Skeptics will point out that everything I have told you about
the quantum properties of black holes - from entropy, temperature,
and Hawking radiation to Black Hole Complementarity
and the Holographic Principle - is pure theory, with not an
ounce of experimental data to confirm it. Unfortunately, they may
be right for a very long time.
That said, a totally unexpected connection has recently turned
up - a connection between black holes, quantum gravity, the Holographic
Principle, and experimental nuclear physics that may once
and for all belie the claim that these theories are beyond scientific
confirmation. On the face of it, nuclear physics seems a most unpromising
place to test ideas such as the Holographic Principle and
Black Hole Complementarity. Nuclear physics is not usually deemed
to be part of the cutting edge. It's an old subject, and most physicists,
including me, thought it had exhausted its capacity to teach us anything
new about fundamental principles. From the viewpoint of
modern physics, nuclei are like soft marshmallows - giant squish
balls that are mostly full of empty space. 1 What could they possibly
teach us about physics at the Planck scale? Surprisingly, it seems,
quite a lot.
String theorists have always had an interest in nuclei. The prehistory
of String Theory was all about hadrons: protons, neutrons,
mesons, and glueballs. Like nuclei, these particles are big, soft
composites made of quarks and gluons. Yet it seems that on a scale
a hundred billion billion times larger than the Planck scale, nature
repeats itself. The mathematics of hadron physics turns out to be almost
the same as the mathematics of String Theory. That seems extremely
surprising in view of the fact that the scales are so different:
nucleons may be 10^20 times larger in size than fundamental strings,
and they oscillate 10^20 times more slowly. How can these theories be
the same, or even remotely similar? Nevertheless, in a way that will
become clear, they are. And if the ordinary subnuclear particles are
really similar to fundamental strings, why not test the ideas of String
Theory in nuclear physics laboratories? In fact, it's been going on
for almost forty years.
The connection between hadrons and strings is one of the pillars
of modern particle physics, but up until very recently, it was not possible
to test the nuclear analog of black hole physics. That situation
is now changing.
Out on Long Island, about seventy miles from Manhattan, nuclear
physicists at the Brookhaven National Laboratory are slamming
heavy atomic nuclei together just to see what happens. The
Relativistic Heavy Ion Collider (RHIC) accelerates gold nuclei to
almost the speed of light - fast enough that when they collide, they
create a huge splash of energy a hundred million times hotter than
the surface of the Sun. The physicists at Brookhaven are not interested
in nuclear weapons or any other nuclear technology. Their
motivation is pure curiosity - curiosity about the properties of a
new form of matter. How does this hot nuclear material behave? Is
it a gas? A liquid? Does it hold together, or does it instantly evaporate
into separate particles? Do jets of extremely high-energy particles
zip out of it?
As I said, nuclear physics and quantum gravity take place on
vastly different scales, so how can they have anything to do with
each other? The best analogy that I know involves one of the worst
movies ever, an old horror flick from the era of the drive-in movie.
The movie features a monstrous fly. I don't know how the film was
made, but I imagine that an ordinary housefly was filmed and then
magnified to fill the entire screen. The image is projected in very
slow motion, which gives the fly the ominous feel of a huge, hideous
bird. The result is horrifying, but more to the point, it almost perfectly
illustrates the connection between gravitons and glueballs.
Both are closed strings, but the graviton is much smaller and faster
than the glueball - about 10^20 times smaller and faster. It seems
that hadrons are a lot like images of fundamental strings blown up
and slowed down not a few hundred times like the fly, but a fantastic
So if we can't collide Planck-sized particles at stupendous energies
to make black holes, perhaps we can collide their blown-up
versions - glueballs, mesons, or nucleons - and create a magnified
version of a black hole. But wait - doesn't that also require
prodigious amounts of energy? No, it does not, and to understand
why, we need to recall from chapter 16 the counterintuitive twentieth-
century connection between size and mass: small is heavy, big is
light. The fact that nuclear physics takes place on an immensely
larger scale than fundamental String Theory implies that the corresponding
phenomena require far less energy concentrated in a
hugely larger volume. When the numbers are plugged in and the
calculations are carried out, something very similar to a slowmotion,
magnified black hole should form when ordinary nuclei collide
To understand in what sense black holes are created by RHIC,
we must return to the Holographic Principle and to Juan Maldacena's
discovery. In a way that no one had foreseen, Maldacena
found that two different mathematical theories were really the
same - "dual to each other," in String Theory jargon. One theory
was String Theory, with its gravitons and black holes, albeit in
(4 + 1)-dimensional anti de Sitter Space (ADS). (In chapter 22, for
purposes of visualization, I took the liberty of decreasing the dimension
of space. In this chapter. I restore the missing dimension.)
Four dimensions of space is one too many for nuclear physics,
but remember the Holographic Principle: everything that takes
place in ADS must be completely describable by a mathematical
theory with one less dimension of space. Because Maldacena started
with four space dimensions, the holographic dual theory has only
three dimensions, the same number as everyday space. Could this
holographic description be similar to any of the theories that we use
to describe conventional physics?
It turns out that the answer is yes: the holographic dual is mathematically
quite similar to Quantum Chromodynamics (QCD), the
theory of quarks, gluons, hadrons, and nuclei.
For me, the main interest in Maldacena's work was the way it
confirmed the Holographic Principle and shed light on the workings
of quantum gravity. But Maldacena and Witten saw another
opportunity. They realized - brilliantly, I must say - that the Holographic
Principle is a two-way street. Why not read it backward?
That is, use what we know about gravity - in this case, gravity in
(4 + 1)-dimensional ADS - to teach us things about ordinary
Quantum Field Theory. For me this was a totally unexpected twist,
a bonus of the Holograpbic Principle that I had never thought of.
A little work was required to accomplish this. QCD is not quite
the same as Maldacena's theory, but the main difference can easily
be taken into account by modifying ADS in a simple way. Let's review
ADS, as seen from a point very near the boundary (where the
last visible devil shrinks to zero). I'm going to call that boundary the
UV-brane. UV stands for ultraviolet - the same term we use for
very short-wavelength light. (Over the years, the term ultraviolet
has come to stand for any phenomenon that takes place on small
scales. In the present context, the word refers to the fact that the angels
and devils near the boundary of Escher's drawing shrink to infinitesimal
size.) The word brane is really a misnomer in UV-brane,
but since it has stuck, I will use it. The UV-brane is a surface close to
Imagine moving away from the UV-brane into the interior,
where the square devils expand and clocks slow down without limit.
Objects that are small and fast near the UV-brane become big and
slow as we move deeper into ADS. But ADS is not quite the right
thing for describing QCD. Although the difference is not great, the
modified space deserves its own name; let's call it Q-space. Like
ADS, Q-space has a UV-brane where things shrink and speed up,
but unlike ADS, there is a second boundary called the IR-brane. (IR
stands for infrared, a term used for very long-wavelength light.) The
IR-brane is a second boundary - a kind of impenetrable barrier
where the angels and devils reach a maximum size. If the UV-brane
is the ceiling of a bottomless chasm, Q-space is an ordinary room
with a ceiling and floor. Ignoring the time direction and drawing
only two space directions, ADS and Q-space look like this:
Imagine putting a stringlike particle into Q-space by first placing
it near the UV-brane. Like the angels and devils surrounding it, it
will appear to be very small - possibly Planck sized - and very
rapidly vibrating. But if the same particle is moved toward the IRbrane,
it will appear to grow, almost as if it were being projected
onto a receding screen. Now watch the string as it vibrates. The vibrations
define a kind of clock, and like all clocks, it runs fast when
it is near the UV-brane and slow as it moves toward the IR-brane.
A string near the IR end will not only look like a gigantic blown-up
version of its shrunken UV self, but it will also oscillate far more
slowly. This difference sounds a lot like the difference between real
flies and their cinematic images - or the difference between fundamental
strings and their nuclear counterparts.
If the super-small Planck-sized particles of String Theory "live"
near the UV-brane and their blown-up versions - the hadrons -
live near the IR-brane,just how far apart are they from each other?
In a certain sense, not so far; you would have to descend through
about 66 square devils to get from Planck-sized objects to hadrons.
But remember that each step is twice as high as the previous
one. Doubling in size 66 times is the same as expanding by a factor
There are two views of the similarity between fundamental
String Theory and nuclear physics. The more conservative view is
that it is accidental, more or less like the similarity between atoms
and the solar system. That similarity was useful in the early days of
atomic physics. Niels Bohr, in his theory of the atom, used the same
mathematics for atoms that Newton had used for the solar system.
But neither Bohr nor anyone else really thought that the solar system
was a blown-up version of an atom. According to this more
conservative view, the connection between quantum gravity and
nuclear physics is also just a mathematical analogy, but a useful
analogy that allows us to use the mathematics of gravity to explain
certain features of nuclear physics.
The more exciting view is that nuclear strings really are the same
objects as fundamental strings, except seen through a distorting lens
that stretches their image and slows them down. According to this
view, when a particle (or string) is located near the UV-brane, it appears
small, energetic, and nlpidly oscillating. It looks like a fundamental
string; it behaves like a fundamental string; so it must be a
fundamental string. For example, a closed string located at the UVbrane
would be a graviton. But the same string, if it moves to the
IR-brane, slows down and grows in size. In every way, it looks and
behaves like a glueball. In this view of things, gravitons and glueballs
are exactly the same objects, except for their location in the
Imagine a pair of gravitons (strings near the UV-brane) about to
collide with each other.
If they have enough energy, when they meet near the UV-brane, an
ordinary small black hole will form: a blob of energy stuck to the
UV-brane. Think of it as a drop of fluid hanging from the ceiling.
The bits of information that make up its horizon are Planck sized.
This is, of course, exactly the experiment that we will probably never
be able to do.
But now replace the gravitons with two nuclei (near the IRbrane)
and smash them together.
Here is where the power of duality makes itself felt. On the one hand,
we can think of it in the four-dimensional version, in which two objects
collide and form a black hole. This time the black hole will be
near the IR-brane - a big puddle on the floor. How much energy is
required? Far less than when the black hole forms near the UV-brane.
In fact, the energy is easily within the range of RHIC.
We can also view it from the three-dimensional viewpoint. In that case,
hadrons or nuclei collide and make a splash of quarks and gluons.
Originally, before anyone realized QCD's potential connection
with black hole physics, QCD experts had expected the energy of
the collision to reappear as a gas of particles that would quickly fly
apart with very little resistance. But what they saw was different:
the energy holds together in what looks much more like a blob of
fluid - call it hot quark soup. Hot quark soup is not just any fluid; it
has some very surprising flow properties that resemble nothing so
much as the horizon of a black hole.
All fluids are viscous. Viscosity is a type of friction that acts between
the layers of a fluid when they slide over each other. Viscosity
is what distinguishes a very viscous fluid such as honey from a much
less viscous fluid such as water. Viscosity is not just a qualitative
concept. Instead, for everyf1uid, there is a precise numerical measure
called shear viscosity.
Theorists had initially applied standard approximation methods
and concluded that hot quark soup would have a very high viscosity.
Everyone was quite surprised when it turned out to have an astonishingly
small viscosity - everyone, that is, except for a few nuclear
physicists who happened to know a bit about string theory.
According to a certain quantitative measure of viscosity, hot
quark soup is the least viscous fluid known to science - much less
viscous than water. Even superfluid liquid helium (the previous
champion of low viscosity) is a good deal more viscous.
Is there anything in nature that might rival the low viscosity of
hot quark soup? There is, but it's not an ordinary fluid. A black hole
horizon behaves like a fluid when it is disturbed. For example, if a
small black hole falls into a bigger black hole, it temporarily creates
a bulge on the horizon, similar to the bulge that a blob of honey
leaves if dropped onto the surface of a pool of honey. The blob on
the horizon spreads out just as a viscous fluid does. Long ago, black
hole physicists calculated the viscosity of a horizon, and when it was
translated to fluid terms, it easily beat out superfluid helium. When
string theorists began to suspect a connection between black holes
and nuclear collisions, they realized that of all things, hot quark
soup is most like the horizon of a black hole.
What eventually becomes of the blob of fluid? Like a black hole,
it evaporates - into a variety of particles, including nucleons, mesons,
photons, electrons, and neutrinos. Viscosity and evaporation
are just two of several properties that horizons and hot quark soup
Nuclear fluid is now under intense study to find out whether
other properties show similar connections to black hole physics. If
that trend continues, it will mean that we have been granted an extraordinary
opportunity - a remarkable window into the world of
quantum gravity, blown up in size and slowed down in frequency, so
that the Planck distance becomes not much smaller than a proton
- to confirm the theories of Hawking and Bekenstein, as well
as Black Hole Complementarity and the Holographic Principle.
It has been said that peace is nothing but the brief interlude between
wars. But in science, Thomas Kuhn has rightly said, the opposite
is true: most "ordinary science" takes place during the long,
peaceful, humdrum periods between upheavals. The Black Hole
War led to a violent restructuring of the laws of physics, but now we
are seeing it work its way into the day-to-day activities of the more
mundane side of physics. Like so many earlier revolutionary ideas,
the Holographic Principle is evolving from radical paradigm shift to
everyday working tool of - surprisingly - nuclear physics.