The Future,

In Theory

Lee Smolin

The Future of the Nature

of the Universe

We are asked to predict the state of our science fifty years from today. Fifty years is a long time, given the pace at which physics and cosmology have progressed over the last several hundred years. But perhaps it is not too long a time to make predictions that will not seem entirely stupid by then. If you look back over the history of science, you will see that often the important questions people were asking had been answered fifty years later. And yet the progress of science has usually been slow enough that people speak roughly the same language as their colleagues working in the same field fifty years earlier.

Let's look back fifty years, then, and note what the big questions were. My own list would include:

1)What is the nature of the strong force that holds atomic nuclei together?

2)What is the nature of the weak force responsible for radioactive decay?

3)Is the Steady State model of the universe right, or might there have been a Big Bang, as speculated by Gamow and other fringe figures?

4)Do protons and neutrons have any internal structure?

5)Why do the proton and neutron have slightly different masses, while the electron is much lighter than either? Why is the neutrino massless? What is the muon and who ordered it?

6)What is the relationship between general relativity and quantum theory?

7)What is the right way to understand the quantum theory?

I think we can confidently assert that now we know the answers to the first four questions. We are still working on the last three. But the first have not been forgotten; indeed, the methods by which those questions were answered form the basis of the training of a theoretical physicist today.

If we look back a hundred years, however, we find that we no longer care about many of the questions people were asking then. I'm not enough of a historian to write a list of questions asked by physicists at the turn of the last century, but they would likely have been more concerned with the properties of the ether than with the properties of atoms. There was no evidence for the existence of physical atoms until a few years later--and, indeed, in 1900 many physicists did not believe that atoms existed. Others, like Ernst Mach, thought the question was not a part of physics because atoms would never be observed. As for astronomy, there was no evidence in 1900 for the existence of galaxies apart from our own Milky Way, nor did anyone have any idea what made the stars shine. So while physicists of the early 1950s would probably have understood the questions that physicists are asking now, no one at the beginning of the twentieth century could have understood even the words that physicists were using in 1950 to talk to each other.

Sometimes science changes so little over fifty years that it makes sense to try to predict what we will know after that span. But there are periods when progress is faster and this is no longer the case. It seems that there is a horizon, somewhere between fifty and a hundred years into the future, beyond which it may be useless to speculate in any detail about the progress of science.

Let's take a moment to consider why this is so. It's probably partly because fifty years is about the length of a scientific career, from the beginning of studies until retirement. This, then, is the span of time over which the conservative tendencies built into the structure of scientific careers act to retard the progress of science. Science is hard, and we scientists prefer to have as good an understanding of what we're doing as possible; thus, unless forced to do otherwise, we prefer to work with techniques and ideas we already understand well. Another factor is that the careers of young scientists are often controlled by senior people nearing retirement, who are in many cases no longer active and therefore unfamiliar with new techniques. Career-savvy graduate students, no matter how imaginative, hesitate to work on something not understood by the powerful old men and women of their field. Thus, in order to think about what my science will be like in fifty years, I imagine what the brightest of my graduate students will be talking about at their retirement parties. My guess is that unless they are forced by data they cannot otherwise explain to make a revolution comparable to that of the early twentieth century, they will be using the language we've taught them. If that's the case, the present exercise may be useful--though the romantics among us would rather anticipate a revolution than confirmation of our own beliefs.

One can also speculate on what was different about the sociology of science in the first half of the twentieth century to enable such enormous progress. Two credible answers come to mind: One is that it was possible for outsiders, such as Albert Einstein and Paul Ehrenfest, to publish in spite of not having university positions; another is that the generation that preceded the inventors of quantum theory was mostly wiped out in World War I, leaving the field open for Heisenberg, Dirac, and their friends.

This said, what will we know about fundamental physics and cosmology in fifty years? Rather than guessing, I propose a method that has a chance of reaching conclusions that won't look silly in the 2050s. I will list the most fundamental questions that are currently unanswered. Then I will ask what developments we may expect in experimental and observational science which will enable answers to them to be checked. I won't worry about theoretical developments, since there are already proposed answers to all of my questions and I assume that over a time span of fifty years we theorists will be able to adjust our theories, or invent new ones, in response to the data.

Here, then, is my list of the seven most important open questions in fundamental physics and cosmology:

1)Is quantum theory true as presently formulated, or will it need to be modified, either to have a sensible physical interpretation or to unify it with relativity and cosmology?

2)What is the quantum theory of gravity? What is the structure of space and time on the Planck scale (10-33 cm, or twenty orders of magnitude smaller than an atomic nucleus)?

3)What explains the exact values of the parameters that determine the properties of the elementary particles, including their masses and the strengths of the forces by which they interact?

4)What explains the large ratios of scales we observe? Why is the gravitational force between two protons forty powers of ten smaller than their electrical repulsion? Why is the universe so big? Why is it at least sixty powers of ten bigger than the fundamental Planck scale? Why is the cosmological constant smaller than any other parameter in physics by roughly the same ratio?

5)What was the Big Bang? What determined the properties of the universe that emerged from it? Was the Big Bang the origin of the universe? If not, what happened before it?

6)What constitutes the dark matter and dark energy that make up between 80 and 95 percent of the density of the universe?

7)How did the galaxies form? What do the patterns we observe in the distribution of the galaxies tell us about the early evolution of the universe?

The first four of these questions continue, and deepen, unanswered questions from fifty years ago. The other three are new. Let us then ask whether the observations and experiments we'll be able to make in 2050 will be sufficient to test answers that theorists may propose to these questions. Of course, anything could be invented in fifty years. If my method is to be believable, we have to be conservative about the development of technology. I will thus consider only technology already existing or under development. In the latter case, I will consider only technology that a sizable fraction of experts believe will work in the next few years. However, for each technology existing or under development, I will assume that over the course of fifty years it will have been developed as far as possible, given only constraints imposed by the laws of physics or economics. Ordinary microscopes have a natural limit given by the wavelength of light, while telescopes have a natural limit imposed by the speed of light in a universe of finite age. Other technologies are more likely to be limited by financial considerations. We can safely assume that no experiment will be done whose cost (at that time) exceeds the defense budget of the United States. Let me hasten to add that I am not an expert in experimental physics or observational cosmology and I have not done a careful study of the limits involved. So my estimates will of necessity be very broad. Based on an extrapolation of current technology to its natural and financial limits, here is what I think we may hope for by mid-century.

We may begin with quantum theory. At present, powerful new techniques are being developed that promise to greatly extend the regime over which the quantum theory has been experimentally tested--techniques chiefly in aid of developing quantum computers. These are macroscopic devices that use quantum effects, such as superposition and entanglement, to do computations impossible for ordinary computers. A quantum computer requires those effects--which have so far been observed only for atomic systems--to work for macroscopic systems like the circuits of a computer, so the devices test those predictions of quantum theory that differ most strongly from classical theory.

Because it has been demonstrated that quantum computers could break all the codes now used by governments, militaries, and business, a lot of money is going into quantum computing. So it is safe to assume that as long as quantum mechanics remains true when extrapolated to macroscopic systems, there will be quantum computers in fifty years, and there are also likely to be quantum communication devices that make use of quantum states extending nonlocally worldwide. And if the present quantum theory is only an approximation of a deeper theory, experiments with quantum computers are likely to reveal this. It is thus reasonable to conclude that fifty years from now we will know the answer to the first question.

Let's turn next to cosmology. By mid-century we will certainly have a detailed picture of the history of the universe, based on observations using the full range of the electromagnetic spectrum, plus neutrinos, cosmic rays, and gravitational waves. The parameters in our current cosmological models will have been measured to high precision, and we will know lots of other facts about the universe, such as the number of black holes, the distributions in space and time of stars, galaxies, black holes, neutron stars, quasars, gamma ray bursters, and other objects. In fact, we will probably know more about the detailed history and properties of the universe than we know now about the history of the surface of our planet. At least in terms of familiarity with the whole range of its phenomena, we will truly be "at home in the universe."

The results will strongly constrain current theories of the early universe, such as inflation. We will also have a detailed picture of how galaxies and the patterns of galactic clusters and superclusters formed. Even without direct observations of dark matter, those observations will strongly constrain theories of the nature of dark matter and dark energy. By mid-century we may or may not have directly observed dark matter and dark energy and learned enough about them to confirm or refute the various theories that have been proposed about them.

Some readers will ask whether all these observations might confirm the Big Bang theory. To answer that question, we must distinguish between two meanings of Big Bang cosmology. I'll call the first the Expanding Universe cosmology: This is the theory that the universe has expanded from a much denser and hotter state for roughly 13 billion years. Among the key events in this story is the decoupling of light and matter when the universe had cooled enough for atoms to be stable. For roughly a million years before that, the universe was filled with a plasma, like the interior of a star. Since the transition, the universe has been filled with a very dilute gas, transparent to light, and all of the structures we see--stars and their planets, galaxies, galactic clusters--were formed. And almost all the chemical elements were formed since the transition, in stars; only helium and other light elements, like deuterium and lithium, were formed beforehand. I think it is unlikely that the outline of this story will be modified fifty years from now. We will know much more about the processes whereby stars, galaxies, and the elements formed, but all the evidence will still support the Expanding Universe theory.

It is also safe to say that we will have observations that strongly constrain our theories about what happened in the very early history of the universe. As we turn back the clock, the density and temperature increase. It is interesting to wonder how far back we can go to constrain theories by observation. By mid-century the part of the theory open to test is likely to extend back at least to the Planck time, a period so small that 1043 of them would fit into one second. Take, for example, the inflation hypothesis. Under a certain set of reasonable assumptions, predictions of this theory are testable in current observations of the fluctuations in the cosmic microwave background. These observations make up one of the greatest achievements of recent science. But even if current observations are compatible with inflation, there remain many open questions: The predictions of inflation are simple and may well be implied by other theories, so measurements that are more detailed may be needed to distinguish inflation from possible rival explanations of the present data. Moreover, there are many different versions of inflation, and further measurements are certainly needed to distinguish them. We hope to have those more detailed measurements of the cosmic microwave background not in fifty years but in five. So it is reasonable, if not certain, to predict that by mid-century it will be old hat to consider theories of the expanding universe testable by detailed observations all the way back from our present era to the Planck time.

But the Planck time is still not the origin of time. Very different from the Expanding Universe theory is the assertion that the Big Bang was the absolute beginning of the universe. Even if we know that the universe is hotter and denser all the way back to some point that may be a fraction of a second after some theoretical beginning, this does not prove that something didn't happen before that to set the expansion in motion. So it remains possible that the universe existed, possibly in some different form, a long time before the moment of the theoretical "Big Bang." To distinguish these kinds of hypotheses, let me refer to them as Origin of the Universe theories.