The first important point is that a multiverse or ensemble is not a unique concept; there can be many quite different realizations. Saying that everything that can happen does happen will not specify a unique multiverse for two reasons.
Given the space of possibilities, realized multiverses are by no means unique. Their description requires the existence of a well defined and physically motivated distribution function f (m) on the space of all possible universes. The mere fact that different distribution functions are possible shows that the concept of an ensemble is not unique. Hence one has to ask: How does the choice of these functions originate and what is their rationale?
These questions can be partially answered scientifically. A really existing ensemble of universes or domains may arise from the operation of a generating process, which adequately explains the origin of its member and their ranges of characteristics and parameter distribution from a more fundamental potential or primordial quantum configuration. That is, there may be a specific generating process which determines f (m).
However, when it comes to the question of what is responsible for the operation of a specific generating process rather than some other one which would generate a different ensemble, an adequate answer cannot be given scientifically. This is the question of why the primordial dynamics leading to the given existing ensemble of universes is of one type rather than another. Even if we could establish f (m) in detail, it is difficult to imagine how we would scientifically explain why one generating process (operating prior to the existence of the universe) was instantiated rather than some other one. The only possible answer comes from philosophical considerations. A priori, there is a complete arbitrariness in the resultant models because the generating process is not uniquely determined by any provable principle.
The 'space of all possible universes' is not an easily delimited or unique concept, but the choice of what is included here determines what kind of multiverse can exist. How wide a variation of properties are we prepared to consider in our class of multiverses? Are we prepared to consider: universes with quite different physics (e.g. not coming from a variational principle and not based on quantum field theory); universes with different kinds of logic and perhaps alternative forms of mathematics; universes allowing magic, such as envisaged in the Harry Potter novels by J. K. Rowling?
If we are not prepared to allow such situations, what is the rationale for this? If the underlying principle is that 'all that can happen does happen', then there should conceivably be universes of all these kinds, as well as theistic and non-theistic ones, universes based in beauty and ethereal vibrations rather than physics, etc. Science fiction provides a fruitful source of such ideas. We surely cannot even conceive of all the alternatives, much less write them down systematically. In many of these classes of universes, no processes generating expanding domains would be possible, because the requisite kind of physics would not be realized. If any such possibilities are to be excluded, we have to be given both a meta-principle excluding them and a justification as to why that meta-principle should apply in the ensemble. Only philosophy can justify such a choice.
We conclude that the idea of a multiverse or ensemble is not a unique concept but a whole package of differing possibilities. This has several implications. First, it is not necessarily true that life can exist in any universe in an ensemble; this may or may not be true. Second, in considering experimental or observational tests for the existence of an ensemble, one needs to be told which specific kind of ensemble is claimed to exist and how to test that claim. How can we observationally or experimentally exclude specific classes of multiverse, e.g. those in which there may be any number of copies of our own universe. Thus one question would be: How do you determine how many copies of our own universe exist in any claimed ensemble?
Was this article helpful?