Book review: The Grand Design (Part 3 of 4: The background physics)

In part 1 of this review I discussed the main issues raised by the book and in part 2 I said that the book by Hawking and Mlodinow argued that M-theory and the no boundary condition can provide answers to the three big questions: Why is there something rather than nothing? Why do we exist? Why this particular set of laws and not some other?

To understand what lies at the basis of M-theory, we need to appreciate a key difference between classical physics (which describes the large-scale structure of the everyday world we live in and from which we draw our intuitions about how the world works) and quantum mechanics (which describes the microscopic atomic and subatomic world).

What classical physics says is that if we release an object at some point A, it will subsequently wander off on some trajectory (or path) that depends on its initial state of motion and the forces that act on it. This is what enables good football quarterbacks to throw passes to receivers with such accuracy. If the ball is poorly thrown on a windy day and/or we stop observing the ball, we may not know or be able to predict which path the ball will take or where it will land but our classical intuition tells us that it will go along some specific path that is determined by the initial throw and the wind conditions.

But quantum mechanics has this counter-intuitive idea that once we stop observing the object, the object takes every conceivable path simultaneously. This means that there is no unique location for the object at any given time, that it is everywhere at the same time and could eventually end up anywhere at all. Another way to say it is that an object has many different histories. This is what boggles most people’s (including scientists’) minds about quantum theory but we have to learn to live and work with it (i.e., develop ‘quantum intuition’, so to speak) because this theory is phenomenally successful and there seems to be no getting around it at this time. Some people are working on developing alternative theories that do not have its strange features but have not had much success so far.

Now if we detect the object at some later time to be at some point B, this eliminates some of the potential paths we started with because they would not have resulted in the object ending up where we detected it. So the act of detection picks out a subset of the initial set of possible histories, limiting the ones of interest to those that began at point A at the specified time and ended at B at the later time, which still includes an infinite number of paths or histories. An elaborate mathematical machinery (called the ‘sum over histories’ or more technically ‘path integrals’) has been created to add up all the possible paths the particle could have taken in going from A to B. The calculated results correctly predict the empirical observations, which is why scientists have confidence in quantum theory despite its counter-intuitive features.

What M-theory does is take this key idea of quantum mechanics and apply the ‘sum over histories’ approach to the universe as a whole. Building on the idea of the inflationary universe (see part 9 and part 13 of my series Big Bang for Beginners for more details), since the net energy of the universe is zero, there is no restriction on the number of new universes that can ‘pinch’ off from previously existing universes. Since the Heisenberg uncertainty principle states that you can never have truly empty and inert space (p. 113) but that space constantly has particles coming into existence and disappearing again, any one of those fluctuations in space could form the seed of a quantum fluctuation that triggers the birth of a new universe.

So universes are being created all the time and there are a vast number of possible histories of the universe, of the order of 10500. They each have their own forms of matter and their own laws. According to the ‘sum over histories’ in quantum mechanics, all these universes exist simultaneously, giving rise to the name ‘multiverse theory’. When we observe our universe, we are picking out just those histories that could produce the present state we see. As Hawking and Mlodinow state:

Quantum physics tells us that no matter how thorough our observation of the present, the (unobserved) past, like the future, is indefinite and exists only as a spectrum of possibilities. The universe, according to quantum physics, has no single past, or history. (p. 82)

We seem to be at a critical point in the history of science, in which we must alter our conception of goals and of what makes a physical theory acceptable. It appears that the fundamental numbers, and even the form, of the apparent laws of nature are not determined by logic or physical principle. The parameters are free to take on many values and the laws to take on any form that leads to a self-consistent mathematical theory, and they do take on different values and forms in different universes. (p. 143)

Given the staggeringly large number of possible histories, it was almost inevitable that one of those universes would have the properties that ours has. It is like rain. If you pick a point on the ground, the probability of it being hit by a raindrop is infinitesimally small. But in a rainstorm, there is such a huge number of drops that it is inevitable that at least one will hit the ground there.

Hawking and Mlodinow’s book does not shy away from making strong claims, such as that the theory they describe has to be the right one. “M-theory is the only candidate for a complete theory of the universe… M-theory is the unified theory Einstein was hoping to find.” (p. 181, emphasis in original.)

That seems hubristic to me. If the history of science teaches us anything it is that theories, however successful at any given time, tend to be later replaced by other theories as the questions that need to be addressed change. However obviously important they may seem, is usually a mistake to think that the questions that concern us now will be the same questions that future generations care about. Also the theory of supersymmetry, which is central to M-theory though not necessarily to the idea of multiverses, has been around since 1970 or so, with none of the exotic partner particles it predicts having been detected as yet. The theory’s supporters are pinning their hopes on the Large Hadron Collider that has just started operations, hoping that its energies will be sufficient to produce these particles.

In the last part of this review, I will look at the implications of M-theory for religion and give some of my reactions to other features of the book.


  1. says


    I appreciate the level-headed pragmatism on display in this review. There is a natural human tendency to become emotionally attached to our theories and worldviews, but if our ultimate objective is attainment of true understanding, we must retain an ability to constantly evaluate the viability of our theories and abandon them for more promising concepts if necessary. I think of this as having an inner ‘Supreme Court’ or ‘super-legislature.’

    This emotional attachment is perhaps most easily diagnosed in religious fundamentalists, but is not just confined to them. Even scientists, the putative apostles of reason, fall prey to this vice. It tends to be bolstered by the desire of competing groups to trounce one another in ideological combat, thereby gaining feelings of superiority over the ‘losers.’ Thus, the disciplined choice to eschew this attachment is somewhat risky for the pragmatist, exposing him to accusations that he doesn’t know what he believes or can’t make his mind up from one day to the next.

    We live in a society that seems to crave simple, hard-and-fast verities that naturally feed Manichean conflicts. So much of this social tension could be defused by cultivating humility in the face of great complexity. Arrogance is the rock under which ignorance likes to crawl.

  2. Steve LaBonne says

    As a mere biologist I get very uncomfortable seeing an enormous, extremely ambitious theoretical edifice built on no experimental foundation whatever (as Mano points out not even supersymmetry has any support decades after it was proposed.) I remember a time in developmental biology when elaborate theories of how embryonic pattern generation worked were proliferating. The actual processes, elucidated by painstaking experimentation, turned out to be far different from anything the “theoretical biologists” had imagined.

    A “theory of everything” that makes few if any testable predictions looks uncomfortably closer to theology than to science.

  3. says


    I share your concern about the lack of data. That is why I am personally leery of string theory and supersymmetry. I think that this lack of a strong base of data makes M-theory very speculative.

    The point is that questions about the origins of the universe used to be a mystery that we had no idea how to even begin addressing. Thanks to some hard work, it has now become a puzzle for which we have some idea how to frame questions, suggest solutions, and make investigations. The solutions may be wrong but there is no mystery any more.

    It is kind of like Darwin’s theory of natural selection. When proposed, the evidence for the underlying mechanism was not really there since he did not even know about genes. But what it did was remove the origin of species from being a mystery to a puzzle and subsequent work provided the needed evidence. Ir was only about 70 years later that natural selection really became well established and about 100 years later with the discovery of DNA that we understood it microscopically.

  4. Steve LaBonne says

    Yes, with your broader point I absolutely agree, and we agree also that it doesn’t depend in any way on Hawking’s (or any other) version of M theory.

  5. says

    I know this is going to sound lame to you, Mano, but I can’t see how you can observe the operation of the process of evolution and the glories of the world without appreciating the MYSTERY of it all. Yes I know…foggy thinking, clouding thinking…etc. But the problem with being relentlessly reductionist is that you end up with something so much less than you started with.

    From mystery to puzzle. That implies that, given perfect knowledge of the present and preceding conditions, one could actually deduce the subsequent conditions, right? A “puzzle” implies a “solution,” right? But if there’s any validity to what you described about quantum theory (that any unobserved history is uncertain (I’m paraphrasing)) and only a sort of probability, then it would be impossible to have perfect knowledge of past and present conditions. Mystery sneaks through the back door.

  6. says


    No, it is an important question and does not sound lame. In fact, I wrote more about mysteries and problems three years ago here. As I say there, understanding does not eliminate awe, it can actually enhance it.

    A solution to a problem can only exist within the parameters provided by the theory used to solve the problem. What constitutes “perfect” knowledge (I think a better word might be “complete”) depends on the theory. Newtonian mechanics allowed for complete knowledge of the position and momenta of any particle at any time, allowing for classical determinism.

    Complete knowledge in quantum mechanics consists of knowing the probability distributions of potential outcomes, allowing for a statistical determinism.

    One can call the absence of Newtonian-type knowledge a mystery if one likes, but it is only “less” in the sense that one has expectations that are outside the framework allowed by quantum theory.

  7. Steve LaBonne says

    But the problem with being relentlessly reductionist is that you end up with something so much less than you started with.

    Ah, if only you had enough appreciation of science to understand just how much things DO NOT look that way to a trained scientist. Reality is so much more awe-inspiring- and so much stranger- than the puny anthropocentric (and anthropomorphic) stories that humans make up about it.

  8. says

    Steve, I find some of your comments really excellent when you focus on the science rather than argumentation. Your comment on a theory of everything which has no way of being tested experimentally, for just one example.

    But when you are arguing, it’s a different story. I am not anthropocentric, and I am not suggesting (puny!) anthropocentric or morphic stories. At some point it might be worth considering your word “puny,” but not now. Mano was right, it was the question of “awe” I was addressing.

    Here is the line to which I was referring: “But what it did was remove the origin of species from being a mystery to a puzzle.” I see that as reductionist, and actually a suggestion that the origin of species is not a mystery, but a CONTAINED puzzle. My point, Mano, was that there is nothing contained about it. Evolution is elegant and sensible, but it can only explain things that have already happened and never actually predict anything with specificity. There is an irreducible mystery about every actual species. Does it give you any pause, Steve, that although the process of evolution can be observed (and I think is clearly the correct explanation of what has happened and is happening), its predictive power is so limited? Doesn’t that leave it susceptible to much of the criticism you raised against the theory of everything?

  9. Steve LaBonne says

    What makes you think that line was directed at you personally, Gibert? I was referring to mythology /religion in general, and my description was apt.

    Your comments about evolution are utterly clueless, though. Retrodictions are logically just as good for validating a theory as forward-looking predictions, and evolutionary theory makes zillions of well-confirmed ones. Do you also think geology, or for that matter history, are like religions?

  10. says

    As I said, Steve, your comments directed towards science, and particularly-perhaps ironically-your skeptical ones seem very insightful. Your comments as acolyte, where your point seems more to silence dissent than to consider the issues, are not, in my opinion, as insightful.

    The only honest reading of your sentence, which quoted a line from my post and began with “if only YOU…” is that your comment was directed at me and not more generally into the ether. Come on.

    I am directing this question at you, Steve. Do you seriously consider that “retrodictions are logically just as good for validating a theory as forward-looking predictions?” Because if so, I would certainly disagree. The stock market is full of “retrodictions,” called for that purpose “back-testing,” and many people have gone broke relying on them. And this is because the use of retrodictions includes a feature of faith. Yes, faith. Faith that the theory has encompassed all factors which may come into play in the future, whereas in fact they only take into account factors that have occurred in the past. New things can happen, and the scientific geniuses who rely on back-testing (most famously the company called Long-term Capital Management) go down in flames.

    As I said, I find evolution an elegant theory which admirably (and I believe truthfully) explains things that have happened. But I do regard its predictive powers as weak in any particular case. And since it is only a backward looking theory, it only describes what has happened and eliminates essentially none of the mystery from what may happen. That is the way it actually works, and my point was that if that theory about the unobserved past being a sort of probability was true, then that is the way the theory of evolution MUST work. Hence evolution doesn’t turn any mystery into a (mere) puzzle.

    How about a serious response?

  11. Steve LaBonne says

    But I do regard its predictive powers as weak in any particular case.

    Again, this would apply equally to geology and history. You just have a fundamental misunderstanding here (one that is an old chestnut of creationist “arguments”), and it does you no credit at all to persist blindly in it rather than making the effort to learn something.

    There is nothing inferior about retrodictions- predictions as to what you’ll find in the fossil record or the genes when you look, compared to predictions about what will happen in the future (which in the case of biological evolution has an irreducible probabilistic component). Evolutionary theory makes an almost uncountable number of such predictions and so far all have been fulfilled by observation. (The classic, albeit rather silly, example of what a disconfirming observation would look like is a well-authenticated rabbit skeleton in a pre-Cambrian stratum.) There is an infinite number of such observations that we could make that would pose a serious challenge to evolution. With the advent of gene sequencing, for example, we could easily have observed any of a huge number of possible patterns in sequence comparisons among species that would be totally incompatible with the core of evolution, descent with modification. Instead we found that gene sequences fit beautifully into exactly the hierarchical arrangement predicted by evolutionary theory. YES, that retrodiction is 100% logically equivalent to a prediction.

    Of course, in organisms with a short generation time, evolution actually CAN be studied in real time and predictions CAN be made, not about exactly what will happen (because, again, of the essential stochastic element in evolution), but of the kinds of things that our theories tell us can or can’t happen. See here for the mother of all such experiments: href=,%20Plant%20Breeding%20Reviews,%20Lenski.pdf

    I’m being as gentle as I can be with someone who is purveying a standard line or creationist BS. But really, you owe it to your self-respect to educate yourself a little.

  12. says

    Really, I’m glad you’re being “gentle.” And I really do appreciate the extent to which you actually cite evidence and address the question of retrodictions.

    But you’re misunderstanding my point entirely.

    I am not creationist, nor am I “purveying” a creationist line. My argument is with the claim that the theory of evolution has removed the mystery from the process of evolution. Specifically, that it has reduced (my word) the process of evolution from mystery to puzzle. This, in my opinion, is a spurious claim for science. It has applied a rationale to what has happened, but its actual predictive power is quite weak. Given perfect knowledge of actual conditions, one still could not predict the size, shape and form of creatures yet to come. I appreciate that you recognize this as irreducibly probabalistic (stochastic?)–I just wish you appreciated what that means in the context of my argument. If evolution is not governed by cause and effect, then what does govern it? and if it is governed by cause and effect, then why can’t the science dedicated to addressing it identify the causes and effects in advance? I’m not saying that this is an argument against evolution, but I do believe it is a fundamental limitation.

    In my view, attaching the word “evolution” to the process and treating that as a full explanation is little different than saying “God did it.” Like the theory of everything apparently does for you, it leaves a little too much unexplained for my comfort.

    You didn’t answer my question about the logical equivalence of retrodiction vs. prediction. You said they are equivalent. Do you seriously maintain that position? Can you just give me a straight answer to that question?

    You seem to have a scorn for any disagreement or discussion not in complete agreement with your position, believing that if only I knew as much as you did, then I would naturally agree with you. I trust it is clear to others, if not to you, how similar that style of argument is to the dreaded “creationists,” who argue that the only way to salvation is through (their kind of) faith.

    You keep mentioning geology and history. Do you actually equate these disciplines?

    How about some head-on, clear answers? I’m willing to be persuaded if you can do it. Or “educated” if you prefer, but in the spirit of this blog, why not engage in a rational discourse?


  13. Steve LaBonne says

    You didn’t answer my question about the logical equivalence of retrodiction vs. prediction. You said they are equivalent. Do you seriously maintain that position? Can you just give me a straight answer to that question?

    Can you read at all? How many different times do I have to keep telling you that the answer is YES YES YES YES? If you predict something as yet unobserved, it matters not at all whether the predicted observation is of the traces left by a past event, or of a future event. It’s still a prediction about something as yet unknown that has the power to confirm or disconfirm your theory. Why is that so hard for you to understand? And evolutionary theory makes an uncountable number of such successful predictions.

    And what kind of silly strawman is it to say that biologists claim “evolution explains everything”? Look, dammit, I AM a biologist(not an evolutionary biologist but a molecular geneticist now working in forensic science) and that comment makes so little sense that it’s “not even wrong”.

    First Rule of Holes, my friend. I have scorn only for willful ignorance, and you’re demonstrating precisely that here. Please stop and go do some serious reading.

  14. says


    Actually, I owe you an apology. I see, upon rereading, that you did maintain that retrodiction is logically equivalent to prediction. I didn’t notice it because you hedged your answer so severely. I disagree with you, though. The retrodiction that one would not find evidence of a rabbit in Precambrian remains is not the equivalent, in my mind, of saying, shortly before the actual evolution of rabbits, that one would soon find…rabbits. A retrodiction may be a fine way of eliminating things that are inconsistent with evolutionary theory, but it imposes few actual limits on what, among possibly consistent variations, will be “selected.” Should I figure out and show the formal logic? or can you take it from here? They are not logically equivalent.

    We could argue about that, and I will demonstrate the formal logic if necessary, but my point continues to address this question of “selection.” As a friend (a PhD in science, incidentally) of mine once put it, evolution doesn’t “drive, it draws.” Which is cool, but it leaves room for mystery.

  15. Steve LaBonne says

    I disagree with you, though.

    You’re wrong. This is not a question of opinion. What’s “in your mind” is not an argument. You are simply unable or unwilling to grasp how historical sciences- again, geology being another example- work. That’s your problem, not a problem for scientists.

    You cannot, by the way, even address this point with formal logic, because the point you’re not grasping is one of the correct INTERPRETATION of your predicates. Again, the key point is that a successful theory must predict things that have not yet been observed. Evolutionary theory has done that many times over.

    And you are also refusing to deal with experiments like Lenski’s where evolution actually can be observed while it happens.

    I don’t care whether you like to be told that you’re peddling discredited creationist horseshit. That is EXACTLY what you’re doing. And you do yourself no favors thereby.

  16. says

    I apologized before seeing your latest. You’re typing too fast for me.

    You say:’And what kind of silly strawman is it to say that biologists claim “evolution explains everything”?’

    Did I say that “biologists claim evolution explains everything”? I looked for, but did not find that statement in my post. Considering that I missed something so obvious in yours, though, I will withhold comment until you tell me where you saw that.

    In any event, my comments are not directed towards a strawman, but towards a very specific part of Mano’s initial blog posting. He argued that evolution had turned the origin of species from a mystery to a puzzle, and I have directed my comments towards that issue. I regard the mystery as intact because, to say it yet again in a slightly different way, evolution cannot choose authoritatively among the probable outcomes of a present situation. I call that “mystery” as opposed to “puzzle.”

  17. Steve LaBonne says

    I regard the mystery as intact because, to say it yet again in a slightly different way, evolution cannot choose authoritatively among the probable outcomes of a present situation.

    Then you’re making the trivial error- and more seriously the assault on language- of confusing a process having a significant stochastic component with a “mystery”. (With that kind of grasp of probability I hope you don’t gamble.) No matter how you twist and turn you have no point. (And you’re dead wrong about retrodiction / postdiction.)

    Oh, and I was objecting to “In my view, attaching the word “evolution” to the process and treating that as a full explanation is little different than saying “God did it.” Like the theory of everything apparently does for you, it leaves a little too much unexplained for my comfort.” If I misinterpreted this, it’s because you’re so unclear about what you’re trying to say (the result of unclarity of understanding).

    I’m done with this- you’re just being silly.

  18. says

    Not to draw too fine a point, you didn’t “misinterpret,” you misquoted. That is, you attributed something to me that was actually an interpretation of yours. Or to put it another way, you created a strawman, attributed it to me, and then accused me of creating a strawman. When I observed that I had missed a part of your post, I apologized. You, by contrast, ignore the essential fact of misattribution and gloss it over with the accusation that I am being unclear.

    I’ll point out also that your original question as to whether I equated “geology and history” to religion has now morphed (once subjected to scrutiny) into “historical sciences- again, geology being another example.”

    I’ll leave others to judge who is being unclear.

    Your statement that prediction and retrodiction are “logically equivalent” can, actually, be phrased in formal logic, and your statement can be logically disproved. I guess I’ll have to do it. As you formulated them, they are not equivalent, logically or otherwise.

  19. Steve LaBonne says

    Your statement that prediction and retrodiction are “logically equivalent” can, actually, be phrased in formal logic, and your statement can be logically disproved.

    I could cite papers by famous philosophers of science that demonstrate otherwise, but I’m not bothered anymore. Bye. I won’t come back to point out the fallacies in whatever you put up here, so you may want to save yourself the trouble.

Leave a Reply

Your email address will not be published. Required fields are marked *