2nd
Stephen Hawking on the univers’s origin
Picture: Starry-Eyed Hubble Celebrates 20 Years of Awe and Discovery / NASA Goddard Space Flight Center
“Spontaneous creation is the reason there is something rather than nothing, why the universe exists, why we exist. It is not necessary to invoke God to light the blue touch paper and set the universe going. (…)”
The universe looks more and more like a quantum phenomenon, in which a multitude of histories diverge. This is what Hawking calls top-down cosmology. Space and time fizzle out, so it can’t be said that there is a time before the big bang — just as you can’t say that there is something north of the North Pole. (I’m talking “north,” not “up.”)
“Gravity is part of the picture because it helps keep the cosmic balance sheet in line. (…) Because gravity shapes space and time, it allows space-time to be locally stable but globally unstable. On the scale of the entire universe, the positive energy of the matter can be balanced by the negative gravitational energy, and so there is no restriction on the creation of whole universes.”
— Hawking says God’s not needed. So?, Cosmic Log on msnbc.com, Sep 2, 2010. See also Stephen Hawking, The Grand Design
Cosmology from the Top Down
“The bottom up approach is more problem in cosmology however, because we do not know what the initial state of the universe was, and we certainly can’t try out different initial states, and see what kinds of universe they produce. (…)
I think the universe may have had an initial de Sitter stage considerably larger than the Planck scale.
I now turn to pre Big Bang scenarios, which are the main alternative to inflation. I shall take them to include the Ekpyrotic and cyclic models, as well as the older pre big bang scenario. The observations of fluctuations in the microwave background, show that there are correlations on scales larger than the horizon size at decoupling. These correlations could be explained if there had been inflation, because the exponential expansion, would have meant that regions that are now widely separated, were once in causal contact with each other. On the other hand, if there were no inflation, the correlations must have been present at the beginning of the expansion of the universe. Presumably, they arose in a previous contracting phase, and somehow survived the singularity, or brane collision. It is not clear if effects can be transmitted through a singularity, or if they will produce the right signature in the microwave background. But even if the answer to both of these questions is yes, the pre big bang scenarios do not answer the central question of cosmology, why is the universe, the way it is. All the pre big bang scenarios can do, is shift the problem of the initial state from 13 point 7 billion years ago, to the infinite past. But a boundary condition is a boundary condition, even if the boundary is at infinity.The present state of the universe, would depend on the boundary condition in the infinite past. The trouble is, there’s no natural boundary condition, like the universe being in its ground state. The universe doesn’t have a ground state. It is unstable, and is either expanding or contracting. The lack of a preferred initial state in the infinite past, means that pre big bang scenarios, are no better at explaining the universe, than supposing that someone wound up the clockwork, and set the universe going at the big bang. (…)
Picture source: João E. Steiner, The origin of the universe
(…) One of the first acts of my research career, was to show with Roger Penrose, that any reasonable classical cosmological solution, has a singularity in the past. This implies that the origin of the universe, was a quantum event. This means that it should be described by the Feynman sum over histories. The universe doesn’t have just a single history, but every possible history, whether or not they satisfy the field equations. Some people make a great mystery of the multi universe, or the many worlds interpretation of quantum theory, but to me, these are just different expressions of the Feynman path integral. (…)
(…) there is no way one can rule out the final surface, from belonging to a different universe to the initial surface. In fact, because there are so many different possible universes, they will dominate, and the final state will be independent of the initial. It will be given by a path integral over all metrics whose only boundary is the final surface. In other words, it is the so called no boundary quantum state.
If one accepts that the no boundary proposal, is the natural prescription for the quantum state of the universe, one is led to a profoundly different view of cosmology, and the relation between cause and effect. One shouldn’t follow the history of the universe from the bottom up, because that assumes there’s a single history, with a well defined starting point and evolution. Instead, one should trace the histories from the top down, in other words, backwards from the measurement surface, S, at the present time. The histories that contribute to the path integral, don’t have an independent existence, but depend on the amplitude that is being measured. As an example of this, consider the apparent dimension of the universe. The usual idea is that spacetime is a four dimension nearly flat metric, cross a small six or seven dimensional internal manifold. But why aren’t theremore large dimensions. Thy are any dimensions compactified. There are good reasons to think that life is possible only in four dimensions, but most physicists are very reluctant to appeal to the anthropic principle. They would rather believe that there is some mechanism that causes all but four of the dimensions to compactify spontaneously. Alternatively, maybe all dimensions started small, but for some reason, four dimensions expanded, and the rest did not. I’m sorry to disappoint these hopes, but I don’ t think there is a dynamical reason for the universe to appear four dimensional. (…)
We live in a universe that appears four dimensional, so we are interested only in amplitudes for surfaces with three large dimensions. This may sound like the anthropic principle argument that the reason we observe the universe to be four dimensional, is that life is possible only in four dimensions. But the argument here is different, because it doesn’t depend on whether four dimensions, is the only arena for life. Rather it is that the probability distribution over dimensions is irrelevant, because we have already measured that we are in four dimensions. (…)
Many physicists believe that string theory, will uniquely predict the standard model, and the values of its 40 or so parameters. The bottom up picture would be that the universe would begin with some grand unified symmetry, like E8 cross E8. As the universe expanded and cooled, the symmetry would break to the standard model, maybe through intermediate stages. The hope would be that String theory, would predict the pattern of breaking, the mass, couplings and mixing angles. Personally, I find it difficult to believe that the standard model, is the unique prediction of fundamental theory. It is so ugly, and the mixing angles etc, seem accidental, rather than part of a grand design. (…)
In string stroke M theory, low energy particle physics is determined by the internal space. It is well known that M theory has solutions with many different internal spaces. If one builds the history of the universe from the bottom up, there is no reason it should end up with the internal space for the standard model. However, if one asks for the amplitude for a space like surface with a given internal space, one is interested only in those histories which end with that internal space. One therefore has to trace the histories from the top down, backwards from the final surface. (…)
How can one get a non zero amplitude for the present state of the universe, if as I claim, the metrics in the path integral, have no boundary apart from the surface at the present time. I can’t claim to have the definitive answer, but one possibility would be if the four dimensional part of the metric, went back to a de Sitter phase. Such a scenario is realized in trace anomaly driven inflation, for example. In the Lorentzian regime, the de Sitter phase would extend back into the infinite past. It would represent a universe that contracted to a minimum radius, and then expanded again. But as we know, Lorentzian de Sitter can be closed off in the past by half the four sphere. One can interpret this in the bottom up picture, as the spontaneous creation of an inflating universe from nothing. Some pre big bang or Ekpyrotic scenarios, involving collapsing and expanding universes, can probably be formulated in no boundary terms, with an orbifold point. However, this would remove the scale free perturbations which, it is claimed, develop during the collapse, and carry on into the expansion. So again it is a no no for pre big bang and Ekpyrotic universes.
In conclusion, the bottom up approach to cosmology, would be appropriate, if one knew that the universe was set going in a particular way, in either the finite, or infinite past. However, in the absence of such knowledge, it is better to work from the top down, by tracing backwards from the final surface, the histories that contribute to the path integral. This means that the histories of the universe, depend on what is being measured, contrary to the usual idea that the universe has an objective, observer independent, history. The Feynman path integral allows every possible history for the universe, and the observation, selects out the sub class of histories that have the property that is being observed. There are histories in which the universe eternally inflates, or is eleven dimensional, but they do not contribute to the amplitudes we measure. I would call this the selection principle, rather than the anthropic principle because it doesn’t depend on intelligent life. Life may after all be possible in eleven dimensions, but we know we live in four. (…)
We can’ t tell whether the universe was likely to have the values we observe, or whether it was just a lucky chance. However, it is note worthy that the parameters we measure seem to lie in the interior of the anthropically allowed range, rather than at one end. This suggests that the probability distribution is fairly flat, not like the exponential dependence on the density parameter, omega, in the open inflation that Neil Turok and I proposed. In that model, omega would have had the minimum value required to form a single galaxy, which is all that is anthropically necessary. All the other galaxies which we see, are superfluous.
Although the theory can not predict the average values of these quantities, it will predict that there will be spatial variations, like the fluctuations in the microwave background. However the size of these variations, will probably depend on moduli or parameters that we can’ t predict. So even when we understand the ultimate theory, it won’t tell us much about how the universe began. It can not predict the dimension of spacetime, the gauge group, or other parameters of the low energy effective theory. On the other hand, the theory will predict that the total energy density, will be exactly the critical density, though it won’ t determine how this energy is divided between conventional matter, and a cosmological constant, or quintessence. The theory will also predict a nearly scale free spectrum of fluctuations. But it won’t determine the amplitude. So to come back to the question with which I began this talk. Does string theory predict the state of the universe. The answer is that it does not. It allows a vast landscape of possible universes, in which we occupy an anthropically permitted location. But I feel we could have selected a better neighbourhood.”
— Stephen Hawking, Cosmology from the Top Down, Department of Applied Mathematics and Theoretical Physics, University of Cambridge, 29 May 2003 (pdf)
See also:
☞ Tim Maudlin, What Happened Before the Big Bang? The New Philosophy of Cosmology, The Atlantic, Jan 2012.
