Skip to main content
Cosmology

Cosmology

A quantum leap for cosmology

05 Nov 2001

A theory that unites quantum mechanics and general relativity claims that there was no first moment in time, but it still agrees with the predictions of classical cosmology.

It's in the stars

One of the most challenging problems in modern physics is the application of quantum theory to the universe as a whole. Progress in this area has been plagued by two types of problem: conceptual and technical. The conceptual problems arise from the old difficulties of interpreting quantum theory. The standard interpretations require that the measuring instruments and observers are outside the quantum system described by the wavefunction. In the late 1950s, however, Hugh Everett proposed an interpretation of quantum theory that might apply to systems that include the observers and measuring instruments, but the adequacy of such interpretations has remained controversial to this day.

The technical problems are no less severe or fundamental. Ever since the pioneering work of Bryce DeWitt, Charles Misner and others in the 1960s, quantum cosmology has basically been studied by applying quantum theory to simple models of the universe. These models typically assume that the universe is completely homogeneous. As a result they only have a few degrees of freedom – the radius of the universe and the value of one or more matter fields. One then makes a quantum-cosmological model by quantizing these simple descriptions of the universe.

Many interesting results related to, for instance, inflation and the initial state of the universe have been obtained from these simple models. However, it has remained very controversial whether any of the results would apply to a real quantum theory of gravity, in which all the many degrees of freedom of the gravitational field are treated quantum mechanically.

A number of authors, including Karel Kuchar, have shown that, in general, there is no reason for results concerning such simple models to apply to a real theory. They argue that quantizing a simplified version of a theory that is restricted only to homogeneous solutions would neglect the real physics. For example, if the theory of strong interactions was restricted this way, then much of the physics concerning quarks and hadrons would be missed.

Technical advances

Recently there has been very significant progress on both the technical and conceptual sides of quantum cosmology. On the technical side it is clear that the right thing to do is to study cosmological solutions of a full quantum theory of gravity. If one is interested in a quantum-mechanical system that has some symmetries, such as a molecule or a solid, it is clear one should quantize the whole system first and then study the symmetric states of the resulting theory.

Because of progress over the last decade towards a theory of quantum gravity, it is now possible to do exactly this. The progress is based on the development of a subject called “loop quantum gravity” (also sometimes called “quantum geometry”) that can be used to study the quantization of theories, such as general relativity and supergravity, without approximations. (Supergravity is an extension of general relativity that has additional symmetries that mix bosons, particles with an integer value of “spin”, and fermions, particles with a half-integer spin.) In spite of this progress, many open questions remain that are being addressed by other approaches to quantum gravity, including string theory.

Contrary to many people’s expectations, it has been found that loop quantum gravity theories are well defined quantum mechanically, and many results have been obtained relating to physics at the so-called Planck scale where quantum gravity effects are important. These results could not have been reached by old-fashion techniques, such as perturbation theory, that make the mistake of studying only the quantum physics of weak gravitational waves around symmetric solutions.

Among the results found in loop quantum gravity are that areas and volumes are discrete, in the same sense that the energy of the hydrogen atom is quantized. According to the theory, the only possible values that any measured area or volume can have come from a certain discrete spectrum. These exact spectra have been computed in general relativity and supergravity, and they have implications for real experiments in which cosmic rays and photons from gamma-ray bursts are used to probe the structure of space-time at the Planck scale. This is because the discrete geometry is expected to modify energy-momentum relations at very high energies, thereby affecting the propagation of particles and photons over cosmological distances.

Thus, it is of interest to see if the techniques of loop quantum gravity can make it possible to study quantum cosmology without making the drastic approximations of earlier approaches to the subject. Now Martin Bojowald of Pennsylvania State University in the US and Chopin Soo of the National Cheng Kung University in Taiwan, have independently shown that this is indeed the case.

Striking predictions

In a series of important papers, Bojowald has shown that one can study quantum states that are exact solutions of the full theory of quantum gravity and have the observed symmetries of our universe (Phys. Rev. Lett. 2001 87 121301). These states behave semiclassically when the universe is much older than the Planck time (10-43 s), and therefore agree with all that is known in classical cosmology. But they also reveal striking new predictions concerning the nature of the universe at Planck times.

In particular, Bojowald has discovered that there is never an initial singularity (i.e. a point where the curvature of space-time becomes infinite) and therefore no first moment to time, as Alex Vilenken and others have hypothesized. Nor is there any excursion into a domain in which the universe has a boundary in “imaginary time”, as hypothesized by Jim Hartle, Stephen Hawking and others. Instead the universe continues back before the moment classical cosmology predicts that it began, to a phase where it was previously expanding. This behaviour has been called a “bounce”; it suggests that the big bang arose from an event in a previous universe, either through the collapse of a black hole in that universe or from the collapse of the whole universe.

The hypothesis that the big bang arose from such a bounce is old – it was suggested by Richard Tolman as early as the 1930s, and has been studied in string theory by Gabrielle Veneziano and collaborators under the name of “string cosmology”. (In string theory, matter is constructed form loops of string 1020 times smaller than an atomic nucleus.) However, this is the first time that the replacement of the initial singularity by a bounce has been shown to be a necessary result of an exact quantum theory of gravity.

Bojowald has made other important discoveries, and it appears that much of the established work in quantum cosmology, based on the application of ad hoc hypotheses to simple models, can now be re-examined using the exact theory.

Another topic of much current interest in cosmology is the role of a cosmological constant – an energy density that can be attributed to empty space (see “Quintessence” by R R Caldwell and P J Steinhardt Physics World 2000 November pp31-37). So one can now ask whether loop quantum gravity has anything to say about quantum cosmology in the presence of a cosmological constant. Another important question is whether the new exact approach to quantum cosmology can predict the spectrum of fluctuations observed in the cosmic microwave radiation.

Results on both of these questions have been reported in a recent paper by Soo. He makes use of an old observation by Hideo Kodama that when the cosmological constant is non-zero, one can find an exact quantum state that solves the equations of quantum gravity. This Kodama state has a semiclassical interpretation that predicts a spectrum of quantum fluctuations in space-time. It also has a precise Planck-scale description, which makes use of very elegant mathematics connected to the invariants of graphs and knots. Soo also finds that this state can be used to predict fluctuations of the gravitational field, which may be observed in the microwave background and the distribution of galaxies.

Conceptual advances

Progress on the conceptual side has been no less dramatic. It is based on a new approach to the problem of how to produce a quantum theory for a closed system, such as the universe, in which the observer must be considered part of the system. The key idea, proposed about 10 years ago by the mathematician Louis Crane, is that the quantum state of the universe should be replaced by a whole array of states. There should be one state for each way of dividing the universe into two regions, one containing the observer and the other containing what the observer sees.

This idea has given rise to what are called “relational” approaches to quantum cosmology. The conceptual structure of these approaches was further developed by Carlo Rovelli and others, while their mathematical structure has been clarified by Christopher Isham, Jeremy Butterfield, Fontini Markopoulou and collaborators. They have given relational quantum cosmology an elegant formalism in terms of a mathematical structure called topos theory. Butterfield and Isham have also shown how an approach to quantum cosmology called “consistent histories” – originally applied to quantum cosmology by Murray Gell-Man and Hartle plus others – may be consistently reformulated as such a relational quantum theory.

Further progress was made by Markopoulou, who showed that the different regions associated with the different states can be specified in terms of the causal structure of a space-time. Basically the outcome is that the array of different states proposed by Crane are connected by the flow of quantum information through the quantum universe.

These ideas may seem abstract, but they have been applied to several important questions. For example, Neil Turok has applied the basic idea that a quantum state is associated with an observer’s past to make progress on the no-boundary proposal for quantum theory. Meanwhile, Tom Banks and Willie Fischler have argued that the same idea may be used to study string theory and M-theory in expanding universes.

There remain many open questions, but it is already clear that, on both the conceptual and technical side, quantum cosmology is waking up from a long period in which it consisted mainly of models that were proposed many years ago.

Modern techniques from quantum gravity, field theory and mathematics are already leading to new predictions concerning the very early universe, and are greatly clarifying what theorists are doing when they apply quantum theory to the universe as a whole.

Copyright © 2024 by IOP Publishing Ltd and individual contributors