Mind Matters Natural and Artificial Intelligence News and Analysis
Piano tuning process. closeup of hand and tools of tuner working on grand piano. Detailed view of Upright Piano during a tuning
Photo licensed via Adobe Stock

Ours Is a Finely Tuned — and No Free Lunch — Universe

Mathematician Ola Hössjer and biostatistician Daniel Andrés Díaz-Pachón explain to Walter Bradley Center director Robert J. Marks why nature works so seamlessly

Our Walter Bradley Center director Robert J. Marks had a chance, recently, to talk with Swedish mathematician Ola Hössjer and University of Miami biostatistician Daniel Andrés Díaz-Pachón on the many ways in which the universe is finely tuned for life. This is Part 1. Part 2 will shortly follow.

This portion begins at 00:00 min. A partial transcript, Show Notes, and Additional Resources follow.

Today on Mind Matters news on the podcast, we’re going to talk about fine-tuning of the universe for life. Scientists know that Earth is is finely tuned for life to come into existence. For example, pronounced atheist Sir Fred Hoyle (1915–2001). Hoyle was a great astronomer, maybe known best for his coining the term, Big Bang to describe the beginning of the universe. Hoyle did not at first believe that the universe had a beginning so he coined the term in a mocking way. He didn’t like it. He tried to make fun of it. But nevertheless, Hoyle’s term stuck Hoyle was also convinced, despite his beliefs, that the science of the universe dictated that it had to be fine tuned.

Robert J. Marks: Hoyle said the following about fine tuning: “A common sense interpretation of the facts suggest that a super intellect has monkeyed with Physics.” I love that word “monkey” with Physics. This monkeying with physics, chemistry and biology is the topic today on Mind Matters News. We have two guests who have published extensively in the literature about fine tuning. Dr. Ola Hössjer, is a Mathematical Statistics professor at Stockholm University and he joins us from Sweden. And our second guest is Daniel Andrés Díaz-Pachón research assistant professor of biostatistics at the University of Miami…

I have to warn the listeners that both of these guys are wickedly smart. So you better listen closely. So let’s start with the chat Fred Hoyle defined fine–tuning and he says, the universe looks monkeyed with to allow life.

Note: Fred Hoyle (1915–2001) actually spent a great deal of time trying to discredit the Big Bang because it sounded like “a party girl jumping out of a cake.” However, he accepted the evidence.

Robert J. Marks: He’s not the only one who noticed. Famous physicist Freeman Dyson said, “The more I examine the universe and study the details of its architecture, the more evidence I find that the universe in some sense must have known we are coming.” Now, the interesting thing about the fine tuned universe is there are different ways to measure the fine tuning of the universe, to put numbers to it that makes sense to us. And the three of us Ola, Daniel, and I agreed on a list of different ways that fine tuning can be measured. And we want to go through these and discuss each one in turn.

The first one is something called active information. Now, Daniel has published quite a bit in terms of active information. Daniel, could you give us a definition of active information and explain why it can be used as a measure of fine tuning?

Daniel Díaz: It is better to talk first about the No Free Lunch theorems. … The No Free Lunch theorems say that, in a big space, you can think of it as if you were shooting at some small target in a big wall. If you start looking at that target at random, then you’re not going to do better than end up in a blind search.… So active information was introduced in order to measure the amount of information that the algorithm is infusing in order to get to a target, the probability better than just that given by a blind search.

Robert J. Marks: The No Free Lunch theorem basically says that you have no knowledge beforehand of any domain expertise of finding the target. So you have nothing that guides you. One technique can be used to find the target as well as any other technique. Right?

Daniel Díaz: Yes. We have some numerical measures, but those numbers are going to change in different settings and different strategies for different values that you’re also considering …

Note: The No Free Lunch theorems, described at NASA, contains such observations as “No device can infer itself” and “No two distinguishable devices can infer each other” – David H. Wolpert,What does dinner cost? No Free Lunch Theorems

Robert J. Marks: Okay. One of the things that you mentioned, Daniel, is this idea of intervals.I know that Ola and you and I have done work on the probability measure of fine tuning and the difference between them and why one is the better. Ola, could you talk about?

Ola Hössjer: Yes. And that’s a very good question because we have some process that generates outcomes and it could be generating the universe by some mechanism, … generating a protein by some evolution or algorithm or something. And as you said, we have a target that could consist of a certain region or a certain interval within this overall space of possible outcomes. And as Daniel said, when we talk about the origin of the universe and look at a particular constant of nature, it’s only a very small life-permitting interval for that constant of nature that corresponds to universe that it meets life.

So, in that case, we could say that this small life-permitting interval, as Daniel said, is the target of this process that generated our universe. And then, so we have a small interval and then we would say, what is the probability that this process of generating this constant of nature, what is the probability that… If we think of this process as random in some way, what is the probability that this constant of nature falls within the life permitting interval?

And then intuitively we think that a smaller life-permitting interval would correspond to small probability of ending up there. Whereas the large life-permitting interval would correspond to a large probability of ending up there.

Robert J. Marks: But you’ve shown that this isn’t true, right?

Ola Hössjer: Not necessarily. Because it depends on the various possibilities of this random outcome of generating the universe. And that’s called the statistical distribution of that constant of nature… how it was generated. So in principle, it could be the case that the distribution is completely within this small life-permitting interval. And then the probability would be one of ending up with a life-permitting universe. At least when we talk about this particular constant of nature. And in the same way, it could be the case that we have a very large life-permitting interval, but this distribution is completely outside of it. And then still the probability is zero, but these are exceptions. So in our paper, we have a general way of choosing this distributional, the value of the constant of nature, which is outside of the universe…

Note: According to growing numbers of scientists, the laws and constants of nature are so “finely-tuned,” and so many “coincidences” have occurred to allow for the possibility of life, the universe must have come into existence through intentional planning and intelligence.

In fact, this “fine-tuning” is so pronounced, and the “coincidences” are so numerous, many scientists have come to espouse The Anthropic Principle, which contends that the universe was brought into existence intentionally for the sake of producing mankind.

Even those who do not accept The Anthropic Principle admit to the “fine-tuning” and conclude that the universe is “too contrived” to be a chance event. – Judaism Online

Robert J. Marks: So a question is you use the term maximum entropy. I think most people are familiar with the idea of maximum entropy associated with thermal dynamics. And I think there’s a relationship here. But as a statistician, you used the term maximum entropy in a more of a mathematical context, is that right? Could you elaborate on that a little bit?

Ola Hössjer: Yes. That is correct. And this is something called Bayesian statistics. So we have a certain parameter, the value of this constant of nature, and we use a certain prior distribution for that constant of nature. And then there is something we observe, the life-permitting interval or physicists have determined it. That’s the likelihood part. And then the question is in Bayesian statistics, how do we choose this prior information? And you could choose this in different ways. It could be chosen on subjective grounds, different people chose it differently.

Next: How Bayesian statistics could help break a deadlock

You may also wish to read: When 900 bytes shut down the world. A great physicist warned us, information precedes matter and energy: Bit before it.

Show Notes

  • 00:11 | A Little Fine Tuning
  • 01:36 Introducing Ola Hössjer and Daniel Díaz
  • 03:19 | No Free Lunch Theorems
  • 05:44 | Formula 409
  • 08:39 | Active Information
  • 09:41 | Intervals
  • 13:53 | Maximum Entropy
  • 21:03 Intervals of Infinite Length
  • 24:26 | Reduction to Practice
  • 29:31 Specified Complexity
  • 32:01 | Irreducible Complexity

Additional Resources

Podcast Transcript Download

Mind Matters News

Breaking and noteworthy news from the exciting world of natural and artificial intelligence at MindMatters.ai.

Ours Is a Finely Tuned — and No Free Lunch — Universe