Looking into a black hole

Cosmologists may not agree on the origins or future of the universe in our own time, or even in Hubble Time, but Joseph Silk finds some theoretical 'gems' in a new compendium of lectures

The scope of astronomy is considerable, ranging from cosmology to comets, and even to the earth's atmosphere, both as a hindrance and as a model for the closest planet. E. A. Milne was a remarkable mathematician, of outstanding originality and breadth, who devoted himself to astronomical problem solving. The Universe Unfolding provides a record of the Milne lecture series, established in his memory, and contains a broad ranging overview of areas where he contributed. The viewpoints expressed are often personal but always authoritative, and invariably original. Most unusually for a lecture compendium, many of the talks include humorous asides that help capture the atmosphere of the original lectures.

Fred Hoyle was the first Milne lecturer, and develops a theme that has fascinated biologists, philosophers, theologians and cosmologists alike. The mystery of the origin of life can be reduced to a mathematical calculation, so Hoyle and many others assert. The simplest living system contains many amino acids which link up to form proteins. The blueprint for reproducing amino acids is carried by DNA. The information content carried by DNA is the key to reproducing life. One can ask how many proteins characteristic of living cells can be constructed by assembling at random the 20 biologically significant amino acids in chains of several hundred. The number of permutations is impressively large, perhaps 100000000 for a complex organism. Yet there are only modest variations of protein structure among plants and animals. The information content of life is vast, and it is very far from random.

To explain such a vast number, Hoyle peers in vain at the clues from the earth, from the chemical elements and from stars. Super-astronomical numbers are not to be found. He resorts to cosmology. Here he runs into a minor problem. The Big Bang, he contends, has limited possibilities, because of the finite time elapsed since creation. Enter the steady state cosmology, with an infinite sequence of past cycles of expansion and contraction (if we accept its latest incarnation). There has been time enough for carbonaceous molecules to propagate on the backs of iron whiskers blown out by stars into the depths of intergalactic space. The organic matter collects in icy blocks in interstellar clouds, and eventually £ the forming earth as a shower of fiery comets. And so we have the information content of life explained: after a long enough time has elapsed. The time since the Big Bang is usually named after the discoverer of the expansion of the universe, Edwin Hubble: a Hubble time is approximately 12 billion years. Only after a hundred million Hubble times have elapsed has sufficient organic matter been generated. Life would be universal.

There is an alternative. Consider an open universe. Space is infinite. Somewhere, therefore, Hoyle's odds are satisfied. But in this location life is unique, or almost so. One can estimate that, if over ten billion years of cosmic history, intelligent life was capable of propagation from star to star and galaxy to galaxy at say 10 per cent of light speed via, for example, the dispersal of robotic probes, life would propagate in random directions and spread out over a distance of perhaps one hundred million light years. This amounts to about one millionth of the volume of the visible universe. Imagine a random time traveller, able to leap from one Hubble volume to another. Once she stumbled into our causally connected horizon volume, she would find that there is one chance in a million that life has emerged in our patch of the universe. Not bad odds really, and rather better than the chances of winning the National Lottery. Of course the difficulty is in finding the right horizon volume. But one could argue that only within the realm of causality could one apply the usual laws of physics and compute the odds of encountering life.

Large numbers also await us as we explore the future of the universe as well as its past. Martin Rees argues that the very laws of physics and the constants of nature may be subject to random variations. In our observable universe, a delicate balance is maintained between the relative strengths of the fundamental forces of physics. If gravity were not the weakest force, our universe would have evolved so rapidly that life would never have developed. The open universe comes to the rescue again: within the infinite range of permutations there must be at least one horizon volume somewhere that leads to a satisfactory outcome. If this patch inflates, it will soon dominate the universe, and all will be well.

There is a price to be paid. An infinite universe expands for ever. Observations indeed suggest that our universe is unlikely to end up in a future big "crunch'', but will continue to expand into a dark, cold infinite space, populated by dim relics of stars. The Earth will be destroyed when the sun swells into a red giant that will swell to about the orbit of our planet, and then subside into a white dwarf. This occurs in about five billion years. Life on earth will cease. But life may continue on planets around other stars. Eventually even the stars will fade away. Stars will die, as their nuclear fuel is exhausted. Gas to form new stars can no longer be resupplied by dying or exploding stars. Eventually there is insufficient raw material for stellar rebirth. The galaxy gradually fades away as stars die, never to be renewed. Our galaxy will largely consist of stellar relics, white dwarfs and neutron stars. Perhaps in this depressingly cold space, life can prepare to hibernate, for worse is to come. Whatever gas remains will accumulate in the central cores of galaxies. Its fate is to be accreted by the massive black hole that lies in wait at the galactic centre. Such massive black holes are commonly found in the central cores of galaxies. Their formation is associated with the early gas-rich stages of galaxy evolution, when the dense cores of galaxies formed. Matter is accreted onto the central massive black hole, which grows and radiates away the energy of the infalling matter. An ultra-luminous phase results, in which the central object can outshine by far all of the stars in the galaxy. A quasar has formed.

Quasars are the primary manifestation of super-massive black holes in their active phase. They are the most luminous objects in the universe. Quasars are short lived: after a few tens of millions of years their fuel supply is exhausted. Observations of nearby galactic nuclei reveal the presence of dead quasars, super-massive black holes, by their perturbing effect on the orbits of nearby stars. Indeed a million-solar-mass black hole even lurks in the core of our own Milky Way galaxy.

Once the gas supply to a quasar is exhausted, the central black hole is inert and almost undetectable. But in the distant future black holes are likely to play an increasingly important role in galactic evolution. After some hundreds of billions of years, all hydrogen burning stars have died. But dynamical processes still occur. Stars interact gravitationally. The timescale for dynamical evolution is long but finite. A cluster of stars relaxes via gravitational encounters to form a dense core surrounded by a more loosely bound halo. The densest agglomeration of stars surrounds the central black hole, which disrupts stars and accretes their debris. Even weak stellar encounters eventually drive in enough stars to ensure that the central black hole grows. Such encounters are at the expense of other stars that are ejected out into the dark depths of intergalactic space. After some billions of billions of years, dynamical evolution has proceeded to the point that the structure of galaxies has dramatically changed. The universe consists of massive black holes surrounded by clouds of dim white dwarfs and neutron stars, the ashes of once luminous stars. Speculation does not stop here. Grand unified theories predict that protons decay after some 1032 years, and the only objects surviving are massive black holes. Even these too eventually evaporate. The vacuum of space consists of pairs of virtual particles that appear and disappear on so short a timescale that no physical laws are violated. Stephen Hawking first demonstrated that black holes do not exist forever but can disrupt the surrounding fabric of space by capturing some of these virtual particles. The consequence is that virtual pairs are disrupted and real particles are emitted by the black hole. There is a net loss of mass by the black hole.

Black holes evaporate: even black holes can die. The more massive the black hole, the longer its lifetime. But sooner or later, its day of reckoning arrives. This takes a mere 1090 years. If we are prepared to face the ultimate future, the universe contains only energy in the form of electromagnetic radiation and light particles or leptons.

Much larger numbers lie in store. There are other possible universes, although there can be no proof of this. For our universe may have been a unique event. One cannot check one's pet theory against rival horizon volumes, and one cannot compute probabilities for one-off events. One must know something more about how those odds were stacked. Inflationary cosmology does provide a bizarre possibility. It tells us, in principle, how much larger than our Hubble volume the universe actually is. But it provides no guidance on allowable variations in the laws of physics: on the contrary, its application presumes their invariance.

If we accept Einstein's theory of gravitation as having always been valid, we can at least assess the odds of finding a universe as simple as our own, by considering the possible structure of the initial singularity. The degree of improbability of our universe can be estimated from the need to have a well-ordered beginning. It would not do to have a universe teeming with super-massive black holes rather than galaxies.

This issue is tackled by Roger Penrose, who estimates the odds against such a universe as we inhabit to have emerged at random to be at least 1010123 to 1. Something is wrong, no doubt a consequence of applying general relativity theory to a regime where quantum theory must play a role. This probability estimate is a symptom of a fatal weakness of the theory in the vicinity of singularities.

One must await the emergence of a Theory of Everything, the long-sought-after quantum cosmology of the beginning of the universe, among other things, before we can expect to have a better understanding of our origins. There are many other gems in this book. Some are outdated, some err on the technical side, but the overall impression is of a grand unfolding of ideas about the cosmos. The lectures are articulate and personal, authoritative and highly readable, and span the breadth of the astronomy to which E. A. Milne brought so many fruitful insights. The ensemble provides a fitting tribute to his work.

Joseph Silk is professor of astronomy, University of California at Berkeley, United States.