Wednesday, May 10, 2006

The Big Bounce

The discovery in 1996 that the universe is not just expanding but accelerating was inconvenient because it meant that cosmologists could no longer ignore the question of the cosmological constant. The acceleration is said to be caused by ‘dark energy’ that makes empty space repulsive, and the most obvious candidate for that is the vacuum energy, due to the constant creation and annihilation of particles and their antiparticles. The problem is that quantum theory implies that this energy should be enormous – too great, in fact, to allow stars and galaxies to form at all. While we could assume that the cosmological constant was zero, it was reasonable to imagine that this energy was somehow cancelled out perfectly by another aspect of physical law, even if we didn’t know what it was. But now it seems that such ‘cancellation’ is not perfect, but is absurdly fine-tuned to within a whisker of zero: to one part in 10^120, in fact. How do we explain that?

A new proposal invokes a cyclic universe. I asked one of its authors, Paul Steinhardt, about the idea, and he made some comments which didn’t find their way into my article but which I think are illuminating. So here they are. Thank you, Paul.


PB: How is a Big Crunch driven, in a universe that has been expanding for a trillion years or so with a positive cosmological constant, i.e. a virtually empty space? I gather this comes from the brane model, where the cyclicity is caused by an attractive potential between the branes that operates regardless of the matter density of the universe - is that right?

PS: Yes, you have it exactly right. The cycles are governed by the spring-like force between branes that causes them to crash into one another at regular intervals.

PB: What is your main objection to explaining the fine-tuning dilemma using the anthropic principle? One might wonder whether it is more extravagant to posit an infinite number of universes, with different fundamental constants, or a (quasi?)infinite series of oscillations of a single universe.

PS: I have many objections to the anthropic principle. Let me name just three:
a) It relies on strong, untestable* assumptions about what the universe is like beyond the horizon, where we are prevented by the laws of physics from performing any empirical tests.
b) In current versions, it relies on the idea that everything we see is a rare/unlikely/bizarre possibility. Most of the universe is completely different - it will never be habitable; it will never have physical properties similar to ours; and so on. So, instead of looking for a fundamental theory that predicts what we observe as being LIKELY, we are asked to accept a fundamental theory that predicts what we see is UNLIKELY. This is rather significant deviation from the kind of scientific methodology that has been so successful for the last 300 years.

*I would like to emphasize that I said "untestable assumptions". Many proponents of the anthropic model like to argue that they make predictions and that those predictions can be tested. But, it is important to appreciate that this is not the standard that must be reached for proper science. You must be able to test the assumptions as well. For example, the Food and Drug Administration (thankfully) follow proper scientific practice in this sense. If I give you a pill and "predict" it will cure your cold; and then you take the pill and your cold is cured; the FDA is not about to give its imprimitur to your pill. You must show that your pill really has the active ingredient that CAUSED the cure. Here, that means proving that there is a multiverse, that the cosmological constant really does vary outside our horizon, that it follows the kind of probability distribution that is postulated, etc. – all things that cannot ever be proved because they entail phenomena that lie outside our allowed realm of observation.

PB: Could you explain how your model of cyclicity and decaying vacuum energy leads to an observable prediction concerning axions - and what this prediction is? (What are axions themselves, for example?)

PS: This may be much for your article, but....
Axions are fields that many particle physicists believe are necessary to explain a well-known difficult of the "standard model" of particles called "the strong CP problem." For cosmological purposes, these are examples of very light, very weakly interacting fields that very slowly relax to the small value required to solve the strong CP problem. In string theory, there are many analogous light fields; they control the size and shape of extra dimensions; they are also light and slowly relax.

A potential problem with inflation is that inflation excites all light fields. It excites the field responsible for inflation itself, which is what give rise to the temperature variations seen in the cosmic microwave background and are responsible for galaxy formation. So this is good.

But what is bad, potentially, is that they also excite the axion and all light degrees of freedom. This acts like a new form of energy in the universe that can overtake the radiation and change the expansion history of the universe in a way that is cosmologically disastrous. So, you have to find some way to quell these fields before they do their damage. There is a vast literature on complex mechanisms for doing this. Even so, some have become so desperate as to turn to the anthropic principle once again (maybe we live in the lucky zone where these fields aren't excited).

In the cyclic models, these fields would only be excited when the cosmological constant was very large, which is a long, long, LONG time ago. There have been so many cycles (and these do not disturb the axions or the other fields) that there has been plenty of time to relax away to negligible values.

In other words, the same concept being used to solve the cosmological constant problem – namely, more time – is also automatically ensuring that axions and other light fields are not problematic.

1 comment:

Unknown said...

a) It relies on strong, untestable* assumptions about what the universe is like beyond the horizon...

WHOA!!!... uh no, actually, YOU are relying on a strong untestable assumption that the universe extends beyond the event horizon.

For starters... but I won't even bother with the rest of the lameness.

Hype... it's the "new-black" of physicists.