Author 
Message 
Lubos Motl science forum beginner
Joined: 02 May 2005
Posts: 38

Posted: Mon Apr 25, 2005 12:32 pm Post subject:
Re: FAQ's



On Mon, 25 Apr 2005 we_pretty@yahoo.com wrote:
Quote:  What is the answer to the question:
Is there an experiment now or in near future such that
the results according to string theory would differ from
the results according to current accepted theories such
that if the predictions of string theory were not observed
it would imply that string theory is false ?

I think that this is a pretty sharply formulated question, and my opinion
is that the answer to this strong question is unfortunately No, there is
no such known experiment, and string theory is not falsifiable in this
strong sense. If someone thinks that I'm wrong, it will be great to learn
why.
Quote:  I know that when the question was asked with relativity the
answer was a definite yes. The answer was not something like
"yes, _if_ the speed of light is low enough or if the hyper
space field is strong enough".

Fair enough. We no longer live in the world where things are as easy as
special relativity. Special relativity is based on 110 papers that are
probably still more important than the 15,000 papers of string theory. But
that's how the life goes. The society is also much richer than it was 100
years ago, so it's not shocking that it typically can afford to fund
research where the expected gain per dollar is smaller than from the
salary for the Swiss patent clerk.
Quote:  Perhaps a question "What is a nontrivial prediction of a theory"
could be in FAQ and the answer something that would result yes in
the previous question.

There are many striking predictions that assume certain assumptions 
assumptions whose validity we're not able to verify with the current
understanding of the theory  but nevertheless these are predictions such
that if they're confirmed, they will go very far to support the theory.
Quote:  I'm almost sure that when physicists cannot agree on whether or
not string theory can make predictions it is not because they do
not understand the theory but because they do not agree on what
is a prediction.

I wish the world were as simple as you think. There are many contexts in
which our understanding is so shallow that even string theorists do not
agree whether something is a prediction of string theory or not  they
disagree whether it follows from string theory, regardless of the
terminological detail whether one calls it a "prediction".
Quote:  I do realize the difference between the answers to the question
about the predictions of string theory. I think that string
theory cannot make nontrivial predictions defined as I did

As I said, I think you're right.
Quote:  but I still would be happy to finance the research of string theory with
my tax dollars if I'd pay my taxes in dollars. I have got an impression
that string theory is cool and therefore should be researched.

I really think that we should be more honest in saying what can be
expected from some direction of research, and what cannot. String theory
is indeed cool. It's a source of great and often unexpected ideas that
seem to connect virtually all good ideas in highenergy theoretical
physics and many branches of mathematics.
All the best
Lubos
______________________________________________________________________________
Email: lumo@matfyz.cz fax: +1617/4960110 Web: http://lumo.matfyz.cz/
eFax: +1801/4541858 work: +1617/3849488 home: +1617/8684487 (call)
Webs: http://schwinger.harvard.edu/~motl/ http://motls.blogspot.com/
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

Back to top 


Peter Woit science forum beginner
Joined: 05 Apr 2005
Posts: 7

Posted: Mon Apr 25, 2005 3:52 pm Post subject:
Re: FAQ's



Urs Schreiber wrote:
Quote:  On Sat, 23 Apr 2005, Peter Woit wrote:
It might be worth noting the inherent incompatibility of the claims
1. string theory can't make any testable predictions because it is
not well enough understood.
2. string theory makes "amazing 'retrodictions'".
True, these two claims are incompatible. If you replace the first one by
1'. One currently cannot use string theory to predict things that
depend on strong coupling behaviour or questions of vacuum selection.
then 1' and 2 are compatible.

Everything depends upon the question of vacuum selection. If you don't
know what the vacuum is, you don't know what
the fundamental excitations about it are, much less anything about their
interactions.
Quote:  There is a greyscale in between "predicts nothing at all" and
"predicts all parameters of the standard model".

Yes, but string theory is not in the grey area.
Quote:  Whatever the vacuum state is, it must be a solution to the background
equations of motions. These include, to lowest order, those of GR.

The problem is not that that you know the equations which the vacuum
state must satisfy and are just having trouble
solving them. The problem is that you don't know what these equations
are. One thing you do know is that
they're not just the ones determined by lowest order string perturbation
theory. If that were the case, you would
know what the vacuum state is (of course it would give you the wrong
physics).
Quote:  It can also be checked nonperturbatively that string theory knows
about gravity. Graviton scattering amplitudes have been computed with
Matrix Theory.
AdS/CFt also demonstrates that strings know about gravity.

AdS/CFT and Matrix theory have nothing to say about the case at issue
here: quantum gravity in four large spacetime dimensions, 6 or 7
compactified (or dealt with
in some braneworld scenario). 

Back to top 


Lubos Motl science forum beginner
Joined: 02 May 2005
Posts: 38

Posted: Mon Apr 25, 2005 4:03 pm Post subject:
Re: FAQ's



On Mon, 25 Apr 2005, Peter Woit wrote:
Quote:  1. string theory can't make any testable predictions because it is not
well enough understood.
2. string theory makes "amazing 'retrodictions'".
True, these two claims are incompatible. If you replace the first one by
1'. One currently cannot use string theory to predict things that depend on
strong coupling behaviour or questions of vacuum selection.

I think that this is a rather accurate description, at least if you added
"except for things that are protected by supersymmetry".
Quote:  AdS/CFT and Matrix theory have nothing to say about the case at issue
here: quantum gravity in four large spacetime dimensions, 6 or 7
compactified (or dealt with in some braneworld scenario).

In the case of AdS/CFT, it is very easy to show that your statement is
incorrect. The simplest example is AdS4 x S7  the near horizon geometry
of an M2brane  that is dual to the 2+1dimensional superconformal field
theory which is the infrared limit of the 2+1dimensional maximally
supersymmetric gauge theory.
Admittedly, that's not a too realistic case. With many people, we have
discussed the conjectured CFT duals of the KKLT AdS4 vacua. There should
be a huge number of these CFT3 theories if holography of AdS spaces is
universal and if the KKLT vacua exist. Various people show examples why
many such theories may exist.
On of the complaints Cumrun had about the CFTs is that they should have
many operators of dimensions comparable to one, which should correspond to
KK modes of the momentum comparable to the inverse AdS radius. Frederik
Denef argues that he has convinced Cumrun that these operators are dual to
the Kahler moduli whose masses are also comparable to the radius... But
I'm getting offtopic now.
Best
Lubos
______________________________________________________________________________
Email: lumo@matfyz.cz fax: +1617/4960110 Web: http://lumo.matfyz.cz/
eFax: +1801/4541858 work: +1617/3849488 home: +1617/8684487 (call)
Webs: http://schwinger.harvard.edu/~motl/ http://motls.blogspot.com/
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

Back to top 


Peter Woit science forum beginner
Joined: 05 Apr 2005
Posts: 7

Posted: Mon Apr 25, 2005 8:02 pm Post subject:
Re: FAQ's



Lubos Motl wrote:
Quote:  In the case of AdS/CFT, it is very easy to show that your statement is
incorrect. The simplest example is AdS4 x S7  the near horizon
geometry of an M2brane  that is dual to the 2+1dimensional
superconformal field theory which is the infrared limit of the
2+1dimensional maximally supersymmetric gauge theory.
Admittedly, that's not a too realistic case. With many people, we have
discussed the conjectured CFT duals of the KKLT AdS4 vacua. There
should be a huge number of these CFT3 theories if holography of AdS
spaces is universal and if the KKLT vacua exist. Various people show
examples why many such theories may exist.

Hi Lubos,
I've always wondered about this. Forgetting KKLT vacua,
if strings on AdS4 really give a 4d theory of quantum gravity
in terms of a holographic dual 3d superconformal field theory
why isn't everyone happily calculating the answer to any
question about 4d quantum gravity using this 3d QFT?
What's not realistic about it?
Peter 

Back to top 


Lubos Motl science forum beginner
Joined: 02 May 2005
Posts: 38

Posted: Mon Apr 25, 2005 8:20 pm Post subject:
Re: FAQ's



On Mon, 25 Apr 2005, Peter Woit wrote:
Quote:  I've always wondered about this. Forgetting KKLT vacua,
if strings on AdS4 really give a 4d theory of quantum gravity
in terms of a holographic dual 3d superconformal field theory
why isn't everyone happily calculating the answer to any
question about 4d quantum gravity using this 3d QFT?
What's not realistic about it?
Peter

Hi Peter, of course that one can in principle calculate any question about
quantum gravity in AdS4 x S7 from the dual CFT  which is the
2+1dimensional superconformal field theory (SCFT) of the M2branes, the
IR limit of the 2+1 dimensional N=8 gauge theory. And in fact, many things
have been checked. Because of the words "IR limit" earlier in this
paragraph, it is not too easy to work with the 3dimensional SCFT. It's
more convenient to work with the fourdimensional conformal field theories
(directly gauge theories) which describe quantum gravity in AdS5. A lot of
stuff has been learned from the dual CFT. Thermodynamics of large black
holes have been verified qualitatively and in some other AdS cases
quantitatively, graviton scattering has been calculated from the CFT
correlators, branes have been found as baryons, various phase transitions
on both sides were matched (HawkingPage vs. confinementdeconfinement)
etc.
Even in AdS4, one can learn a lot from the 3dimensional conformal field
theory. What's not realistic about AdS4 x S7? For example, the radius of
AdS4 is (up to factors of two) equal to the radius of S7, so claiming that
S7 is negligibly small also implies assuming that the AdS space is
infinitely curved. AdS4 x S7 has 32 supercharges  too much for a
realistic theory. Realistic theories can start from 4 supercharges at
most.
"Quantum gravity in d=4" is not such a unique notion. Quantum gravity 
namely string theory  has many vacua (and our friends who believe the
anthropic haystack would say that 10^{350} vacua) that describe four large
dimensions, but the details of the remaining degrees of freedom (the extra
6 or 7 dimensions, speaking about the most famous examples) may be found
in many solutions (compactifications, inclusion of branes etc.)  and many
detailed results of quantum gravity will of course depend on the
compactification of the other dimensions etc.
I think that the only universal features of 4D theories of quantum gravity
are those that can be extracted from the classical and semiclassical
treatment of the EinsteinHilbert action and nonstringy considerations.
Everything else is backgrounddependent.
Also, it is pretty difficult to extract very local physics (of quantum
gravity) from the boundary CFT (such as subplanckian behavior)  it's
easier to use the CFT to calculate the properties of "large" objects.
Nevertheless, a lot of local physics (like in the ppwave limits) has been
extracted from the CFTs, too.
______________________________________________________________________________
Email: lumo@matfyz.cz fax: +1617/4960110 Web: http://lumo.matfyz.cz/
eFax: +1801/4541858 work: +1617/3849488 home: +1617/8684487 (call)
Webs: http://schwinger.harvard.edu/~motl/ http://motls.blogspot.com/
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^ 

Back to top 


Google


Back to top 



The time now is Sat Nov 17, 2018 5:37 pm  All times are GMT

