חדש באתר: עוזר בינה מלאכותית המבוסס על כתביו ושיעוריו של הרב מיכאל אברהם

Q&A: God Plays Dice, The Improbable Mountain

Back to list  |  🌐 עברית  |  ℹ About
Originally published:
This is an English translation (via GPT-5.4). Read the original Hebrew version.

God Plays Dice, The Improbable Mountain

Question

Hello Rabbi, I’m stuck in your book between pages 177 and 191. I’m not managing to understand why we need to add the time axis here. How can’t one talk about “the probability of something coming into being”? After all, that’s what we did on the previous pages when we talked about the probability of a “living” codon forming in the ocean, and that was without needing the time axis. Also in the example you give on page 180 with the die, there’s no need for the time axis. I’d be happy if you could clarify a bit the claims about the “breakdown into stages.” Intuitively, that דווקא seems understandable. Many times people look at an exceptional case, but when you understand the chain of events step by step it becomes more plausible. Thank you very much

Answer

I explained it there well.
The probability of something coming into being is always per unit of time. The probability of getting a result of 5 on a die depends on how many throws you make per second/minute. The probability of a codon forming depends on the question of how many attempts are made per second.
I also explained the breakdown. Probability has nothing to do with understanding. The breakdown does indeed help one understand what happened and how it happened. But the probability is the product of the probabilities of the different stages. If you throw a die a hundred times and get a sequence of numbers: 1,2,3,4,5,6,1,2,3,…
Does breaking up the chain and discussing each stage separately change the probability? The probability is the product of the probabilities of each and every stage.

Discussion on Answer

Sam (2019-06-23)

That helped me, thanks

Regarding the first point, I didn’t understand what the problem is with simply saying that the probability of getting a result of 5 is 1/6.
Just like in the example you brought about a sequence of sixes, there’s no time axis there. We talked about probabilities without a time axis.

Michi (2019-06-23)

There are probabilities that don’t depend on time, but the probability of something coming into being always depends on time. Even when I ask you what the probability is of getting a 5 on a die, the answer is not 1/6. That is the probability of getting a 5 in one throw. But will there even be a throw? And how many such throws will there be? And over how much time? The probability is per throw, that is, per unit of time or per action.

Sam (2019-06-24)

Hello Rabbi,
I went over the material again and it really is clearer, but I’d still like to sharpen the point a little more.
You say that the probability of something coming into being depends on two things: 1. the duration of time, 2. the rate of formation per unit time.
On page 180 you say that it depends on two criteria that indeed do not change (the amount of genetic information and the age of the world). So there’s no difference whether we break it down or not. But that’s not the whole story. What determines this discussion is not only the basic data, but also the rate per unit time. That’s what the slope of the graph depends on.
After that you give the example of a mountain a thousand kilometers high and compare breaking it down versus not breaking it down. But of course that won’t change anything, because it’s the same rate either way: one kilometer per second.
And here, it seems to me, is the critical point. The question is the question of the rate. Dawkins’ claim is that treating the process as broken into stages increases the probability, and with it the rate as well. Meaning: all at once it’s unlikely to happen, so the rate would be very slow; but if you divide the process, it’s more likely to happen and so the rate is faster.
Here you come and argue: no, the division into stages does not significantly change the rate; at most it actually does the opposite (p. 188).
The critical point in this discussion (without taking into account the claim on page 189) is whether the rate—which is what changes the probabilistic slope—changes because we treat the process as divided into stages.
What undermines Dawkins’ claim is not that we’re talking about the same time span (because that’s obvious; without that there’s nothing to compare), but rather that 1. breaking it into stages does not change the probability, and then the rate is the same rate.
2. In our case it even makes the probability harder.
The important assumption is that the breakdown does not change the probability, because otherwise it would change the rate.
Thank you very much

Neil Jordan (2019-06-25)

Here’s a little experiment you can do.

Take three dice. Roll the first until it lands on 6. When that happens, set it aside and start rolling the second die, until it too lands on 6. Once that happens, set it aside and roll the third until it too comes up 6. There—you now have three fair dice showing 6.

Did this take about 216 rolls, which is the expected time for rolling all three dice together until they all come up 6? What if we used 100 dice—wouldn’t it be obvious that we’d need only about 600 rolls for all hundred to show 6, as opposed to six to the hundredth power?

The theory of evolution talks about the *fixation* of traits over time. That is much more like rolling one die at a time and leaving it on the “right” number once you got it, than rolling all the dice together every time.

Neil Jordan (2019-06-25)

(Or alternatively, it’s even more like rolling all the dice together but continuing to roll only those that are not showing the optimal number—that would be even faster than the process described above.)

Michi (2019-06-25)

I answered this common (and mistaken) claim in great detail.

Neil Jordan (2019-06-25)

I’d be happy for a reference to that, because I’m not familiar with any answer on the subject.

It’s obviously not in the relevant chapter of your book—which I just went through again to verify the point. There you only mention at the beginning of the chapter Hoyle’s mistake—which is, of course, the assumption of independence and uniform distribution—and then throughout the rest of the chapter you go on answering other issues and raising arguments that do not deal at all with the fixation I mentioned, or with Hoyle’s assumptions (which are mistaken by every view). Then on page 205 you simply state, “This is despite the fact that, as we have seen, his claim is entirely correct, and all the flaws found in it are the result of atheists confusing the scientific plane with the philosophical plane.” That is without actually having answered the criticisms raised against Hoyle or addressed the subject of my response, about which you say you already answered, and at length.

So as I said, in light of that, I’d appreciate a page reference (or a blog post) where you addressed the fact that fixation of traits completely destroys your dice analogy, as well as the ridiculous probabilities Hoyle derived from assumptions that no one in evolutionary biology ever claimed.

Michi (2019-06-25)

You can see a direct discussion of it here:

מבט שיטתי על יחסי אבולוציה ואמונה

Neil Jordan (2019-06-25)

I read the post in question. To sum up, there you are forced to accept that given a selecting factor and a stabilizing factor, there is indeed no substance to the claim that you yourself still make here—that the probabilities are still very low—and instead you only argue there that this does not destroy your physico-theological argument.

So first of all, fine, it doesn’t destroy it—but it still does destroy the answer you gave here (and in the book). It shows that there is no conceptual barrier here to evolution producing order and complexity, merely through replicating creatures and an external environment that imposes different selection conditions on them. It completely destroys everything you wrote here regarding the rate not changing in climbing the improbable mountain. That is very far from what was presented—a response showing that this is a mistaken claim.

(As an aside, it also spoils your claim that this very improbability is what constitutes evidence for the need for some designer who intentionally brings things about. If it’s good enough for you to replace that with “but who created the laws of reality,” then fine—but at least let’s not pretend that the old argument still stands as is, that Hoyle was not mistaken, or that this is a “common (and mistaken) claim.”)

I won’t wait for you to retract, and on the other hand I don’t expect you’ll manage to come up with a new reference where you actually show the claim is mistaken (as opposed to admitting that it’s a valid objection but that you can build the broader argument around it); so this will probably end here. But I hope the questioner at least got a good answer to his question.

K (2019-06-25)

Neil, I don’t understand this smugness. He writes there explicitly that he moves to an argument above the laws.
Also, he mentions in passing that despite this, there are calculations showing that the resulting probability is still very low, but we won’t need them here. This objection can be rejected even without them.

Neil Jordan (2019-06-25)

By all means, K, if so I’d be happy for a reference to those calculations. That only means the reference given dodges the question.

K (2019-06-25)

You’re welcome to ask him. I remember that at some point it was said who calculated it, but I don’t even know where to look for it.
By the way, I’m curious to know whether you believe, and if so, why? Because from what you’re saying it sounds like you also don’t accept any proof of God, and you’re also a determinist, etc.

Neil Jordan (2019-06-25)

Well, I did ask him—the reference he gave did not lead to the expected answer, but rather to the answer that there exists an answer…

Determinism does not contradict belief in God, and specifically I’m criticizing the claim that evolution has difficulty arriving by natural means (already known to us) at the existing diversity of species. As someone who studied biology, that bothers me. But as the Rabbi himself would also tell you, there is no real connection between that and belief in God.

K (2019-06-25)

I meant that he told me who the researcher was who studied it, I just forgot where that was said—not that he said an answer exists, etc.
If so, what kind of proofs do you accept?
Of course determinism doesn’t contradict belief in God, but statistically it still shows that the person standing opposite you is probably also a materialist atheist.

Neil Jordan (2019-06-26)

It seems to me that this is too broad a topic for something so complex, and that it doesn’t belong in the Q&A of another topic.

I do accept the point about the statistical correlation.

K (2019-06-26)

Neil, there used to be a forum here on the site—do you know where it disappeared to?

Neil Jordan (2019-06-26)

I’m not familiar with it. I only started writing here recently.

Michi (2019-06-26)

Hello Neil.
If you want to insist—that is of course your right. I explained there very clearly the physico-theological argument and the probabilities that underlie it (those outside the laws). I explained why the calculation within the laws is unimportant, and if I remember correctly it is done by some Dutch biologist whom I cited in the book. But you prefer to cling to this after it became clear to you that your claims against the physico-theological argument are unfounded.
What I especially liked was the passive-aggressive remark that you don’t expect me to retract (even though you didn’t raise a shred of argument against my words). More power to you! Poland and all Polishness are behind you.

Neil Jordan (2019-06-26)

First, I apologize if the tone I took was smug. It’s just that in light of a claim that can’t be defended, and a reference that didn’t even try to defend it, it was hard to take the matter seriously. Your last reply also doesn’t help, and for one simple reason: my claim was not against the physico-theological argument at all, and I made that clear.

No. My claim was meant to challenge only what was presented here in this specific Q&A: that the dice argument demonstrates some problem in the accepted evolutionary position according to which evolution is the source of complexity in the living world. It does not. A situation in which there is heredity, selection, and fixation is enough for the resulting outcome to be expected. You still have no answer to that—not here, and not in the post you linked to (because there you only deal with the impact of this counter-argument on the physico-theological argument, as opposed to the validity of the dice argument that you again brought here).

If the calculation within the laws is unimportant, then why do you keep repeating the mistaken claim, even after you know of the demonstration that it is mistaken? Why repeat the example of the dice when we know it is not parallel to the accepted conception of evolution? Why continue to claim that Hoyle was right when we know the assumptions underlying his calculation are not accepted by biologists? Why once again declare that the breakdown into stages does not affect the probability, when the mechanisms are known that make this clearly false?

The reference to “some Dutch biologist” appearing somewhere in the book is very vague. At first I thought perhaps it referred to the footnote on page 144 about the Belgian Christian de Duve, but apparently not, because de Duve’s writings do not support that at all. So I ask for the third time for a reference to something that would show that even given fixation, there should still be a low probability for the formation of life’s complexity / the accumulation of the number of mutations; I did not find this in the book. At the very least, an admission that your original argument (as well as Hoyle’s) misrepresents evolutionary theory would be in order.

Michi (2019-06-26)

I was not speaking now or previously about smugness. I was speaking about the repeated insistence on not addressing my arguments.

I do indeed use the dice example again and again, and I explained very clearly why it is similar to evolution. Evolution only describes the holder of the dice. On the contrary, the demonstrations you brought are irrelevant to my argument. But again and again you ignore my explanation. Breaking it down into stages does not help at all on the probabilistic level. Only the involvement of the programmer / dice-thrower changes the picture (perhaps).
And in parentheses, yes, I did mean de Duve, but as I said, that really is not important. De Duve pointed out that without constraints imposed from outside there is no chance of the spontaneous formation of life, nor of its evolutionary development. That is exactly what moves me from an argument within the laws to an argument outside them. See about this in Iris Fry’s book, The Emergence of Life on Earth, p. 156 onward, and especially p. 159. Obviously once the constraints are there, it becomes plausible, but the question is who set the constraints. But I’m stopping here. If you want to argue about de Duve’s view and the meaning of his calculations, that is really of no interest to me and, as I explained, also not relevant to the physico-theological argument.

Neil Jordan (2019-06-27)

I am addressing only the arguments relevant here. Since you keep repeating the dice example, and continue to claim that it is relevant to evolution, those arguments require defending. You can keep saying it doesn’t interest you because it is not relevant to the larger argument you want to defend, and that would have been perfectly fine—if only you would stop continuing to make the mistaken claims.

You can also keep repeating ad nauseam the claim that the breakdown into stages does not help on the probabilistic level, but the dice example that *I* brought shows very well how one can—given a process that includes fixation similar to evolution (as opposed to your example, which does not include fixation)—arrive at much higher probabilities. Only in the absence of fixation does breaking it down into stages not help. That is the point you ignore, and therefore you simply keep repeating your silly example (which echoes Hoyle’s mistake, which you also failed to defend despite your repeated claims that you did).

I read de Duve’s relevant article (“Constraints on the Origin and Evolution of Life,” Proceedings of the American Philosophical Society, December 1998, p. 525 onward), and he does not say there—not what you claim in the book that he says, and not what you claim here that he says. Since you’re not interested in discussing that, we don’t have to; but just so it’s clear, the specific claim you raise here in the Q&A is simply not correct. Evolution includes fixation, it operates in light of environmental constraints that bias the outcome, and in general it is not at all similar to your dice example. That is the bottom line, and the implication for the physico-theological argument neither adds nor subtracts.

(If you really want to get into the implications for the physico-theological argument, one could spill many words on that. For example, whereas the example you brought speaks of aiming at a specific result—which made it easier for you to argue that a designer is needed to dictate that result in advance—my example contained a very simple rule: aiming for “6.” And it is already known that simple rules, for example ones that could naturally result from processes without intention, can lead to increasing complexity—see cellular automata. But even if we assume that your broader argument still stands and that a designer is necessary for the existence of those laws in the first place, your original argument about the improbability of life under a non-intentional process is still not correct. To my mind that is the heart of the matter here and the issue I wanted to clarify.)

Neil Jordan (2019-06-27)

It’s possible that we have a misunderstanding here, and I’d like to clarify it.

I also read the sections you refer to in Dr. Iris Fry’s book and de Duve’s other relevant article, “The Beginnings of Life on Earth.” Both of them talk about the origin of life, the source of the first self-replicating molecule; that is different from climbing evolution’s improbable mountain, which is the one you talk about in your book.

Is it possible that the dice example you keep bringing refers only to the origin of life? I got the impression that it referred to evolution as a whole, but perhaps I misunderstood.

I would appreciate clarification.

Michi (2019-06-27)

Not true. It explicitly speaks also about evolution and not only about the origin of life.
As for your previous message, either you don’t understand what I’m talking about or you’re just being stubborn. Either way, it seems to me we’ve exhausted the issue. Let the reader decide.

Neil Jordan (2019-06-27)

Well then, it simply doesn’t work with regard to evolution, and I explained why—for the simple reason that there is fixation.

None of the references given undermine that (nor could they). You can keep pretending that this argument works, but that is all it will be—pretending.

Leave a Reply

Back to top button