Paul Davies's forthcoming book Cosmic Jackpot is subtitled "Why Our Universe Is Just Right for Life," so you know that he's not going after small questions, here. The book is a lengthy and detailed discussion of what he terms the "Goldilocks Enigma," and what others refer to as "fine-tuning"-- basically, how do you account for the fact that the universe allows us to exist? A small change in the values of any of the constants of nature would very likely make it impossible for life as we know it to exist. And yet, here we are-- so how did that happen?
Though this book won't be released for a while yet-- the press materials say April-- I finished it a little while ago. I've been dragging my feet on writing up a review, though, because it's a difficult task. I like having publishers send me free advance copies of books, and I would like to encourage that behavior by writing a good review.
The problem is, trying to review this book gives me some sympathy for Richard Dawkins. It's very hard to write responsibly about a topic you just can't take seriously.
It's not an exact parallel, of course, because I don't have the utter contempt for Davies's subject matter that Dawkins does for religion. But this is fundamentally a book about anthropic principle sorts of arguments, and I have a very hard time treating those as anything more than dorm-room bull session material. They're certainly not what I would call science in the normal sense.
The central issue, as I said, is the observation that life as we know it is absurdly sensitive to the precise values of the constants of nature. If the ratio of proton to neutron masses were different by a percent or so, there would be no chance of life as we know it. If the strength of the weak force were a little different, we could've ended up with a universe containing nothing but helium, and stars would've been unable to form. If gravity were a little stronger, everything would be black holes. And so on, for all of the fundamental forces and particles.
The question Davies is trying to address is why did it turn out that way? He gives the obligatory run-down of modern particle physics, and offers a set of explanations for why the universe is hospitable to life: it might be just a big coincidence, with no deeper reason ("The Absurd Universe" in Davies's term); it might be that the values we see are the inevitable result of some deeper theory ("The Unique Universe"); it might be the handiwork of some cosmic designer ("Intelligent Design"); it might be that we're just seeing the result of an anthropic selection among the infinite number of possible universes in the multiverse ("The Multiverse"); or, there might be something inherent in the nature of life that predisposes the universe to encourage the development of sentient observers ("The Life Principle"). Davies favors the last of these, and spends most of the book dismissing the other possibilities and building up to a sketch of his preferred version, in which the constants of nature did not start with well-defined values, but rather evolved to their present values through a sort of wavefunction collapse thanks to the observing powers of intelligent beings (namely us).
On a purely mechanical level, this is a very well-done book. Davies provides a concise and readable summary of the state of modern physics, and the issues he raises. He has clearly thought deeply about the subject of the origins of life in the universe, and he provides as clear a presentation of the central idea as I've seen.
I have a number of problems with the whole concept, though, starting with the fact that some of the key premises seem misguided. He spends a great deal of time talking about the improbability of our universe arising from a multiverse scenario, and arguing that the inevitable conclusion of certain theories is that we're living in a fake universe, either one constructed by hyper-intelligent beings from some other part of the multiverse, or a computer simualtion of a universe constructed by hyper-intelligent beings from some other part of the multiverse. He seems to find this compelling, but in my opinion, this is just what happens when you get too carried away with the principle of mediocrity.
The "principle of mediocrity" is one of those meta-scientific rules like "Occam's razor," and holds that there shouldn't be anything particularly exceptional about our situation. The Earth is an unremarkable planet around an unremarkable star in an unremarkable galaxy, in an unremarkable universe. Davies interprets this to mean that we ought to be living in the most likely of all possible universes. From there the argument goes that an infinite multiverse ought to include at least one universe which evolves beings with technology sufficient to either create universes, or create simulated universes of sufficient detail to contain us. As it takes relatively little effort to create lots of fake universes, these should vastly outnumber real universes, and since our universe should be unremarkable, we're most likely living in a fake universe.
This is a great basis for a Greg Egan story, and probably a terrific way to mess with the heads of stoned college students, but I have a major problem with treating it as a serious argument, that I can demonstrate by pounding my fist on the number pad of my keyboard: 863846153456864153132. There is only one chance in 1,000,000,000,000,000,000,000 of me typing that specific combination of digits, and yet, there it is.
The problem I have with this line of argument is that it elevates the "principle of mediocrity" to a level that it doesn't really merit. The principle of mediocrity isn't a scientific law, like conservation of energy-- it's a meta-scientific principle, like Occam's Razor. There's nothing that says it can't be wrong, in the same way that there's no actual rule that says the simplest explanation is always right. Sometimes the simple explanation is wrong, and sometimes, the improbable happens.
The chance of producing that specific sequence of numbers is absurdly small, but the chance of producing some21-digit sequence of numbers by banging my fist on the keyboard was pretty close to 100%. Similarly, the chance of any specific individual winning the lottery is vanishingly small, but the chance of somebody winning it is pretty darn good. While the odds of this happening to be the universe in which somebody I know wins a quarter of a million in the lottery are pretty low, the fact is, that's the universe we live in.
I have a deeper problem with the book, though, which is that I just don't find the question that compelling. As I've said before, this is partly a matter of my normal field of research-- I'm a low-energy physicist, and we're used to taking things for granted. Any time I do a calculation, I end up with a bunch of arbitrary parameters whose specific values I read out of a table. The mass of a proton is the mass of a proton, and for my purposes, it really doesn't matter why it's a tiny bit less than the mass of a neutron.
As a result, I'm not particularly bothered by the need to specify certain values. I'm certainly not bothered enough to embark on huge flights of meta-science, talking about multiverses and Life Principles and other untestable hypotheses. If you ask me what the constants of nature are, that's a well-formed question. I can do a measurement, and find a value. If you ask me why the constants of nature have the values they do, that's not a well-formed question. There's no measurement I can do to answer that, and as a result, I just don't find it that compelling. If it scratches your teleological itch, great, but as far as I'm concerned, it's vaguely interesting to debate in a casual bull-session manner, but not really worth the effort of writing a three hundred page book.
That said, if you're more enthusiatic about this sort of thing than I am, this book is well worth a read. Davies writes clearly and persuasively, and while his preferred explanation may have a slight whiff of theology (he did win a prize from the Templeton Foundation, which probably makes him suspect in some circles), it's undeniably a provocative idea (and it would probably be interesting to read this paired with Leonard Susskind's The Cosmic Landscape).
I just have a hard time treating it seriously as science.
- Log in to post comments
This is the sort of book that could only be written by a theorist (or a mathematician), because only a theorist can possibly be deluded enough to think that we understand physics well enough to make predictions about what would be possible (or impossible) in a universe with different physical laws than our own.
I second Chris G's remark! You said it. We know we don't understand the universe we live in now, yet we are supposed to know that changing some law or constant would have certain effects. We can play with models, but that is a far cry from saying we know the full ramifications of them.
I third it. I'm often bemused by the leap from, "Change this parameter and LIFE AS WE KNOW IT COULD NOT EXIST!" (which I can concede) to "Change this parameter and ALL CHANCE OF ANY DYNAMIC COMPLEXITY IS ELIMINATED!" which seems a stretch, since I can write some pretty damn simple rules on a cheap-assed PC that give shocking and non-obvious dynamic complexities.
What you note contradicts both designer (anti-anthropic universes) and, it seems, life principle universes (strong anthropic universes) should be a variant of the sharpshooter fallacy in probabilistic terms. Seems tautological ("absurd" or "unique") or weak anthropic ("multiverse") universes are 'just right'. :-)
Speaking of unsupported speculations, I don't get how fake universes can be considered so easily. Doesn't Bell tests results say that they must work with another physics (since no local hidden variables)? So now we get another consistent but quite different physical theory to explain too. Is that plausible? (A related problem in philosophy should be that an indistinguishable result when allowing things akin to faking is solipsism - another implausible outcome.) "Carried away" seems like a fair judgment.
I'm willing to accept Davies's claims regarding some of the parameter changes that are considered. Changing the neutron/ proton mass ratio so that isolated protons are unstable seems pretty unlikely to result in a universe that does much of anything interesting... As a general matter, though, I agree that most "fine tuning" anthropic arguments probably say more about the lack of imagination of the people making them than they do about the actual probability of life in circumstances other than our own.
Speaking of unsupported speculations, I don't get how fake universes can be considered so easily. Doesn't Bell tests results say that they must work with another physics (since no local hidden variables)?
Well, maybe we're being simualted on a quantum computer...
It's certainly possible to simulate quantum effects on a computer (via Monte Carlo methods if nothing else). The resources required are vastly greater than for a purely classical system, but it can be done.
The created 'fake' universes may have the same type of physics, of course. What I don't get is the simulated ones.
Chad, that is a thought. Though I don't see how the to make a quantum computer completely devoid of classical parts. And I'm not completely sure how to make a massive program compatible with "no local hidden variables". (I'm reminded of the déjà vu cat in "Matrix" now. :-)
I am profoundly uninterested in the question this book engages with, so a trivial comment:
Did anyone else try hitting their number pad with their fist? I got mostly the center column, because I have a small hand.
What if they're all equally likely (as in your 21 digit number example)?
We know essentially nothing about what's required to create or simulate an entire universe. How can we possibly say that a multiverse ought to contain beings who can? After all, the set of all integers is infinite. That doesn't mean the set ought to include at least one integer between 1 and 2.
We have no idea how much effort it would take in either case. How can we be so sure that the fake ones are easier?
Make enough unfounded assumptions, and you can conclude whatever you like.
We have no idea how much effort it would take in either case. How can we be so sure that the fake ones are easier?
I don't have a huge problem with this bit of the argument, particularly as regards simulations. I think you could sensibly argue that if the simulation can be run on a computer contained within a real universe, then it has less information content than the universe that contains it, and is thus "easier" to produce, in some sense.
As a general matter, though, I agree that there are a lot of assumptions creeping around through these arguments, just waiting to leap up and make an ass of u and mption, as the saying goes.
Did anyone else try hitting their number pad with their fist? I got mostly the center column, because I have a small hand.
My fist covers pretty much the whole keypad, but the keyboard in my office at work has some sort of anti-cat thing activated, so it doesn't really allow me to generate random sequences by hitting all the keys at once.
Perhaps. How much larger would a universe need to be, to contain a computer that could simulate our universe, not to mention the aliens that could create it? Are there any bounds on the largest possible universe within the postulated multi-verse? Even if there are an infinite number of universes, that doesn't automatically mean they can be arbitrarily large, does it?
I'm not saying it's an outrageous assumption. I am saying that we know so little here that I doubt we're in any position to judge which assumptions are sensible and which are not.
This is a great basis for a Greg Egan story, and probably a terrific way to mess with the heads of stoned college students, but I have a major problem with treating it as a serious argument
You took the words right out of my mouth.
The "principle of mediocrity" is one of those meta-scientific rules like "Occam's razor," and holds that there shouldn't be anything particularly exceptional about our situation.
That's what the "meta-scientific" Cosmological Principle says, as well, but there IS theoretical presidence for this that does not apply to the observed universe, called the Principle of Relativity.
The Earth is an unremarkable planet around an unremarkable star in an unremarkable galaxy, in an unremarkable universe.
"but"... this isn't what is observed. Ergo the anthropic cosmological principle, which makes testable predictions about the observed universe from the average of extreme runaway tendencies that are inherent to all of the anthropic coincidences.
The Goldilocks Enigma
It's not an exact parallel, of course, because I don't have the utter contempt for Davies's subject matter that Dawkins does for religion. But this is fundamentally a book about anthropic principle sorts of arguments, and I have a very hard time treating those as anything more than dorm-room bull session material. They're certainly not what I would call science in the normal sense.
Which really cracks me up since the universe is observed to be STRONLY anthropically constrained until somebody proves that the multiverse is real science with a valid proven ToE, or at the very least MAYBE a real, tested, proven theory of quantum gravity.
Unless Richard Dawkins is lying to us when he says that the universe "appears designed for life"...
OR... unless Lenny Susskind is lying to us when he says that "we will be hard-pressed to answer the IDists if the landscape fails"
Not that I buy the intelligent design misnomer, but...
It's very hard to write responsibly about a topic you just can't take seriously.
And they wonder why we have no ToE... *eyeroll*
Trying this link, again...
The Goldilocks Enigma
http://evolutionarydesign.blogspot.com/2006/09/goldilocks-enigma.html
I'm so tired of Paul Davies.
I smell a rat. Isn't it convenient that every book of his he's just alittle bit more sensative to the idea of a divinely created universe?
Either he's really that much more impressed every time he decides to write a book, or he's a phoney. Think about it. Davies acts like he doesn't believe in God but admits that some evidence from physics could be used to support a religious claim. Then he writes another books. Still says that he disagrees with the religious view but is not as critical to that view. Then the NEXT time around he's even less critical, maybe slightly accepting of it.
I don't buy it. I think he knows damn well what he's doing. 3 books later if he's converted to some form of theism no one should be surprised. I have more than a feeling that this is all planned out.