|Eternal triangles: in a universe containing
only three particles, measure the distance between them and use those three
numbers to plot a point in space. All possible arrangements inhabit an infinite
pyramid in this space, a region called Triangle Land
Surely nothing is possible without time? But according to physicist
Julian Barbour, it doesn't even exist
TIME seems to be the most powerful
force, an irresistible river carrying us from birth to death. To most people
it is an inescapable part of life, a fundamental element of the Universe.
But I think that time is an illusion. Physicists
struggling to unify quantum mechanics and Einstein's general theory of relativity
have found hints that the Universe is timeless. I believe that this idea
should be taken seriously. Paradoxically, we might be able to explain the
mysterious "arrow of time"-the difference between
past and future by abandoning time. But to understand how, we need to change
radically our ideas of how the Universe works.
Let's start with Newton's picture of absolute time. He argued that objects
exist in an immense immobile space, stretching like a block of glass from
infinity to infinity. His time is an invisible river that "flows equably
without relation to anything external". Newton's absolute space and time
form a framework that exists at a deeper level than the objects in it.
To see how it works, imagine a universe containing only three particles.
To describe its history in Newton's terms, you specify a succession of sets
of 10 numbers: one for time and three for the spatial
coordinates of each of the three particles. But this picture is suspect.
As the space-time framework is invisible, how can you determine all the numbers?
As far back as 1872, the Austrian physicist Ernst Mach argued that the Universe
should be described solely in terms of observable things, the separations
between its objects. With that in mind, we can use a very different framework
for the three-particle Universe-a strange, abstract realm called Triangle
Land. Think of the three particles as the corners of a triangle. This
triangle is completely defined by the lengths of its three sides-just three
numbers. You can take these three numbers and use them as coordinates, to
mark a point in an abstract "configuration space" (see Diagram, above).
"It is utterly beyond our power to measure the changes
of things by time.Quite the contrary,time is an abstraction at which we arrive
by means of the changes of things"-Ernst Mach
Each possible arrangement of three particles corresponds to a point in this
space. There are geometrical restrictions-no triangle has one side longer
than the other two put together-so it turns out that all the points lie in
or on a pyramid. At the apex of Triangle Land,
where all three coordinates are zero, is a point that I call Alpha. It represents
the triangle that has sides all of zero length (in other words, all three
particles are in the same place).
In the same way, the configurations of a four-particle universe form
Tetrahedron Land. It has
six dimensions, corresponding to the six
separations between pairs of particles-hard to conceive,
but it exists as a mathematical entity. And even for the stupendous number
of particles that make up our own Universe, we can envisage a vast
multidimensional structure representing its configurations. In collaboration
with Bruno Bertotti of Pavia University in Italy, I have shown that conventional
physics still works in this strange world. As
Plato taught that reality exists as perfect forms,
I think of the patterns of particles as Platonic forms, and call their totality
Platonia is an image of eternity. It is all the arrangements of matter that
can be. Looking at it as a whole, there seems to be no more river of time.
But could time be hiding? Perhaps there is some sort of local time that makes
sense to inhabitants of Platonia.
In classical physics, something like time can indeed creep back in. If you
were to lay out all the instants of an evolving Newtonian universe, it would
look like a path drawn in Platonia. As a godlike being, outside Platonia,
you could run your finger along the path, touching points that correspond
to each different arrangement of matter, and see a universe that continuously
changes from one state to another. Any point on this path still has something
that looks like a definite past and future.
Now's the place
But we know that classical physics is wrong. The world is described by
quantum mechanics-and in the arena of Platonia,
quantum mechanics kills time.
In the quantum wave theory created by Schrödinger, a particle has no
definite position, instead it has a fuzzy
probability of being at each possible position. And for three particles,
say, there is a certain probability of their forming a triangle in a particular
orientation with its centre of mass at some absolute position. The deepest
quantum mysteries arise because of holistic statements of this kind. The
probabilities are for the whole, not the parts.
What probabilities could quantum mechanics specify for the complete Universe
that has Platonia as its arena? There cannot be probabilities at different
times because Platonia itself is timeless. There can only be once-and-for-all
probabilities for each possible configuration.
In this picture, there are no definite paths. We are not beings progressing
from one instant to another. Rather, there are many "Nows" in which
a version of us exists-not in any past or future, but scattered in our region
This may sound like the "many worlds" interpretation
of quantum mechanics, published in 1957 by Hugh Everett of Princeton University.
But in that scheme time still exists: history is a path that branches whenever
some quantum decision has to be made. In my picture there are no paths. Each
point of Platonia has a probability, and that's the end of the story.
A similar position was reached by much more sophisticated arguments more
than 30 years ago. Americans Bryce DeWitt and John Wheeler
combined quantum mechanics and Einstein's theory
of general relativity to produce an equation that describes the whole
Universe. Put into the equation a configuration of the Universe, and out
comes a probability for that configuration. There is no mention of time.
Admittedly, the Wheeler-DeWitt equation is controversial and fraught with
mathematical difficulties, but if quantum cosmology is anything like it-if
it is about probabilities-the timeless picture is plausible.
"All this seems a far cry from the reality of our
lives.Where is the history we read about? Where are our memories? Where is
the bustling,changing world of our experience?"
So let's take seriously the idea of a "probability mist" that covers the
timeless Platonic landscape. The density of the mist is just the relative
probability of the corresponding configuration being realised, or experienced,
as an instantaneous state of the Universe-as a Now. If some Nows in Platonia
have much higher probabilities than others, they are the ones that are actually
experienced. This is like ordinary statistical physics: a glass of water
could boil spontaneously, but the probability is so low that we never see
All this seems a far cry from the reality of our lives. Where is the history
we read about? Where are our memories? Where is the bustling, changing world
of our experience? Those configurations of the Universe for which the probability
mist has a high density, and so are likely to be experienced, must have within
them an appearance of history-a set of mutually consistent records that suggests
we have a past. I call these configurations "time capsules".
An arbitrary matter distribution, like dots distributed
at random, will not have any meaning. It will not tell a story. Almost
all imaginable matter distributions are of this kind; only the tiniest fraction
seem to carry meaningful information.
One of the most remarkable facts about our Universe is that it does have
a meaningful structure. All the matter we can observe in any way is found
to contain records of a past. The first scientists to realise this were
geologists. Examining the structure of rocks and fossils, they constructed
a long history of the Earth. Modern cosmology has extended this to a history
of the Universe right back to the big bang.
What is more, we are somehow directly aware of the passing of time, and we
see motion-a change of position over time. You may feel these are such powerful
sensations that any attempt to deny them is ridiculous. But imagine yourself
frozen in time. You are simply a static arrangement of matter, yet all your
memories and experience are still there, represented by physical patterns
within your brain-probably as the strengths of the
synapse connections between neurons. Just as the
structure of geological strata and fossils seem to be evidence of a past,
our brains contain physical structures consistent with the appearance of
recent and distant events. These structures could surely lead to the impression
of time passing. Even the direct perception of motion could arise through
the presence in the brain of information about several different positions
of the objects we see in motion.
And that is the essence of my proposal. There is no history laid out along
a path, there are only records contained within Nows. This timeless vision
may seem perverse. But it turns out to have one great potential strength:
it could explain the arrow of time.
We are so accustomed to history that we forget how peculiar it is. According
to conventional cosmology, our Universe must have started out in an
extraordinarily special state to give rise to the highly ordered Universe
we find around us, with its arrow of time and records of a past. All matter
and energy must have originated at a single point, and had an almost perfectly
uniform distribution immediately after the big bang.
Hitherto, the only explanation that science has provided is the anthropic
argument: we experience configurations of the Universe that seem to have
a history because only these configurations have the characteristics to produce
beings who can experience anything. I believe that timeless quantum cosmology
provides a far more satisfying explanation.
In Platonia, there are no initial conditions. Only two factors determine
where the probability mist is dense: the form of some equation (like the
Wheeler-DeWitt equation) and the shape of Platonia. And by sheer logical
necessity, Platonia is profoundly asymmetric. Like Triangle Land, it is a
lopsided continent with a special point Alpha corresponding to the configuration
in which every particle is at the same place.
From this singular point, the timeless landscape opens out, flower-like,
to points that represent configurations of the Universe of arbitrary size
and complexity. My conjecture is that the shape of Platonia cannot fail to
influence the distribution of the quantum probability mist. It could funnel
the mist onto time capsules, those meaningful arrangements that seem to contain
records of a past that began at Alpha.
"The notion of time as an invisible framework that contains
and constrains the Universe is not unlike the crystal spheres invented centuries
ago to carry planets"
This is of course, only speculation, but quantum mechanics
supports it. In 1929, the British physicist Nevill Mott and Werner Heisenberg
from Germany explained how alpha particles, emitted by radioactive nuclei,
form straight tracks in cloud chambers. Mott pointed out that, quantum
mechanically, the emitted alpha particle is a spherical wave which slowly
leaks out of the nucleus. It is difficult to picture how it is that an outgoing
spherical wave can produce a straight line," he argued. We think intuitively
that it should ionise atoms at random throughout space.
Mott noted that we think this way because we imagine that quantum processes
take place in ordinary three-dimensional space. In fact, the possible
configurations of the alpha particle and the particles in the detecting chamber
must be regarded as the points of a hugely
multidimensional configuration space, a miniature
Platonia, with the position of the radioactive nucleus playing the role of
When Mott viewed the chamber from this perspective, his equations predicted
the existence of the tracks. The basic fact that
quantum mechanics treats configurations as whole entities leads to track
formation. And a track is just a point in configuration space but one that
creates the appearance of a past, just like our own memories.
There is one more reason to embrace the timeless view. Many theoretical
physicists now recognise that the usual notions of time and space must break
down near the big bang. They find themselves
forced to seek a timeless description of the
"beginning" of the Universe, even though they use
time elsewhere. It seems more consistent and economical to use an entirely
But for these ideas to be more than speculation, they should have concrete,
measurable results. Fortunately, Stephen Hawking and
other theorists have shown that the Wheeler-DeWitt equation can lead to
verifiable predictions. For example, established physical theories cannot
predict a value for the cosmological
constant, which measures the gravitational repulsion of empty space.
But calculations based on the WheelerDeWitt equation suggest that it should
have a very small value. It should soon be possible to measure the cosmological
constant, either by taking the brightness of far-off supernovae and using
that to track the expansion of the Universe, or by analysing the shape of
humps and bumps in the cosmic microwave background. And a definitive equation
of quantum cosmology should give us a precise prediction for the value of
the constant. It is a distant prospect, but the nonexistence of time could
be confirmed by experiment.
The notion of time as an invisible framework that contains and constrains
the Universe is not unlike the crystal spheres invented centuries ago to
carry the planets. After the spheres had been shattered by Tycho Brahe's
observations, Kepler said: "We must philosophise
about these things differently." Much of modern physics stems from this
insight. We need a new notion of time.
Julian Barbour claims time should be banished from the catalogue
of reality (16 October, p 28). But he may be making an unwarranted
assumption-that to understand reality is the same as to experience it. Physics
is concerned exclusively with the first. But, as Bertrand Russell said some
70 years ago, all physics can ever do is to reveal the
causal/structural/relational features of reality. As the philosopher David
Chalmers puts it: "The picture of the world that (physics) reveals is that
of a giant causal flux, but the picture tells us nothing about what all this
A famous line from Stephen Hawking sums this up well: "What is it that breathes
fire into the equations?" There may be some sort of an answer here as to
why time is so fundamental a part of reality in our "normal" lives, yet
apparently redundant when we try to understand the nature of that reality.
It is arguable that what we know more surely and immediately than anything
else is our conscious experience, and a fundamental element in the nature
of that experience is a sense of before and after. All other knowledge including
all of physics is a kind of story that we tell about reality, more or less
accurate (science being far more accurate than any other tale), but always
at one remove from it. This is a view that is as yet far from widely accepted,
especially in conventional scientific circles. Perhaps only if and when it
is will we be in a position to start thinking more deeply about the paradox
that time seems to represent.
David Boothroyd Ely, Cambridgeshire
It is good to know that science and religion are
converging. In his article Barbour suggests that time is simply a human
interpretative device and writes: "There are many 'Nows' in which a version
of us exists." St Augustine of Hippo anticipated
this by some sixteen hundred years, when he said that only "Now" exists.
Past and future exist in our present; time is subjective. A rather perceptive
Susan Lampitt Charlecote, Warwickshire
I found Barbour's article "Timeless" difficult to follow. Was I reading it
in the wrong direction?
John Martin King's College, London
New Scientist 6/11/1999