Multiverses and Super-Turtles

Last night’s talks were really interesting and it was good to get three different opinions being discussed in a (semi-)public setting. Usually these sort of lectures are one-sided in their opinions, which tends to hide the fact that the hot topics are also the most controversial. So as I mentioned earlier there were three speakers, Bernard Carr, George Ellis and Paul Davies. On the night Chris Isham was not able to attend, so one of the representatives of the Templeton Foundation acted as chairperson.

Bernard introduced the idea of the multiverse, after having plugged the new book and thanked all the contributors. With only thirty minutes allotted to each speaker there wasn’t enough time for a detailed explanation, but he explained some of the different multiverse ideas, in particular using this picture by Max Tegmark to illustrate the hierarchy of multiverse structures. With time running out (and some quite surprising heckling when he asked for some leeway), Bernard had to race through the history of physics on one slide, imploring us to consider it as “an artistic journey”.

The main thrust of the later sections was that throughout history science has expanded the realm of its applicability, from the Earth, to the Solar System, to the galaxy and beyond. The next step in this progression will be to extend our thinking beyond the particle horizon to consider what might lie beyond.

George Ellis immediately made clear that he was not enamoured of the multiverse as a scientific investigation, but later conceded that as a philosophical idea it might have merit as an explanation. But he insisted that it cannot make any predictions and at least currently is not verifiable. In one particularly insightful comment he destroyed notions of using statistics to show how probable a multiverse must be, countering that these statistical analyses presuppose the existence of the multiverse by virtue of their nature. If there is only one universe, then a statistical analysis has nothing useful to say about this one data point!

Prof Ellis also showed how much extrapolation must take place by considering the past light cone of our observable universe, at first filling the whole wall, but then being reduced in stages to a small lonely triangle surrounded by the unobservable region beyond. The idea that we can talk knowledgeably about these regions was rejected, as was the tendency of scientists to talk about “infinity” as he put it. He implored scientists not to open the door to a topic that cannot be verified, insisting that it would lead to a rush of similarly unverifiable ideas claiming scientific validity (presumably referring to ID/creationism).

Paul Davies spoke last and started by lamenting that most discussions of this type revolved around choosing between ignorance, a multiverse, or Intelligent Design as an explanation for the universe’s existence. He quoted the famous “turtles all the way down” explanation of the universe, but expanded it to include a god/multiverse as a “super-levitating turtle”, a description which also appears in his new book. This certainly brought a few chuckles from the audience, but the deeper argument was that it would be more intellectually pleasing to not have to resort to any outside cause to explain the universe and it’s “bio-friendliness”.

The second half of the talk then turned to an explanation of this helpful property of the universe using the uncertainty inherent in a time-reversible quantum mechanical description of the universe. I can’t say I followed exactly what was meant, and Paul Davies freely admitted that Mike Duff called this part of the talk “mumbo-jumbo”. The main idea was that the universe should be considered as an information processor, with the maximum amount of processing possible determined by the volume and hence changing with time. Weird things should happen if something exceeds this processing power, although what would constitute weird wasn’t mentioned. On top of this the universe would be in a feedback loop in which observers shape the course of physical laws in the past, with ever greater precision leading on from “better” observations. This would not be causality violating but rather a later decision would give context to earlier events as in the delayed choice experiment of Wheeler. In this way the tower of turtles would be turned into a turtle loop in which the existence of the universe explains itself. As Davies noted in response to a question, this kind of closed loop history is familiar in any time-travelling story, with no external influence needed to set the events in motion (for example in the film Twelve Monkeys).

After the three talks there was an audience Q&A session, with most of the questions being quite considered and helpful. There was however the usual feature of events like this, where someone stood up and complained that his paper on similar ideas had been censored by the powers-that-be on the arXiv. Paul Davies actually took up this poin, agreeing that sometimes physicists are afraid of questioning the basis on which they rely on physical laws.

All in all it was a good night, with some nice food and wine afterwards and a few of us headed to a local to continue the discussion.