91av

What do we really know about black holes? That may sound like an odd question. Aren’t black holes and all their well-known attributes – the singularity, the event horizon, the ability to swallow light and matter – just part of the furniture of astrophysics? Strangely, no. Astronomers know of massive bodies that fit the bill, but for now black holes remain largely theoretical. So much so that some researchers even claim that they don’t exist.

The debate over the existence of black holes has been rumbling on since about 1939, when Albert Einstein published a arguing that for a black hole to form, a collection of stars would have to orbit each other faster than the speed of light, which special relativity prohibited. That turned out to be wrong, but the point is that black holes are so weird, Einstein didn’t believe in them – even though his own 1915 theory of general relativity predicted their existence.

Since then the field has matured. Astrophysicists have solidified their models and identified some promising black hole candidates in the Milky Way and nearby galaxies. So far, however, no one has produced any incontrovertible evidence that these objects fit the classical notion of a black hole: a spinning point of infinite density surrounded by an event horizon from which nothing can escape, not even light.

Such evidence could be coming soon. The technological capability to observe black hole candidates directly and model in detail the complex signals they give off is fast approaching. We will be able to test, for the first time, whether black holes are what we think they are – whether event horizons actually form, what happens to space-time around a black hole and so forth. The stakes are high: far from being merely cosmological curiosities, black holes may hold the key to understanding how galaxies form and how we might merge relativity and quantum mechanics into a “theory of everything”.

Indeed, researchers are confident that we will soon know the true nature of black holes. “Before 2005, you’d go to a conference and it would be really depressing. You’d hear one talk after another that went into excruciating detail about why some technique failed,” says Frans Pretorius, a physicist at Princeton University who specialises in general relativity simulations. “Now everyone’s excited about black holes again.”

In the dark

If you count a paper by the Reverend John Michell in 1769, people have known about the possibility of black holes for nearly two-and-a-half centuries. The modern debate began in the 1930s, however, when Subrahmanyan Chandrasekhar and J. Robert Oppenheimer independently showed that under the laws of general relativity these weird objects were necessary to explain what happens when a large star collapses at the end of its life.

So why is it that, seven decades later, we are still in the dark about what black holes actually are? After all, we have ground telescopes that dwarf what was available in the 1930s and space telescopes that see with crystal clarity. We also can see in parts of the spectrum that our grandparents couldn’t dream of – X-rays, gamma rays, infrared, radio – where black holes show some of their most impressive features.

The problem is that black holes are so uncooperatively, well, black. That’s because a black hole is a concentration of matter so great that its central point, or singularity, has infinite density. The curvature of space around the singularity prevents anything, even a beam of light, from escaping. To make matters worse, researchers showed in the 1960s that in theory all you can ever measure about a black hole are its mass, electric charge and angular momentum, or spin; there is no way to know what originally collapsed to form the hole, and no other way to tell one from another. This led to physicist John Wheeler’s quip that “black holes have no hair” – they have no distinguishing features or structure.

So, while most astrophysicists are pretty sure that black holes exist, their blackness and lack of structure means that the observational evidence adds up to only a very strong circumstantial case. “What we do know,” says Avery Broderick of the Harvard-Smithsonian Center for Astrophysics in Cambridge, Massachusetts, “is that there are objects within our galaxy and inside nearby galaxies that are too compact and dense to be anything else, barring some unforeseen exotic physics.”

Why can’t they be anything else? Put simply, they are too dense. Normally, massive objects such as planets and stars don’t collapse under their own gravity because there are other forces pushing in the opposite direction – chemical bonds between atoms in the Earth, for example, or the outward pressure from electromagnetic radiation in the sun. Classical black holes are so dense that gravity overwhelms all other forces, with dramatic consequences.

Since the 1970s, at least 10 stars in our galaxy have been found orbiting such dense, unseen objects. In addition, many stars and gas clouds are whipping around something at the core of the Milky Way that is generally considered to be a supermassive black hole, and half a dozen nearby galaxies have something similar (see “Black hole contenders”). Astrophysicists have measured the velocities of the stars and gas clouds at the centre of the Milky Way, and using the laws of orbital motion deduced by Johannes Kepler half a millennium ago, they have calculated the mass of the central object to be some 3.7 million times the mass of the sun, all concentrated into a volume much smaller than our solar system. No known material could remain intact at this density without collapsing to a singularity.

In theory, anyway. Now researchers are trying to pin down with observations whether such objects really are black holes in the classical way we imagine them. For the past few years Broderick, who works with Ramesh Narayan’s group at Harvard University, has been trying to measure the spin of space-time surrounding the black hole at the centre of our galaxy. Since black holes presumably form from stars or other objects that were spinning in the first place, they should still be spinning – and even faster than the original object, to conserve angular momentum. According to general relativity, spin like this drags the local space-time around with it, and that can be measured, says Broderick, even from far away. If the researchers find the spin effect they are looking for, it would all but confirm the hole’s classical nature.

“Measuring the spin of space-time could confirm our classical notions of black holes”

The way they measure the spin effect is by examining the “accretion disc” of gas that builds up around a black hole (see “Are black holes what we think they are?”). This material, superheated by friction as it orbits the hole and tries to fall in, can throw out huge jets of debris so bright they can be seen across the universe. These jets are the hallmarks of massive black hole candidates, with quasars the prime example: they are thought to be jets that happen to point in our direction.

Once jets draw attention to a particular candidate, researchers need to look more closely to pin down its spin. Using high-resolution telescopes, astronomers have previously detected and modelled “hotspots” – bright patches of ultra-hot material – within accretion discs (). Now Narayan’s group is planning to look at them in a way that takes advantage of the weird warping of space around a black hole. They know that a light beam aimed horizontally just above the event horizon should whip around the black hole, like a satellite getting a gravity assist from a planet, and fly off in another direction.

But hotspots shine in all directions, not just one. “You see some light that comes directly toward you,” says Broderick, “but you also see light that goes around the back of the black hole and comes at you that way.” In short, you see a double image of the hotspot, and depending on where in the accretion disc the spot is, you’ll see different relative brightnesses between them, which allows you to figure out the local space-time curvature, and from that the spin of the black hole.

The best parts of the electromagnetic spectrum for such observations are the near-infrared and sub-millimetre (between infrared and microwave) bands: plans are afoot to use the Very Large Telescope array in Chile to detect the former, and several widely spaced radio telescopes for the latter. Both sets of observations could be completed within two or three years. If they reveal that space-time is distorted as relativity predicts, it would show definitively that the object in question behaves as a classical black hole should.

Next on the checklist is the event horizon – the point of no return beyond which neither matter nor light can escape. Since a black hole sits in the middle of a glowing accretion disc, part of that disc will lie behind the hole, from Earth’s point of view. The light from that part should be blocked, creating a dim patch in the overall glow; the event horizon should be silhouetted against the brightness of the disc. This silhouette would ordinarily be too small to see, like trying to resolve a human hair 200 kilometres away, but a black hole’s strong gravity should act as a lens to magnify its apparent size. A non-rotating black hole would magnify the silhouette about seven times, says Broderick, and a rotating one up to 27 times. At that size, he says, we could spot the silhouette with an array of radio telescopes – an experiment likely to be performed within a year or two.

Other researchers argue, however, that we should not expect to see event horizons – because they don’t actually exist. Earlier this year, physicists Lawrence Krauss, Tanmay Vachaspati and Dejan Stojkovic from Case Western Reserve University in Cleveland, Ohio, made the controversial claim that, in theory, event horizons need infinite time to form. That means black holes as we understand them could never be observed – an assertion that has garnered a fair amount of attention in the field.

The idea is the latest respectable alternative to classical black holes (see “Black hole pretenders”). According to relativity, time slows down for an object, from the point of view of an outside observer, as it accelerates close to the speed of light. Anything falling into a black hole approaches that velocity as it crosses the event horizon. So while someone riding a spaceship across the horizon would feel that he or she was moving at a terrific speed, someone watching from outside would see the ship slow and eventually stop at the horizon, never quite falling in.

If that’s true, the same argument should apply to gas, stars or whatever was collapsing to form the black hole in the first place. To an external observer, it would take infinitely long for the black hole to come into being. This is actually a long-standing problem that has never been fully addressed. “People have just assumed it’s one of those weird general relativity things and don’t discuss it very much,” says Krauss. When you add in quantum mechanics, which says that black holes actually radiate particles, the problem becomes even more acute. “If quantum theory says black holes must evaporate in finite time,” says Krauss, “and general relativity says they take an infinite time to form, you’ve got something disappearing before it exists.”

Quantum hole

Krauss’s work, which will appear in the journal Physical Review D, began as an exercise in particle physics (). Researchers had suggested that experiments at the Large Hadron Collider at CERN, the European particle physics lab near Geneva, Switzerland, scheduled to go online next year, could produce energy densities great enough to create a quantum black hole, much tinier than the tiniest subatomic particle. The smaller a black hole, the more quickly it should radiate away its mass, and Krauss and his colleagues were trying to figure out what that radiation might look like.

According to their model, quantum black holes should emit light, X-rays and other electromagnetic radiation at a rate so high that they never fully form in the first place. This piqued the researchers’ interest, so they tried the same calculations for cosmic black holes. They found that as a spherical shell of mass collapses inward, its gravity disrupts the quantum vacuum, giving rise to radiation similar to the quantum black hole’s. It, too, leaks so much energy that the mass never gets dense enough to form a black hole with an event horizon. Instead it forms what the researchers call a “black star”, which never completely swallows any surrounding matter from an external observer’s point of view.

That doesn’t mean the model is right, of course. “We’ve discussed it at lots of colloquia and seminars, and there has been lots of interest. The first reaction is incredulity; people are sceptical, but nobody’s poked a hole in it yet,” says Krauss. “For now I’m happy just that it’s spurring debate.”

The astrophysical community has been receptive but lukewarm. “When people ask me if I think black holes exist,” says Broderick, “it really depends on what you mean by the term. A black hole is not just this thing inside an event horizon, it’s an entire region of curved space-time. So I prefer to talk about ‘black hole space-time’ rather than black holes.” Still, Broderick thinks it’s going to be very difficult for one of these objects not to have an event horizon. “The only way to know for sure is to drop an undergraduate in and see whether his cellphone signal cuts off,” he says.

In fact, there is one other serious way to probe what black holes really are. “Since they’re made from warped space and time rather than of matter, the only truly convincing proof will come from radiation made from the same stuff,” says Kip Thorne of the California Institute of Technology in Pasadena. He is talking about gravitational waves, also predicted by Einstein. A massive body, such as a star, warps the space-time around it. If that body is accelerating, the warping will send ripples out into space-time like a moving boat sends a wake across a pond.

The warping is so subtle that only the most violent events will create a detectable ripple – and only the most sensitive instruments will be able to detect it. That’s the idea behind LIGO, the Laser Interferometric Gravity-Wave Observatory, which has stations in Louisiana and Washington state. LIGO isn’t yet sensitive enough to detect even the most powerful of gravitational waves, but by 2013 its sensitivity should be 10 times finer. A space-based version of the experiment, known as LISA (Laser Interferometric Space Array), is planned for after 2015, and will be still more sensitive. The most powerful gravitational waves would come from a merger of two black holes, and this is what researchers plan to look for.

First, though, the researchers need to know what specific patterns of gravitational waves say about the collision that created them. Until recently, the calculations were so complicated that nobody knew how to do them. It was Pretorius who made the breakthrough two years ago by reworking Einstein’s relativity equations (91av, 2 June, p 34). Armed with this insight, Pretorius and others have been churning out simulations of the gravitational-wave signal from merging black holes.

Thorne and his Caltech colleagues plan to compare the frequencies of the gravitational waves with what general relativity says you should get from a classical black hole merging with another object (). The result should say whether the black hole in question is what we think it is, or whether there is something surprising.

Of course, no gravitational waves have been detected anywhere yet, and it’s not certain that they ever will be. Given the difficulty of detecting a colossal collision between two black holes, even LISA probably won’t be able to detect the much fainter signal from a black hole spinning quietly on its own.

So the final answer to whether our black-hole ideas are correct may have to wait until we can send a probe to a nearby candidate and have it transmit data as it makes its final plunge. Meantime, will researchers find what they’re looking for? Maybe, but it will be even more exciting if they find something else.