Knocking on Heaven's Door: How Physics and Scientific Thinking Illuminate the Universe and the Modern World - Lisa Randall (2011)



For quite some time physicists had been looking forward to the LHC turning on. Data are essential to scientific progress, and particle physicists had been starved for high-energy data for years. Until the LHC provides answers, no one can know which of the many suggestions for what might underlie the Standard Model are on the right track. But before this book explores several of the more intriguing possibilities, we’ll take a detour in these next few chapters to consider some important questions about risk and uncertainty that are critical both to understanding how to interpret the LHC’s experimental studies and to many issues that are relevant in the modern world. We’ll begin this excursion with the topic of LHC black holes, and how they just might have received a bit more attention than they deserved.


Physicists are currently considering many suggestions for what the LHC might ultimately find. In the 1990s, theorists and experimenters first got excited about a particular newly identified class of scenarios in which not just particle physics, but gravity itself, is modified, and would produce new phenomena at LHC energies. One interesting potential consequence of these theories attracted a good deal of attention, especially from people outside the physics community. This was the possibility of microscopic low-energy black holes. Such tiny extra-dimensional black holes might actually be produced if ideas about additional dimensions of space, such as those that Raman Sundrum and I had proposed, turn out to be correct. Physicists had optimistically predicted that such black holes—if created—could provide one verification of such ideas about modified gravity.

Mind you, not everyone was so enthusiastic about this possibility. Some people in the United States and elsewhere worried that the black holes that could be created might suck in everything on Earth. I was often asked about this potential scenario after my public lectures. Most questioners were satisfied when I explained why there was no danger. Unfortunately, however, not everyone had the opportunity to learn the whole story.

Walter Wagner, a high school teacher and a botanical garden manager in Hawaii, who is also a lawyer and was a nuclear safety officer, together with the Spaniard Luis Sancho, an author and self-described researcher on time theory, were among the most militant of the alarmists. These two went so far as to file a lawsuit in Hawaii against CERN, the U.S. Department of Energy, the National Science Foundation, and the American accelerator center Fermilab, in order to hinder the LHC’s start. If the goal had been simply to delay the LHC, you might think sending a pigeon to drop a piece of baguette to gum up the works would have been simpler (this actually happened, though the bird was ostensibly an independent agent). But Wagner and Sancho were interested in a more permanent forestalling of the LHC’s operation, so they pressed on.

Wagner and Sanchez were not the only ones who worried about a black hole crisis. A book by public interest trial attorney Harry V. Lehmann, which seemed to concisely summarize the concerns, was entitled No Canary in the Quanta: Who Gets to Decide If the Large Hadron Collider Is Worth Gambling Our Planet? A blog about it concentrated on fears from the September 2008 explosion and questioned whether the LHC could safely start again. The chief concern, however, centered not on the technology failure that was responsible for the September 19 mishap, but on the actual physical phenomena that the LHC might produce.

The purported threats that Lehmann and many others described about the “Doomsday machine” focused on black holes that they suggested could lead to the implosion of the planet. They worried about a lack of reliable risk assessment in light of the reliance on quantum mechanics in the LHC Safety Assessment Group’s study—given claims by Richard Feynman and others that “no one understands quantum mechanics”—as well as uncertainties due to the many unknowns in string theory, which they thought to be relevant. Their questions involved whether it is permissible to risk the Earth for any reason, even when risks are supposed to be tiny, and who should take charge of deciding.

Though the instantaneous destruction of the Earth is certainly more apocalyptic a concern, in reality, the latter questions are more appropriate to other discussions—such as those concerning global warming. Hopefully this chapter and the next will convince you that your time is better spent worrying about the depletion of the contents of your 401(k) than fretting about the disappearance of the Earth by black holes. Although schedules and budgets posed a risk for the LHC, theoretical considerations, supplemented by careful scrutiny and investigations, demonstrated that black holes did not.

To be clear, this doesn’t mean the questions shouldn’t have been asked. Scientists, like everyone else, need to anticipate possible dangerous consequences of their actions. But for the question of black holes, physicists built on existing scientific theories and data to evaluate the risk, and thereby determined there was no worrisome threat. Before moving on to a more general discussion of risk in the chapter that follows, this chapter explores why anyone even considered the possibility of LHC black holes, and why the doomsday fears about them that some suggested were ultimately misguided. The details this chapter discusses aren’t going to be important for the general discussion following, or even for the next part’s outline of what the LHC will explore. But it serves as an example of how physicists think and anticipate, and sets the stage for the broader considerations of risk that follow.


Black holes are objects with such strong gravitational attraction that they trap anything that approaches too closely. Whatever comes within a radius known as the event horizon of the black hole gets engulfed and becomes imprisoned inside. Even light, which seems rather inconspicuous, succumbs to a black hole’s enormous gravitational field. Nothing can escape a black hole. A Trekkie friend jokes that they are the “perfect Borgs.” Any object that encounters a black hole gets assimilated, since the laws of gravity dictate that “resistance is futile.”

Black holes form when enough matter gets concentrated inside a small enough region that the force of gravity becomes indomitable. The size of the region required to make a black hole depends on the amount of mass. Smaller mass must congregate in a proportionately smaller region, while larger mass can be distributed over a larger region. Either way, when the density is enormous and a critical mass is within the required volume, the gravitational force becomes irresistible and a black hole is formed. Classically (which means according to calculations that ignore quantum mechanics), these black holes grow as they accrete nearby matter. Also according to such classical calculations, these black holes wouldn’t decay.

Before the 1990s, no one thought about creating black holes in a laboratory since the minimum mass required to make a black hole is enormous compared to a typical particle mass or the energies of current colliders. After all, black holes embody very strong gravity, whereas the gravitational force of any individual particle that we know of is negligible—far less than other forces such as electromagnetism. If gravity jibes with our expectations, then in a universe composed of three dimensions of space, particle collisions at accessible energies fall far short of the requisite energy. Black holes do, however, exist throughout the universe—in fact they seem to sit at the center of most large galaxies. But the energy required to create a black hole is at least fifteen orders of magnitude—a one followed by fifteen zeroes—bigger than anything a lab will create.

So why did anyone even mention the possibility of black hole creation at the LHC? The reason is that physicists realized that space and gravity could be very different from what we have observed so far. Gravity might spread not just in the three spatial dimensions we know, but also in as-yet-invisible additional dimensions that have so far eluded detection. Those dimensions have had no identifiable effect on any measurement made so far. But it could be that when we reach the energies of the LHC, extra-dimensional gravity—if it exists—could manifest itself in a detectable manner.

As we will explore further in Chapter 17, the extra dimensions that were briefly introduced in Chapter 7 are an exotic idea—but have reasonable theoretical underpinnings and might even explain the extraordinary feebleness of the gravitational force we know. Gravity can be strong in the higher-dimensional world but diluted and extremely weak in the three-dimensional world that we observe, or—according to the idea Raman Sundrum and I worked out—it could vary in an extra dimension so that it is strong elsewhere but weak in our location in higher-dimensional space. We don’t know yet whether such ideas are correct. They are far from certain, but as Chapter 17 will explain, they are among the leading contenders for what experimenters at the LHC might discover.

Such scenarios would imply that when we explore smaller distances at which the effect of the extra dimensions can in principle appear, a very different face of gravity could emerge. Theories involving additional dimensions suggest that the physical properties of the universe should change at the larger energies and smaller distances that we will soon explore. If extra-dimensional reality is indeed responsible for observed phenomena, then gravitational effects could be much stronger at LHC energies than previously thought. In this case, LHC results would not simply depend on gravity as we know it, but also on the stronger gravity of a higher-dimensional universe.

With such strong gravity, protons could conceivably collide in a sufficiently tiny region to trap the amount of energy necessary to create higher-dimensional black holes. These black holes, if they lasted long enough, would suck in mass and energy. If they did this forever, they would indeed be dangerous. This was the catastrophic scenario that the worriers envisioned.

Fortunately, however, classical black hole calculations—those that rely solely on Einstein’s theory of gravity—are not the whole story. Stephen Hawking has many accomplishments to his name, but one of his signature discoveries was that quantum mechanics provides an escape hatch for matter trapped in black holes. Quantum mechanics allows black holes to decay.

The surface of a black hole is “hot,” with a temperature that depends on its mass. Black holes radiate like hot coals, sending off energy in all directions. They still absorb everything that comes too close, but quantum mechanics tells us that particles evaporate from a black hole’s surface through this Hawking radiation, carrying away energy so that it slowly goes back out. The process allows even a large black hole to eventually radiate away all its energy and disappear.

Because the LHC would have at best just barely enough energy to make a black hole, the only black holes it could conceivably form would be small ones. If a black hole started off small and hot, such as one that could potentially be produced at the LHC, it would pretty much disappear immediately. The decay due to Hawking radiation would very efficiently deplete it to nothing. So even if higher-dimensional black holes did form (assuming this whole story is correct in the first place), they wouldn’t stick around long enough to do any damage. Big black holes evaporate slowly, but tiny black holes are very hot and lose their energy almost right away. In this respect, black holes are rather strange. Most objects, coals for instance, cool down as they radiate. Black holes, on the other hand, heat up. The smallest ones are the hottest, and therefore radiate the most efficiently.

Now I’m a scientist—so I have to insist on rigor. Technically, a potential caveat to the above argument based on Hawking radiation and black hole decay does exist. We understand black holes only when they are sufficiently big, in which case we know precisely the equations that describe their gravitational system. The well-tested laws of gravity give a reliable mathematical description for black holes. However, we have no such credible formulation of what extremely small black holes would look like. For these very tiny black holes, quantum mechanics would come into play—not just for their evaporation, but in describing the nature of the objects themselves.

No one really knows how to solve systems in which both quantum mechanics and gravity play an essential role. String theory is physicists’ best attempt, but we don’t yet understand all its implications. This means that in principle there could be a loophole. Extremely tiny black holes, which we will understand only with a theory of quantum gravity, are unlikely to behave the same way as the big black holes we derive using classical gravity. Perhaps such very tiny black holes don’t decay at the rates we expect.

Even this isn’t a serious loophole however. Few people, if any, worried about these objects. Only black holes that can grow to be big can possibly be dangerous. Small black holes can’t accrete enough matter to pose any problem. The only potential risk is that the tiny objects could grow to a dangerous size before evaporating. Yet even without knowing exactly what these objects are, we can estimate how long they should last. These estimates yield lifetimes that are so significantly less than would be required for a black hole to be dangerous that even the very unlikely events on the tails of distributions would still be extremely safe. Small black holes wouldn’t behave very differently from familiar unstable heavy particles. Like these short-lived particles, small black holes would very rapidly decay.

Some did, however, still worry that Hawking’s derivation, although consistent with all known laws of physics, could be wrong and that black holes might be completely stable. After all, Hawking radiation has never been tested by observations since the radiation from known black holes is too weak to see. Physicists are rightfully skeptical of these objections since they would then have to throw away not only Hawking radiation, but also many other independent and well-tested aspects of our physical theories. Furthermore, the logic underlying Hawking radiation directly predicts other phenomena that have been observed, giving us further confidence in its validity.

Nonetheless, Hawking radiation has never been seen. So to be super-safe, physicists asked the question: If Hawking radiation was somehow not correct and the black holes the LHC might create were stable and never decayed, would they be dangerous then?

Fortunately, even stronger proof exists that black holes pose no danger. The argument makes no assumptions about black hole decay and is not theoretical but is based instead solely on observations of the cosmos. In June 2008, two physicists, Steve Giddings and Michelangelo Mangano,39 and soon afterward, the LHC Safety Assessment Group,40 wrote explicit empirically based papers that convincingly ruled out any black hole disaster scenario. Giddings and Mangano calculated the rate at which black holes could form and what their impact would already have been in the universe if they were indeed stable and didn’t decay. They observed that even though we haven’t yet produced the energies required to create black holes—even higher-dimensional black holes—at accelerators here on Earth, the requisite energies are reached quite frequently in the cosmos. Cosmic rays—highly energetic particles—travel through space all the time, and they often collide with other objects. Although we have no way to study their consequences in detail as we can with experiments on Earth, these collisions frequently have energies at least as high as that which the LHC will achieve.

So if extra-dimensional theories are correct, black holes might then form in astrophysical objects—even the Earth or the Sun. Giddings and Mangano calculated that for some models (the rate depends on the number of additional dimensions), black holes simply grow too slowly to be dangerous: even over the course of billions of years, most black holes would remain extremely small. In other cases, black holes could indeed accrete enough matter to grow big—but they often carried charge. If these had indeed been dangerous, they would have been trapped in the Earth and in the Sun, and both of the objects would have disappeared long ago. Since the Earth and Sun seem to have remained intact, the charged black holes—even those that rapidly accrete matter—can’t have dangerous consequences.

So the only possibly dangerous scenario that remains is that black holes don’t carry charge but could grow big sufficiently quickly to be a threat. In that case, the Earth’s gravitational pull—the only force that could slow them down—wouldn’t be sufficiently strong to stop them. Such black holes would pass right through the Earth so we couldn’t use the Earth’s existence to draw any conclusions about their potential danger.

However, Giddings and Mangano ruled out even that case too, since the gravitational attraction of much denser astrophysical objects—namely, neutron stars and white dwarfs—is sufficiently strong to stop black holes before they could escape. Ultra-high-energy cosmic rays hitting dense stars with strong gravitational interactions would have already produced exactly the sorts of black holes that are potentially possible at the LHC. Neutron stars and white dwarfs are much denser than the Earth—so very dense that their gravity alone would suffice to stop black holes in their interior. If the black holes had been produced and had been dangerous, they would have already destroyed these objects that we know have lasted billions of years. The number of them in the sky tells us that even if black holes exist, they certainly are not dangerous. Even if black holes were formed, they must have disappeared almost immediately—or at worst left tiny innocuous stable remnants. They wouldn’t have had sufficient time to do any damage.

On top of that, in the process of accreting matter and destroying such objects, black holes would have released large amounts of visible light, which no one has ever seen. The existence of the universe as we know it and the absence of any signal of white dwarf destruction is very convincing proof that any black holes the LHC could possibly make cannot be dangerous. Given the state of the universe, we can conclude that the Earth is in no danger from LHC black holes.

I’ll now give you a moment to breathe a sigh of relief. But I’ll nonetheless briefly continue with the black hole story—this time from my perspective as someone who works on related topics such as the extra dimensions of space necessary for low-energy black holes to be created.

Before the black hole controversy blew up in the news, I’d already become interested in the topic. I have a colleague and friend in France who used to work at CERN but now works on an experiment called Auger, which studies cosmic rays as they descend through our atmosphere toward Earth. He complained to me that the LHC takes away resources that can be used to study the same energy scales in his cosmic rays. Since his experiment is far less precise, the only type of events it might find would be those with dramatic signatures such as decaying black holes.

So along with a postdoctoral fellow at Harvard at the time, Patrick Meade, I set out to calculate the number of such events they might observe. With a more careful calculation, we found that the number was much less than physicists had originally optimistically predicted. I say “optimistic” since we are always excited about the idea of evidence for new physics. We weren’t concerned about disasters on the Earth—or in the cosmos, which I hope you now agree were not a real threat.

After recognizing that Auger wouldn’t discover tiny black holes, even if higher-dimensional explanations of particle physics phenomena were correct, our calculations made us curious about the claims other physicists had made that black holes could be produced in abundance at the LHC. We found that those rates were overestimates as well. Although the rough ballpark estimates had indicated that in these scenarios, the LHC would copiously produce black holes, our more detailed calculations demonstrated that this was not the case.

Patrick and I had not been concerned about dangerous black holes. We had wanted to know whether small, harmless, rapidly decaying higher-dimensional black holes could be produced and thereby signal the presence of higher-dimensional gravity. We calculated this could rarely happen, if at all. Of course, if possible, the production of small black holes could have been a fantastic verification of the theory Raman and I had proposed. But as a scientist, I’m obliged to pay attention to calculations. Given our results, we couldn’t entertain false expectations. Patrick and I (and most other physicists) don’t expect even small black holes to appear.

That’s how science works. People have ideas, work them out roughly, and then they or others go back and check the details. The fact that the initial idea had to be modified after further scrutiny is not a mark of ineptitude—it’s just a sign that science is difficult and progress is often incremental. Intermediate stages involve forward and backward adjustments until we settle theoretically and experimentally on the best ideas. Sadly, Patrick and I didn’t finish our calculations in time to prevent the black hole controversy from permeating the newspapers and leading to a lawsuit.

We did realize, however, that whether or not black holes could ultimately be produced, other interesting signatures of strongly interacting particles at the LHC might provide important clues about the underlying nature of forces and gravity. And we would see these other signals of higher dimensions at lower energies. Until we see these other exotic signals, we know there is no chance for making black holes. But these other signals themselves might eventually illuminate some aspects of gravity.

This work exemplifies another important aspect of science. Even though paradigms might shift dramatically at different ranges of scales, we rarely suddenly encounter such abrupt shifts in the data itself. Data that was already available sometimes precipitated changes in paradigms, such as when quantum mechanics ultimately explained known spectral lines. But often small deviations from predictions in active experiments are preludes to more dramatic evidence to come. Even dangerous applications of science take time to develop. Scientists might be held accountable in some respects for the nuclear weapons era, but none of them suddenly discovered a bomb by surprise. Understanding the equivalence of mass and energy wasn’t enough. Physicists had to work very hard to configure matter into its dangerous explosive form.

Black holes could even possibly be worthy of worry if they could grow to be large, which calculations and observations demonstrated won’t happen. But even if they could, small ones—or at least the gravitational effects on particle interactions just discussed—would nonetheless signal the presence of a shift in gravity first.

In the end, black holes don’t pose any danger. But just in case, I’ll promise to take full responsibility if the LHC creates a black hole that gobbles up the planet. Meanwhile, you can do what my freshman seminar students suggested and check out http://hasthelargehadroncollider