April 11, 2005

Sir Martin Rees and Cosmological Speculation

Posted by Arcane Gazebo at April 11, 2005 7:55 PM

I just got back from the annual Oppenheimer Lecture, given this year by Sir Martin Rees of Trinity College, Cambridge, under the title "Scanning Cosmic Horizons". I was trying to recall where I'd seen him in the general media before, and a Google search after the talk revealed that he was the guy suggesting that humanity has only a 50% chance of surviving the next century. His talk today was concerned instead with cosmology, but Rees' penchant for speculation was evident here as well.

Most of the lecture was devoted to explaining what is known about the parameters of the universe: its rate of expansion, how much matter there is and how much dark energy, and how uniformly the matter is distributed. He then moved on to counterfactuals: how much can various parameters change while still allowing for an interesting universe? The subtext of this sort of discussion is always the fine-tuning question: some claim that the range of physical parameters that allow for intelligent life is very narrow, and this is frequently used as an argument* for the existence of God. Rees kept these concerns at a subtextual level, even while his subsequent speculation was essentially an attack on the fine-tuning argument.

His favored explanation is very interesting: he suggests that, since we can only put a lower bound on the size of the universe, it could be far, far larger than the (roughly 15 billion light-year) observable region, and while the laws of nature and parameters of the universe seem to be locally uniform, they could in fact be varying over very large distances. Thus, some of what we consider laws could really be "local bylaws", and the only explanation for fine-tuning would be a version of the weak anthropic principle. In effect, he said, cosmology becomes an environmental science.

The question period after the talk was relatively unenlightening; people seemed to take Rees' speculations as license to deluge him with their own crackpot theories. (This is actually not unusual for the Oppenheimer lecture anyway.) Rees took the "smile and nod" approach to most of these. One questioner asked if science or philosophy had made any progress on Hume's problem of induction; normally one of my favorite topics, but it's hard to imagine any response to this particular question other than "no", which was basically how Rees answered.

Anyway, the lecture was entertaining, both due to its speculative aspect and to Rees' skill as a lecturer. If any readers get the chance to hear him speak I would recommend it.

*In my opinion this argument is unconvincing: when confronted with a strange and improbable phenomenon, I do not find the invention of an even more strange and complex entity to be a very satisfying explanation.


And 65% of all statistics are made up on the spot.

Posted by: Mason | April 11, 2005 11:15 PM

Was Rees suggesting that the laws of nature could vary in really fundamental ways in other parts of the universe? That, say, the speed of light might be different thirty billion light years from here? Or was he suggesting only tiny variations from what we know? I guess even the latter would be pretty mind-boggling unless you could understand why what we know and experience would not be uniform even throughout an infinite universe. I'm just wondering if he gave examples, or had any math to back up his speculation, or whether he was just trying to keep the audience awake.

Posted by: Dad | April 12, 2005 9:39 AM

Well, Rees concentrated on a few parameters which one might or might not consider fundamental: \Omega, the mass density of the universe, which determines whether it keeps expanding or starts to contract; \Lambda, the cosmological constant that drives the expansion, and Q, the unevenness of the distribution of matter that allows it to clump together into stars, galaxies, etc. \Omega and Q in particular seem less fundamental than the speed of light, in that they are just properties of how matter was distributed at the beginning of the universe. The implication however was that any of the physical constants could vary.

In particular he alluded to experiments that look at spectral lines from very far away, hence very old, atoms, where there are suggestions of small variations in the speed of light. However, he acknowledged that this data is not really conclusive.

He also made a vague reference to string theories predicting things like this, but since it was a "general public" lecture he didn't present anything quantitative.

One questioner accused him of doing metaphysics. He responded that what he is talking about is "speculative physics", in that if certain unified theories are confirmed we would expect to see these things as corollaries.

Posted by: Arcane Gazebo | April 12, 2005 11:57 AM
Post a comment