Paul Davies is out promoting a new book, and gets a long interview in Salon to explain his ideas. It explains his views better than the last article I linked to.
Here's the key sections:
Now we're into another variant of the anthropic principle -- which is sometimes called the "final anthropic principle" -- where, somehow, the emergence of life and observers link back to the early universe. Now, Wheeler didn't flesh out this idea terribly well, but I've had a go at trying to extend it...The mechanism by which they are changeable over time seems rather vague speculation to me, and he doesn't seem to suggest a way to test the idea. (Although there has been mention recently that whether changes to certain laws of physics have taken place over time is testable.)
It's part of conventional quantum mechanics that you can make observations now that will affect the nature of reality as it was in the past. You can't use it to send signals back into the past. You can't send information back into the past. But the nature of the quantum state in the past can't be separated from the nature of the quantum state in the present.
What we're saying is that as we go back into the past, there are many, many quantum histories that could have led up to this point. And the existence of observers today will select a subset of those histories which will inevitably, by definition, lead to the existence of life. Now, I don't think anybody would really dispute that fact.
What I'm suggesting -- this is where things depart from the conventional view -- is that the laws of physics themselves are subject to the same quantum uncertainty. So that an observation performed today will select not only a number of histories from an infinite number of possible past histories, but will also select a subset of the laws of physics which are consistent with the emergence of life. That's the radical departure. It's not the backward-in-time aspect, which has been established by experiment. There's really no doubt that quantum mechanics opens the way to linking future with past. I'm suggesting that we extend those notions from the state of the universe to the underlying laws of physics themselves. That's the radical step, because most physicists regard the laws as God-given, imprinted on the universe, fixed and immutable. But Wheeler -- and I follow him on this -- suggested that the laws of physics are not immutable.
One area in which I think is a bit inadequately addressed in the interview is the odd "Platonic world" feeling of mathematics.
In my previous post about Davies, I suggested that it was a bit of a stretch for him to say that there was "ultimate meaning" to the universe when he doesn't seem to believe in eternal life of any kind. However, maybe he is a secret admirer of Tipler's Omega Point after all:
Ultimately, it may not be living intelligence or embodied intelligence but some sort of intelligent information-processing system that could become omniscient and fill the entire universe. That's a grand vision that I rather like. Whether it's true or not is another matter entirely.The whole interview is worth reading.