The Trouble with Physics

SpaceAdmiral writes "You've likely heard of Lee Smolin's book The Trouble with Physics. It has created a lot of controversy because it argues that string theory gets far too much attention and money, despite a complete lack of evidence. It accuses string theorists of groupthink. Smolin has dabbled in string theory from time to time but he's a proponent of the alternative loop quantum gravity. Although irrelevant to this book review, he has also suggested that it is possible that universes reproduce via black holes, making them prone to pressure similar to natural selection (universes that produce a lot of black holes are more successful spawners than those that don't). In The God Delusion, Richard Dawkins quotes Nobel-winner Murray Gell-Mann as once saying, "Smolin? Is he that young guy with those crazy ideas? He may not be wrong."" Read the rest of SpaceAdmiral's review. The Trouble with Physics author Lee Smolin pages 392pp publisher Houghton Mifflin Company rating 9 reviewer Fane Henderson ISBN 0618551050 summary The Rise of String Theory, the Fall of a Science, and What Comes Next

The Trouble with Physics is very unlike most pop-physics books not only in its criticism of string theory, but in its open adulation of Einstein and skepticism of the Copenhagen interpretation of quantum theory. Having said that, it does provide a very decent summary of 20th century physics (including string theory) for laypeople, not unlike more traditional pop-physics books (e.g. by Hawking and Greene).

The book's main criticisms of string theory are that it makes no testable predictions and that some things string theorists take for granted haven't been rigorously proven mathematically. Smolin is highly skeptical of many string theorists' reliance on the Anthropomorphic Principle.

The book becomes most interesting somewhere in the middle where he discuses truly controversial approaches to physics. This includes things like MOND, which, interestingly enough, Smolin is skeptical of.

In case you've forgotten your high-school physics, I'm going to use this paragraph to refresh your memory of special relativity to prepare you for the next couple paragraphs. The basic idea of special relativity is that the speed of light is constant. Pretend that I am shining a light at you while (A) standing still relative to you; (B) moving towards you at half the speed of light, and; (C) moving away from you at half the speed of light. In all three scenarios, I will accurately measure the light moving away from me at 3,000,000 km/s and you will accurately measure the light moving toward you at 3,000,000 km/s. To ensure this result, distances and times will have to be different for me than they are for you, except in case (A).

Now I'll quickly remind you of the Planck length: This is a theoretical limit on how small something can be. According to Smolin, all versions of quantum gravity seem to suggest the Planck length as a limit. But would observers moving relative to each other disagree about the Planck length?

I used to be a big fan of MOND (in a layperson sense) until Smolin introduced me to DSR (doubly special relativity) and DSR II. The basic idea is that it may be possible to modify the theories of relativity such that observers agree not only on a constant speed of light, but also on a constant Planck length. It's not unreasonable to guess that a modification of this sort could solve some of the same problems MOND does (e.g. explain astronomical observations without resorting to dark matter and dark energy). Furthermore, since DSR in its current incarnation predicts that more energetic photons are slightly faster than less energetic photons (only the speed of the least energetic photons is constant in DSR), it could also explain away, for example, inflation in the Big Bang model. (Immediately after the Big Bang, everything was hotter and more energetic, so the average speed of light would have been faster than it is now if DSR is correct.) Although I'm not qualified to judge the actual mathematics of such a theory, I find it very appealing for reasons of consilience.

I was slightly disappointed with the final chapters of Smolin's book since, despite an obvious effort to the contrary, it struck me as awfully bitter and reeked of sour grapes. Leaving physics in favor of sociology, he lambasted the current tenure and peer review systems (particularly in the United States) as favoring Master Craftspeople (like those scientists who developed the standard model of particle physics) over Seers (like Einstein, Bohr, and de Broglie) who look at the deep questions of physics that border on the philosophical rather than the latest technical problem. A few interesting things do emerge in these chapters. One such thing is that Smolin seems to have a soft spot for Paul Feyerabend as a philosopher of science (despite describing himself as a proud Popperazzo in an endnote). Another is that Smolin thinks a scientist who is hated by half his senior colleagues and loved by the other half is likely better than a scientist who is liked by all his senior colleagues. I strongly recommend this book.

You can purchase The Trouble with Physics from bn.com. Slashdot welcomes readers' book reviews -- to see your own review here, read the book review guidelines, then visit the submission page.

Comments

Anthropomorphic Principle?

[tylersoze]

Really? I didn't realize giving human characteristics to subatomic particles was a part of any current mainstream physics theory. :) I'm assume you mean the *anthropic* principle.

Accurate?

[Whalou]

In all three scenarios, I will accurately measure the light moving away from me at 3,000,000 km/s and you will accurately measure the light moving toward you at 3,000,000 km/s.

Not very accurate. It should be 300,000 km/s. Or 299,792.458 km/s to be precise.

I think you meant "Anthropic"

[olclops]

Smolin is highly skeptical of many string theorists' reliance on the Anthropomorphic Principle.

That's the Anthropic Principle: the idea that the constants we observe in this universe which are ostensibly crucial for the formation of life, are that way because if they were any other way we wouldn't be here to observe them.

actually

[giampy]

A quick search on arxiv.org will show you that he is indeed very active, since he is still publishing very technical papers.
Not only that, Lee Smolin seems one of the very very few physicists who understands BOTH string theory AND other approaches (that is _the_ other approach, loop quantum gravity).

In any case, it seems that many predictions of loop quantum gravity will be actually tested within the next couple of years trough the GLAST satellite, so, we will get news relatively soon ...

Re:Two sides

[gardyloo]

Smolin himself has not been research active in a long time. And it is unlikely that he understands enough mathematics to judge string theory - like most people.


      Not sure what you mean by "research active". His contributions to xarchiv (many published in hack journals like Physical Review and The Journal of Quantum Gravity) are prolific as recently as 2006 and 2004 (noticeable lack of submissions in 2005). And I would NOT call him mathematically illiterate, even in an "esoteric" field like string theory. (Yes, I am a physicist.)

Vilenkin says...

[Cally]

Vilenkin has published an interesting paper which suggests a problem with Smolin's "natural selection of life-friendly universes via black holes" theory; OTOH Smolin strikes back! Ahhh, I love it when cosmologists attack ;)

String theory, pros and cons

[Ambitwistor]

Below I'd like to repost something I once wrote trying to explain why string theorists think string theory is an important approach, to counter the inevitable "it's not science" claims I see on string-related threads.

(I would like to note first that Smolin himself has written string theory papers, and historically has advocated combining string theory with loop quantum gravity, so even he doesn't think string theory is nonsense — he just would like to see it mesh with his own theories and doesn't like the attention it gets relative to them.)

Anyway, my two cents on string theory and its justification and testability:

First, string theory could certainly be tested if we could probe the Planck scale. We will never be able to build an accelerator to do
that directly. There is some chance we might eventually do it indirectly by measuring fluctuations in the cosmic gravitational wave
background. In addition, string theory encompasses many scenarios in which the string scale could be probed at much lower energies, but nobody is very confident that those scenarios are likely to be correct.

That being said, there is a serious possibility that string theory might not be testable in practice, at least in the foreseeable future. I don't believe that puts string theory totally outside the realm of science altogether. String theory does at least make predictions, even if we can't test them. But that is a weak argument. More strongly, string theory is motivated by reason of consistency with known physics. Gravity has to be reconciled with quantum theory somehow. There are strong reasons to believe that string theory overcomes obstacles to quantizing gravity in a unique way that all other approaches can't duplicate, although this can't be proven. That is one of the main reasons why string theory is taken so seriously despite its experimental shortcomings (which are not surpassed by its alternatives, either).

Here are a couple of arguments in favor of string theory put forth by string theorists which I have begun to agree with:

In particle physics, it has been possible to write down theories of the non-gravitational forces while being ignorant of high energy
Planck scale physics. This is essentially due to the Applequist-Carrazone "decoupling" theorem, which uses renormalization
group arguments to show that low-energy physics can be made independent of high energy physics, because at sufficiently low
energies you can't excite the higher-energy modes; therefore, their contribution is irrelevant.

This decoupling breaks down for gravity. Because gravity is a universal interaction, it couples to everything (because everything
has mass-energy); the low energy effects of quantum gravity are never independent of high-energy physics. So you can't write down a theory of quantum gravity unless you purport to know everything about particle physics up to arbitrarily high energies — which of course you can't possibly say, unless you can do experiments at the Planck scale.

This is a criticism that string theorists level against loop quantum gravity. LQG is usually attempted ignoring all realistic particle
physics, and even if that approach succeeded, you'd have to write down a different LQG theory to take into account real particles, which might work completely differently than a vacuum LQG theory. LQGers respond by saying that they want to start by just proving it's possible to quantize *any* kind of gravity using this approach, and then worry about "realistic gravity".

String theory, on the other hand, evades the whole problem. It has a very unique mathematical structure which provides "mysterious" exact cancellations at all orders, rendering low energy physics decoupled from high energy physics despite the universal coupling of gravity.
Thus, it can make predictions about high energy physics even without our being able to make measurements at that scale. No other approach to quantum gravity has shown any signs of being abl

Re:Science is prediction, not explaination

[bcrowell]

The serious problem with string theory is that it doesn't yield falsifiable predictions.
One of Smolin's arguments is that string theory has yielded falsifiable predictions, at those predictions were later disproved. For instance, one of the early predictions of string theory was that the cosmological constant had to be less than or equal to zero; there seemed to be no reasonable way to make the theory produce a positive value. Then it turned out that the cosmological constant was nonzero and positive. The string theorists came up with mechanisms to allow the positive value, but Smolin argues that they're unnatural, and it's not clear that they really work in general. Another example is that string theory appears to require exact supersymmetry, and nobody knows how to relax that requirement to agree with reality. Another is that (at least according to Smolin) there are fundamental problems in string theory that prevent the assumed statically flat spacetime background from being made dynamical.

Re:Maybe quantum theory is wrong too...

[2short]

Heisenberg's Uncertainty Principle says it's impossible to know both the position and velocity of a particle; and particularly that increasing the precision whith which you know one will decrease the precision with which you know the other. It is related to, but not quite the same as, the assertion of quantum mechanics that at the smallest scales, reality is not deterministic.

Einstein thought this risiculous, as expressed in the famous quote, but most physicists now beleive he was wrong.

Either position is clearly saying quite a lot about Fate, God and Free-will!

Re:Maybe quantum theory is wrong too...

[Deviant Q]

The Bell Inequality rules out local counterfactually definite hidden variable theories. Nonlocal theories in particular are quite doable, and David Bohm worked on those for a long time before he died. I've been reading some of his books (but not scientific papers), and they seem fairly reasonable; however, I think they fall in to the "not mainstream enough to take seriously" category.

I don't know what 't Hooft's theory is though.

On another note, I've written a paper on why the Bell Inequality does not falsify local counterfactually definite hidden variable theories, but I'm 95 % sure my conclusions must be based on some kind of misunderstanding because I can't have been the first person to see this.

MOND

[soliptic]

Was I the only one to think it wouldn't have hurt to have spelled out what MOND is the first time it is used?

Before you say, "Well, anyone who knows ANYTHING about physics knows that, you retard, this book is not for you..." - well, I did think this was supposed to be a layperson's book. So, I clicked to read this review despite having an effectively non-existent knowledge of physics.

Well, anyway, here's your answer, at least according to Wikipedia (obviously, not being my field, I can't vouch for its accuracy):

In physics, Modified Newtonian Dynamics (MOND) is a theory that explains the galaxy rotation problem without assuming the existence of dark matter. Currently, the most widely accepted galactic rotation theory assumes that a halo of dark matter surrounds each galaxy, causing all the stars in the galaxy disc to orbit with the same velocity. When this uniform velocity was first observed it was unexpected because the Newtonian theory of gravity predicted that objects that are farther out will have lower velocities. For example, planets in our Solar System orbit with velocities that decrease as their distance from the Sun increases.

MOND was proposed by Mordehai Milgrom in 1981 to model the observed uniform velocity data without the dark matter assumption. His key insight was that Newton's Second Law ( F = ma ) for gravitational force has only been verified when gravitational acceleration is large.

Source: http://en.wikipedia.org/wiki/MOND

Re:How long is a piece of string?

[The_Wilschon]

What do you mean real? Call me back when you see a real electron. Or better yet, when you see a real quark. Superstrings are just as real as either of these. In fact, unless I misunderstand, electrons and quarks would merely be a special case of strings, that is, strings carrying particular vibrational modes.

Ernst Mach raised precisely this objection against atomic theory. He said that atoms were not real because we could not, and would never be able to, see them. They were just a convenient mathematical model which happened to make reasonable predictions, but they were not actually real. Well, as it turns out, theories which utilize these "unobservable, unreal, mathematical constructs" are often very successful, and, where they have been successful, we have later found ways to observe precisely the objects described.

So, I would say that strings, if the theory turns out to produce useful, accurate, precise results, are just as real as photons, atoms, rocks, and stars.

That's not to say I like string theory. I hope string theory doesn't win. I think that it would put us in actually a worse position than the Standard Model has us in right now. The standard model has umpteen different parameters which must be fine tuned by experiment. This is generally regarded as a serious shortcoming, as the values of those parameters ought to be predicted by a good theory. String theory is "parameterless". This is a wonderful thing, until you consider that those extra spacial dimensions can be wrapped up around each other in an enormous number of ways, and each way produces a completely different set of particles and natural laws. So now, rather than measuring a few values, we must instead investigate every possible way of wrapping up the extra dimensions, until we find one which matches our own universe. So, in short, the topology of space is the parameters of string theory, and a much nastier parameter space than for the standard model it is.

Re:Why string theory is stupid

[Goaway]

If I explained to every person mouthing off on Slashdot why they are wrong, I'd never have time to leave the house. Sometimes, you have to keep it short.

In short: He's confusing string theory and old-fashioned quantum mechanics, while understanding neither. Schrödinger's cat predates string theory several decades, and it is actually meant as a criticism of early quantum theories.

http://en.wikipedia.org/wiki/Schr%C3%B6dinger's_ca t
http://en.wikipedia.org/wiki/Quantum_mechanics
http://en.wikipedia.org/wiki/String_theory

IBM

[ObsessiveMathsFreak]

Ernst Mach raised precisely this objection against atomic theory. He said that atoms were not real because we could not, and would never be able to, see them.

IBM disagrees.