June 27, 2008
BOY, NO ONE BELIEVES IN GALILEO ANYMORE:
Stephen Hawking's explosive new theory (Roger Highfield, 26/06/2008, Daily Telegraph)
By quantum lore, when a particle of light travels from A to B, it does not take one path but explores every one simultaneously, with the more direct routes being used more heavily.This is called a sum over histories and Prof Hawking and Prof Hertog propose the same thing for the cosmos.
In this theory, the early universe can be described by a mathematical object called a wave function and, in a similar way to the light particle, the team proposed two years ago that this means that there was no unique origin to the cosmos: instead the wave function of the universe embraced a multitude of means to develop.
This is very counter intuitive: they argued the universe began in just about every way imaginable (and perhaps even some that are not). Out of this profusion of beginnings, like a blend of a God’s eye view of every conceivable kind of creation, the vast majority of the baby universes withered away to leave the mature cosmos that we can see today.
But, like any new idea, there were problems. [...]
Most models of the universe are bottom-up, that is, you start from well-defined initial conditions of the Big Bang and work forward. However, Prof Hertog and Prof Hawking say that we do not and cannot know the initial conditions present at the beginning of the universe. Instead, we only know the final state - the one we are in now.
Their idea is therefore to start with the conditions we observe today - like the fact that at large scales one does not need to adopt quantum lore to explain how the universe (it behaves classically, as scientists say) - and work backwards in time to determine what the initial conditions might have looked like.
The Hawkingcentric universe. Posted by Orrin Judd at June 27, 2008 7:19 PM
Not once the wave function is collapsed. Someone chose this history.
Posted by: oj at June 27, 2008 11:03 PMThe article says this links up with string theory and may actually produce testable predictions. That would be a good thing.
Posted by: PapayaSF at June 27, 2008 11:33 PMIt's an imaginary thing.
Posted by: oj at June 28, 2008 10:06 AMCollapsed wavefunctions are irrelevant.
"Collapse of the wavefunction" is by definition the transition of a quantum system from a mixed state to a pure state. But "mixed" and "pure" are defined relative to some physical quantity, such as position. The mathematics of QM require that a "pure" state relative to one quantity be a mixed state relative to another, such as momentum (the "canonically conjugate" one, in general). This is where the Uncertainty Principle comes from.
So all states are mixed. It's just a question of which physical quantities are mixed, and how much. A collapsed wavefunction may be less mixed, but it's still mixed.
If you accept QM, you accept that physical systems exist in a multiplicity of states. In the past, in the future, in the present. You don't have to admit it, and indeed this has been the traditional approach. The incoherence of the traditional approach has been recognized since at least the 1930s, but coming up with something better has taken a lot longer.
You could always reject QM, but we need it to design semiconductors.
Bingo! Creation is the transition. Someone chose.
Posted by: oj at June 28, 2008 3:14 PM
They're halfway there.
The "bottom-up" model is Newtonian classical mechanics quantized. The sum-over-histories model is Hamiltonian classical mechanics quantized. The thing is, Newton and Hamilton have the same content, in different forms. One is "2+2" and the other is "3+1". Drawing conclusions from the differences in form is like saying "Hey, 2+2 has only even numbers and 3+1 only odd numbers! They couldn't be more different!"
So it's taken about 80 years for someone to realize that Hamilton is just as good or bad a guide as Newton. More precisely, just as meaningless. If you draw a different conclusion from 2+2 than from 3+1, the only thing you know is that your conclusion is invalid.
The only conclusion that could possibly be meaningful is one that works with both. In the case under discussion, I suspect this would involve recognizing that the universe is still traversing all possible histories.
Posted by: Bob Hawkins at June 27, 2008 10:00 PM