Header 1

Our future, our universe, and other weighty topics

Friday, December 15, 2017

Fake Physics Does Not Explain Real Cosmic Fine-Tuning

Marcus Du Sautoy is a mathematician who has taken over Richard Dawkin's chair as Charles Simonyi Professor for the Public Understanding of Science at Oxford University. While holding that chair, Dawkins committed many an error in reasoning, some of which are described here. Some equally bad errors of reasoning are committed in Du Sautoy's book The Great Unknown.

On page 221 Du Sautoy marvels at how fine-tuned the fundamental constants of our universe seem to be. He states:

What is particularly striking is how sensitive the possibility of life in our universe is to a small change in these constants. For example, if the constant that controls the way the electromagnetic field behaves in a vacuum is changed by four percent, then fusion in stars could not produce carbon....Change the cosmological constant in the 123rd decimal place and suddenly it's impossible to have a habitable galaxy.

Du Sautoy suggests what he calls an “explanation” for this mystery – the idea of the multiverse, that there exists some vast collection of universes. He says, “In the multiverse model there are lots of different universes, and in each of these universes the fundamental constants could be randomly assigned.” That is indeed the idea of the multiverse, although Du Sautoy misspeaks by referring to “the multiverse model.” In science a model is a simplified representation of a known physical reality. Speculating about unknown or unobserved universes is not at all a case of making a scientific model.

Next Du Sautoy goes off the rails, both by the very fact of referring to the multiverse idea as an explanation, and also by claiming, most absurdly, that such an idea is a simple and economical explanation. He states the following on page 222:

The multiverse theory at least has that sense of economy we are after in a good theory. The explanation terminates and does not require further explanation. The addition of these other universes is just more of the same with variations, and once you accept all these other universes you get a complete solution to the fine-tuning problem....A good scientific theory should make sense of how everything is put together, and shouldn't need to introduce too many extra characters into the story to get the narrative we experience. There is a simplicity and naturalness about the multiverse theory that makes it a strong candidate theory.

Reading Du Sautoy claim that a multiverse is a simple explanation to the problem of why our universe is so fine-tuned, I am reminded of Voltaire's famous quip about the Holy Roman Empire: that it was neither holy nor Roman nor an empire. The multiverse is not at all an explanation and not at all simple. You present an explanation when you either discuss some causal factor that produced something, or you present some circumstances that allow you to say, “Now what seemed so surprising is not surprising.” In neither of these senses is the multiverse idea an explanation. Imagining some vast collection of other universes does not involve some causal factor that preceded our universe, making it have fine-tuned constants. If we imagine such a vast collection of universes, it is still every bit as surprising that our particular universe should so improbably have fine-tuned characteristics it was so unlikely to have.

Imagine if some teenager buries all the furniture in his house in buckets of honey. You could try to explain this by claiming there is an infinity of other universes, and so in at least one such universe we would expect such behavior. But that is in no sense an explanation, because it does nothing to increase the likelihood that this particular teenager would have acted the way he did. Our surprise about the teenager's behavior is not reduced, so in no sense has an explanation occurred. We can say the same thing about a multiverse as an attempted explanation for our universe's incredibly improbable fine-tuned constants. Imagining a multiverse does nothing to make it more likely that our particular universe would have such fine-tuned constants, so it is not correct to say that any explanation has occurred for our universe's characteristics.

Although he is a mathematician, Du Sautoy has ignored a mathematical rule very relevant to these considerations. The rule is that you do not increase the likelihood of a success on any one random trial no matter how much you increase the number of random trials. Each universe is like a random trial in which the success is the appearance of intelligent life. You do not increase the chance of success on any one trial by increasing the number of trials. So even if you assume an infinity of universes, that does not increase by even .00000000000000001 percent the likelihood that our particular universe would have coincidentally been consistent with life. And similarly, if I assume that there are an infinity of casinos, and an infinity of gamblers at such casinos, this does not increase by even .00000001 percent the likelihood that I will be a winner the next time I gamble at a casino.

As for Du Sautoy's claim that the multiverse idea is a simple and economical theory,  this is certainly the exact opposite of the truth, for nothing could be less simple and less economical than imagining some vast collection of other universes, each with a different set of fundamental constants. Du Sautoy has previously told us that the cosmological constant seems to be fine-tuned to something like one part in 10 to the 123rd power. So his multiverse presumably needs to consist of more than 1,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000,000 universes. Calling a theory requiring such a thing simple and economical is every bit as fatuous as calling the Pacific Ocean “dry” or the Sahara Desert “wet.”


On page 223 Du Sautoy attempts to suggest that a multiverse theory is better than imagining a designer who fine-tuned the universe's constants, because “a designer who fine tunes the constants raises as many questions as it answers.” But, of course, exactly the same objection can be made to the infinite clutter of a multiverse, which raises far more questions than it answers, such as the question of where this vast collection of universes came from, and why all these universes happened to have the characteristics they have. And it is not a valid procedure to judge an explanation on whether it raises as many questions as it answers. All kinds of successful theories (such as the atomic theory) raised as many questions as they answered.

Du Sautoy suggests that a multiverse is a better explanation because it's not supernatural. But he's wrong – there's nothing more supernatural than a multiverse. The Merriam-Webster defines supernatural as “of or relating to an order of existence beyond the visible observable universe.” A vast collection of universes (each with different physical constants) outside of our universe is just such a thing.

On page 225, Du Sautoy states, “Although at the moment there is no way of testing the multiverse theory, there is no a priori reason why it will always remain untestable.” This is completely false; there is exactly such an a priori reason. It is the fact that we could never verify that there existed even one other universe with some set of physical constants different from our own.

On the same page, Du Sautoy states the following:

The multiverse theory, although potentially untestable, does come with a mechanism, inflation, by how these multiverses arise. And we do at least have evidence for one of these multiverses: our own universe.

The second statement is a clumsy misstatement, and the first statement is misleading. According to multiverse terminology, a multiverse is some collection of many individual universes. Our universe is therefore not a multiverse, but a universe. Our universe is no more evidence for a multiverse than a single tree is evidence for a forest.

As for the claim that inflation theory provides a mechanism for creating universes, this is not true in any relevant sense. Certain versions of the theory of cosmic inflation (itself a speculation designed mainly to explain evidence for a certain type of cosmic fine-tuning) imagine that our universe is kind of like a bubble in a larger cosmic reality that pops out bubble universes, kind of like a hot soup pops out bubbles. But this cosmic inflation theory has no mechanism for creating other universes with random physics and random physical constants. The “bubble universes” of the theory of eternal cosmic inflation should have the same physical constants as our universe. This point was made by Columbia University multiverse critic Peter Woit, who states the following:

Claims are often made that the theory of inflation provides evidence for a multiverse with different physics in each universe. If one looks into actual models of inflation one finds that again, no theory of the sort has been claimed.

Woit reiterates the same point here, stating “models of inflation...are not models that lead to the kind of multiverse of different physical laws.” Over the years, Woit's “Not Even Wrong” blog has been great about exposing the misstatements and malarkey of multiverse theorists. Woit now uses the term “Fake Physics” to describe such theories.

Moreover, since the cosmic inflation theory is a speculative theory for which there is no solid evidence, it can hardly be used to back up claims of a multiverse, no matter what it predicts or describes. Trying to do that is very much like a person trying to substantiate his claim of a vast horde of invisible fairies by saying that his other theory of a fourth-dimensional magic kingdom provides a mechanism for the appearance of such fairies. You do not substantiate one speculation by referring to another speculation.

Although perhaps sympathetic to his naturalism, Woit was scolding biologist Jerry Coyne, who recently had a post enthusing that maybe the multiverse can help us explain the appearance of human beings such as us. In a post with the misleading title "New evidence for the multiverse -- and its implications," Coyne states:

Further, it means that the evolution of humans was inevitable somewhere. In one of those universes that permitted the evolution of life, it was inevitable that a thinking hominin would evolve.

We may ask here: why does Coyne feel the need to drag in a multiverse (a vast collection of universes) to help explain the appearance of humans such as us? This is not at all what anyone should be doing if he felt that he had an adequate Darwinian explanation for human beings. On the other hand, someone who did not have an adequate Darwinian explanation for human beings might, in an act of utter desperation, try to imagine an infinity of universes to help explain human origins. But the problem is that exactly the same “it would happen at least once in an infinity of universes” reasoning can be used to justify a belief in fairies and leprechauns.

All these attempts to explain our universe's fine-tuned constants through imagining a vast collection of universes are guilty of a fallacy we may call the lucky numbers fallacy. This is the fallacy of assuming that some favorable set of physical constants would be sufficient to make the universe habitable for beings such as us. Having favorable physical constants is a necessary condition for a habitable universe, but not a sufficient condition. You need something much more than such lucky numbers. For a universe to be habitable, you also need favorable laws of nature, which act programmatically to help create favorable conditions for life. Such laws act like intelligent programming. If there were an infinite set of random universes, we would not expect that any of them would have such a favorable set of laws like those in our universe. There would be no reason why any random universe would have laws resembling intelligent programming.

As for Coyne's idea that a multiverse might help explain humans, it is completely mistaken. Human minds have many characteristics that we cannot explain as being a result of natural selection or brain activity. As discussed here, human mental characteristics such as empathy, spirituality, mathematical ability, abstract thinking, musical ability, aesthetic appreciation, and artistic creativity are ones that do not have any survival benefit to organisms in the wild, so we cannot explain them by evoking natural selection. The most basic things such as consciousness and life-long memory cannot be explained by brain mechanisms. Besides having no plausible theory to explain human memories lasting for decades, scientists lack any explanation for how neurons could be producing human consciousness. Human minds have characteristics that can only be explained by imagining something beyond the brain: something such as soul or spirit. So no matter how many combinations of matter that might occur in an infinity of universes, not one single one of them would ever be sufficient to explain the human mind. You can't multiverse your way to explaining minds such as ours.

As shown in this post's table, we see many astonishing examples of fine-tuning both in the universe's laws and fundamental constants. We cannot explain away these things through infinitely extravagant multiverse speculations that Woit has recently labeled “Fake Physics.” Fake Physics does not explain real cosmic fine-tuning.