probability loses all of its meaning when you are dealing with infinite time and infinite possible universes.


    I wonder if those who use this argument ever believe that the past is beginningless. Those I have seen use the argument seem to assume the universe had just this “one chance” to come into existence, which is precisely what gives their probability argument its weight–but also weakens it because it is an unprovable premise. To discuss whether or not there was only one chance, you would have to define the mechanism by which the universe came to be. I think we end up at the cosmological argument again because it deals with perhaps the most basic of premises, which are buried in the FTA.

  7. Mike

    Ryan, I would assume you’re correct that most people who use the FTA would not believe in a beginningless past. If they have done much studying, their reasons for denying it will likely be the types of reasons given by William Lane Craig (those arguments I pointed out above). So, going into the conversation they may not grant a beginningless past, but they can’t rule it out because their objections don’t work like I hoped to show. It might go like this:

    Theist: Actual infinites cannot exist in reality.
    Atheist: Why not?
    Theist: Because you would never get to now.
    Atheist: Get to now from where/when?

    Now, what do they say? They can’t give a moment in the past because you could get to now from a defined moment. And they can’t say the beginning of the infinite set because that doesn’t make sense.

  8. Ryan

    They might not have an answer to that question, but they might point out that accepting an infinity in reality even though we cannot really understand it opens up the door to all manner of other concepts beyond our reasoning, including God. This would be more of an attack on a strictly logical view of the universe rather than a scientific one, of course, but the two are often conflated and, in this case, we’re not dealing with empirical data anyway.

  9. andrew

    @michael: Inevitable, the way I see it. If time has effectively stopped, then all probabilities are either 0 (can’t possibly happen) or 1 (must happen). Probablility is a statement of time — given x chance per Y amount of time, how long would it take to become probable that something would exist. But if there is no time, then you either get all your dice rolls at once or you get no dice rolls at all. Once a universe is possible — that is, once the chance of its existence becomes nonzero, it exists. Until it exists, it wasn’t possible.

    @ryan: Our universe has a finite number of particles, so I believe it must have a finite number of possible configurations or states of existence, and a finite number of paths through them.

    I should point out though that I’m not a scientist or mathematician, just taking a philosophical approach. Reality could prove me wrong.

  10. Ryan


    I suppose it all comes down to whether or not the particles in our current universe represent all possible particles in the first place. If so, then I agree with you. But it’s hard to say without being certain about the nature of the Prime Cause.

  11. Chris


    It seems like you and Mike might be proposing different things. In your first comment, you mentioned “before there was stuff” but Mike is talking about a beginningless past. Do you think that time began to exist? I’m not sure what to make of time having no meaning. I’ve read Mike talk about the Kalam Cosmological Argument in a few posts and I think he wants to argue that time did not begin to exist as a rebuttal to the KCA.

    Although, it doesn’t seem like he’s not necessarily arguing for the correctness of any one position, just that the infinity critiques are a sort of straw man. I haven’t really thought about it before, but that actually seems on the money.

  12. Mike

    Chris, I can’t speak for Andrew but that’s about right for me. I want to keep the possibility of actual infinites open for a counter to both the KCA and FTA.

  13. Josh

    I wonder what prevents your argument from generalizing in unwanted ways. For example, suppose we discovered a a star cluster that was arranged (from our perspective here) into the exact pattern of the Hebrew words in Genesis. Couldn’t we reason that this pattern is less likely to occur within our view on design than on chance, whether or not the universe is beginningless? If so, I wonder what makes this case different from the fine-tuning case. I also wonder if the considerations on probability given by Roger White in connection to FTAs applies here…

  14. Josh

    I meant: “Couldn’t we reason that this pattern is more likely to occur within our view on design than on chance, whether or not the universe is beginningless?”

  15. Mike


    I’ve downloaded Roger White’s article Fine Tuning and Multiple Universes from Nous (2000). Is that what you are referencing? I’ll have to read it and see what I think because I don’t know his argument offhand.

    For your other point, are you saying you think we should be concluding that the star pattern is design or we should not?

  16. Josh

    I’m thinking the star pattern would provide evidence of design. Another example would be if we were in an infinite desert with bugs that make random carvings and we came across “come look for me —>” scratched into the sand. The probability of its existence somewhere on the desert on chance may be 1, yet we may think its existence in any particular region is more likely due to design.

  17. Mike

    I see what you mean. It does seem to be a concern that the force of infinity could overcome any odds. I too would want to be able to say that, if we found a star pattern as described, then it should be considered positive evidence. The obvious answer would be to say that maybe that’s just the way it is and too bad if we don’t like it, but that isn’t very satisfying.

    If the number of possible universes like our own is finite, then I think I would have a pretty good answer even if the past was beginningless. I’m not sure if that’s the case (it seems potentially infinite). But maybe we only need to be concerned with certain characteristics (not what t-shirt is john wearing) and those could be finite. Then, it seems like we would still need enough occurances within each possible world to give us a very low probability.

  18. Mike

    Actually, see what you think if I put it this way:

    I don’t want to necessarily say that a beginningless past automatically makes occurrences within each universe likely by the same virtue. It seems like there is a probability that the particular universe exists, and then there are further probabilities derived within each universe based on how said universe behaves, etc. So, just because I’m saying that, given a beginningless past, we don’t have to be surprised that an improbable universe can arise doesn’t mean I’m not saying that things within that universe cannot still be considered improbable and judged from a within-universe perspective. So, we could still say that the rise of intelligent humans through natural means is very unlikely and we can be surprised by it and think it requires explanation.

    It seems appropriate to judge universe creation from an extra-universe perspective and internal events from an intra-universe perspective. Do you agree?

  19. Josh


    I’m not sure. But those are good thoughts worth considering!

  20. Mike


    I got around to reading White’s paper. It was very good. Thanks for the suggestion!

    I think his arguments could be relevant to extensions of my claims, but I also think his targets in the paper are going further than I am. I think he hits it on the head in his last section on Improbable versus Surprising. I don’t want to say that a long or infinite past makes the universe less improbable, but I do want to say that it isn’t surprising to see a universe like ours, if such a thing is true. That’s really my goal here because I don’t at the moment have any good argument in favor of a beginningless past. I’m just trying to rescue it’s possibility from counter-arguments and then say that, if it does happen to be true, then I don’t think FTAs are a problem.

    I think that distinction protects me from the inverse gambler’s fallacy noted by White and proposed by Hacking because I am not saying that the multiverse or an infinite past is more likely because of the fine tuning. I also tend to fall into the camp that he mentions early on, but does not cover. I think it’s very hard to give an accurate idea of probability for the initial conditions, etc.

    In the end, this post is probably mis-named. My real intent is to show that arguments against an actual infinite are actually arguments that a potential infinite cannot become an actual infinite, thus, don’t really compel against actual infinites. That just happens to have some implications for the FTA or KCA.

    As an aside, I think you might be interested in this interview from the Common Sense Atheism Podcast: http://commonsenseatheism.com/?p=8109

    It covers some of the common arguments against Fine Tuning with astronomer Luke Barnes and is mostly about why they aren’t very good objections to the FTA. It’s one of the best discussions of the subject I’ve heard and certainly gives me pause about some things I used to think were good objections.

    Thanks for the comments.

    [Later Edit] Leslie’s argument that fine tuning makes the multiverse more likely also reminds me of evolution. Darwin’s theory needed a really old Earth to make it work (and genetics). So, we really could have said the same thing because no one, as I understand it, really thought the Earth was older than perhaps a few milion years (with most probably going with thousands). But we did find an older age. I don’t know if that will ever happen regarding spacetime beyond our universe, but if that’s the case, I think we have a good reason to think our hurtle is no longer so big with respect to fine tuning.

  1. Current Thoughts on the Kalam Cosmological Argument | Foxhole Atheism

    […] my thoughts on the philosophical perspective, you can read my previous article here or philosopher Wes Morriston on the subject here and here. There are several more papers by both […]

  2. God and Time: A Dilemma | Foxhole Atheism

    […] If you think that time has always existed, and God has existed within time, then you actually may be helping the atheist case. Two of the most popular arguments in favor of God’s existence—the Kalam (or other cosmological arguments) and the Fine Tuning Argument—rest upon an assumption that the past is not infinite. In the case of the Kalam, it’s pretty easy to see this because one of the premises states, “The Universe began to exist.” Obviously, if the Universe is infinitely old, this is false. The objection to the Fine Tuning Argument takes a bit more explanation. This is an extension of the point I made in the comment section of this article. […]

Leave a Reply

Your email address will not be published. Required fields are marked *

You may use these HTML tags and attributes: <a href="" title=""> <abbr title=""> <acronym title=""> <b> <blockquote cite=""> <cite> <code> <del datetime=""> <em> <i> <q cite=""> <s> <strike> <strong>

Notify me of followup comments via e-mail. You can also subscribe without commenting.