Slashdot Mirror
String Theory a Disaster for Physics?
BlueCup writes "Mathematician Peter Woit of Columbia University describes string theory in his book Not Even Wrong,. He calls the theory 'a disaster for physics.' Which would have been a fringe opinion a few years ago, but now, after years of string theory books reaching the best sellers list, he has company."
String Theory attempts an actual prediction and then gets it correct.
Till then, it's a bunch of fancy gobbedly gook as far as I'm concerned.
*Sigh* One of the biggest problems of string theory is it is damn near unprovable. It could be true. It might not be. But if the facts don't fit, you just modify the theory again. And yes, this is oversimplification, but not by much.
Makes me wonder if we are near the edge of what humans can know. Growing up, I took it on fiath that it was just a matter of time before we knew it all. Now, I am not so sure. Perhaps our monkey brains simply can't conceive of the true nature of reality.
See my journal for slashdot ID's by year. Mine created in 2005. http://slashdot.org/journal/289875/slashdot-ids-by-year
I've never felt very comfortable with string theory. Not that it threatens some deep-held belief (I have few of those), but that it seems mostly like conjecture, trying to shoehorn increasingly complex theories to fit some phenomena that is probably explainable in a simpler manner which we just yet haven't found. Of course, physics often doesn't adhere to common sense.
"I would say that 99 per cent of what my father has written about his own life is false." - L. Ron Hubbard Jr.
Well, yes, because that's not how science works. Theories have to be put forward which make predictions which can be meaningfully tested. If there is no way to actually test it, then it is, in effect, impossible to develop - by definition it cannot be wrong, and therefore is effectively complete, and science is 'finished', more or less. If such an idea ends up being the dominant trend, then yes, it would be something of a disaster.
I suppose we should stop looking for what started the universe, since we can't disprove the existance of God or anything. What a load of BS.
Would you class the statement "Everything happens because God says so" as scientific? I would hope not - such a statement is inherently impossible to scrutinise or critique, as "God did it" is, in this theory, a perfectly valid response. Science does not advance by pursuing ideas of that sort, but rather putting forward testable theories, and working out ways to stress them and see if their predictions hold, and refining them as a result.
That some ideas cannot be disproven is not a problem for science - instead we content ourselves with studying those that can be.
It's not a theory yet until it makes a testable prediction. The difference is it has the potential to be one whereas intelligent design does not.
Maybe Intelligent Design can get some respect if other hard-to-test and long-shot hypotheses are allowed to be called "science". Just because backers tend to be religious does not by itself make it wrong. If a Darwin or SETI cult formed, would evolution or SETI's hypothesis grow less likely? Human bias does not change the truth values of the universe. If biased people want to hunt down evidence for long-shot hypotheses, so be it.
Table-ized A.I.
If we were to accept the existance of God it would also be a disaster for modern science. It's not science. It allows for every answer and is not testible.
How does this follow? "God did it" and "how did God do it?" are two different things. I find it really interesting that the former excludes the latter. Some historians of science have argued that it was because of the idea of a rational God that the idea arose that nature was ordered and could be fathomed, particularly through observation and testing. Wasn't it Kepler who studied the heavens in order to "think God's thoughts after Him"? So the idea that God and science are incompatible is, to borrow a phrase, "not even wrong". It's simply a by product of the anti-intellectual, anti-historical, "don't offend me" nonsense that passes for thinking these days.
Science flourished within a theistic worldview in Europe and elsewhere, so I don't see how you can support the idea that God is a disaster when it comes to science. But maybe by "modern" you mean "completely materialistic". And of course that's true. It's the age old dilemma: which came first? The naturalist says, "In the beginning were the particles...". The Christian says, "In the beginning was the Word..." And never the 'twain shall meet.
Oh, what the hell, let's go for troll moderation. I'll go so far as to argue that denying the existence of God is actually hindering science. Why? Because atheism a priori denies the existence of an intelligence far greater than man's and therefore denies the possibility of design in nature. Therefore, the notion of using science to identify and measure design isn't considered (except, perhaps, with SETI. But aliens are metaphysically less scary than God. There's no evidence for them yet the search goes on.) Nor are information theory, computer science, and complexity theory being applied to natural, especially living, systems. Why not? Who knows what we'll find?
A real concern is that the "landscape" (the fact the string theory is really a collection of theories that could have something like 10^500 (yes, that's a googol to the fifth power) possible vacua as solutions) renders string theory nearly unfalsifiable. It's not that they can't predict anything. Indeed, they've predicted everything. If the LHC at CERN started up tomorrow and found a Higgs boson with a mass of 220 GeV, and some kind of light supersymmetric partner at 260 GeV, they could claim that's consistent with string theory. Heck, if the Tevatron folks at Fermilab found a fourth family of leptons next week, the string community could claim to understand that, too. I would love to see just one example of something that could credibly be found at the LHC that string theory can't explain. Just one.
ST is an attempt to unify QFT with GR. In that setting QFT and GR are a consequence of ST. However string theorists have discovered something called the landscape. In which not only is QFT and GR possible but so are 10^500 other types of universes and the theories that describe them.
This is a huge problem. Of course there is the possibility that finding the ultimate theory of everything is impossible and this would be the physics dual to Godel's incompleteness result. Which I'm sure sure scares the shit out of many physicists.
The best education consists in immunizing people against systematic attempts at education. - Paul Feyerabend
Time to confess, boy.
Infuriate left and right
Science did not flourish until the Deists decided that God was an honest pinball player. He built the machine, and he flipped the levers, but he didn't tilt the box or otherwise "miraculously" influence the flight of the ball. He left that all up to the initial starting conditions (and how far he pulled back the shot lever).
OK, that's an oversimplification of what they were claiming...but it's got the essence. And with those restricitions on "God did it!", scientific research became feasible. Without it... sorry. Can't do it.
I think we've pushed this "anyone can grow up to be president" thing too far.
They're two totally different things. If I tell you all cats have tails you can bring me as many cats with tails as you like and it will not prove that all cats have tails. You can bring me a cat without a tail and it will prove that not all cats have tails, but it still won't prove that all cats have tails. Nothing can prove that all cats have tails. That's science buddy.
How we know is more important than what we know.
The real issue here is that particle physicists have received no "surprises" in many years -- perhaps the only genuinely unexpected recent data point being the non-zero value of the cosmological constant.
Excuse me? You should try keeping up with experiment if you're going to make broad statements like this. Minos, up here at fermilab, recently discovered that neutrinos do in fact have mass. This was suspected a year or few ago, which was why Minos was built, but is nonetheless quite surprising. It is surprising because it is really the first definitive measurement which is nearly unquestionably outside the standard model. (I don't need to tell you this, I suppose, but others will read this too: The standard model assumes explicitly that neutrinos have no mass at all.)
Anyway, the problem that most experimentalists, such as myself, see with String Theory is that in some ways it is a step backwards from the standard model. It is purported to be "parameterless", which contrasts with the plethora of unconstrained parameters that the standard model contains. However, this is really only a bit of sleight of hand. Instead of numerical parameters, which are (relatively) easy to measure, and continuous, we now have the topology of space, which is discrete (no smooth change from one topology to the next) and quite difficult to measure, and embodies immensely more variation than the parameters of the standard model.
SIGSEGV caught, terminating
wait... not that kind of sig.
It very much is science, it's just not a proper theory. Perhaps "not yet" or "not ever" a proper theory, no one can say which is correct at the moment.
Science, on the other hand, does not require one wait for the finished product. Working on string theory is working on science. It's just not complete, nor even all that useful currently. It's still in the early stages--a stage that is rarely so long and drawn out as it is in this case.
For example, when devising special relativity, Einstein's theory was, at some point, still in the state string theory is in currently--that is, significantly conceptual, with a lot of math and refining yet to be done, and early on was entirely untestable making no real predictions. He was still engaged in science during that stage. That doesn't mean that special relativity was useful yet, nor do I mean to imply that string theory is correct or will bear fruit, just that even at this early stage it is legitimate to call it science.
"Untestable! Unfalsifiable!" This is a common refrain from string theory critics, and even from many string theory fans - and it really bugs me. I'm a philosopher of science, and I say: forget falsifiability!
Well, don't completely forget falsifiability - but don't let it be the whole story. Falsifiability was already outdated philosophy of science 50 years ago. Its main problem is what's sometimes called the "Quine-Duhem Thesis" - roughly speaking, any treasured theory can be made to fit any evidence, as long as you're willing to adjust enough auxiliary hypotheses. Here's an ordinary example: when your high school science lab experiments didn't fit predictions, your results didn't get published in Science for falsifying the theory at hand. Instead, quite reasonably, you drew the conclusion that something wasn't quite right with the instruments, etc., and you kept the original theory. The tricky question is to figure out when it's reasonable to excuse recalcitrant data, and when you're unreasonably trying to rescue a theory that's just wrong. Intelligent design advocates can lay out all sorts of falsification criteria, and then make similar excuses should unhappy data come their way. So does that make ID a science? (If on the other hand you insist that only actual falsification makes something a science, then only theories we no longer believe can count as scientific!)
It's too easy in all sciences - not just string theory - to make theories "supported" by the data. Given this problem, the name of the science game is to find the simplest explanation that fits the data. It's very hard to say exactly what counts as a simpler theory, but some theories are clearly less simple. Compare the hypothesis that "the butler did it" to the hypothesis that "unknown sneaky aliens planted all that evidence to make it look like the butler did it." Both hypotheses fit the evidence equally well, but the latter is clearly less simple, and we normally never even consider it for a moment.
String theory explains all the data, from quantum physics to relativity, with a simplicity that's hard to beat. (Its elegance is so good, we're apparently willing to posit 11 dimensions for it!) That's what makes it a legitimate scientific theory. Of course it would be great to have more relevant data, to see if string theory can accommodate them simply too. But just because we can't get such data (now, or maybe ever) doesn't spoil the current scientific status of the theory.
Humanity has been thinking it has got an almost complete understanding of the rules of physics for a few centuries now.
// MD_Update(&m,buf,j);
Um sorry, but even amateur scientists know that a scientific LAW is merely a THEORY that has not been successfully falsified for such a long period of time that it is assumed that a single, even several, experiment(s) controverting it is a botched experiment(s).
But if enough people can prove it wrong, it goes. Even Sir Issac Newton's most basic equations are discarded in the scientific canon, replaced by Einstein's more complete set. Scientific LAW tends to get amended rather than scrapped, and that's what seperates it from theory.
The SCIENTIFIC METHOD, what science is *supposed* to be based upon, relies upon repeatable results based upon a given set of assumptions: the theory. If you can't test it, what's the point?
The writer of the article doesn't claim that string theorists are WRONG, he claims that they are WASTING THEIR TIME. Why? Because he feels they are more interested in "elegance" and "beauty" than they are in finding a relevant way to describe the universe that is useful and TESTABLE.
Whether that is right is a matter of some debate, but get the premise right.
--
Torodung
If on the other hand you insist that only actual falsification makes something a science, then only theories we no longer believe can count as scientific!
That's actually a reasonable and sound position. It's more or less impossible to prove something correct in science - we've discovered many times that if a theory looks right on one level, when you go deeper it's just a good approximation. So you can never really be sure whether you've found the right answer, or just something close to it. And from our experience of science the odds are in favour of the 'good approximation' side.
On the other hand, when you've proven something wrong, you're pretty damned sure that it's wrong. It's then a scientific fact. You can't do that for something that appears right - it's more like a scientific guess.
It's important to keep this in mind - remember that something you believe to be correct is almost certainly not correct in every particlar. And follow up on those "Hmm, that's odd" moments, because those are how progress is made.
I am a theoretical physicist. I am educated in Supersymmetry, Riemannian geometry, and the original Kaluza-Klein theory. I am educated in field theory in curved space time, but not Supergravity, nor do I care to be.
I have only one thing to say about all of this. Everytime I sit in a talk on strings or branes all I can think is one thing.
Extra dimensions are the epicycles of Modern Physics
That's all I have to say. If you understand this, it is profound.
American's shun science careers because they are so punishingly expensive. Conside the lifecycle of someone who want's to go into String Theory:
4 years undergrad ($40-$120k in cost)
5-7 years grad school (making $15k per year)
2-6 years postdoc (making $40k per year)
7 years pre-tenure (making $60-80k per year)
tenure (making $80-$100k per year)
Oh, and if you fail at any point along the line, you have no career. Since you are looking at 18-24 years to get to tenure, that's a HUGE
investmet to make. You are basically looking at not knowing whether you have a career or not until you are 36-42 years old.
Compare this with the career track of an equally bright student going into CS, and getting a job in tech.
4 years undergrad ($40k-$120k in cost)
starting job ($60k-80k)
5 years experience ($80k-$120k)
move into management ($100k-$150k)
etc... notice how the CS grad going into IT hits the $60-$80k range 7-13 *years* before the scientist?
Plus, while there is less risk of being laid off as a tenured professor, the risk of having your career evaporate as an IT person (please note, IT person who could have hacked being an academic scientist) is MUCH lower. Sure, you may loose your job, but there are plenty of other jobs.
Few American's go into science because the economics of science is so bad. I don't know how to fix that, but the cause is pretty clear.
Oh... by the way, I am an American who was on the academic track in String Theory and got off the merry-go-round. Everytime I talk to my friends who stuck it out I am overjoyed I left. My friends are at the mid-potsdoc stage right now. They are trying to scrape by living on $40k a year in ultra expensive locals like Boston, and live in terror that the only jobs they will be available will be in middle of no-where universities in unpleasant places. By way of contrast there are tech jobs to be had in a variety of nice locals to meet most peoples tastes. It's particularly hard on the women, who are starting to hear their biological clocks tick VERY loudly, and who are still years away from being settled in enough to take a break to have children. It's also very hard for both genders to find a long term mate, as they face the aforementioned prospect of having to move a lot to unpleasant places (like Norma, OK, middle of nowhere PA, middle of nowhere plains states, etc) if they want to continue their careers. And let's not even start talking about the two body problem (two romantically entwined academics).
So basically, if you want more Americans to go into science, make it suck less.
The analogy is thus.
Q "Look, the orbits aren't exactly circular"
A "Try adding more circles"
Q "..."
A "Try adding more dimensions"
I don't mean to say that the true answer is simpler in a sense. I honestly have no idea.
I came to this conclusion after sitting through dozens of talks and realizing that so many physicists were using extra dimensions like a tool in their toolbox (like renormalization or something), but that tool has never actually fixed anything yet. It's their hammer and everything looks like a nail. Only no nails are proven to exist.