Introduction

The parameters we call constants of nature may, in fact, be stochastic variables taking different values in different parts of the Universe. The observed values of these parameters are then determined by chance and by anthropic selection. It has been argued, at least for some of the constants, that only a narrow range of their values is consistent with the existence of life [1—5].

These arguments have not been taken very seriously and have often been ridiculed as handwaving and unpredictive. For one thing, the anthropic worldview assumes some sort of a ‘multiverse’ ensemble, consisting of multiple universes or distant regions of the same Universe, with the constants of nature varying from one member of this ensemble to another. Quantitative results cannot be obtained without a theory of the multiverse. Another criticism is that the anthropic approach does not make testable predictions; thus it is not falsifiable, and therefore not scientific.

While both of these criticisms had some force a couple of decades ago, much progress has been made since then, and the situation is now completely different. The first criticism no longer applies, because we now do have a theory of the multiverse. It is the theory of inflation. A remarkable feature of inflation is that, generically, it never ends completely. The end of inflation is a stochastic process; it occurs at different times in different parts of the Universe, and at any time there are regions which are still inflating [6,7].