by Lee Smolin
If proposals for unification are so shocking to our previous ways of thinking, how is it that people come to believe them? This is in many ways the crux of our story, for it is a story of several proposed unifications, some of which have come to be strongly believed by some scientists. But none of them have achieved consensus among all scientists. As a consequence, we have lively controversy and, at times, emotional debate, the result of the attempted radical alteration of worldviews. So when someone proposes a new unification, how do we tell whether it is true or not?
As you might imagine, not all proposals for unification turn out to be true. At one time, chemists proposed that heat was a substance, like matter. It was called phlogiston. This concept unified heat and matter. But it was wrong. The right proposal for the unification of heat and matter is that heat is the energy in random motion of atoms. But although atomism had been proposed by ancient Indian and Greek philosophers, it took until the late nineteenth century before the theory of heat as random motion of atoms was properly developed.
In the history of physics, there have been many proposals for unified theories that turned out to be wrong. A famous one was the idea that light and sound were essentially the same thing: They were both thought to be vibrations in matter. Since sound is vibrations in air, light was proposed to be vibrations in a new kind of matter called the aether. Just as the space around us is filled with air, the universe is filled with aether. Einstein killed this particular idea with his own proposal for unification.
All the important ideas that theorists have studied in the last thirty years—such as string theory, supersymmetry, higher dimensions, loops, and others—are proposals for unification. How do we tell which are right and which are not?
I have already mentioned two features that successful unifications tend to share. The first, surprise, cannot be underestimated. If there is no surprise, then the idea is either uninteresting or something we knew before. Second, the consequences must be dramatic: The unification must lead quickly to new insights and hypotheses, becoming the engine that drives progress in understanding.
But there is a third factor that trumps both of these. A good unified theory must offer predictions that no one would have thought to make before. It may even suggest new kinds of experiments that make sense only in light of the new theory. Most important of all, the predictions must be confirmed by experiment.
These three criteria—surprise, new insights, and new predictions confirmed by experiment—are what we will be looking for when we come to judge the promise of current efforts at unification.
Physicists seem to feel a deep need for unification, and some speak as if any step toward further unification must be a step toward the truth. But life is not that simple. At any one time, there can be more than one possible way to unify the things we know—ways that lead science in different directions. In the sixteenth century, there were two very different proposals for unification on the table. There was the old theory, of Aristotle and Ptolemy, according to which the planets were unified with the sun and the moon as part of the celestial spheres. But there was also the new proposal of Copernicus, which unified the planets with Earth. Each had great consequences for science. But at most only one could be right.
We can see here the cost of choosing the wrong unification. If Earth is at the center of the universe, that has tremendous implications for our understanding of motion. In the sky, planets change direction because they are attached to circles whose nature is to rotate eternally. This never happens to things on Earth: Anything we push or throw quickly comes to rest. That is the natural state of things that aren’t attached to cosmic circles. Thus in Ptolemy and Aristotle’s universe there is a big distinction between being in motion and being at rest.
In their world, there is also a big distinction between the heavens and the earth—things on the earth follow laws different from those that obtain in the sky. Ptolemy proposed that certain bodies in the sky—the sun, the moon, and the five known planets—move on circles that themselves move on circles. These so-called epicycles enabled predictions of eclipses and the motions of the planets—predictions that were accurate to 1 part in 1,000, thus showing the fruitfulness of the unification of sun, moon, and planets. Aristotle gave a natural explanation for Earth’s being at the center of the universe: It was composed of Earth-stuff, whose nature was not to move on circles but to seek the center.
To someone educated in that point of view and familiar with how powerfully it explained what we saw around us, Copernicus’s proposal that the planets should be considered one with the earth and not the sun must have been profoundly unsettling. If the earth is a planet, then it and everything on it is in continuous motion. How could that be? This violates Aristotle’s law that everything not on a celestial circle must come to rest. It also violates experience, for if the earth is moving, how come we don’t feel it?
The answer to this puzzle was the greatest unification in all of science: the unification of motion and rest. It was proposed by Galileo and codified in Newton’s first law of motion, also called the principle of inertia: A body at rest or in uniform motion remains in that state of rest or uniform motion unless it is disturbed by forces.
By uniform motion, Newton means motion at a constant speed, in a single direction. Being at rest becomes merely a special case of uniform motion—it is just motion at zero speed.
How can it be that there is no distinction between motion and rest? The key is to realize that whether a body is moving or not has no absolute meaning. Motion is defined only with respect to an observer, who can be moving or not. If you are moving past me at a steady rate, then the cup of coffee I perceive to be at rest on my table is moving with respect to you.
But can’t an observer tell whether he is moving or not? To Aristotle, the answer was obviously yes. Galileo and Newton were forced to reply no. If the earth is moving and we do not feel it, then it must be that observers moving at a constant speed do not feel any effect of their motion. Hence we cannot tell whether we are at rest or not, and motion must be defined purely as a relative quantity.
There is an important caveat here: We are talking about uniform motion—motion in a straight line. (While the earth of course doesn’t move in a straight line, the deviations from it are too small to feel directly.) When we change the speed or direction of our motion, we do feel it. Such changes are what we call acceleration, and acceleration can have an absolute meaning.
Galileo and Newton achieved here a subtle and beautiful intellectual triumph. To others, it was obvious that motion and rest were completely different phenomena, easily distinguished. But the principle of inertia unifies them. To explain how it is that they seem different, Galileo invented the principle of relativity. This tells us that the distinction between moving and being at rest is meaningful only relative to an observer. Since different observers move differently, they distinguish which objects are moving and which are at rest differently. So the fact that each observer makes a distinction is maintained, as it must be. Thus, whether something is moving or not ceases to be a phenomenon that needs to be explained. For Aristotle, if anything moved, there must be a force acting on it. For Newton, if the motion is uniform, it will persist forever; no force is needed to explain it.
This is a powerful strategy that was repeated in later theories. One way to unify things that appear different is to show that the apparent difference is due to the difference in the perspective of the observers. A distinction that was previously considered absolute becomes relative. This kind of unification is rare and represents the highest form of scientific creativity. When it is achieved, it radically alters our view of the world.
Proposals that two apparently very different things are the same often require a lot of explaining. Only sometimes can you get away with explaining the apparent difference as a consequence of different perspectives. Other times, the two things you choose to unify are just different. The need to then explain how things that seem different are really in some
way the same can land a theorist in a lot of trouble.
Let us look at the consequences of Bruno’s proposal that the stars are just like our sun. Stars appear much dimmer than the sun. If they are nevertheless the same, then they must be very far away. The distances he had to invoke were much, much larger than the universe was then thought to be. So Bruno’s proposal seemed at first absurd.
Of course, this was an opportunity to make a novel prediction: If you could measure the distances to the stars, you would find they were in fact much farther away than the planets. Had it been possible to do this in Bruno’s day, he might have escaped the fire. But it was centuries before the distance to a star could be measured. What Bruno had done, in practical terms, was to make an assertion that was untestable, given the technology of the time. Bruno’s proposal conveniently put the stars at such a distance that no one could check his idea.
So sometimes the need to explain how things are unified forces you to posit new hypotheses you simply cannot test. This, as we have seen, does not mean you are wrong, but it does mean that originators of new unifications can easily find themselves on dangerous ground.
And it can get worse. Such hypotheses have a habit of compounding themselves. Copernicus, in fact, needed the stars to be very far away. If the stars were as close as Aristotle believed, you could have disproved the motion of the earth—because as the earth moved, the apparent positions of the stars relative to one another would change. To explain why this effect was not seen, Copernicus and his followers had to believe that the stars were very distant. (Of course, we know now that the stars also move, but they are at such tremendous distances that their positions in our sky change extremely slowly.)
But if the stars were so far away, how could we see them? They must be very bright, perhaps as bright as the sun. Hence Bruno’s proposal for a universe filled with an infinitude of stars fit naturally with Copernicus’s proposal that the earth moved as a planet does.
We see here that different proposals for unification often go together. The proposal that the stars are unified with the sun goes with the proposal that the planets are unified with the earth, and these both require that motion and rest be unified.
These ideas, new in the sixteenth century, opposed another cluster of ideas. Ptolemy’s proposal that the planets be unified with the sun and moon and that all move in epicycles went with Aristotle’s theory of motion, which unified all known phenomena on the earth.
So we end up with two clusters of ideas, each consisting of several proposals for unification. What is at stake, therefore, is often a whole group of ideas, in which different things are unified at different levels. Before the debate is resolved, there can be good reasons for believing each side. Each side can be supported by observation. Sometimes even the same experiment can be interpreted as evidence for competing theories of unification.
To see how this can happen, consider a ball dropped from the top of a tower. What happens? It falls to the ground and lands at the tower’s base. It does not fly off in a westerly direction. Well, you could say, Copernicus and his followers are clearly wrong, for this proves that the earth is not rotating on its axis. Were the earth rotating, the ball would land well away from the base of the tower.
But Galileo and Newton could also claim that the falling ball proves their theory. The principle of inertia tells us that if the ball is moving eastward along with the earth when it is dropped, it will continue to move eastward as it falls. But the ball is moving eastward at the same speed as the tower, so it falls at the tower’s base. The same evidence that an Aristotelian philosopher might have used to prove Galileo wrong was taken by Galileo as proof that his theory was correct.
How do we nevertheless decide which proposed unifications are right and which are wrong? At some point, there is a preponderance of evidence. One hypothesis is shown to be so much more fruitful than the other that a rational person has no choice but to agree that the case is proved. With regard to the Newtonian revolution, there was eventually genuine evidence from observation that the earth moved relative to the stars. But before that happened, Newton’s laws had proved to be correct in so many instances that there was no going back.
However, in the midst of a scientific revolution there are often rational cases to be made for supporting rival hypotheses. We are in such a period now, and we’ll examine conflicting claims for unification in the chapters to come. I will do my best to explain the arguments that support the various sides, while showing why scientists have yet to reach a consensus.
Of course, we do have to exercise caution. Not all evidence said to support a view is solidly based. Sometimes the claims invented to support a theory in trouble are just rationalizations. I recently met a lively group of people standing in the aisle on a flight from London to Toronto. They said hello and asked me where I was coming from, and when I told them I was returning from a cosmology conference, they immediately asked my view on evolution. “Oh no,” I thought, then proceeded to tell them that natural selection had been proved true beyond a doubt. They introduced themselves as members of a Bible college on the way back from a mission to Africa, one purpose of which, it turned out, had been to test some of the tenets of creationism. As they sought to engage me in discussion, I warned them that they would lose, as I knew the evidence pretty well. “No,” they insisted, “you don’t know all the facts.” So we got into it. When I said, “But of course you accept the fact that we have fossils of many creatures that no longer live,” they responded, “No!”
“What do you mean, ‘no’? What about the dinosaurs?”
“The dinosaurs are still alive and roaming the earth!”
“That’s ridiculous! Where?”
“In Africa.”
“In Africa? Africa is full of people. Dinosaurs are really big. How come no one has seen one?”
“They live deep in the jungle.”
“Someone would still have seen one. Do you claim to know someone who has seen one?”
“The pygmies tell us they see them every once in a while. We looked and we didn’t see any, but we saw the scratch marks they make eighteen to twenty feet up on the trunks of trees.”
“So you agree they are huge animals. And the fossil evidence is that they live in big herds. How could it be that nobody but these pygmies have seen them?”
“That’s easy. They spend most of their time hibernating in caves.”
“In the jungle? There are caves in the jungle?”
“Yes, of course, why not?”
“Caves big enough for a huge dinosaur to enter? If the caves are so big, they should be easy to find, and you can look inside and see them sleeping.”
“To protect themselves while they hibernate, the dinosaurs close up the mouths of their caves with dirt so no one can tell they’re there.”
“How do they close up the caves so well they can’t be seen? Do they use their paws, or perhaps they push the dirt with their noses?”
At this point, the creationists admitted they didn’t know, but they told me that “biblical biologists” from their school were in the jungles now, looking for the dinosaurs.
“Be sure to let me know if they bring out a live one,” I said, and went back to my seat.
I am not making this up, and I’m not telling this just for your amusement. It illustrates that rationality is not always a simple exercise. Usually it is rational to disbelieve a theory that predicts something that has never been seen. But sometimes there is a good reason for something never having been seen. After all, if there are dinosaurs, they must be hiding somewhere. Why not in caves in the African jungle?
This may seem silly, but particle physicists have more than once felt the need to invent an unseen particle, such as the neutrino, in order to make sense of certain theoretical or mathematical results. To explain why it was difficult to detect, they had to make the neutrino interact very weakly. In this case, it was the right strategy, for many years later someone was able to devise an experiment that did find
neutrinos. And they did interact very weakly.
So sometimes it is rational not to throw a good theory away when it predicts things that haven’t been seen. Sometimes the hypotheses you are forced to invent turn out to be right. By inventing such ad hoc hypotheses, you can not only keep an idea plausible but also sometimes predict new phenomena. But at some point you begin to stretch credulity. The cave-inhabiting dinosaurs probably qualify here. Exactly when you pass the point where a once good idea becomes not worth the trouble is at first a matter of judgment. There certainly have been cases in which well-trained, smart people disagreed. But eventually a point is reached where there is such a preponderance of evidence that no rational, fair-minded person will think the idea plausible.
One way to assess whether you’ve reached that point is to look at uniqueness. During a scientific revolution, several proposals for unification are often on the table at any given time, threatening to take science in incompatible directions. This is normal, and in the midst of the revolution there does not need to be a rational reason to choose one over the others. At such times, even very smart people who choose between competing views too soon will often be wrong.
But one proposal for unification may end up explaining far more than the others, and it is usually the simplest. At this point, when a single proposal is vastly superior to others in terms of generation of new insights, agreement with experiment, explanatory power, and simplicity, it takes on an appearance of uniqueness. We say it has the ring of truth.
To see how this can happen, let us consider three unifications proposed by one person, the German astronomer Johannes Kepler (1571–1630). Kepler’s lifelong obsession was the planets. Since he believed that the earth was a planet, he knew of six, Mercury out to Saturn. Their motions on the sky had been observed for thousands of years, so there was a lot of data. The most accurate came from Tycho Brahe, a Danish astronomer. Kepler eventually went to work for Tycho to get hold of his data (and after Tycho died he stole it, but that is another story).