String theory and Lee Smolin.

[ollypenrice]

I'm reading Lee Smolin's The Trouble with Physics and finding that, as ever, I really do admire his way of thinking. As an amateur I can hardly comment sensibly on string theory but as a human being I can hardly believe in it either. It simply reeks of epicycles to me. Complex mathematical attempts to make strings into something that are not strings, just as circles were tortured by Ptolemy into replicating things which were not circles.

Here's Lee Smolin dealing graciously and with prodigious compusure with all that is fatuous about television. For me Lee Smolin is a real thinker.

Olly

[michael.h.f.wilkinson]

String theory has morphed into brane theory, I gather. One problem with many of these theories is that it is hard to make falsifiable predictions. That is a SERIOUS problem in science

[JamesF]

I'm some way through the same book.  As a layman he does seem to make some convincing arguments.  Unfortunately he does also sometimes dismiss things out of hand without adequate explanation which doesn't leave me entirely satisfied.

He's definitely worth reading or listening to though.  Very thought-provoking.

James

[Knight of Clear Skies]

Yes, theory running ahead of experimentation is a problem. Look at the Higgs boson, theorized in the 1960s and detected in 2012. I do wonder how close we are to the horizon of discoverable physics, the LHC has just had a considerable upgrade but it's only a doubling of power. What is the next really big advance requires 100 times more energy, or a 1,000?

Going back to the Greeks they did consider heliocentrism and that the stars might be too distant to show measurable parallax, but without any means of testing this the more 'natural' theory took hold.

[JamesF]

I must have a cast about to see if he has anything to say about the Higgs boson discovery, actually.  I think it's in this book, and in the video above, he says that where theories are shown to be correct then the experimental evidence tends to come very soon (a matter of years) after the theory.  This appears to be one occasion where it didn't.

James

[Ouroboros]

Atoms are another example of a hypothetical explanation. One that was not verified for 2000 years. I am not sure we should be too worried that theory runs ahead of experimental verification. In fact it often does, and is one way in which science makes progress. Nevertheless I do think Lee Smolin makes some good points in his book. I share his concern that string theory became to some extent the only theoretical game in town, attracting much of the funding for theoretical physics. Also that some physicists were ostracised and their careers adversely affected for choosing not to work in that area.

[michael.h.f.wilkinson]

I am not worried that theory runs ahead of experiment, I am worried if theory makes no falsifiable prediction. Now physicists are certainly aware of it, and many are trying to come up with such predictions, but there are some voices that claim falsifiability isn't necessary. That flies in the face of the scientific method, I would say.

[Tiki]

Thanks for posting the video Ollie, I felt that both the interviewer and interviewee were excellent. Twenty minutes well spent for me.

I am not worried that theory runs ahead of experiment, I am worried if theory makes no falsifiable prediction.

cf. 6 min, 11.20 and 14.50 in the video.

It is interesting to note that Lee Smolin's other main beef with string theory is that is simply not beautiful in the way that the relativity and quantum mechanics are. (Much follows from the Principle of Relativity and the Uncertainty Principle).

[ollypenrice]

I'm glad the thread has attracted comment.

I also agree with Michael that alarm bells should ring when mathematical physicists feel that the making of falsifiable predictions is no longer necessary. This is not science, it's maths.

The observations which follow are those of in interested outsider and I post them simply as such, without pretension to any greater ambition.

My hunch is that the problem's origin is to be found in quantum theory which is, itself, possibly formulated in a way new to science. That's to say, it is entirely non-descriptive. You cannot draw the inside of an atom. You cannot say whether a particle is a particle or a wave. The maths races ahead because it can make predictions and it is satisfied with 'wave-like' and 'particle-like' but at some point I feel that this kind of conceptual fudge must have its come-uppance and that come-uppance may be string theory which may prove to be no more than an orgy of epicycles. But, on the other hand, perhaps we've been there before. Newton produced his gravitiational equations but told Halley that he did not want the idea of forces acting instantaneously at a distance to be attributed to him. So Newton gave mathematical predictions without offering a conceptual mechanism to go with them, and this lasted till Einstein did away with the idea of forces acting at a distance and replaced them with geometric distortions of spacetime. (Round of applause for humble Michael Farraday here, maybe, for introduing the concept of fields?)

I think we'd do better if we had a conceptual quantum theory and that might well require a new theory of time, or spacetime.

Of course all this may really boil down to the possibility that a point arrives at which we no longer have the right brains for the task. (I know I don't!)

Olly

[michael.h.f.wilkinson]

This is not science, it's maths.

Which is also a science

I would not call it maths because mathematicians feel no need to develop something that is supposed to describe the world around us. Pure mathematicians positively scorn the idea. The only thing that counts is to make a self-consistent system of axioms and theorems (or if the coffee was weak, lemmas ). Here the idea is to develop mathematical models that should describe nature at some level, but do not allow us to check whether the model is valid (for a given value of "valid" )

[ollypenrice]

I don't think maths is science. Maths is several thousand years older than science. It was, possibly, Gallileo who identified that nature was written in the language of mathematics but that is not the same as saying that nature is mathematics. (I know you're not saying this.) Science is partially - I insist partially - written in the language of mathematics but science has also described nature very well indeed in the language of words - until the early twentieth century and quantum theory. I wonder if the divergence of verbal science from mathematical science will, in the future, be identified as a sea-change? (And maybe as a wrong turning?  ) Words have served us well. We are using them now...

Olly

[michael.h.f.wilkinson]

I work for the Faculty of mathematics and Natural Sciences, I think that makes a clear distinction. Mathematics is considered a science, but not a natural science, because it does not attempt to describe nature. In the same vein, computing science is very much a science, but also not a natural one, even though ALL sciences seem to need our tools (just like mathematics )

[JamesF]

I don't think I have a major problem with (some of, at least) Quantum Theory.  Bizarre as it may be, it can be used to usefully explain stuff in the "real" world and make falsifiable predictions about how things will behave.  Some of its strangeness surely does come though from the fact that it can be very hard to do anything with except on a macro level.

If a branch of science can't make predictions though what value does it have?  One may be able to describe some aspect of the cosmos in a mathematical way, but unless it's also possible to use that same maths to predict how something else will behave, is it any more useful than describing some aspect of the cosmos as a manifestation of hyper-intelligent shades of the colour blue?  Until it really can make potentially falsifiable predictions, isn't it really just a hypothesis?

James

[michael.h.f.wilkinson]

BTW I agree with you that words are important, though we may differ about the reason for that importance. Words are important because they allow us to tell ourselves stories about the world, and we are very much Pan narrans - the story-telling chimp, much more than Homo sapiens, the wise man (from Terry Pratchett). Mathematics is a beautiful, concise language. It is much easier to be very precise in it than English. Therefore scientists turned to mathematics, because they were too lazy to writ down all the required words, when a handful of symbols do a much better job.

[michael.h.f.wilkinson]

I don't think I have a major problem with (some of, at least) Quantum Theory.  Bizarre as it may be, it can be used to usefully explain stuff in the "real" world and make falsifiable predictions about how things will behave.  Some of its strangeness surely does come though from the fact that it can be very hard to do anything with except on a macro level.

If a branch of science can't make predictions though what value does it have?  One may be able to describe some aspect of the cosmos in a mathematical way, but unless it's also possible to use that same maths to predict how something else will behave, is it any more useful than describing some aspect of the cosmos as a manifestation of hyper-intelligent shades of the colour blue?  Until it really can make potentially falsifiable predictions, isn't it really just a hypothesis?

James

QM is very powerful, even beautiful. We wouldn't have transistors or computers without it. I took three QM courses at university, and thoroughly enjoyed them (mind-expanding without any illegal substances ).

By contrast, I would argue that any description of reality that cannot make falsifiable predictions isn't even a hypothesis. It is essentially a "Just so story" (Wolfgang Pauli might have called such descriptions "nicht einmal falsch" (not even wrong))

[JamesF]

I work for the Faculty of mathematics and Natural Sciences, I think that makes a clear distinction. Mathematics is considered a science, but not a natural science, because it does not attempt to describe nature. In the same vein, computing science is very much a science, but also not a natural one, even though ALL sciences seem to need our tools (just like mathematics )

I do wonder if computer science isn't almost entirely a subset of branches of maths.  Formal Language Theory, Lambda Calculus, Graph Theory,  Information Theory, formally provable programming languages; they're all maths really.  I guess it depends where you draw the line between science and engineering.

Perhaps the same is true of maths and science.  I've always thought of maths as being a science, but perhaps a little apart from those that attempt to describe and explain the real world.  The same rigour applies to its functioning, but in maths the experimentation and observation part of the scientific method involves white boards, marker pens and large quantities of coffee...

James

[michael.h.f.wilkinson]

I do wonder if computer science isn't almost entirely a subset of branches of maths.  Formal Language Theory, Lambda Calculus, Graph Theory,  Information Theory, formally provable programming languages; they're all maths really.  I guess it depends where you draw the line between science and engineering.

Perhaps the same is true of maths and science.  I've always thought of maths as being a science, but perhaps a little apart from those that attempt to describe and explain the real world.  The same rigour applies to its functioning, but in maths the experimentation and observation part of the scientific method involves white boards, marker pens and large quantities of coffee...

James

It did certainly start as a branch of mathematics, but fields like signal and image processing, computer vision, pattern recognition, artificial intelligence, etc have far too much to do with the messy real world out there to be comfortable to mathematicians

[ollypenrice]

I didn't know Pauli's phrase, 'Not even wrong,' but it's beautiful. I'm going to find it very useful!!

I can't see why we use the term 'string theory' at all. Why isn't it called 'String hypothesis?'

I remain convinced that, until we can have a conceptual articulation of quantum theory, it will remain unfinished. Didn't Eddington say that Waves and Particles might just as well be called Slithy Toves? And Bohr said, 'I feel somehow suspended in language.'  Although I doubt that I'll live to read it I do think that a new insight will one day render the quantum world 'envisageable.' But it will be a new insight, not a further refinement of what we have.

Olly

[JB80]

I have mentioned before in a thread past my somewhat cynical feelings on the subject of falsifiability(maybe not so much the subject but more so the argument surrounding it), especially in the realms of this particular part of academia and every time I look further into it my view doesn't change.

It's always the same protagonists, some of whom have been mentioned in this thread or are maybe the subject of  and likewise the same guard roll out to defend themselves and it really is a bit of a cat fight. Truth of the matter is they both have valid points but you have to look into why some people have gotten their noses so out of joint about this  and I can only postulate that it is a matter of funding and moreover livelihoods and it's possibly a natural defense mechanism when you feel threatened. After all we are talking about some brilliant minds here who are quite clearly distracted by 'String Theory' specifically with other equally non-falsifiable(bleurgh) theories mentioned as an after thought, or it at least reads like that anyway.

And well to be fair String Theory has had a massive impact over the last 20-30 years and certainly has almost dominated the theoretical landscape and the funding. I suspect if they manage to kill this off then they'll tackle the inflationists next. 

It's probably not fair to call it sour grapes especially as there is lots of merit to being able to falsify a theory or not but it is hard to find an objective view on the matter. This blog somewhat although not entirely shares my views on the matter http://www.scottaaronson.com/blog/?p=1653 as well as this article http://blogs.scientificamerican.com/the-curious-wavefunction/falsification-and-its-discontents/ both were in response to a Sean Carroll piece.

I would also like to think, perhaps foolishly that the vast majority of theorists can realise the limitations of falsification, lets face it I think you would be hard pressed to find a cosmologist or theorist who thinks that the observable universe is all there is and frankly if you subscribe to certain mainstream theories you have already given in to assumption and manipulation to get the desired results and it's probably why most don't throw stones at others.

[Tiki]

I don't think maths is science....

Olly

For me at least, maths is the science of all possible patterns.

[ollypenrice]

For me at least, maths is the science of all possible patterns.

OK, but for me science is the investigation of nature as it's observed and reflected upon. This doesn't mean we have to brawl!

lly

[Earl]

im not a fan of String Theory, (i have no formal education but) it come accross to me like this.

Here is an idea about stuff, if the stuff does this it could do lots of other stuff but if it does not we need to make up some more stuff and see if that stuff works.

I do appreciate pushing boundries etc, but i feel its making up things for the sake of making up things rather than actually doing the dirty working out, is it more Art than Science?

Maths is nothing but a man made concept, planets dont go orbiting knowing a load of Mathematical rules that have to be followed, its our interpritatrion of what is going on... and it the math dont fit we make some new math.... humm

[nameunknown]

Is maths a science? - ohhhh hard question! I'm slowly working through Suskind's book on the absolute basics of quantum physics - which complements his video series on UTube (from Stamford). Physics is modeled by a lot of very clever maths, but is it physics - or just a description of "theory" in another language?

P

[Macavity]

A quite reasonable (known) critique. An argument that surfaces quite regularly.

But, to me, the *eminent*, Smolin, still sounds like a guy "chucking toilet rolls

from the touchline"... At an easy (v.broad!) target? Also none too clear where

his "Einstein on the subs bench" comes from, to score a winner in extra time.

If you want to see an [iMO] rather good counter (balanced) argument try:

http://www.theguardian.com/science/2013/jun/16/has-physics-gone-too-far (I did "lol" at the Spock-like rebuttal!)

Mike Duff, professor of theoretical physics at Imperial College London:

Finally, you offer no credible alternative. If you don't like string theory the answer is simple: come up with a better one. The battle for the correct theory will not be won on Amazon or on the blogosphere, however. It will be won in the pages of scholarly scientific journals. Sadly, many critics of string theory, having lost their case in the court of science, try to win it in the court of popular opinion. A science writer calling the theorists who are actually doing the research "confidence tricksters'' or Stephen Hawking "a fairytale physicist'' doesn't cut the mustard.

But pretty much what this sometime experimentalist thinks too. Why so bad

that theorists study string theory? Many of them *already* study other things.

Theoreticians are not expensive. Just a whiteboard and a box of markers?

There is no rule that theory and experiment have to run at similar pace. It

has long been quasi-impossible anyway. Maybe experiment will throw up

something entirely revolutionary. Maybe theory will INFER something that

can never be proved experimentally.  But that doesn't "deny science" IMO.

If scientists do popular (modestly lucrative) things, are they worse than the

rest of us? But that raises deeper questions re. general attitudes to scientists

and (UK) science. Read what a sneering "intelligent" public write about "us"?

Something "science publicists" (with different paymasters) haven't cured.  

[Piero]

Unless one is strict with the meaning that a proof has to be empirical, mathematics is a science. Mathematics is not a natural science (where proofs are based on experiments), but is a formal science where proofs are based on logic. Similarly theoretical computer science.

And because of its logical formalism, mathematics is used extensively by all natural sciences (and also engineering).