Emergent Complexity, Teleology, and the Arrow of Time

For example, during the 1960s it was suggested by the cosmologist Thomas Gold8that one day the expanding universe may start to recontract, and that during the contraction phase, the Seco

10 Emergent Complexity, Teleology, and the Arrow of Time

Paul Davies

1. the dying universe

In 1854, in one of the bleakest pronouncements in the history of science, the German physicist Hermann von Helmholtz claimed that the universe must be dying He based his prediction on the Second Law of Thermody-namics, according to which there is a natural tendency for order to give way to chaos It is not hard to find examples in the world about us: peo-ple grow old, snowmen melt, houses fall down, cars rust, and stars burn out Although islands of order may appear in restricted regions (e.g., the birth of a baby, crystals emerging from a solute), the disorder of the envi-ronment will always increase by an amount sufficient to compensate This one-way slide into disorder is measured by a quantity called entropy A state

of maximum disorder corresponds to thermodynamic equilibrium, from which no change or escape is possible (except in the sense of rare statisti-cal fluctuations) Helmholtz reasoned that the quantity of entropy in the universe as a whole remorselessly rises, presaging an end state in the far future characterized by universal equilibrium, following which nothing of interest will happen This state was soon dubbed the “heat death of the universe.”

Almost from the outset, the prediction of cosmic heat death after an ex-tended period of slow decay and degeneration was subjected to theological interpretation The most famous commentary was given by the philosopher

Bertrand Russell in his book Why I Am Not a Christian, in the following terms:1

All the labors of the ages, all the devotion, all the inspiration, all the noonday bright-ness of human genius are destined to extinction in the vast death of the solar system, and the whole temple of man’s achievement must inevitably be buried beneath the debris of a universe in ruins All these things, if not quite beyond dispute, are yet so nearly certain that no philosophy which rejects them can hope to stand Only within the scaffolding of these truths, only on the firm foundation of unyielding despair, can the soul’s habitation henceforth be safely built

191

The association of the Second Law of Thermodynamics with atheism and cosmic pointlessness has been an enduring theme Consider, for example, this assessment by the British chemist Peter Atkins:2

We have looked through the window on to the world provided by the Second Law, and have seen the naked purposelessness of nature The deep structure of change is decay; the spring of change in all its forms is the corruption of the quality of energy

as it spreads chaotically, irreversibly and purposelessly in time All change, and time’s arrow, point in the direction of corruption The experience of time is the gearing

of the electrochemical processes in our brains to this purposeless drift into chaos as

we sink into equilibrium and the grave

As Atkins points out, the increase in entropy imprints upon the universe

an arrow of time, which manifests itself in many physical processes, the most conspicuous of which is the flow of heat from hot to cold; we do not encounter cold bodies getting colder and spontaneously giving up their heat

to warm environments The irreversible flow of heat and light from stars into the cold depths of space provides a cosmic manifestation of this simple “hot

to cold” principle On the face of it, it appears that this process will continue until the stars burn out and the universe reaches a uniform temperature Our own existence depends crucially on a state of thermodynamic disequilibrium occasioned by this irreversible heat flow, since much life on Earth is sustained

by the temperature gradient produced by sunshine Microbes that live under the ground or on the sea bed utilize thermal and chemical gradients from the Earth’s crust These too are destined to diminish over time, as thermal and chemical gradients equilibrate Other sources of energy might provide

a basis for life, but according to the Second Law, the supply of free energy continually diminishes until, eventually, it is all exhausted Thus the death

of the universe implies the death of all life, sentient and otherwise It is probably this gloomy prognosis that led Steven Weinberg to pen the famous phrase, “The more the universe seems comprehensible, the more it also seems pointless.”3

The fundamental basis for the Second Law is the inexorable logic of chance To illustrate the principle involved, consider the simple example

of a hot body in contact with a cold body The heat energy of a material substance is due to the random agitation of its molecules The molecules

of the hot body move on average faster than those of the cold body When the two bodies are in contact, the fast-moving molecules communicate some

of their energy to the adjacent slow-moving molecules, speeding them up After a while, the higher energy of agitation of the hot body spreads across into the cold body, heating it up In the end, this flow of heat brings the two bodies to a uniform temperature, and the average energy of agitation is the same throughout The flow of heat from hot to cold arises entirely because chaotic molecular motions cause the energy to diffuse democratically among all the participating particles The initial state, with the energy distributed

in a lopsided way between the two bodies, is relatively more ordered than the final state, in which the energy is spread uniformly throughout the system One way to see this is to say that more information is needed to describe the initial state – namely, two numbers, the temperatures of the two bodies – whereas the final state can be described with only one number – the common final temperature The loss of information occasioned by this transition may be quantified by the entropy of the system, which is roughly equal to the negative of the information content Thus as information goes down, entropy, or disorder, goes up

The transition of a collection of molecules from a low to a high entropy state is analogous to the shuffling of a deck of cards Imagine that the cards are extracted from the package in suit and numerical order After a period of random shuffling, the cards will very probably be jumbled up The transition from the initial ordered state to the final disordered one is due to the chaotic nature of the shuffling process So the Second Law is really just a statistical effect of a rather trivial kind It essentially declares that a disordered state

is much more probable than an ordered one – for the simple reason that there are numerically many more disordered states than ordered ones, so that when a system in an ordered state is randomly rearranged, it is very probably going to end up less ordered than it was before Thus blind chance lies at the basis of the Second Law of Thermodynamics, just as it lies at the basis of Darwin’s theory of evolution Since chance – or contingency, as philosophers call it – is the opposite of law and order, and hence of purpose,

it seems to offer powerful ammunition to atheists who wish to deny any overall cosmic purpose or design If the universe is nothing but a physical system that began (for some mysterious reason) in a relatively ordered state, and is inexorably shuffling itself into a chaotic one by the irresistible logic

of probability theory, then it is hard to discern any overall plan or point

law of thermodynamics

Reaction to the theme of the dying universe began to set in the nine-teenth century Philosophers such as Henri Bergson4and theologians such

as Teilhard de Chardin5 sought ways to evade or even refute the Second Law of Thermodynamics They cited evidence that the universe was in some sense getting better and better rather than worse and worse In Teilhard de Chardin’s rather mystical vision, the cosmic destiny lay not in an inglorious heat death but in an enigmatic “Omega Point” of perfection The progressive school of philosophy saw the universe as unfolding to ever greater richness and potential Soon after, the philosopher Alfred North Whitehead6

(curi-ously, the coauthor with Bertrand Russell of Principia Mathematica) founded

the school of process theology on the notion that God and the universe are evolving together in a progressive rather than a degenerative manner

Much of this reaction to the Second Law had an element of wishful think-ing Many philosophers quite simply hoped and expected the law to be wrong If the universe is apparently running down – like a heat engine run-ning out of steam, or a clock unwinding – then perhaps, they thought, nature has some process up its sleeve that can serve to wind the universe up again Some sought this countervailing tendency in specific systems For example,

it was commonly supposed at the turn of the twentieth century that life somehow circumvents the strictures of thermodynamics and brings about increasing order This was initially sought through the concept of vitalism – the existence of a life force that somehow bestowes order on the material contents of living systems Vitalism eventually developed into a more sci-entific version, what became known as organicism – the idea that complex organic wholes might have organizing properties that somehow override the trend into chaos predicted by thermodynamics.7 Others imagined that order could come out of chaos on a cosmic scale This extended to periodic resurrections of the cyclic universe theory, according to which the entire cos-mos eventually returns to some sort of pristine initial state after a long period

of decay and degeneration For example, during the 1960s it was suggested

by the cosmologist Thomas Gold8that one day the expanding universe may start to recontract, and that during the contraction phase, the Second Law of Thermodynamics would be reversed (“time will run backwards”), returning the universe to a state of low entropy and high order The speculation was based on a subtle misconception about the role of the expanding universe

in the cosmic operation of the Second Law (see the following discussion)

It turns out that the expansion of the universe crucially serves to provide the necessary thermodynamic disequilibrium that permits the entropy in

the universe to rise, but this does not mean that a reversal of the

expan-sion will cause a reversal of the entropic arrow Quite the reverse: a rapidly contracting universe would drive the entropy level upward as effectively as

a rapidly expanding one In spite of this blind alley, the hypothesis that the directionality of physical processes might flip in a contracting universe was also proposed briefly by Hawking,9who then abandoned the idea,10calling

it his “greatest mistake.” Yet the theory refuses to lie down Only this year, it was revived yet again by L S Schulman.11

The notion of a cyclic universe is, of course, an appealing one, and one that is deeply rooted in many ancient cultures; it persists today in Hinduism, Buddhism, and Aboriginal creation myths The anthropologist Mircea Eliade12 termed it “the myth of the eternal return.” In spite of detailed scrutiny, however, the Second Law of Thermodynamics remains

on solid scientific ground So solid, in fact, that the astronomer Arthur Eddington felt moved to write,13“if your theory is found to be against the second law of thermodynamics I can give you no hope; there is nothing for

it but to collapse in deepest humiliation.” Today, we know that there is noth-ing anti-thermodynamic about life As for the cyclic universe theory, there is

no observational evidence to support it (indeed, there is some rather strong evidence to refute it).14

In this chapter I wish to argue, not that the Second Law is in any way sus-pect, but that its significance for both theology and human destiny has been overstated Some decades after Helmholtz’s dying universe prediction, as-tronomers discovered that the universe is expanding This changes the rules

of the game somewhat To give a simple example, there is good evidence that 300,000 years after the Big Bang that started the universe off, the cos-mic matter was in a state close to thermodynacos-mic equilibrium This evidence comes from the detection of a background of thermal radiation that per-vades the universe, thought to be the fading afterglow of the primeval heat The spectrum of this radiation conforms exactly to that of equilibrium at a common temperature Had the universe remained static at the state it had reached after 300,000 years, it would in some respects have resembled the state of heat death described by Helmholtz However, the expansion of the universe pulled the material out of equilibrium, allowing heat to flow and driving complex physical processes The universe cooled as it expanded, but the radiation cooled more slowly than the matter, opening up a tempera-ture gap and allowing heat to flow from one to the other (The temperatempera-ture

of radiation when expanded varies inversely in proportion to the scale fac-tor, whereas the temperature of nonrelativistic matter varies as the inverse square of the scale factor.) In many other ways too, thermodynamic dise-quilibrium emerged from edise-quilibrium, most notably in the formation of stars, which radiate their heat into the darkness of space This direction-ality is the “wrong way” from the point of view of a na¨ıve application of the Second Law (which predicts a transition from disequilibrium to equilib-rium), and it shows that even as entropy rises, new sources of free energy are created

I must stress that this “wrong way” tendency in no way conflicts with the letter of the Second Law To see why this is so, an analogy may be helpful Imagine a gas confined in a cylinder beneath a piston, as in a heat engine The gas is in thermodynamic equilibrium at a uniform temperature The entropy of the gas is at a maximum Now suppose that the gas is compressed

by driving the piston forward; it will heat up, as a consequence of Boyle’s Law

If the piston is now withdrawn again, restoring the gas to its original volume, the temperature will fall once more In a reversible cycle of contraction and expansion, the final state of the gas will be the same as the initial state What happens is that the piston must perform some work in order to compress the gas against its pressure, and this work appears as heat energy in the gas, raising its temperature In the second part of the cycle, when the piston is withdrawn, the pressure of the gas pushes the piston out and returns exactly

the same amount of energy as the piston had injected The temperature

of the gas therefore falls to its starting value when the piston returns to its starting position

However, in order for the cycle to be reversible, the piston must move very slowly relative to the average speed of the gas molecules If the pis-ton is moved suddenly, the gas will lag behind in its response, and this will cause a breakdown of reversibility This is easy to understand If the piston moves fast when it compresses the gas, there will be a tendency for the gas molecules to crowd up beneath the piston As a result, the pressure of the gas beneath the piston will be slightly greater than the pressure within the body

of the gas, and so the piston will have to do rather more work to compress the gas than would have been the case had it moved more slowly This will result in more energy being transferred from the advancing piston to the gas than would otherwise have been the case Conversely, when the piston

is suddenly withdrawn, the molecules have trouble keeping pace and lag back somewhat, thus reducing the density and pressure of the gas adjacent

to the piston The upshot is that the work done by the gas on the piston during the outstroke is somewhat less than the work done by the piston on the gas during the instroke The overall effect is a net transfer of energy from the piston to the gas, and the temperature, hence the entropy, of the gas rises with each cycle Thus, although the gas was initially in a state of uniform temperature and maximum entropy, after the piston moves the entropy nevertheless rises The point is, of course, that to say the entropy of the

gas is a maximum is to say that it has the highest value consistent with the

external constraints of the system But if those constraints change – because

of the rapid motion of the piston, for example – then the entropy can go higher During the movement phase, then, the gas will change from a state of equilibrium to a state of disequilibrium This comes about not because the entropy of the gas falls – it never does – but because the maximum entropy

of the gas increases, and, moreover, it increases faster than the actual en-tropy The gas then races to “catch up” with the new constraints

We can understand what is going on here by appreciating the fact that the gas within a movable piston and cylinder is not an isolated system To make the cycle run, there has to be an external energy source to drive the piston, and it is this source that supplies the energy that raises the temperature of the gas If the total system – gas plus external energy source – is considered, then the system is clearly not in thermodynamic equilibrium to start with, and the rise in entropy of the gas is unproblematic The entropy of the gas cannot go on rising forever Eventually, the energy source will run out and the piston and cylinder device will stabilize in a final state of maximum entropy for the total system

The confusion sets in when the piston-and-cylinder expansion and con-traction is replaced by the cosmological case of an expanding and (maybe, one day) contracting universe Here the role of the piston-and-cylinder

arrangement is played by the gravitational field The external energy supply

is provided by the gravitational energy of the universe This has some odd features, because gravitational energy is actually negative Think, for exam-ple, of the solar system One would have to do work to pluck a planet from its orbit around the sun The more material concentrates, the lower the gravi-tational energy becomes Imagine a star that contracts under gravity; it will heat up and radiate more strongly, thereby losing heat energy and making its gravitational energy more negative in order to pay for it Thus the principle that a system will seek out its lowest energy state causes gravitating systems

to grow more and more inhomogeneous with time A smooth distribution

of gas, for example, will grow clumpier with time under the influence of gravitational forces Note that this is the opposite trend from the case of a gas, in which gravitation may be ignored In that case, the Second Law of Thermodynamics predicts a transition toward uniformity This is only one sense in which gravitation somehow goes “the wrong way.”

It is tempting to think of the growth of clumpiness in gravitating systems

as a special case of the Second Law of Thermodynamics – that is, to regard the initial smooth state as a low-entropy (or ordered) state, and the final clumpy state as a high-entropy (or disordered) one It turns out that there are some serious theoretical obstacles to this simple characterization One such obstacle is that there seems to be no lower bound on the energy of the gravitational field Matter can just go on shrinking to a singular state

of infinite density, liberating an infinite amount of energy on the way This fundamental instability in the nature of the gravitational field forbids any straightforward treatment of the thermodynamics of self-gravitating systems

In practice, an imploding ball of matter would form a black hole, masking the ultimate fate of the collapsing matter from view So from the outside, there is a bound on the growth of clumpiness We can think of a black hole

as the equilibrium end state of a self-gravitating system This interpretation has been confirmed by Stephen Hawking, who proved that black holes are not strictly black, but glow with thermal radiation.15The Hawking radiation has exactly the form corresponding to thermodynamic equilibrium at a characteristic temperature

If we sidestep the theoretical difficulties of defining a rigorous notion

of entropy for the gravitational field and take some sort of clumpiness as

a measure of disorder, then it is clear that a smooth distribution of matter represents a low-entropy state as far as the gravitational field is concerned, whereas a clumpy state, perhaps including black holes, is a high-entropy state Returning to the theme of the cosmic arrow of time, and remem-bering the observed fact that the universe began in a remarkably smooth state, we may conclude that the matter was close to its maximum entropy state, but that the gravitational field was in a low-entropy state The expla-nation for the arrow of time that describes the Second Law of Thermody-namics lies therefore in an explanation of how the universe attained the

smooth state it had at the Big Bang Penrose16has attempted to quantify the degree of surprise associated with this smooth initial state In the case

of, say, a normal gas, there is a basic relationship between the entropy of its state and the probability that the state would be selected from a ran-dom list of all possible states The lower the entropy, the less probable would be the state This link is exponential in nature, so that as soon as one departs from a state close to equilibrium (i.e., maximum entropy), the probability plummets If one ignores the theoretical obstacles and just goes ahead and applies this same exponential statistical relationship to the grav-itational field, it is possible to assess the “degree of improbability” that the universe should be found initially in such a smooth gravitational state In order to do this, Penrose compared the actual entropy of the universe to the value it would have had if the Big Bang had coughed out giant black holes rather than smooth gas Using Hawking’s formula for the entropy of

a black hole, Penrose was able to derive a discrepancy of 1030between the actual entropy and the maximum possible entropy of the observable uni-verse Once this huge number is exponentiated, it implies a truly colossal improbability that the universe should start out in the observed relatively smooth state In other words, the initial state of the universe is staggeringly improbable

What should we make of this result? Should it be seen as evidence of design? Unfortunately, the situation is complicated by the inflationary uni-verse scenario, which postulates that the uniuni-verse jumped in size by a huge factor during the first split second This would have the effect of smoothing out initial clumpiness But this simply puts back the chain of explanation one step, because at some stage one must assume that the universe is in a less-than-maximum entropy state, and hence in an exceedingly improbable state The alternative – that the universe began in its maximum entropy state – is clearly absurd, because it would then already have suffered heat death

The most plausible physical explanation for the improbable initial state of the universe comes from quantum cosmology, as expounded by Hawking, Hartle, and Gell-Mann.17 In this program, quantum mechanics is applied

to the universe as a whole The resulting “wave function of the universe” then describes its evolution Quantum cosmology is beset with technical mathematical and interpretational problems, not the least of which is what

to make of the infinite number of different branches of the wave function, which describes a superposition of possible universes The favored resolu-tion is the many-universes interpretaresolu-tion, according to which each branch of the wave function represents a really existing parallel reality, or alternative universe

The many-universes theory neatly solves the problem of the origin of the arrow of time The wave function as a whole can be completely time-symmetric, but individual branches of the wave function will represent uni-verses with temporal directionality This has been made explicit in the time-symmetric quantum cosmology of Hartle and Gell-Mann,18 according to which the wave function of the universe is symmetric in time and describes

a set of recontracting universes that start out with a Big Bang and end up with a big crunch The wave function is the same at each temporal extremity (bang and crunch) However, this does not mean that time runs backward

in the recontracting phase of each branch, `a la Gold To be sure, there are some branches of the wave function in which entropy falls in the recon-tracting phase, but these are exceedingly rare among the total ensemble of universes The overwhelming majority of branches correspond to universes that either start out with low entropy and end up with high entropy, or vice versa Because of the overall time symmetry, there will be equal proportions

of universes with each direction of asymmetry However, an observer in any one of these universes will by definition call the low-entropy end of the universe the Big Bang and the high-entropy end the big crunch Without the temporal asymmetry implied, life and observers would be impossible, so there is an anthropic selection effect, with those branches of the universe that are thermodynamically bizarre (starting and ending in equilibrium) going unseen Thus the ensemble of all possible universes shows no fa-vored temporal directionality, although many individual branches do, and within those branches observers regard the “initial” cosmic state as exceed-ingly improbable Although the Hartle–Gell-Mann model offers a convinc-ing first step in explainconvinc-ing the origin of the arrow of time, it is not without its problems.19

To return to the description of our own universe (or our particular branch

of the cosmological wave function), it is clear that the state of the universe

in its early stages was one in which the matter and radiation were close to thermodynamic equilibrium, but the gravitational field was very far from equilibrium The universe started, so to speak, with its gravitational clock wound up, but with the rest in an unwound state As the universe expanded, there was a transfer of energy from the gravitational field to the matter, simi-lar to that in the piston-and-cylinder arrangement In effect, gravity “wound up” the rest of the universe The matter and radiation started out close to maximum entropy consistent with the constraints, but then the constraints changed (the universe expanded) Because the rate of expansion was very rapid relative to the physical processes concerned, a lag opened up between the maximum possible entropy and the actual entropy, both of which were rising In this way, the universe was pulled away from thermodynamic equi-librium by the expansion Note that the same effect would occur if the universe contracted again, just as the instroke of the piston serves to raise the entropy of the confined gas So there is no thermodynamic basis for

supposing that the arrow of time will reverse should the universe start to contract

The history of the universe, then, is one of entropy rising but chasing

a moving target, because the expanding universe is raising the maximum possible entropy at the same time The size of the entropy gap varies sharply

as a function of time Consider the situation one second after the big bang (I ignore here the situation before the first second, which is complicated but crucial in determining some important factors, such as the asymmetry between matter and antimatter in the universe.) The universe consisted of a soup of subatomic particles – such as electrons, protons, and neutrons – and radiation Apart from gravitons and neutrinos, which decoupled from the soup well before the first second owing to the weakness of their interactions, the rest of the cosmic stuff was more or less in equilibrium However, all of this changed dramatically during the first 1,000 seconds or so As the tem-perature fell, it became energetically favorable for protons and neutrons to stick together to form the nuclei of the element helium All of the neu-trons got gobbled up in this way, and about 25 percent of the matter was turned into helium However, protons outnumbered neutrons, and most

of the remaining 75 percent of the nuclear matter was in the form of iso-lated protons – the nuclei of hydrogen Hydrogen is the fuel of the stars It drives the processes that generate most of the entropy in the universe to-day, mainly by converting slowly into helium So the lag behind equilibrium conditions is this: the universe would really “prefer” to be made of helium (it is more stable), but most of it is trapped in the form of hydrogen I say

“trapped” because, after a few minutes, the temperature of the universe fell below that required for nuclear reactions to proceed, and it had to wait un-til stars were formed before the conversion of hydrogen into helium could

be resumed Thus the expansion of the universe generated a huge entropy gap – a gap between the actual and the maximum possible entropy – during the first few minutes, when the equilibrium form of matter changed (due to the changing constraints occasioned by the cosmological expansion and the concomitant fall in temperature) from a soup of unattached particles to that

of composite nuclei like helium It was this initial few minutes that effectively

“wound up” the universe, giving it the stock of free energy and establishing the crucial entropy gap needed to run all the physical processes, such as star burning, that we see today – processes that sustain interesting activity, such as life The effect of starlight emission is to slightly close the entropy gap, but all the while the expanding universe serves to widen it However, the rate of increase of the maximum possible entropy during our epoch is modest compared to what it was in the first few minutes after the Big Bang – partly because the rate of expansion is much less, but also because the cru-cial nuclear story was all over in a matter of minutes (The gap-generating processes occasioned by the expansion of the universe today are all of a less significant nature.) I haven’t done the calculation, but I suspect that today