In 2017, the influential online platform Edge.org asked leading academics as its question of the year: “What scientific term or concept should be more widely known?” Martin Rees, former President of The Royal Society and Emeritus Professor of Cosmology and Astrophysics at the University of Cambridge, had a clear answer: the “multiverse” idea. As to what this idea entails, he explains:
Physical reality could be hugely more extensive than the patch of space and time traditionally called “the universe.” Our cosmic environment could be richly textured, but on scales so vast that our astronomical purview is restricted to a tiny fraction: we’re not aware of the “grand design,” any more than a plankton whose “universe” was a spoonful of water would be aware of the world’s topography and biosphere. We may inhabit a “multiverse.”
Rees and others have two main motivations for taking the multiverse idea very seriously. First, it offers a candidate response to the observation that several parameters of our universe seem fine-tuned for life: had those parameters been even slightly different, life might not have been able to exist in it. This seems puzzling and, in the eyes of many, calls for an explanation. But if there is a multiverse, it is not at all surprising that the parameters are right for life at least somewhere. As a matter of triviality, we could not have found ourselves anywhere else. So, some physicists and philosophers suggest, we should infer from fine-tuning for life that we likely live in a multiverse.
The second motivation for taking the multiverse idea very seriously is that some of the more advanced and speculative theoretical ideas in high energy physics and cosmology are widely taken to suggest it. Notably, a combination of string theory and the “eternal” variant of inflationary cosmology provide a natural framework for it in form of the so-called “landscape multiverse.”
When one expresses the multiverse idea as Rees does, it sounds daring and humbling. But many of Rees’ physicist colleagues have a completely different reaction: as they see it, the multiverse idea is unscientific and perhaps even “dangerous” in that it may lead to misdirected scientific efforts. Three years before Rees advocated it on Edge.org, Paul Steinhardt, Albert Einstein Professor in Science at Princeton University, already submitted “the multiverse” as his answer to Edge’s 2014 question of the year, but that question was “What scientific idea is ready for retirement?”! In his piece, Steinhardt derides the multiverse as a “Theory of Anything”, compatible with arbitrary observations and therefore without any empirical bite.
My new book “Multiverse Theories: A Philosophical Perspective” analyses the fine-tuning motivation for the multiverse idea and the proposals to make concrete multiverse theories testable that have been made in the literature. I believe that both Rees and Steinhardt are right in important ways: Rees is right that we might indeed live in a multiverse – it is a serious possibility, far from unscientific, and there are versions of the fine-tuning motivation that are at least defensible.
But Steinhardt is right that we may never be able to rigorously test specific multiverse theories, for the predictions derived from them simply have too many “moving parts.” Observers in different sub-universes of the overall multiverse will in general see very different physical laws. If we assume that a multiverse theory predicts what a “typical” observer in the multiverse measures, the results will depend on how we model observers and count them. Unfortunately, there is so much freedom here that almost arbitrary “predictions” can be derived.
The upshot of my discussion is a call for humility: we should be prepared for the possibility that we live in a multiverse, but even the most intellectually advanced future scientists may not be able to obtain conclusive evidence for it. Whether we live in a multiverse and, if so, what our universe’s companion universes are like may forever remain unknown.