From the earliest AP discussions, it was recognized that there are both weak (WAP) and strong (SAP) formulations. The weak versions assert that, since there are observers in our universe, its characteristics, including the values of the fundamental constants and the initial conditions, must be consistent with the presence of such observers (see ref. , p. 372). Thus the existence of observers acts a posteriori to select values of the fundamental constants and other important parameters. These versions of the AP just specify the conditions which have been fulfilled for complexity and life to arise - they do not explain how or why those conditions have been realized. In fact, some writers describe the WAP as just a selection effect.
Strong versions of the AP go much further: they asssert that our universe -right from the start - had to be such that the appearance of observers is inevitable. That is, they purport to account in a basic way for our universe being life-bearing. For instance, one version of the SAP would be as follows: 'The universe must be such as to admit the existence of observers within it at some stage' (see ref. , p. 376). Here, the words 'must be' indicate a priori necessity - not the consequence of there being observers now . The eventual emergence of observers somehow explains why the universe possessed its initial characteristics - it has these characteristics in order that observers will appear.
From this, it is clear that some evidence or justification for the requirement of having observers must be provided. Many - but not all - such formulations do this by incorporating an explicit or implicit finality or purposiveness in our universe, which goes considerably beyond what can be concluded from the natural sciences themselves. Sometimes this is done on philosophical grounds, sometimes on theological ones.
Over the past decade, two very different - and certainly inequivalent - versions of the SAP have been discussed. The first is essentially the way it was first formulated: the characteristics of our universe are chosen to ensure the appearance of life and observers. But this raises the issue of what or who tailors the laws of nature and the fundamental constants in this way, which immediately goes beyond the domain of science.
Thus, a second version of the SAP has become popular, which - at first sight - keeps it within the realm of the natural sciences. This asserts that our universe - or our domain - is one of a large, actually existing, ensemble of universes or domains, each having different laws, fundamental constants and initial conditions. In fact, a frequent, but much less adequate, specification of this ensemble is that it contains all possible universes. The presupposition here is that there exist universes or domains representing the full range of possibilities . There is then some probability that any one of the these really existing universes will allow the emergence of life and observers.
This, in one sense, does explain why our universe is life-bearing, providing the presuppositions can be justified and providing a meaningful probability measure can be defined on the space of the ensemble [7,8]. But this explanation is obviously incomplete. It immediately invites further understanding of the process by which this particular cosmic ensemble emerged and why it contains universes or domains which allow for the emergence of complexity. And if we can substantiate the operation of such a fertile cosmogonic process, then we may certainly want to seek an explanation for its origin, however we have come to understand and model its scientifically accessible underpinnings.
Thus, this formulation of the second version of the SAP clearly manifests its inequivalence with the first version, as well as the extraordinarily strong presuppositions on which it rests. In fact, if we use it to argue from the presence of observers in our universe to the existence of a certain type of ensemble of universes, then it seems to reduce to the WAP. However, the characteristic feature of this version - the existence of at least a large subset of possible domains, some of which are life-bearing - really takes us beyond the WAP. It solves the fine-tuning problem, but does not explain in any a priori way why our universe should have observers in it at some stage, much less why the ensemble of all existing universes should include some which admit their emergence.
It is certainly true that, if the ensemble exists, then our universe itself must exist. But this is obviously a very weak form of the SAP. The 'necessity' of the existence of our life-bearing universe rests on the presupposed existence of all possible universes, or at least of a large number of universes of a broad range of types. Clearly - to achieve equivalence with the first version of the SAP - we require an adequate explanation for the necessity of the encompassing ensemble - or at least some explanation or justification for its de facto realization. Anchoring this version of the SAP really requires some compelling cosmological account of the ensemble, which is by no means unique , or - even better - of why it must exist. This would make the multiverse version of the SAP equivalent to the first version. However, the multiverse version will never be able to go that far, since it strives to avoid scientifically inaccessible causes and explanations.
Another strong reason for stressing the multiverse version of the SAP is that we now recognize that there are a number of natural ways in which an ensemble of actually existing 'universes' could arise in quantum cosmology: for example, Andrei Linde's chaotic or eternal inflation scenario . However, as we shall see later, such suggestions are not yet very secure. Furthermore, there are serious physical and philosophical issues which need to be resolved before they can be regarded as evidential [7,8]. Until then, this version of the SAP, despite its popularity, must be relegated to the category of (at best) informed cosmological speculation.
In discussing the multiverse version of the SAP more fully, several points should be emphasized.
• As it now stands, it does not provide either an adequate or complete - let alone an ultimate - scientific explanation. Only strong evidence for - and an adequate description of - the process by which the ensemble emerges can do that. To constitute an ultimate explanation, that process must further be shown to be necessary, an understandable accident that was always a possibility, or intended by some transcendent agent for a specific reason. But the scenarios by which the ensembles may have originated are still very uncertain and ad hoc, so it is impossible to envision them as necessary or providing any fundamental or final philosophical explanation.
• Once we grant that the ensemble embracing our universe really does exist, then saying this ensemble explains how our universe is fine-tuned for life does have some meaning and validity - in terms of the probability of any one of the universes being like our own. But this requires that there be some well defined distribution function on the space of all possible universes, with an associated probability measure [7,8].
• It is very difficult, if not impossible, to define a really existing ensemble of all possible universes in a meaningful way which avoids infinities [7,8]. Also, in order for the ensemble idea to work, it cannot just be an ensemble of conceptually possible universes - it must really exist. Any power this version possesses relies on the universes or domains having a bona fide existence. Possible or potential existence has no a posteriori implications and explains nothing (see ref. , p. 371).
Was this article helpful?