The Fabric of Reality David Deutch

2 
Shadows
 
There is no better, there is no more open door by which you can enter into
the study of natural philosophy, than by considering the physical phenomena
of a candle.
Michael Faraday 
(A Course of Six Lectures on the Chemical History of a
Candle)
In his popular Royal Institution lectures on science, Michael Faraday used to
urge his audiences to learn about the world by considering what happens
when a candle burns. I am going to consider an electric torch (or flashlight)
instead. This is quite fitting, for much of the technology of an electric torch is
based on Faraday’s discoveries.
I am going to describe some experiments which demonstrate phenomena
that are at the core of quantum physics. Experiments of this sort, with many
variations and refinements, have been the bread and butter of quantum
optics for many years. There is no controversy about the results, yet even
now some of them are hard to believe. The basic experiments are
remarkably austere. They require neither specialized scientific instruments
nor any great knowledge of mathematics or physics — essentially, they
involve nothing but casting shadows. But the patterns of light and shadow
that an ordinary electric torch can cast are very strange. When considered
carefully they have extraordinary ramifications. Explaining them requires not
just new physical laws but a new 
level of description and explanation that
goes beyond what was previously regarded as being the scope of science.
But first, it reveals the existence of parallel universes. How can it? What
conceivable pattern of shadows could have implications like that?
Imagine an electric torch switched on in an otherwise dark room. Light
emanates from the filament of the torch’s bulb and fills out part of a cone. In
order not to complicate the experiment with reflected light, the walls of the
room should be totally absorbent, matt black. Alternatively, since we are only
imagining these experiments, we could imagine a room of astronomical size,
so that there is no time for any light to reach the walls and return before the
experiment is completed. Figure 2.1 illustrates the situation. But it is
somewhat misleading: if we were observing the torch from the side we
should be able to see neither it nor, of course, its light. Invisibility is one of
the more straightforward properties of light. We see light only if it enters our
eyes (though we usually speak of seeing the object in our line of sight that
last affected that light).

---

FIGURE 2.1 
Light from an electric torch (flashlight).
We cannot see light that is just passing by. If there were a reflective object in
the beam, or even some dust or water droplets to scatter the light, we could
see where it was. But there is nothing in the beam, and we are observing
from outside it, so none of its light reaches us. An accurate representation of
what we should see would be a completely black picture. If there were a
second source of light we might be able to see the torch, but still not its light.
Beams of light, even the most intense light that we can generate (from
lasers), pass through each other as if nothing were there at all.
Figure 2.1 does show that the light is brightest near the torch, and gets
dimmer farther away as the beam spreads out to illuminate an ever larger
area. To an observer within the beam, backing steadily away from the torch,
the reflector would appear ever smaller and then, when it could only be seen
as a single point, ever fainter. Or would it? Can light really be spread more
and more thinly without limit? The answer is no. At a distance of
approximately ten thousand kilometres from the torch, its light would be too
faint for the human eye to detect and the observer would see nothing. That
is, a human observer would see nothing; but what about an animal with
more sensitive vision? Frogs’ eyes are several times more sensitive than
human eyes — just enough to make a significant difference in this
experiment. If the observer were a frog, and it kept moving ever farther away
from the torch, the moment at which it entirely lost sight of the torch would
never come. Instead, the frog would see the torch begin to flicker. The
flickers would come at irregular intervals that would become longer as the
frog moved farther away. But the brightness of the individual flickers would
not diminish. At a distance of one hundred million kilometres from the torch,
the frog would see on average only one flicker of light per day, but that
flicker would be as bright as any that it observed at any other distance.
Frogs cannot tell us what they see. So in real experiments we use
photomultipliers (light detectors which are even more sensitive than frogs’

---

eyes), and we thin out the light by passing it through dark filters, rather than
by observing it from a hundred million kilometres away. But the principle is
the same, and so is the result: neither apparent darkness nor uniform
dimness, but flickering, with the individual flickers equally bright no matter
how dark a filter we use. This flickering indicates that there is a limit to how
thinly light can be evenly spread. Borrowing the terminology of goldsmiths,
one might say that light is not infinitely ‘malleable’. Like gold, a small amount
of light can be evenly spread over a very large area, but eventually if one
tries to spread it out further it gets lumpy. Even if gold atoms could somehow
be prevented from clumping together, there is a point beyond which they
cannot be subdivided without ceasing to be gold. So the only way in which
one can make a one-atom-thick gold sheet even thinner is to space the
atoms farther apart, with empty space between them. When they are
sufficiently far apart it becomes misleading to think of them as forming a
continuous sheet. For example, if each gold atom were on average several
centimetres from its nearest neighbour, one might pass one’s hand through
the ‘sheet’ without touching any gold at all. Similarly, there is an ultimate
lump or ‘atom’ of light, a 
photon. Each flicker seen by the frog is caused by a
photon striking the retina of its eye. What happens when a beam of light gets
fainter is not that the photons themselves get fainter, but that they get farther
apart, with empty space between them (Figure 2.2). When the beam is very
faint it can be misleading to call it a ‘beam’, for it is not continuous. During
periods when the frog sees nothing it is not because the light entering its eye
is too weak to affect the retina, but because no light has entered its eye at
all.
This property of appearing only in lumps of discrete sizes is called
quantization. An individual lump, such as a photon, is called a quantum
(plural 
quanta). Quantum theory gets its name from this property, which it
attributes to all measurable physical quantities — not just to things like the
amount of light, or the mass of gold, which are quantized because the
entities concerned, though apparently continuous, are really made of