Cosmological Natural Selection

I have been reading this paper, Scientific alternatives to the anthropic principle, by Lee Smolin.

It will take some time to analyze this paper—it has more words than equations, which makes it harder (for me) to evaluate.

The claim of the paper is that a certain type of multiverse theory is falsifiable, one called “cosmological natural selection.”

Before a description of this cosmological natural selection, I want to point out that a falsifiable multiverse theory is a good thing—a very good thing. If another universe with different physics is ever observed, then I for one will abandon cosmological ID. On the other hand, if there is a legitimate falsification test for a multiverse model that produces a negative result—well that would indirectly strengthen cosmological ID.

The cosmological natural selection idea is indeed provocative. I will attempt to explain the idea without, at this time, evaluating its merits.

We begin with a mechanism for creating new universes: the black hole bounce.

The black hole bounce results from quantum modifications on a “classical” black hole collapse. Instead of collapsing down to a singularity, the black hole, at some point, begins to expand—producing a new region of spacetime that is not causally connected with the universe in which the black hole was originally formed. It is, in fact, a new universe.

Smolin writes:

A multiverse formed by black holes bouncing looks like a family tree. Each universe has an ancestor, which is another universe. Our universe has at least 10¹⁸ children, if they are like ours they have each roughly the same number of their own.

In setting up the case for cosmological natural selection, Smolin (see p. 29 of his paper) presents three hypotheses:

1. A physical process produces a multiverse with long chains of descendents
   
2. Let P be the space of dimensionless parameters of the standard models of physics and cosmology, and let the parameters be denoted by p. There is a fitness function F(p) on P which is equal to the average number of descendents of a universe with parameters p.
   
3. The dimensionless parameters p_new of each new universe differ, on average by a small random change from those of its immediate ancestor. Small here means with small with respect to the change that would be required to significantly change F(p).

Let me paraphrase. In this model, there is a process for creating children (new universes.) That process is black hole bouncing. Furthermore—there is a fitness function that is being maximized: the average number of children (black holes) produced by the universe. Finally, the set of physical constants (such as the cosmological constant) get passed to the descendent universes, like cosmic DNA, but are slightly modified (mutated) in the process.

The ultimate result, from natural selection seeking to maximize the fitness function, is that universes that are very good at producing blackholes will emerge as the “fittest.” What types of universes are good at producing black holes? Universes such as ours with an improbably small values for their cosmological constant and the right low energy physics and chemistry for star production. Which, as an aside, are also the types of universes that can produce intelligent life.

Note that this model absolutely requires that the physics in a child universe differs only slightly from its ancestor. Smolin admits:

The hypothesis that the parameters p change, on average by small random amounts, should be ultimately grounded in fundamental physics. We note that this is compatible with string theory, in the sense that there are a great many string vacua, which likely populate the space of low energy parameters well. It is plausible that when a region of the universe is squeezed to Planck densities and heated to Planck temperatures, phase transitions may occur leading to a transition from one string vacua to another. But there have so far been no detailed studies of these processes which would check the hypothesis that the change in each generation is small.

I will continue studying this paper and will comment further in the near future.