Monday, June 08, 2009

Lee Smolin and the multiverse

Lee Smolin argues in Physics World against the notion that there exists a multiverse of timeless universes. Smolin believes that the need to invoke a multiverse is rooted in the dichotomy between laws and initial conditions in existing theoretical physics, and suggests moving beyond this paradigm.

A choice of initial conditions, however, is merely one of the means by which particular solutions to the laws of physics are identified. More generally, there are boundary conditions, and free parameters in the equations, which have no special relationship to the nature of time. Each theory in physics represents (a part of) the physical universe by a mathematical structure; the laws associated with that theory select a particular sub-class of models with that structure; and the application of a theory to explain or predict a particular empirical phenomenon requires the selection of a particular solution, i.e., a particular model. The choice of initial conditions, or boundary conditions, or the choice of particular values for the free parameters in the equations, is simply a way of picking out a particular model of a mathematical structure. For example, in general relativity, the structure is that of a 4-dimensional Lorentzian manifold, the Einstein field equations select a sub-class of all the possible 4-dimensional Lorentzian manifolds, and the choice of boundary conditions or initial conditions selects a particular 4-dimensional Lorentzian manifold within that sub-class.

As a consequence, any theory whose domain extends to the entire universe, (i.e. any cosmological theory), has a multiverse associated with it: namely, the class of all models of that theory. Irrespective of whether a future theory abolishes the dictotomy between laws and initial conditions, the application of that theory will require a means of identifying particular models of the mathematical structure selected by the theory. If there is only one physical universe, as Smolin claims, then the problem of contingency will remain: why does this particular model exist and not any one of the other possibilities? The invocation of a multiverse solves the problem of contingency by postulating that all the possible models physically exist.

12 comments:

Sean said...

Is not Smolins general theory/guess is that the laws of physics are evolving?

So in Smolins world, the universe is the way it is now because that's how it evolved. The other possibilities could have existed and might still do.

Nice pic btw, what is it?

Gordon McCabe said...

Yes, you're quite right Sean. However, unless Smolin is proposing to abandon mathematics entirely, specifying the evolutionary history of the laws of physics will simply be a means of picking out one particular model of a mathematical structure.

Gordon McCabe said...

The picture is Multiverse II (2006), by Jon Barwick.

Bob said...

[the problem of contingency will remain: why does this particular model exist and not any one of the other possibilities?]

It just does (exist). It is a coincedence, that things are the way they are. Why is that such an interesting question?

Oh, and if other universes exist, where are they? Next to our own universe, just a 'little bit' further away?

Gordon McCabe said...

There's no contradiction, Bob, between the existence of multiple entities and the absence of a distance between those entities.

I take your first point, however, and it is quite plausible to say that there is a point at which the explanatory regress of science must cease; at that point, one would merely have a description of the ways things are, and there would be no reason for the way things are.

Shawna said...

I'm a fan of the multiverse version. Maybe because that's what I've always had an inclination to believe.

This is all very dense material. Scintillating, but head-achingly dense business.

Will I ever have the tools to fully comprehend the universe?

Most likely not, unfortunately.

Shawna's Study Abroad

Gordon McCabe said...

Welcome to McCabism Shawna!

Sean said...

I have to say that as a non scientist the multiverse has BIG appeal too.

I lost a uncle a couple of months ago and the idea that in another dimension we are having a couple of rounds of golf arguing about mundane matters is very satisfying as well as reassuring, But is that what science tells us?

Without getting side tracked or changing the issue, let me explain it like this. A couple of months ago I had a todo with Sir Bryan Appleyard and his mates about torture in gitmo.

my general view is that water boarding is torture. And if I waterborded you I would have the info I wanted within the hour. its torture and it works sort of thing.

Now if I had to waterborard you 180 times, that tells me that what happened was something different. either they were not doing it right OR they were doing it differently as not to be torture.

There seems to me to be a point when something becomes something else, a tipping point so to speak.

That seems to me is what Smolin is on about. in evolution there are only certain points that can be mapped or in your view given a mathematical models. But at certain points,the rules start behaving differently.

At the end of the day, Math is a language, the most precise that we have, but is it really explaining this bit of the model? what appears to be one thing but in reality is another? the badlands, the part of the model that behaves differently?

This idea of the perfect model might appeal to scientifically minded, schooled in math, but is if how the univesrs works?

Gordon McCabe said...

But I take Smolin to be proposing that the laws of physics evolve, not that they break down into the mathematically ineffable.

Sean said...

Flux is the word I think I am struggling to describe, A state of potential to be something else, but from our viewpoint it still appears to be the same thing as before?

So to evolve they need to be in a state of flux, a state of transfer from one form to the next.

Sorry for the mad rantings, but this stuff is pure mind food, even for those of us who have no idea about what we are eating....but then again I am very suspicious of "models"

just seems to me without time then there is no atomic decay.

Mark Vernon said...

Hi there - This one is definitely for you: Tim Folger's piece about quantum mechanics in the latest issue of Science

Gordon McCabe said...

Cheers Mark. Valentini seems to be working on a Bohmian approach to quantum mechanics (i.e., one in which the probabilities are all epistemic, and the theory is deterministic but nonlocal).