Header 1

Our future, our universe, and other weighty topics

Tuesday, January 24, 2017

With Overwhelming Likelihood, a Random Universe Would Be Lifeless and Light-Less

Shortly after publishing an essay on the topic of cosmic fine-tuning that had some of the worst reasoning I have ever read on the topic (which I discuss here), the Nautilus web site is out with another essay on this weighty topic that I have often discussed on this blog. The new essay by scientist Fred Adams is entitled “The Not So-Fine Tuning of the Universe.” Adams pulls some misleading tricks to try to make you believe his very wrong conclusion that “our universe does not seem to be particularly fine-tuned.”

Here are the main fallacies Adams is guilty of:

  • The “ant near the needle hole” fallacy of visually representing something incredibly unlikely to make it look as if it is likely
  • The fallacy of considering any type of star allowing life when considering stars and cosmic fine-tuning, while not considering the equally important likelihood of stars as suitable for life's evolution as our own sun
  • The fallacy of considering only less sensitive requirements when considering the likelihood of stars existing, and ignoring a vastly more sensitive requirement which makes the existence of stars incredibly unlikely in random universes
  • The fallacy of ignoring the universe's most dramatic cases of cosmic fine-tuning, and focusing only on less dramatic cases

Adams' scientific specialty is stars. He gives us a graph that plots possible strengths of the electromagnetic force and the gravitational force in hypothetical possible universes. A shaded portion taking up a fairly large part of the graph is described as an area “consistent with life.” The graph makes clear that stars require an unlikely balance between the gravitational force and the electromagnetic force, but from looking at the graph you might think that such a thing wasn't all that unlikely. 

Adams here is guilty of a fallacy that we might call the fallacy of the “ant near the needle hole.” Consider an ant that somehow wanders into your sewing kit. If it were smart enough to talk, the ant might look at the eye of a needle hole in your sewing kit, and say, “Wow, that's a big needle hole!” Such an observation will only be made if you have a perspective looking a few millimeters away from the needle hole. 

 Similarly, Adams has given us a graph in which his “camera” is placed a few millimeters from the needle hole that must be threaded for stars to exist. He has graphed a parameter space in which two fundamental constants vary by only a few times. But physicists routinely deal with a difference of 40 orders of magnitude (10,000,000,000,000,000,000,000,000,000,000,000,000,000), which, for example, is roughly the difference between the strength of the strong nuclear force and the gravitational force. So if we are imagining a parameter space of alternate universes, we must imagine a parameter space vastly larger than the relatively microscopic parameter space Adams has graphed. Rather than just visualizing something like the small changes in the fundamental constants Adams graphs, we should imagine that any of them could vary by a trillion times or a quadrillion times or a quintillion times. 

Given such a parameter space, a realistic visual representation of the chance of a random universe having parameters allowing stars to exist would be one like the visual below, which shows a tiny needle hole somewhere in the Grand Canyon. It is therefore correct to say that with overwhelming likelihood, a random universe would not have stars. Since stars are necessary for both light and life to exist, it is correct to say that with overwhelming likelihood, a random universe would be lifeless and light-less.


The misleading nature of Adams' graph is discussed on page 40 of this excellent scientific paper by physicist Luke Barnes, who concludes (contrary to Adams) that “the existence of stable stars is indeed a fine-tuned property of our universe.”

The second fallacy Adams commits is the fallacy of merely considering the likelihood of stars in relation to cosmic fine-tuning, when he should also be considering the likelihood of stars as suitable for life as our own star.

Among the stars in our universe are short-lived blue stars, long-lived yellow stars like our sun, and much less bright “red dwarf” stars that are very long-lived. It could be that life exists on planets around red dwarf stars, but it is almost universally recognized that life is much less probable to arise on planets revolving around such stars. There are two main reasons, discussed fully here. One is that since red dwarf stars are much dimmer, a planet would have to be fairly close to a red dwarf star for life to exist on the planet; and at such closer distances the planet would be subjected to very troublesome tidal effects that might make it uninhabitable. The second reason is that red dwarf stars are more unstable than stars like our sun; as a wikepedia.org article says, “Red dwarfs are far more variable and violent than their more stable, larger cousins,” such as our sun. Such variability would make a planet near a red star much more likely to get zapped by crippling radiation.

So it's kind of like this: yellow stars like our sun are good for the evolution of life, but red dwarf stars are not-so-good (kind of what we may call borderline possibilities). But when considering how much cosmic-fine tuning our universe has, we should consider the odds of getting the best thing we have, not just the odds of getting some “just barely works” borderline possibility. In fact, the requirements for sun-like stars are much more stringent than for red dwarf stars. The physicist Paul Davies says this on page 73 of The Accidental Universe: 

If gravity were very slightly weaker, or electromagnetism very slightly stronger (or the electron slightly less massive relative to the proton), all stars would be red dwarfs. A correspondingly tiny change the other way, and they would all be blue giants.

So we can put it this way: it is incredibly unlikely that a random universe would have any stars, and super-incredibly unlikely that a random universe would have sun-like stars. Clearly we should pay attention to both of these probabilities when judging how fine-tuned the universe is.

I can give an analogy. Suppose you walk deeply into the wooded wilderness of a national park with your friend, and come across a log cabin. You may say, “That must have been fine-tuned” or “That must have been designed.” Now your friend may say, “Not so, because trees might have fallen in such a way to provide you with some type of shelter from the rain.” This is fallacious, because the relevant thing to consider is the most fine-tuned thing you see, not some other less suitable thing that luck might have given you. And similarly, when considering fine-tuning in regard to stars, we should be noting that the requirements of the most suitable types of stars (stars like our sun) are much, much more stringent than the requirements of “some type of stars.” Adams ignores these more stringent requirements.

The third fallacy Adams commits is the fallacy of considering only some of the less stringent requirements of stars, while ignoring the most stringent requirement for stars. The most stringent requirement of stars is that the proton charge exactly balance the electron charge. This requirement has been pointed out by the astronomer Greenstein, who pointed out that no stars could exist if the proton charge did not exactly match the electron charge.

If there was a very small difference between the electron charge and the proton charge, you would either have (1) an electrical imbalance between particles which would completely overwhelm gravity, making it impossible for stars to hold together, or (2) an electrical imbalance between particles that would completely preclude the possibility of the thermonuclear reactions we observe in stars.

In our universe each proton has a mass 1836 times larger than each electron, but the charge of the proton exactly matches the charge of the electron to at least eighteen decimal places, as measured here (the only difference being that the proton has a positive charge and the electron has a negative charge). Stars could not possibly exist if this precise fine-tuning did not exist. Adams has simply ignored this ultra-stringent requirement, focusing on less stringent requirements. Were he to consider this requirement, he might realize that stars are trillions of times less probable to exist in random universe than he imagines.

I may note that this requirement is an entirely different requirement than the one previously considered. So for a random universe to have stars, it must not only “thread the needle” involving the balance of the gravitational and electromagnetic force (the balance that Adams has considered), but a random universe would also have to “thread the microscopic needle” of having the proton charge exactly match the electron charge. So it is as if the arrow of the blind archer must hit not just one very distant bulls-eye for stars to exist, but two very distant bulls-eyes.

We are then doubly justified in saying: with overwhelming likelihood, a random universe would be both lifeless and light-less.

The fourth fallacy that Adam commits is the fallacy of ignoring the universe's most dramatic cases of fine-tuning, and focusing only on less dramatic cases. The three most dramatic cases of cosmic fine-tuning all seemingly involve fine-tuning more precise than 1 part in 1,000,000,000,000,000,000,000,000. They are:

  • the exact match of the absolute magnitude of the proton charge and the electron charge, to more than 18 decimal places
  • the fine-tuning of the vacuum energy density, discussed here, by which we have a cosmological constant more than 1050 times smaller than the amount predicted by quantum field theory (such as we would have if opposing parameters of nature accidentally canceled out each other to more than fifty decimal places)
  • the fine-tuning of the universe's initial expansion rate (in which the universe's initial critical density matched the actual density to something like 1 part in 1050).

Which of these does Adams discuss in his Nautilus essay? None of them. Of course, he does not want to discuss such things as they would obliterate his claim that “our universe does not seem to be particularly fine-tuned.” 

Adams is very well aware of the cosmological constant problem (also known as the vacuum density problem and the “vacuum catastrophe” problem), because he discusses it at length in a scientific paper he co-authored. There he gives us some reasoning that is as off-the-mark as his insinuations about the likelihood of accidental universes having stars.

The issue in regard to the cosmological constant is that quantum field theory predicts the cosmological constant should be 1060 or 10120 times larger than the value we observe. This prediction (which you can find discussion of by doing a Google search for “worst prediction in the history of physics”) is that the vacuum of space should be super-dense – much denser than steel. But the actual vacuum of space has very little energy or density – it's almost empty.

We know that life could never exist if the vacuum was anything like that predicted by quantum field theory. Obviously you can't have life if the space between a star and a planet is thicker than steel – light cannot even travel through that. But an interesting question is: by how much could the cosmological constant differ from its current value and still allow life to exist?

Adams concludes that the cosmological constant could be up to 1030 times larger and still allow life to exist. This is almost certainly a far-too-generous estimate, and other estimates have estimated much greater sensitivity. He uses this estimate to support a conclusion in the paper that “the universe is not overly fine-tuned.” But he should be reaching exactly the opposite conclusion from these facts. If the cosmological constant is supposed to be 1060 or 10120 times larger than the value we observe because of quantum considerations, and  a value 1030 times larger than the observed value would have prevented life, then how much luck did we have in this regard to have a habitable universe? The answer is: luck with a probability of about 1 part in 1030 or 1 part in 1090. Adams should have reached the conclusion that the universe is astonishingly fine-tuned, in a way that less than 1 universe in a billion trillion should have by chance.

Adams also gives some misinformation about the fine-tuning issue involving nuclear resonances and the triple-alpha process, a process by which stars produce energy. He claims that this fine-tuning issue “goes away,” because there's some particular way in which an alternate physics could allow carbon to exist. He's using some fallacious reasoning he uses in this scientific paper. The fine-tuning issue involving the triple-alpha process and resonances is that the physics of the universe must be fine-tuned for both carbon and oxygen to exist in abundant qualities, as it does in our universe. But in his paragraph claiming a way to make this fine-tuning “go away,” he does not discuss oxygen. And on page 26 of his paper, he says, “This set of simulations does not include nuclear reactions that produce oxygen, neon, and heavier elements.” So he cannot truthfully claim to have made this fine-tuning issue “go away.” The difficulty is explaining how a random universe could have abundant amounts of both oxygen and carbon, not just carbon.

This fine-tuning requirement is correctly stated in a 2014 scientific paper which tells us on page 16 that in order for you to have abundant quantities of oxygen and carbon, you need for the quark masses to be within 2 to 3 percent of their current values, and you also need for the fine-structure constant to be within 2.5% of its current value. You could therefore say nature has to hit two different “holes in one,” and these aren't the only “holes in one” nature has to hit in order to end up with intelligent life. Because these two “holes in one” that nature must hit are different from the two other “holes in one” I discussed before, while discussing stars.

Another bit of sloppy thinking Adams gives in his Nautilus essay is when he attempts to explain away a fine-tuning of the strong nuclear force by claiming that if some alternate physics were true,  "The longest-lived stars could shine with a power output roughly comparable to the sun for up to 1 billion years, perhaps long enough for biological evolution to take place."  This is laughable, since earthly life is believed to have required 3.5 billion years to have appeared; and obviously a universe in which stars like ours can burn brightly for 10 billion years is greatly preferable to one in which they can only burn for 1 billion years.  Again, I may note that you do not explain a more favorable case of fine-tuning by imagining some much less favorable situation requiring less fine-tuning.

In his Nautilus essay, Adams misreads what nature is telling us, and his conclusion that “our universe does not seem to be particularly fine-tuned” is very much at odds with both the facts and the statements of numerous other scientists with a variety of philosophical standpoints, who have again and again stated the opposite

Postscript: I may note based purely on Adam's graph plotting the electromagnetic force versus the gravitational force and a life-compatible region, and the fact that the potential parameter space in random universes is more than a billion trillion times larger than the parameter space he has graphed, we should conclude that the chance of stars in a random universe is less than 1 in a billion trillion (less than 1 in 1,000,000,000,000,000,000,000).  The requirement of the proton charge matching the electron charge is simply a second reason for drawing the same conclusion.