Among all the possible alternative universes, only one is completely uniform and regular. Calculating the probability of this sort of universe, we find that it is very likely indeed. In fact, it is the most probable universe of all, but it isn’t our universe. A universe like that, without small irregularities in the early universe that show up now as small variations in the CMBR, could never be a home for us. Ours has to be a universe with some regions slightly more dense than others, so that gravitational attraction can draw matter together to form galaxies, stars, planets and, maybe, us. As Hawking had put it in his 2006 Caltech lecture, ‘The [CMBR] map of the microwave sky is the blueprint for all the structure in the universe. We are the product of the fluctuations in the early universe.’3 Luckily there are many universe histories that are only slightly un-uniform and irregular. These are almost as probable as the one that is completely uniform and regular. We don’t know how many alternative universes end up producing something like ‘us’, but we do know it did happen once.
Another familiar concept that is significant in Hawking’s thinking about M-theory is the fact that on the quantum level of the universe we cannot observe without interfering, without changing the very thing we are trying to observe. More important and less familiar to most of us, no matter how carefully and completely we observe the present, the part of the past that we cannot observe is, like the future, indefinite. It exists as a range of possibilities, some more probable than others. Putting that together with Feynman’s sums-over-histories, Hawking concludes that ‘the universe doesn’t have just a single history, but every possible history, each with its own probability; and our observations of its current state affect its past and determine the different histories of the universe’.4 That should not sound entirely unfamiliar. We saw earlier how Hawking and Hartle used sums-over-histories when they were developing their no-boundary proposal. What has occurred in Hawking’s thinking is a change of emphasis, the realization that the ability of our observations of the present to decide among those histories has enormously significant implications for our understanding of the universe.
Go back to thinking about Feynman’s method of considering all the possible paths a particle might follow from its starting point to its end point. It’s not so easy to do with the history of a universe. We don’t know point A (the beginning), but we do, in the case of our own universe, know quite a lot about point B, where we are today. Hawking asks us to consider all the histories that satisfy the no-boundary condition (histories that are closed surfaces without boundaries – recall the globe of the Earth) and that end with the universe we know today (point B). There is a vast range of point As, though we can’t say they include universe histories starting off in ‘every possible way’ because we are limiting them to those that satisfy the no-boundary condition. If we were to start our thinking at point A, we’d end up with many possible point Bs, some of which are similar to our universe today, but most of which are not.
Hawking is recommending, instead, what he calls his ‘top-down’ approach to cosmology, tracing the alternative histories from the top down, backwards from the present time. It is a new view of cosmology, and, for that matter, a new view of cause and effect. The universe doesn’t have a unique observer-independent history. We create the history of our universe by being here and observing it. History doesn’t create us.
Take, for example, the question of why there are only four un-curled-up dimensions in our universe. In M-theory there is no overall rule that a universe must have four observable dimensions. Top-down cosmology says that there will be a range of possibilities that includes every number of large space dimensions from zero to ten. Three dimensions of space and one of time may not be the most probable situation, but that is the only sort of situation that is of interest to us.
Considering the universe in the old way, from the ‘bottom up’, there seems to be no discoverable reason why the laws of nature are what they are and not something different, why the universe is fine-tuned for our existence. But we do observe the laws of nature to be what they are, and we are here. Why not start with that? Our presence is hugely significant. Out of the enormous array of possible universes, our presence ‘selects’ those universes that are compatible with our existence, and makes all the rest of them almost irrelevant (though we shall see about that as Hawking continues).
With the no-boundary universe we no longer needed to ask how the universe began. There was no beginning. With M-theory we no longer need to ask why the universe is fine-tuned for our existence. It is our existence that ‘chooses’ the universe we live in. In effect, we fine-tune it ourselves. The anthropic principle has come to its full strength indeed. As Hawking puts it, ‘Although we are puny and insignificant on the scale of the cosmos, this makes us in a sense the lords of creation.’5
Now, the question arises: can we test this theory? Hawking writes that there may be measurements capable of differentiating the top-down theory from others, to support it or refute it. Perhaps future satellites can take such measurements. In his 2006 Caltech lecture, Hawking had mentioned the ‘new window on the very early universe’ that the detection and measurement of gravitational waves would open for us. Unlike light, which was scattered many times by free electrons before freezing out when the universe was 380,000 years old, gravitational waves reach us from the earliest universe without interference from any intervening material.6
Hawking extends top-down thinking to the emergence of intelligent life on Earth. He offers an eloquent account of the manner in which our universe, our solar system, and our world are incredibly fine-tuned to allow our existence, far far beyond any reasonable expectation. Nevertheless, in a restatement of the anthropic principle in a simple and unarguable way, he tells us that ‘Obviously, when the beings on a planet that supports life examine the world around them, they are bound to find that their environment satisfies the conditions they require to exist.’7 Just as we, by the fact of our presence, choose our universe, we choose a history of this Earth and our cosmic environment that allows us to exist.
In The Grand Design, Hawking seems no longer to entertain doubts that everything is determined. The information paradox, by whatever means it was banished, has ceased to be a bother. He states, unequivocally, that ‘The scientific determinism that Laplace formulated is … in fact, the basis of all modern science.’8 That, of course, he had never called into question. His earlier suggestion, regarding the implications of information loss, was that all of modern science might be wrong. Such fears apparently have been put to rest, for he goes on to say that scientific determinism is ‘a principle that is important throughout this book’.9 And, later, ‘This book is rooted in the concept of scientific determinism.’10
Scientific determinism applies to us humans too: ‘It seems,’ he writes, ‘that we are no more than biological machines and that free will is just an illusion … Since we cannot solve the equations that determine our behaviour, we use the effective theory that people have free will.’11 We might wish Hawking had spent a little more time in the book with this issue. Important scientific work has taken place having to do with human free will – some of it supporting his view and some of it not – but Hawking does not discuss it. He has made his own choice. His comment that the world is in a mess because, ‘as we all know, decisions are often not rational or are based on a defective analysis of the consequences of the choice’, also leaves one wishing for more discussion. The comment comes across as uncharacteristically ‘throw-away’, compared with Hawking’s thoughtful comments about the world situation in his lectures and public statements.
Determinism turns out, however, to be a somewhat complicated concept and not as rigid as we might suppose. As we saw earlier in this book, on the quantum level of the universe we have to accept a somewhat revised version of determinism in which, given the state of a system at any one time, the laws of nature determine the probabilities of various different futures and pasts rather than dictating the future and past precisely. As Haw
king puts it, ‘Nature allows a number of different eventualities, each with a certain likelihood of being realized.’12 You can test a quantum theory by repeating an experiment many times, noting how frequently different results occur and whether the frequency of their occurrence fits the probabilities the theory predicted.
Hawking mentions again the ideas that he and I discussed in the old DAMTP common room back in 1996, ideas for which some people were criticizing him at that time. His words to me back then were ‘We never have a model-independent view of reality. But that doesn’t mean there is no model-independent reality. If I didn’t think there is, I couldn’t go on doing science.’ Now, in The Grand Design, he writes, in italics for emphasis, ‘There is no picture- or theory-independent concept of reality.’ He goes on to say that this is ‘a conclusion that will be important in this book’. This statement rephrases the first part of his statement to me, substituting ‘concept’ for ‘view’, but not the second part. We are left to wonder whether the rest still holds.
Hawking lists two other ways of thinking about ‘reality’ that he is rejecting. One is the ‘realist’ viewpoint of classical science based on the belief that a real, external world exists, a world that can be measured and analysed – that is the same for every observer who studies it. The other is what Hawking calls the ‘anti-realist’ viewpoint. This viewpoint is so insistent on confining itself to empirical knowledge gleaned through experiment and observation that it has little use for theory and ends up self-destructing with the notion that because anything we learn is filtered through our brains, we can’t really count on there being such a thing as empirical knowledge.
Hawking believes that his own ‘model-dependent realism’ makes the argument between realism and anti-realism unnecessary. He insists it is only meaningful to ask whether a model agrees with observation, not whether it is ‘real’. If more than one model agrees with observation, you don’t have to argue which is more ‘real’ or ‘right’. ‘Our perception – and hence the observations upon which our theories are based – is not direct, but rather is shaped by a kind of lens, the interpretive structure of our human brains.’13 That goes for everyday experience, he says, not just in science. On that level too, whether we are consciously devising models or not, we never have a model-independent view of reality. Nevertheless, our model-dependent views of reality are not worthless. They are the way human beings come to understand and manage their world. Models stand and fall as they continue or cease to match observation and experience.
It isn’t difficult to agree with Hawking. Unless I am in a state of denial – which we all probably are in part, sometimes – I do make my learning-progress through life in precisely that way. You and I come from different sets of experiences. Perhaps we might be able to agree to disagree without making claims about who is ‘right’ and who is ‘wrong’. Would Hawking go so far as to apply his philosophy to the more extreme views that divide our world? That, possibly, is where he would invoke something like the second, decidedly Platonic, part of his statement, perhaps to say, ‘But that does not mean there are no such things as “right” and “wrong”; if I didn’t think there are, I couldn’t go on living in any meaningful way.’ On the other hand, there are claims that human values are products of our evolutionary history. In this way of thinking, ‘right’ is what has aided the survival of our species – nothing more profound or fundamental than that. If that is true (and what, after all, does ‘truth’ come down to in a discussion like this?), then model-independent morality is perhaps as illusive as model-independent reality.
Be that as it may, Hawking’s discussion of ‘reality’ helps with something you may have been wondering about since Chapter 2. If no one has actually seen, for example, an electron, how do we know electrons are ‘real’?
Though it’s true that no one has ever seen an electron, electrons are a useful ‘model’ that makes sense of observations of tracks in a cloud chamber or spots of light on a television tube. The model has been applied with enormous success in both fundamental science and engineering. But are electrons real? Though a great many physicists would say, yes, of course they are, that question according to Hawking is meaningless.
‘Model-dependent realism’, as he calls it, is a useful way of thinking about dualities – those situations in which two different, perhaps mutually exclusive, descriptions are necessary to gain a better understanding than either description alone can provide. Neither theory is ‘better’ or more ‘real’ than the other. Recall the most familiar example, wave–particle duality, which emerged in the early twentieth century with the discovery that when light interacts with matter it acts as though it must be particles, while experiments with the way light travels show that it acts as though it were waves.
All of which brings us back to think more knowledge-ably about M-theory. As we’ve said, it appears that no mathematical model is able to describe every aspect of the universe. Each theory in the M-theory family can describe a certain range of phenomena. When these ranges overlap, the theories agree. In this manner they are all parts of the same theory, just as the smaller sections of the map in Hawking’s analogy were all parts of the same map. But no single theory in the family is capable of describing all the forces of nature and the particles that we mentioned in Chapter 2, plus the framework of time and space where the universe game plays out. If this seemingly fragmented map is where the great quest must end, so be it, ‘it is acceptable within the framework of model-dependent reality’.14 We have no more fundamental theory that we can claim is independent of the models we know.
Hawking and Mlodinow write that all the universes in the multiverse were created out of nothing, arising naturally from physical law, and that they require no creator. They have oversimplified a bit to make their point. In eternal-inflation theory, which Hawking favours, universes don’t arise from nothing. They arise from other universes. Somewhere in the past, there may have been a first universe and a first inflation sequence, where it all started off, or the repeating self-replication process may stretch back eternally into the past. Presumably the origin of that first universe (if there was a ‘first universe’) can be explained by the no-boundary-proposal, which leaves us precisely where A Brief History of Time stopped, asking those same profound questions that left plenty of room for God.
The Grand Design, however, addresses another puzzle, the fine-tuning mystery. Some who believe in God – still not warned off the God-of-the-Gaps theology which clings to instances where something seems unexplainable without God – will no doubt find it distressing that Hawking and Mlodinow have very successfully shown another plausible explanation, using the top-down method and multiverses. If you believe in God only as a necessary explanation, Hawking has once again cut you adrift. More interesting than the media attention Hawking’s books gain for God/science issues is the fact that, for careful, thoughtful readers, they do lead to some profound inner debates. Those don’t always end the way Hawking might expect.
In their final chapter, Hawking and Mlodinow address the question of where the physical laws come from, introducing the discussion with the following comment: ‘The laws of nature tell us how the universe behaves, but they don’t answer why.’ At the end of A Brief History of Time Hawking had written that the answer to that question would be to know the mind of God. Now he has broken the question into three parts: ‘Why this particular set of laws and not some other?’, ‘Why is there something rather than nothing?’ (the laws being part of the ‘something’) and ‘Why do we exist?’
To help address the first of those questions, Hawking and Mlodinow list the laws that are necessary in a physical universe that looks like our own. They must be a set of laws that have a concept of energy in which the amount of energy is constant, not changing over time. Another requirement is that the laws must dictate that the energy of any isolated body surrounded by empty space will be positive. And there must be a law like gravity. The theory of this gravity must have supersymmetry between the forces of na
ture and the particles of matter they govern. Adopting the top-down method, the answer to ‘Why this particular set of laws and not some other?’ can be simply ‘Because any other set of laws would make it impossible for us to be here asking that question.’ That would be an answer invoking the anthropic principle, but M-theory has a little more to say on the issue than that: because of all the different ways the extra dimensions curl up, with each universe having laws determined by how they curl in that universe, there will certainly be a universe around that has these laws.
To help address the third of those questions (‘Why do we exist?’), Hawking and Mlodinow introduce us to a computer game known as ‘The Game of Life’. It’s a fascinating game, invented back in 1970 by John Horton Conway, then a Cambridge mathematician. The layout looks like a chessboard, with some squares ‘alive’ and some ‘dead’. A very simple set of rules dictate ‘deaths’, ‘births’ and ‘survival’ as the game moves from ‘generation to generation’. It soon becomes evident that extremely simple rules can play out in very complicated ways. Remember the ‘alien who has never experienced our universe’ in Chapter 2. Someone coming in on this game after it’s been going a while will be in a similar position, able to deduce ‘laws’ from what’s going on, laws that seem to govern the formation and behaviour of elaborate groupings of the live and dead squares – laws that however are not among the simple original laws at all but that arise out of them. The game is a simple example of ‘emergent complexity’ or ‘self-organizing systems’. It helps us comprehend, for example, how the stripes on a zebra or patterns on a flower petal occur from a tissue of cells growing together.
Conway invented this game as an attempt to find out whether in a ‘universe’ with extremely simple fundamental rules, objects would emerge that were complex enough to replicate themselves. In the game, they do. They could even, in a sense, be thought of as ‘intelligent’. The bottom line is that a very simple set of laws is capable of producing complexity similar to that of intelligent life. In Hawking’s words, ‘It is easy to imagine that slightly more complicated laws would allow complex systems with all the attributes of life.’15 There is disagreement as to whether such life would be self-aware.
Stephen Hawking, His Life and Work Page 34