An ensemble may include singular and non-singular, as well as special and general, universes. Which is more fundamental and so more likely to occur: fine-tuning to produce the very special initial conditions for our observed universe3 or generic primordial conditions? While the introduction of ensembles has in effect been driven by the second view, they can equally well fulfil the first, the difference between them coming through the choice of distribution function on the space of possibilities. Hence, the mere existence of an ensemble does not by itself support either view. A generating mechanism might spit out numerous copies of one (successful) universe4 rather than numerous universes with greatly different properties. Whichever is the case will determine the form of the distribution function . Neither possibility can be excluded without some understanding of the creation mechanism. That mechanism cannot be tested in the true multiverse case and may not be testable in practice even in the multi-domain case. Until it is tested, any specific form assigned to it - in particular, the assumption that it creates generality rather than speciality - is metaphysical rather than physical.
This underlines the point that the present philosophical predilection for generality does not necessarily reflect the nature of physical reality. Indeed, philosophical predilection in cosmology has oscillated from assuming the universe is very special (in the sense postulated by the Cosmological Principle [26,27]) to assuming that it has attained its present nature through operation of standard physics on generic initial conditions [28,29]. The present tendency to assume that only the latter assumption is allowed is a philosophical supposition rather than a physical requirement. Either assumption may be allowed by appropriate assumptions about the pre-physics that leads to universe generation.
Was this article helpful?