Suppose that there was a theory that explained why certain physical constants are the way that they are, the same constants that purportedly seem to be “fine tuned” for life in the sense that if they were any different, life would cease to materialize.

Now, in a debate between Robin Collins (a primary proponent of this argument) and Peter Millican, which can be accessed here, Collins claims that even with such a theory, it would not solve the fine tuning problem. He claims that it would beg the question of why there is a theory in the first place that creates constants destined to be suitable for life instead of many other imaginable theories that wouldn’t produce these constants. In a sense, this would still be improbable.

He writes in his paper,

Besides being entirely speculative, the problem with postulating such a law is that it simply moves the improbability of the fine-tuning up one level, to that of the postulated physical law itself. Under this hypothesis, what is improbable is that all the conceivable fundamental physical laws there could be, the universe just happens to have the one that constrains the parameters of physics in a life-permitting way. Thus, trying to explain the fine-tuning by postulating this sort of fundamental law is like trying to explain why the pattern of rocks below a cliff spell "Welcome to the mountains Robin Collins" by postulating that an earthquake occurred and that all the rocks on the cliff face were arranged in just the right configuration to fall into the pattern in question. Clearly this explanation merely transfers the improbability up one level, since now it seems enormously improbable that of all the possible configurations the rocks could be in on the cliff face, they are in the one which results in the pattern "Welcome to the mountains Robin Collins."

I’m having trouble understanding this objection or at the very least having trouble understanding whether this contention is legitimate. First, doesn’t the notion of what’s improbable depend upon a probability space? If so, what are the rules for creating this probability space? I fail to see how these rules can be objective given our complete lack of knowledge of what theories are possible or not. In the case of ricks, other formations don’t violate known physical laws. In the case of a correct theory, other theories would by definition not be possible…since they wouldn’t be the correct theory.

Second, cannot a person raise the same contention against God? If God designed the constants for life, one can ask: why did God not design a universe using any of the other constants that aren’t designed for life? Of course, a natural answer to this question may be to say that God cares about us or that in His very nature is propelled to create constants that are suited for life. But then why can’t one just say that the universe is destined to have constants that lead to life by its very nature? This might seem unintuitive but I fail to see how this explanation is any worse than God. In fact, I think one can make a similar argument to address his rock example. Why couldn’t nature just happened to be set in a way such that it produced a formation saying “Hi Robin Collins.” given that this actually happened even if the probability of this happening is very low apriori. Why is this naturalistic explanation any less plausible than a God happening to exist that was ultimately also set to create that rock formation instead of all the other states of affairs reality could have been ordained as?

Furthermore, there are other questions that one may ask about God that can also introduce an improbability. For example, why did God not design the universe in a way where He chose another set of values to be the special ones that lead to life? Lastly, why does God need to fine tune in the first place?

Ultimately, Robin Collins states that any naturalistic explanation of fine tuning does not resolve the surprisingness of the coincidence of fine tuning, but that God does. Would he be correct?

  • Related.
    – Mark
    Commented Mar 29 at 1:49
  • At this point, I would find the existence of God very surprising. What are the chances that all those crazy-sounding people and seemingly random arrangements of words in books were actually right? Would a correct theory of God solve the problem of believing something we had no evidence for?
    – Scott Rowe
    Commented Mar 29 at 2:25
  • Why tuning and not fine-tuning as the tag? It looks like the only other user who has ever used the tag missed the opportunity of wording it correctly.
    – Mark
    Commented Mar 29 at 5:19
  • I changed the tag to intelligent design Commented Mar 29 at 12:11
  • TVs used to have a "Fine Tuning" knob on them. See? It's built in to the universe. TV snow was said to be the background noise left over from the Big Bang. You could see the evidence of it right there in your living room.
    – Scott Rowe
    Commented Mar 30 at 2:23

2 Answers 2


"Fine tuning" is questionable

Range and distribution of values

I've heard apologists argue something like the following: We have a physical constant 123456789. If this constant were 123456788 or 123456790, reality as we know it wouldn't be the same. Therefore that's a 1/123456789 probability of having this particular constant.

This would be like taking some essay written by a university student, measuring the length of the essay in millimeters when you put all the characters next to one another in some font size, and saying the probability of getting that essay is 1 over that.

It's saying that the possible range of values is anywhere between 1 and that value, and each of those are exactly equally likely, which is highly questionable in both cases. For university essays, there are probably word or character requirements and standards that would greatly affect essay length, you can look at some distribution of lengths to see how likely different lengths are, and millimeters is completely the wrong scale to measure that at. For physical constants, we generally have no idea what the actual range of possible values nor the distribution is, so asserting some range and distribution is pretty much just pure speculation. It might be that the only possibility is for it to be that very specific value (or there might be an even bigger range). Since fine-tuning relies on the physical constants being unlikely, the argument would be invalid if we don't actually know the range or distribution.

* Collins seems to go quite a few steps beyond this (in the wrong direction), and says "any number between 0 and infinity".

If things were different, they'd be different

Another problem is that if the universe were different... the universe would be different. Well, yeah, duh. Someone might say e.g. "carbon-based life can't survive under X circumstances", but how do they know there can't be some other form of life under those circumstances? If those were the circumstances and there was, I don't know, hydrogen-based life, the hydrogen-based lifeforms would probably also be saying "hydrogen-based can't survive under Y circumstances", and some hydrogen-based skeptic might be pointing out the possibility of carbon-based life. And if no life were possible, there'd be no-one to say "we can't survive under these circumstances", so such a universe would come and go without anyone's notice. How many such universes has come and gone? Or maybe we just got lucky?

Collins does address this by saying if the strong nuclear force were different, only hydrogen could exist. But, still, the problem with the argument remains, as this takes how atoms look like in our universe, and projects that onto that universe. But we really have no idea what such a universe would look like.

He also responds to atheists saying "fine-tuning is not really improbable or surprising at all under atheism, but simply follows from the fact that we exist", by replacing "fine tuning" with "our existence", but... our existence also isn't improbable from the fact that we exist. I think he entirely missed the point of that objection, because his response doesn't seem to do anything about the objection from atheists.

We don't know how likely God is

Yet another problem is that we have absolutely no idea how likely God is, and we need to know that for this argument to work, so that renders invalid any sort of "argument from improbability", no matter how improbable. There's something we don't know, and a theist comes along and just slaps a "God did it" sticker on it. How did God do it? Why did God do it? Why did God do it in this particular way? What evidence do we have for any of this? How do you know it wasn't any other being? And how do you link any of this to any sort of personal god? *Shrug*... "but atheists don't have a better explanation"... at least until they do, and then theists just slap their "God did it" sticker onto something else. God doesn't "solve" the fine tuning problem - it's just an ad-hoc explanation that doesn't do much explaining.

Collins is very specifically comparing how probable things are under both theism and atheism, so this is highly relevant (because you also need to consider the probabilities of both of those for that comparison to be valid).

Curiously, he responds to "[God is complex], who designed God", which is a response to "Complex things need a designer" (which is a different argument). But he fails to address the obvious parallel of "God is improbable, who designed God", and instead just says that's about complexity so it doesn't apply here.

To provide a brief analogy of the problem with ad-hoc explanations, let's say I've had a bunch of bad Tuesdays in a row. Now I could say this is the result of the complex interactions of thousands to billions of people, and possibly also some complex biological interactions. From all of that, it seems really unlikely that those specific Tuesdays would've been bad in those specific ways, right? So, instead, I might propose a "bad Tuesday" goblin, which specifically makes sure that I have bad Tuesdays. Under that hypothesis, my bad Tuesdays is extremely likely. So... does a "bad Tuesday" goblin exist? No, that seems unlikely, given that it's entirely an ad-hoc explanation that "fits perfectly" onto whatever evidence there is. God is like that, but on steroids, as one can explain pretty much anything with some God who specifically cares about humans, but who will also test us, and everything has some ultimately purpose. I've heard theists attribute pretty much anything to God.

There may also be some other external forces that could potentially have affected the physical constants, like a multiverse or some other being, or some other physical force that we just don't understand yet.

Collins' argument is ... uh... not good

Side note: He starts off with "besides being entirely speculative" and I'm just screaming at my monitor, "No, that's the problem with what you're proposing!" ... as I argued above. Theists need their speculation to be right for fine tuning to make sense as an argument, whereas atheists can challenge that by merely pointing out that we don't know what the range or distribution is. Imagine there's a jar of gumballs, and someone comes along and says, "there are exactly 534 gumballs in that jar". You respond, "uhh... I'm not so sure about that. Why do you think that?". They respond, "you're just speculating that there aren't 534 gumballs in the jar!". That's the sense I get from the above response.

Anyway, as I argued above, the argument would be invalid if we don't actually know the range or distribution of the physical constants (among other problems).

Collins kind of concedes that point, but also kind of tries to argue against it, by shifting the focus to whatever made those constants be what they are.

The whole point of fine tuning is that those constants are really unlikely (a questionable claim, as I pointed out). So if we find something that shows these constants are an inevitable result of physical circumstances, the entire argument falls apart.

It's not clear whether he misunderstands this to mean that there's yet another physical law that's subject to similar probabilities, whether he concedes that the probabilities may indeed be a miniscule fraction of what "fine tuning" traditionally proposes (even though the supposed strength of fine tuning are those dubious extreme improbabilities), or whether he's just saying "what if physics itself wasn't what it is", at which point this "what if" game has gotten way too far past the point of reality and reason.

In any case, (in my opinion) Collins' argument seems to betray his intent - he has no interest in conceding even a hypothetical where God doesn't exist. He "knows" that God exist, with no doubt, so every path must lead to God existing. For him, God will and must always be at the edge of our knowledge. I linked to "God of the gaps" above, and this paper is a live demonstration of that - he slapped the "God did it" sticker on the physical constants being what they are, and even if we find some natural explanation for why those physical constants are what they are, he says he'll just slap the "God did it" sticker onto that explanation.

I don't imagine that there exists even any hypothetical evidence that could convince him that God doesn't exist. Or, at least, he believes that God exists for reasons that has nothing to do with fine-tuning, but he wants some way to rationalise his belief, so he'll hang on to fine-tuning as long as possible.

He also uses merely the standard of "conceivable" to base a lot of his argument on, whereas I'm much more concerned with and interested in what's actually possible. Speculating about what's conceivable - that's called fiction.

Side note 2: He also seems to conflate "theism" with his particular belief in God ("since God is an all good being ... Thus ... under theism"), which subtly circumvents the problem of the possibility of other religions.

  • "but atheists don't have a better explanation"... at least until they do, and then theists just slap their "God did it" sticker onto something else." - - - Theists did try that with the Higg's Boson... remember the term "God Particle". It didn't really stick. I think they are running out of stickers, and probably won't have any left to put on the final answer. Commented Mar 29 at 5:16
  • What Collins, and I suspect many theists and most atheists, don’t realize is that God Himself would have to either have His decisions be random or His decisions be determined. So we’re left with either a brute coincidence or the world as it is being necessary. But if our current state of affairs such as the constants is a brute coincidence or necessary, one does not need God in this picture. If God makes random decisions, one can posit that our constants are random without God. If God’s decision to make our constants is determined, then one can posit our constants are necessary without God Commented Mar 29 at 13:18

Is Robin Collins correct that a Theory of Everything doesn’t solve the fine tuning problem?

Fact 1: Robin Collins does not know the Theory of Everything, or The Answer to Everything.

Fact 2: Robin Collins does not know the attributes or properties or characteristics of the Theory of Everything, or the Answer to Everything.

Fact 3: Therefore, his claims are not supportable, not justifiable. His opinion is under-informed.

  • 1
    We absolutely do know some attributes of a Theory of Everything, though. If it does not unify physics, it's not a Theory of Everything, by definition! We can say that with confidence despite not yet having a Theory of Everything.
    – Michael W.
    Commented Mar 29 at 16:24
  • 1
    Be specific. I may know all of the attributes of the final answer to everything. Humanity collectively, excluding myself, does not. You yourself do not. And none of those are what I said.... What I said is "Robin Collins does not know". Being specific matters. "a" versus "the" kinda deal. Like from the beginning of Corner Gas... with the guy from the IRS. Brett calls him "the taxman". The taxman complains and says "I am A taxman, not THE taxman". IT becomes a running joke throughout that episode. My statement stands as factually correct... Robin Collins does not know. Commented Mar 29 at 17:21
  • If you require all attributes to be known before you can describe a thing, then we cannot actually describe anything. This is silly on it's face.
    – Michael W.
    Commented Apr 1 at 16:20
  • @MichaelW.we arent assessing "a thing"... we are sssessing whether we know enough about the not yet known Theory of Everything to allow the not yet known Theory of Everything be used as evidence for or against a fine tuning argument. Commented Apr 1 at 18:37
  • A theory is a thing.
    – Michael W.
    Commented Apr 2 at 15:22

You must log in to answer this question.

Not the answer you're looking for? Browse other questions tagged .