Ugly theories are good, says Anthropic Susskind!

17 views
Skip to first unread message

Serenus Zeitblom

unread,
Mar 24, 2003, 2:23:42 AM3/24/03
to sci-physic...@moderators.isc.org

In http://arxiv.org/abs/hep-th/0302219, Leonard Susskind declares that
the
recent attempt by Kachru et al,
http://arxiv.org/abs/hep-th/0301240,
to get hold of a positive cosmological constant in string theory,
involves
extremely unnatural, artificial models ["Rube Goldberg contraptions".]
However,
he doesn't mean this as criticism: he solemnly goes on to state that
this is
ok, because it is just what one would expect in an anthropic context;
and
he believes that string theory is anthropic. Of course he's right:
anthropicists
*should* expect successful physical theories to look really ugly and
contrived. And this is the most powerful argument against anthropicism
that I have ever seen. [I just hope that he is wrong about string
theory being anthropic!]
I can now justify my misanthropicism. Yay!

Dirk Bruere at Neopax

unread,
Mar 28, 2003, 2:54:18 AM3/28/03
to sci-physic...@moderators.isc.org

"Serenus Zeitblom" <serenusze...@yahoo.com> wrote in message
news:c7fd6c7a.03032...@posting.google.com...

I disagree.
I would expect that our universe to be a particular example selected from an
ensemble.
The theory describing the ensemble need not be ugly, and probably isn't.
However, curve fitting almost always is.

Dirk


Jeffery

unread,
Mar 29, 2003, 11:57:04 AM3/29/03
to
serenusze...@yahoo.com (Serenus Zeitblom) wrote in message news:<c7fd6c7a.03032...@posting.google.com>...


Lenny Susskind supports both string theory and the anthropic
principle, and says that string theory is anthropic, and anthropic
theories may be ugly but this is not a criticism. I don't why you
would interprete that as a reason to justify your misanthropicism. Let
me explain. He says the string theorists are looking for an extremely
beautiful simple underlying theory, and really there is no such
theory. You have the five superstring theories, and these along with
11d supergravity are all limiting points on the moduli space of an
underlying M-theory. String theories are hoping M-theory is the simple
beautiful underlying theory. Well, it may be underlying but it's
neither simple nor beautiful, at least in terms of how it reduces to
the real world. Susskind says that there are a googleplex or an
infinite number of other points in the M-theory moduli space, and our
universe must correspond to one of these other points, and not any of
the five known superstring theories, because they don't allow for
deSitter space. The likelihood of finding the point in the M-theory
moduli space that corresponds to the real world is infinitesimal, so
it's hopeless to use M-theory to actually calculate the details we
observe about the real world. In that sense, you can call it ugly.
However, there's nothing wrong with that. Try combining the Susskind
Landscape with inflationary cosmology. Each bubble in the inflating
universe would correspond to a different point in the moduli space of
the underlying M-theory, and give rise to totally different laws of
physics. Most would not be compatible with life. However, using the
anthropic principle, we would be in a bubble that would be condusive
to life, and so here we are. So, this seems to suggest that
superstring theory or M-theory is probably true, and the anthropic
principle is probably true, despite the "ugliness" of theory in the
sense that it encompasses a googleplex or an infinite number of
different point points in the moduli space, which are like different
theories, so we can't find which one is ours, so we can't calcuate the
laws of physics of our universe.

Jeffery Winkler

http://www.geocities.com/jefferywinkler

Boris Borcic

unread,
Mar 29, 2003, 8:14:47 PM3/29/03
to sci-physic...@moderators.isc.org

Serenus Zeitblom wrote:
> In http://arxiv.org/abs/hep-th/0302219, Leonard Susskind ...

> Of course he's right:
> anthropicists *should* expect successful physical theories to look really
> ugly and contrived.

Why ?

> And this is the most powerful argument against anthropicism
> that I have ever seen. [I just hope that he is wrong about string
> theory being anthropic!]
> I can now justify my misanthropicism. Yay!

I am really not clear about the status of the anthropic principle, my belief
is that there is an unspoken consensus about it that turns it into a form of
flag or boundary stone treated with scorn. It is difficult to debate because
it is unclear what positive value physicists admit for it; they usually allude
to it with negative undertones more than they tell what they think it is or
should be and why. The above is an example.

From my memory of a few sentences by Weinberg on AP - who clearly doesn't
like it either - my vigorous perception is that AP is grudgingly allowed a
position that's easier to explain as a consequence of the sociology and
genealogy of natural sciences than as the result of a properly reasoned
assessment. The metaphor of the boundary stone is I guess quite apt : at one
level, it is viewed as a boundary stone between natural science and religion
or faith, and the distinctively scornful conventional attitude appears at that
level to be something of a "patriotic obligation" of natural scientists not to
allow ground to what they view as the pathological anthropocentrism central to
revealed religion.

At another level, it is meant as a boundary stone internal to natural
sciences, a triple point between chemistry, physics and biology. That's the
position Weinberg ascribes to it when he quite directly reduces it to a
(pretty useless in terms of predictive power) apology of carbon chemistry.
This equation of AP with an apology of carbon chemistry has imo very little
rational justification and quite purely reflects the intention to reduce the
family of intuitions it represents, to practical insignificance - while
clothing the gesture as a tribute to darwinian reasoning in a minimal alliance
with biologists (studying the "ugly and contrived" forms living beings take)
against the common enemy of creationism.

To get at my point by a shortcut, the problem with equating AP to an apology
of carbon chemistry is imo made clear when one notes that carbon chemistry is
the widest imaginable encircling boundary one can draw around the known
observers of nature relevant to the anthropic argument, while the name of
"anthropic" itself alludes to a much closer boundary that's nevertheless much
wider than the one implied by the argument itself - if one means to take
it seriously. The question then becomes : having the more restricted anthropic
populations in view, can't we think of "conjugate" features just as reasonable
than a carbon-based biochemistry, and with distinct predictive power, to
abstract from them as candidate conditions of possibility of scientific
observers of a physical Universe ?

Otoh, one can't help noticing a correlation between this domination of
"anthropism-as-carbonism" in physics and the sadly observable fact that
physics-as-a-cultural-asset serves more often than (I presume) most physicist
would wish, to reduce human beings to (ugly) inanimate carbon.

Regards, Boris Borcic
--
"Archimedes did weapons research - while his life was irrefutably at risk"

Serenus Zeitblom

unread,
Mar 29, 2003, 8:18:14 PM3/29/03
to sci-physic...@moderators.isc.org

"Dirk Bruere at Neopax" <di...@neopax.com> wrote in message news:<b5uko1$2di47v$2...@ID-120108.news.dfncis.de>...

> "Serenus Zeitblom" <serenusze...@yahoo.com> wrote in message
> news:c7fd6c7a.03032...@posting.google.com...
> >However,
> > he doesn't mean this as criticism: he solemnly goes on to state that
> > this is
> > ok, because it is just what one would expect in an anthropic context;
> > and
> > he believes that string theory is anthropic. Of course he's right:
> > anthropicists
> > *should* expect successful physical theories to look really ugly and
> > contrived. And this is the most powerful argument against anthropicism
> > that I have ever seen.
>
> I disagree.
> I would expect that our universe to be a particular example selected from an
> ensemble.
> The theory describing the ensemble need not be ugly, and probably isn't.
> However, curve fitting almost always is.

But think of what Susskind is opening the door to here. Basically, now anyone
can put forward a ridiculously artificial model "motivated" by string theory,
and claim that this is ok because of anthropic selection.
PROFESSOR NAIBA: TIME HAS INERTIA!
PROFESSOR ZEAB: What a ridiculous, Rube Goldberg contraption!
PROFESSOR NAIBA: Yes, I agree. Isn't it wonderful?

Peter Woit

unread,
Apr 1, 2003, 2:24:16 AM4/1/03
to
Jeffery wrote:

>So, this seems to suggest that
>superstring theory or M-theory is probably true, and the anthropic
>principle is probably true, despite the "ugliness" of theory in the
>sense that it encompasses a googleplex or an infinite number of
>different point points in the moduli space, which are like different
>theories, so we can't find which one is ours, so we can't calcuate the
>laws of physics of our universe.

The reason no one can calculate anything in string/M theory has
nothing to with not knowing which vacuum state history has landed us
in amidst a googleplex of possibilities. No one understands what the
underlying M-theory is, and that is why no one can calculate anything.
If one had a well-defined theory and what it told us about the early
universe was that we could only get probabilistic information about
the late-time vacuum state, then talking about anthropic principles
might make sense. Right now this is just one more excuse for not
dealing with the fact that there is no theory.

Peter


Urs Schreiber

unread,
Apr 1, 2003, 4:14:38 PM4/1/03
to
[I had posted the following on 03/24 but haven't seen it appear yet.]

Serenus Zeitblom schrieb:


>
> In http://arxiv.org/abs/hep-th/0302219, Leonard Susskind declares that
> the recent attempt by Kachru et al,
> http://arxiv.org/abs/hep-th/0301240,
> to get hold of a positive cosmological constant in string theory,
> involves extremely unnatural, artificial models ["Rube Goldberg contraptions".]

[...]


> [I just hope that he is wrong about string theory being anthropic!]

You have probably seen the long paper hep-th/0303194 by Michael R.
Douglas today on the server, which contains attempts to understand the
statistics of string vacua, so to say, motivated by the idea that
instead of studying individual vacua it might be more helpful to have an
idea about which volume in parameter space a class of vacua occupies.

On p. 63 it says:

"Not having a reliable estimate, we would still conjecture that the
qualitative structure of the Standard Model is the result of discrete
choices which are not hard to realize, and that the fraction of models
which meet the qualitative tests (ignoring values of couplings) is
closer to O(10^-10) than to O(10^-100)."

The last sentence is:

"We even argued that depending on what comes out [of the study of string
vacuum distribution], we might find that string/M theory has much less
predictive power than we thought, perhaps none. At present it is
reasonabe to think that string/M theory will have predictive power, but
we should admit that we do not really know, and try to find out."

Alfred Einstead

unread,
Apr 1, 2003, 4:17:46 PM4/1/03
to
Boris Borcic <bor...@users.ch> wrote:
> I am really not clear about the status of the anthropic principle,

It's a theory of initial/boundary conditions which says that the
initial condition on that part of a suitable Cauchy surface that
intersects the Earth at January 1, 2001 is that there is carbon
based sentient life on that part of the Cauchy surface that
intersects the Earth; and that all theories are constrained
by this boundary condition.

It's perfectly valid to impose boundary conditions on a physical
system: even when the system is the entire universe.

Thomas Larsson

unread,
Apr 1, 2003, 8:22:30 PM4/1/03
to sci-physic...@moderators.isc.org

serenusze...@yahoo.com (Serenus Zeitblom) wrote in message news:<c7fd6c7a.03032...@posting.google.com>...
> Of course he's right: anthropicists
> *should* expect successful physical theories to look really ugly and
> contrived. And this is the most powerful argument against anthropicism
> that I have ever seen. [I just hope that he is wrong about string
> theory being anthropic!]
> I can now justify my misanthropicism. Yay!

Why string theory should not be anthropic - I had the impression that the
antropic principle was invented precisely because string theory failed to explain
certain experiments, notably a positive cosmological constant and the emergence
of the standard model.

However, not everyone seems to think that making predictions up to a googleplex
of alternatives is the best we can hope for. Recently (in the abstract to
http://www.arxiv.org/abs/hep-th/0212247), Witten wrote about string theory that
"It is also more predictive than conventional quantum field theory".

Aaron Bergman

unread,
Apr 1, 2003, 10:05:43 PM4/1/03
to
In article <4b8cc0a6.03040...@posting.google.com>,
thomas....@hdd.se (Thomas Larsson) wrote:

> serenusze...@yahoo.com (Serenus Zeitblom) wrote in message
> news:<c7fd6c7a.03032...@posting.google.com>...

> > Of course he's right: anthropicists
> > *should* expect successful physical theories to look really ugly and
> > contrived. And this is the most powerful argument against anthropicism
> > that I have ever seen. [I just hope that he is wrong about string
> > theory being anthropic!]
> > I can now justify my misanthropicism. Yay!

> Why string theory should not be anthropic - I had the impression that

> the anthropic principle was invented precisely because string theory

> failed to explain certain experiments, notably a positive
> cosmological constant and the emergence of the standard model.

Ack, no. (Yeesh.) The anthropic principle goes way back. I don't have a
cite, but it's a fairly obvious idea. It has probably been
re"discovered" a number of times.

Aaron
--
Aaron Bergman
<http://www.princeton.edu/~abergman/>
<http://aleph.blogspot.com>

[Moderator's note: The classic reference, which also contains a history
of earlier thinking about the anthropic principle, is

John D. Barrow and Frank J. Tipler, The anthropic cosmological principle
Oxford University Press, Oxford, 1986.

- jb]


Serenus Zeitblom

unread,
Apr 4, 2003, 2:50:58 AM4/4/03
to sci-physic...@moderators.isc.org

thomas....@hdd.se (Thomas Larsson) wrote in message news:<4b8cc0a6.03040...@posting.google.com>...

> serenusze...@yahoo.com (Serenus Zeitblom) wrote in message news:<c7fd6c7a.03032...@posting.google.com>...
> > Of course he's right: anthropicists
> > *should* expect successful physical theories to look really ugly and
> > contrived. And this is the most powerful argument against anthropicism
> > that I have ever seen. [I just hope that he is wrong about string
> > theory being anthropic!]
> > I can now justify my misanthropicism. Yay!
>
> Why string theory should not be anthropic - I had the impression that the
> antropic principle was invented precisely because string theory failed to explain
> certain experiments, notably a positive cosmological constant and the emergence
> of the standard model.

I guess my main point is just this. Theories rarely get destroyed
overnight. The
evidence against them gradually piles up, and their adherents are
gradually forced to make more and more unnatural assumptions to get
out of trouble until finally everybody gives up because the theory
looks so artificial and ugly. Think of what happened to the Steady
State Cosmology for example---those guys were never *proved* wrong,
but the initially simple and beautiful theory had to be patched up so
many times that it ended up looking like a Rube Goldberg contraption.
That's how you know a theory is wrong---when too many excuses have to
be made. Now El Susskind comes along and says that no, the extreme
artificiality of the string models with positive cosmological constant
is ok, it is just what one would expect anthropically. My objection to
this is that you can get away with *anything* this way. The anthropic
principle is like a licence to kill. I'm *not* saying that string/M
theory is wrong. I'm just saying that things are looking bad right
now, but the Anthropic Principle is *not* the way to salvation.
And no, the AP was NOT invented to save string theory! Thank God.
Speaking of whom, Ed Witten has often come out in the past against
anthropic ideas in string theory, and I have not heard that he has
changed his mind.

Serenus Zeitblom

unread,
Apr 3, 2003, 6:12:06 AM4/3/03
to
Peter Woit <wo...@cpw.math.columbia.edu> wrote in message

> The reason no one can calculate anything in string/M theory has
> nothing to with not knowing which vacuum state history has landed us
> in amidst a googleplex of possibilities. No one understands what the
> underlying M-theory is, and that is why no one can calculate anything.
> If one had a well-defined theory and what it told us about the early
> universe was that we could only get probabilistic information about
> the late-time vacuum state, then talking about anthropic principles
> might make sense. Right now this is just one more excuse for not
> dealing with the fact that there is no theory.

I think that's not really fair. Take a look, for example, at Bobby
Acharya's recent preprint and his papers cited there. He shows that
there are interpretations of M theory that make pretty definite
predictions about the cosmological constant. Unfortunately, those
predictions are almost certainly wrong. But they are predictions.

If you look at the recent papers of Acharya, Douglas, and Susskind,
three things become rather clear:

[1] These guys are even smarter than I thought they were.

[2] The discovery of a positive cosmological constant is getting
string theory into really serious trouble

[3] This last fact shows that string/M theory is real science.

Reading the literature now, you can detect a change of attitude.
People are more open to new ideas, and they are more willing to
accept that the theory could be in trouble; they don't have such a
tendency to weasel their way around criticism any more. The view of
string theory as a kind of catapult for launching bandwagons at
regular intervals is also dying down--see e.g. the notable lack of
enthusiasm for the latest candidate, the Dijkgraaf-Vafa stuff.

It's true that there is unfortunately a lot of anthropic talk, but I
guess that nobody really takes that very seriously---think of it as
a cry of pain.

Things are looking up for string theory! People are paying attention
to the serious stuff again.

Jeffery

unread,
Apr 4, 2003, 6:51:58 PM4/4/03
to sci-physic...@moderators.isc.org

serenusze...@yahoo.com (Serenus Zeitblom) wrote in message news:<c7fd6c7a.03040...@posting.google.com>...

>
> I guess my main point is just this. Theories rarely get destroyed
> overnight. The
> evidence against them gradually piles up, and their adherents are
> gradually forced to make more and more unnatural assumptions to get
> out of trouble until finally everybody gives up because the theory
> looks so artificial and ugly. Think of what happened to the Steady
> State Cosmology for example---those guys were never *proved* wrong,
> but the initially simple and beautiful theory had to be patched up so
> many times that it ended up looking like a Rube Goldberg contraption.
> That's how you know a theory is wrong---when too many excuses have to
> be made. Now El Susskind comes along and says that no, the extreme
> artificiality of the string models with positive cosmological constant
> is ok, it is just what one would expect anthropically. My objection to
> this is that you can get away with *anything* this way. The anthropic
> principle is like a licence to kill. I'm *not* saying that string/M
> theory is wrong. I'm just saying that things are looking bad right
> now, but the Anthropic Principle is *not* the way to salvation.
> And no, the AP was NOT invented to save string theory! Thank God.
> Speaking of whom, Ed Witten has often come out in the past against
> anthropic ideas in string theory, and I have not heard that he has
> changed his mind.

Most physicists have a very negative reaction to the anthropic
principle because they think it's a cop out, like "I can't figure out
the theory so we'll slap on the anthropic principle". Some people
think that physicists are using the anthropic principle as a crutch,
that they don't have to work as hard to come up with a theory because
they can just fall back on the anthropic principle. I don't think we
have a problem of physicists misusing it in this way. I think string
theorists are trying just as hard to figure out M-theory as if there
was no anthropic principle. Also, there is nothing wrong with the
anthropic principle itself. A point in the Universe chosen at random
would be very unlikely to be as close to a star as we are. How do you
explain the apparent unlikelihood of us being so close to a star? You
explain it with the enthropic principle, meaning life would be much
more likely to arise near a star. We are not located at a totally
random point in the Universe because we are here to ask the question.
Similarly, the anthropic principle applied to M-theory, might actually
be (or may not be) the best explanation for why we are in the point in
the quantum vacua, or in the type of Universe, we are in.

Jeffery Winkler

http://www.geocities.com/jefferywinkler

Peter Woit

unread,
Apr 6, 2003, 4:23:01 PM4/6/03
to sci-physic...@moderators.isc.org

Serenus Zeitblom wrote:

>>The reason no one can calculate anything in string/M theory has
>>nothing to with not knowing which vacuum state history has landed us
>>in amidst a googleplex of possibilities. No one understands what the
>>underlying M-theory is, and that is why no one can calculate anything.
>>If one had a well-defined theory and what it told us about the early
>>universe was that we could only get probabilistic information about
>>the late-time vacuum state, then talking about anthropic principles
>>might make sense. Right now this is just one more excuse for not
>>dealing with the fact that there is no theory.
>>
>>
>
>I think that's not really fair. Take a look, for example, at Bobby
>Acharya's recent preprint and his papers cited there. He shows that
>there are interpretations of M theory that make pretty definite
>predictions about the cosmological constant. Unfortunately, those
>predictions are almost certainly wrong. But they are predictions.
>
>

The word "prediction" seems to mean something different to me and to
a lot of string theorists. I don't see anything I would call a
prediction of the cosmological constant in Acharya's recent paper.
He's studying
compactifications on singular G2 manifolds, and claiming that for some
choice of what to put at the singularities, he can get a superpotential
whose minima fix moduli. He notes that he gets supersymmetric
vacua with negative cosmological constant this way, and that if
one somehow adds in supersymmetry breaking to make this all realistic,
"In principle, therefore, these models could have positive cosmological
constants."

He also notes that depending on how one chooses the singularities
and what one does at them, one can get a "vast number of possibilities
for the low energy physics" and he suggests that there are lots of
M-theory vacua that look like the standard model.

>If you look at the recent papers of Acharya, Douglas, and Susskind,
>three things become rather clear:
>
>[1] These guys are even smarter than I thought they were.
>

I don't know much about Acharya, but personally I found the Susskind
and Douglas recent "anthropic" papers sad and disturbing. Both of
them have a history of real accomplishments but to me these two
papers seem to represent an explicit abandonment of the idea that
theoretical physicsists should try and predict anything about the
real world.

>
>[2] The discovery of a positive cosmological constant is getting
>string theory into really serious trouble
>
>[3] This last fact shows that string/M theory is real science.
>
>

String/M-theory has always had a huge problem with the cosmological
constant and nothing has really changed on that front, although the
fact that it is non-zero may cause some people to take the problem
more seriously. I'd like to believe you that string theory is getting
into really serious trouble.

>Things are looking up for string theory! People are paying attention
>to the serious stuff again.
>
>

The Susskind/Douglas papers seem to me to be part of the final death-spiral
of the theory, as they and others finally recognize that the framework they
are working in can only lead two places:

1. There are no stable non-supersymmetric M-theory vacua that look anything
like the real world, so M-theory is wrong.

2. There are an infinity of M-theory vacua that look like the real
world, as well
as like anything else you might imagine, so M-theory is vacuous.

Peter


Urs Schreiber

unread,
Apr 7, 2003, 1:04:17 AM4/7/03
to sci-physic...@moderators.isc.org

Peter Woit schrieb:

[...]

> 2. There are an infinity of M-theory vacua that look like the real
> world, as well as like anything else you might imagine, so M-theory
> is vacuous.

I do not quite understand this reasoning: Why should a theory be vacuous
that does not predict its own initial conditions?

Why should we expect a theory of quantum gravity to predict not only how
the world works but in addition why it works that way (e.g. by
predicting that it's the unique stable vacuum)?

I recall that in his recent paper (hep-th/0303185) Lee Smolin advertises
the fact that loop quantum gravity can consistentently be formulated for
many kinds of matter couplings, for arbitrary dimensions and for
supersymmetric as well as for non-supersymmetric scenarios. Why should
versatility be a virtue for LQG but be a flaw for strings?

What should we demand a quantum gravity theory to predict? If it
correctly predicts Plack scale physics but does not uniquely fix the
low-energy gauge group and particle spectrum, does that mean its
worthless (or vacuous)?

Kevin A. Scaldeferri

unread,
Apr 7, 2003, 9:46:40 PM4/7/03
to
In article <3E90A600...@uni-essen.de>,
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote:

>Peter Woit schrieb:

>[...]
>> 2. There are an infinity of M-theory vacua that look like the real
>> world, as well as like anything else you might imagine, so M-theory
>> is vacuous.

...

>What should we demand a quantum gravity theory to predict? If it
>correctly predicts Plack scale physics but does not uniquely fix the
>low-energy gauge group and particle spectrum, does that mean its
>worthless (or vacuous)?

A QG theory that did this would be unsatifying, but would have value.
However, I think the problem is if you have a large number of vacua,
which all reproduce the low-energy physics, then you don't have any
predictive power, because you can't fill in the gap between low-energy
and the Planck scale.

--
======================================================================
Kevin Scaldeferri Calif. Institute of Technology
The INTJ's Prayer:
Lord keep me open to others' ideas, WRONG though they may be.

Thomas Larsson

unread,
Apr 7, 2003, 9:46:30 PM4/7/03
to
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote in message
news:<3E90A600...@uni-essen.de>...

> What should we demand a quantum gravity theory to predict? If it
> correctly predicts Plack scale physics but does not uniquely fix the
> low-energy gauge group and particle spectrum, does that mean its
> worthless (or vacuous)?

It might be somewhat difficult to evaluate whether Planck scale physics is
correctly predicted or not, given the difficulty to conduct Planck scale
experiments. And it is hard to tell what M-theory's Planck scale predictions
are, since it lacks a mathematical formulation.

Peter Woit

unread,
Apr 8, 2003, 7:20:42 PM4/8/03
to
Urs Schreiber wrote:

>Peter Woit schrieb:
>
>[...]
>
>
>>2. There are an infinity of M-theory vacua that look like the real
>>world, as well as like anything else you might imagine, so M-theory
>>is vacuous.
>>
>>
>

>I recall that in his recent paper (hep-th/0303185) Lee Smolin advertised


>the fact that loop quantum gravity can consistentently be formulated for
>many kinds of matter couplings, for arbitrary dimensions and for
>supersymmetric as well as for non-supersymmetric scenarios. Why should
>versatility be a virtue for LQG but be a flaw for strings?
>
>

I don't think the people who do LQG think of this versatility as a
virtue, but they
have never claimed that they have a theory which can explain the
standard model.
String/M-theory has been claiming this for nearly twenty years, but not
actually
ever coming up with a theory that actually explains anything at all
about the
standard model.

>What should we demand a quantum gravity theory to predict? If it
>correctly predicts Plack scale physics but does not uniquely fix the
>low-energy gauge group and particle spectrum, does that mean its
>worthless (or vacuous)?
>
>

If string/M-theory actually did this, it would be unsatisfying, but not
vacuous.
The fact of the matter is that it not only doesn't predict low energy
physics,
it also doesn't predict Planck scale physics. Ask a string/M-theorist
to write
down the spectrum of excitations of the theory at or above the Planck scale.
They can't do this. Despite claims one often sees, the
problem with string/M-theory is not that one
can't compute its vacuum state either because the calculation is too
difficult or because this is determined by an initial condition at the
big bang.
The problem is that there is no theory, just some fragmentary ideas which
many take as evidence for the existence of a wonderful unknown theory.

The hope of string/M-theorists recently seems to have been that it doesn't
matter that they don't have a theory, that perhaps the ideas about
string/M-theory they do have are
enough to constrain the possible vacuum states of the unknown theory to
a finite number, one of which would be the standard model. All the evidence
I've seen so far is that this doesn't work. It appears that if you can
write down a vacuum state
consistent with the Standard model, you can write down an infinity of
them, with whatever couplings, masses, etc. you want, predicting nothing.

Lawrence Foard

unread,
Apr 8, 2003, 7:32:47 PM4/8/03
to sci-physic...@moderators.isc.org
In article <325dbaf1.03040...@posting.google.com>,

Jeffery <jeffery...@mail.com> wrote:
>
>Most physicists have a very negative reaction to the anthropic
>principle because they think it's a cop out, like "I can't figure out
>the theory so we'll slap on the anthropic principle". Some people
>think that physicists are using the anthropic principle as a crutch,
>that they don't have to work as hard to come up with a theory because
>they can just fall back on the anthropic principle. I don't think we
>have a problem of physicists misusing it in this way. I think string
>theorists are trying just as hard to figure out M-theory as if there
>was no anthropic principle. Also, there is nothing wrong with the
>anthropic principle itself. A point in the Universe chosen at random
>would be very unlikely to be as close to a star as we are. How do you
>explain the apparent unlikelihood of us being so close to a star? You
>explain it with the enthropic principle, meaning life would be much
>more likely to arise near a star. We are not located at a totally
>random point in the Universe because we are here to ask the question.
>Similarly, the anthropic principle applied to M-theory, might actually
>be (or may not be) the best explanation for why we are in the point in
>the quantum vacua, or in the type of Universe, we are in.

Would it make sense to say the anthropic principle says one of
several things:

1) We are where we are because we can't be anywhere else. In this case
where refering not only to position is space, but of one of many possible
universes.

2) Any combination of flexible parameters will give rise to a universe with
life of some sort. This seems not to be the case in general, but might be
the case for atleast some variation.

3) The weirdest possibility, bordering on religion and requiring time
travel :) That intelligent life somehow plays a role in the existance
of the universe. I'm thinking of some sort of odd bootstrapping,
where adjustments influencing the past made by intelligent life
allow the universe to exist.

The interesting thing is all of the possibilities I can think of have
implications which run beyond whats currently observable. I believe the
fact that we exist is infact an important piece of data, but one which
is hard to interpret.
--
Be a counter terrorist perpetrate random senseless acts of kindness
Rave: Immanentization of the Eschaton in a Temporary Autonomous Zone.
"Anyone who trades liberty for security deserves neither liberty nor security"
-Benjamin Franklin

Urs Schreiber

unread,
Apr 9, 2003, 6:14:00 PM4/9/03
to sci-physic...@moderators.isc.org

Peter Woit schrieb:
>
> Urs Schreiber wrote:

> >I recall that in his recent paper (hep-th/0303185) Lee Smolin advertised
> >the fact that loop quantum gravity can consistentently be formulated for
> >many kinds of matter couplings, for arbitrary dimensions and for
> >supersymmetric as well as for non-supersymmetric scenarios. Why should
> >versatility be a virtue for LQG but be a flaw for strings?
> >
> I don't think the people who do LQG think of this versatility as a
> virtue,

Lee Smolin emphasizes it because it means that LQG would not be
invalidated by possible future discovery of such phenomena. It's not
regarded as a "problem" because for some reason (that still escapes me)
people do not demand of LQG (or any other approach to QG) what they
demand of strings: Namely (among other things) that it fixes the why and
not just the how of the world that we perceive.

The point seems to be that in string theory it looks as if there are, or
might be, mechanisms to actually say something about why-questions,
which is something, I'd think, completely beyond of what we could
reasonably hope a theory does for us. This looks like a bonus for me,
not like something that makes the theory vacuous if it does not exist.

> but they
> have never claimed that they have a theory which can explain the
> standard model. String/M-theory has been claiming this for nearly
> twenty years, but not actually ever coming up with a theory that
> actually explains anything at all about the standard model.

OK. So what should then be criticized in this regard are the claims
about the theory, not the theory itself. It's not a problem of string
theory, as a theory of quantum gravity, not to uniquely predict the
standard model (partly because it shares this characteristic with all
other approaches to QG), but a failure that people have argued
otherwise. Right?

> >What should we demand a quantum gravity theory to predict? If it
> >correctly predicts Plack scale physics but does not uniquely fix the
> >low-energy gauge group and particle spectrum, does that mean its
> >worthless (or vacuous)?
> >
> If string/M-theory actually did this, it would be unsatisfying, but not
> vacuous. The fact of the matter is that it not only doesn't predict low
> energy physics, it also doesn't predict Planck scale physics. Ask a
> string/M-theorist to write down the spectrum of excitations of the theory
> at or above the Planck scale. They can't do this.

I am not sure what you mean by this. At the Planck scale string theory
predicts that you see the massive modes of the string, the string's
spectrum. Is that not what you are referring to here? There are lots of
papers on what signatures of Planck scale physics to expect if string
theory applies, as Lubos Motl has recently emphasized in another thread.
Here are two examples that predict stringy signatures that could be seen
in coming experiments:

Kingman Cheung: Black hole, string ball, and p-brane production at
hadronic supercolliders, hep-ph/0205033

Abstract:
In models of large extra dimensions, the string and Planck scales become
accessible at future colliders. When the energy scale is above the
string scale or Planck scale a number of interesting phenomena occur,
namely, production of stringy states, p-branes, string balls, black
hole, etc. In this work, we systematically study the production cross
sections of black holes, string balls, and p-branes at hadronic
supercolliders. We also discuss their signatures. At the energy scale
between the string scale M_s and M_s/g_s^2, where g_s is the string
coupling, the production is dominated by string balls, while beyond
M_s/g_s^2 it is dominated by black holes. The production of a p-brane is
only comparable to black holes when the p-brane wraps entirely on small
extra dimensions. Rough estimates on the sensitivity reaches on the
fundamental Planck scale M_D are also obtained, based on the number of
raw events.


Kin-ya Oda: Alternative Signature of TeV Strings, hep-ph/0111298

Abstract:
In string theory, it is well known that any hard scattering amplitude
inevitably suffers exponential suppression. We demonstrate that, if the
string scale is M_s < 2TeV, this intrinsically stringy behavior leads to
a dramatic reduction in the QCD jet production rate with very high
transverse momenta p_T > 2TeV at LHC. This suppression is sufficient to
be observed in the first year of low-luminosity running. Our prediction
is based on the universal behavior of string theory, and therefore is
qualitatively model-independent. This signature is alternative and
complementary to conventional ones such as Regge resonance (or string
ball/black hole) production.
(end of abstract)


These papers of course assume large extra dimensions. If extra
dimensions are small this simply moves the signatures beyond the reach
of present and upcoming technology. But they still can be tested in
principle. The problem of actually measuring Plack scale physics is
inherent to QG in general. Once it was also hard to test classical GR.
In fact it still is hard to check for gravitational waves.

Ralph E. Frost

unread,
Apr 9, 2003, 8:40:23 PM4/9/03
to

Peter Woit <wo...@cpw.math.columbia.edu> wrote in message
news:b64t3b$4nm$1...@newsmaster.cc.columbia.edu...

> Jeffery wrote:

Isn't the real situation one where the competing trial theory to the
one running on the Anthropic Principle asserts that, in theory, given
complete knowledge of all initial conditions (hmmm, where've we heard
of that ambitious ability), then in theory all downstream stuff
including creative acts can be calculated, er, except for the
non-computable stuff? However, that program of perfect scientific
omniscence has not materialized on the Earthly plane and the calcs are
chaotic so while folks assert there IS a theory, it is a
non-functional one leaving us to still be guided by the functional AP.

Thomas Larsson

unread,
Apr 10, 2003, 3:57:13 PM4/10/03
to
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote in message
news:3E93DBB3...@uni-essen.de...

> Lee Smolin emphasizes it because it means that LQG would not be
> invalidated by possible future discovery of such phenomena. It's not
> regarded as a "problem" because for some reason (that still escapes me)
> people do not demand of LQG (or any other approach to QG) what they
> demand of strings: Namely (among other things) that it fixes the why and
> not just the how of the world that we perceive.

It would be truly great if string theory could explain the why, and even if
it only could explain the how. However, most people would be quite happy
with a much more modest achievement, namely that it would make *some*
testable prediction, and that this prediction is confirmed experimentally
within a reasonable amount of time. Witten recently claimed that string
theory is more predictive that conventional QFT, e.g. because it predicts
supersymmetry. However, few string theorists seem willing to stand up and
unambigously predict that sparticles will be seen at the LHC. (To me the
word "prediction" means that non-confirmation proves the theory wrong).

For a very clear discussion on why physics must produce reliable knowledge
about the real world, based on experiments, see Section 1.4 of
http://www.arxiv.org/abs/hep-th/0204131. Being a string theory pioneer, the
author should know what he talks about.

> These papers of course assume large extra dimensions. If extra
> dimensions are small this simply moves the signatures beyond the reach
> of present and upcoming technology.

For tests of large (> 0.01 mm) extra dimensions, see

Nature 421 922 (2003)
http://arXiv.org/abs/hep-ph/0210004
http://arxiv.org/abs/hep-ph/0303057

(thanks to Uncle Al).

Peter Woit

unread,
Apr 10, 2003, 5:15:13 PM4/10/03
to
Urs Schreiber wrote:

>OK. So what should then be criticized in this regard are the claims
>about the theory, not the theory itself. It's not a problem of string
>theory, as a theory of quantum gravity, not to uniquely predict the
>standard model (partly because it shares this characteristic with all
>other approaches to QG), but a failure that people have argued
>otherwise. Right?
>
>

String/M-theory doesn't just not uniquely predict the standard
model. It doesn't predict anything. If one wants to evaluate it
purely as an explanation of how to reconcile quantum mechanics
with general relativity, see Smolin for details of why this is an
unfinished project, one that LQG arguably does better at.

>
>
>>>What should we demand a quantum gravity theory to predict? If it
>>>correctly predicts Plack scale physics but does not uniquely fix the
>>>low-energy gauge group and particle spectrum, does that mean its
>>>worthless (or vacuous)?
>>>
>>>
>>>
>>If string/M-theory actually did this, it would be unsatisfying, but not
>>vacuous. The fact of the matter is that it not only doesn't predict low
>>energy physics, it also doesn't predict Planck scale physics. Ask a
>>string/M-theorist to write down the spectrum of excitations of the theory
>>at or above the Planck scale. They can't do this.
>>
>>
>
>I am not sure what you mean by this. At the Planck scale string theory
>predicts that you see the massive modes of the string, the string's
>spectrum. Is that not what you are referring to here? There are lots of
>papers on what signatures of Planck scale physics to expect if string
>theory applies, as Lubos Motl has recently emphasized in another thread.
>

What exactly is the string spectrum you expect to see at high
energy? Remember the diagram with six cusps that M-theorists like
to draw at the start of their talks. The only thing you know how
to calculate involves what happens at the points of the cusps and
even there you have the compactification problem. Where are we
in this diagram and what is the theory at any point in this
diagram? M-theorists seem to claim that the underlying theory is
some sort of eleven dimensional supersymmetric theory whose
fundamental excitations are higher-dimensional branes, but have
no idea what the dynamics of these is supposed to be.

>Here are two examples that predict stringy signatures that could be seen
>in coming experiments:
>
>

A prediction is not something that tells you what "could" be seen
in coming experiments, it is something that tells you what "will"
be seen in coming experiments. A scientific prediction is
supposed to be something that can be used to falsify a theory.
If black holes don't come out of the interaction regions at the
LHC, no one is going to think this says anything one way or
another about string/M-theory.

The large extra dimension scenario has all the problems of
string/M-theory, together with the added one of not even knowing
what the overall scale of the theory is. It also throws out the
only two arguments for supersymmetry (hierarchy problem and
unification of coupling constants).

Have to run now and get ready to watch the TV program "Einstein's
Dream" tonight. The advertisement indicates that the physics
faculty of the Institute for Advanced Study (Seiberg, Witten,
Maldacena) will be explaining their pioneering work on a theory
to explain all the forces of nature in the same terms. Freeman
Dyson is also supposed to appear in the role of the skeptic.

Peter

Aaron Bergman

unread,
Apr 11, 2003, 8:41:10 PM4/11/03
to sci-physic...@moderators.isc.org

In article <4b8cc0a6.03040...@posting.google.com>,
thomas....@hdd.se (Thomas Larsson) wrote:

> However, few string theorists seem willing to stand up and
> unambigously predict that sparticles will be seen at the LHC. (To me the
> word "prediction" means that non-confirmation proves the theory wrong).

The reason is that TeV scale symmetry is nice for solving the hierarchy
problem, but one can easily imagine breaking SUSY at a higher scale.
Still, the lack of low energy SUSY would be a fair sized blow, IMO.

Urs Schreiber

unread,
Apr 11, 2003, 10:47:16 PM4/11/03
to sci-physic...@moderators.isc.org

Peter Woit schrieb:

> String/M-theory doesn't just not uniquely predict the standard
> model. It doesn't predict anything.

I can well understand dissatisfaction with the achievements of string
theory. But annoyance about the habit of some string proponents to
overemphasize the success of the theory should probably not justify
copying this habit under reversed sign and deliberately underestimating
the success to the point of claiming that "there is no theory" and that
"it does not predict anything". To take the most banal counter example
(and there are less banal ones as you know better than I do): As soon as
accelerators see a total of 12 or more spacetime dimensions (yes, that's
not likely to happen anywhere soon, but that's how it goes with quantum
gravity) string theory is physically falsified. On the other hand, LQG
would apparently still be a possible candidate.

> If one wants to evaluate it
> purely as an explanation of how to reconcile quantum mechanics
> with general relativity, see Smolin for details of why this is an
> unfinished project,

Without any doubt is it unfinished. It seems that this is one point that
really everybody agrees on.

> one that LQG arguably does better at.

In the present state it is certainly a matter of individual opinion on
which approach to put more personal hopes, if on any. I am prepared to
seriously consider your and Thomas Larsson's point of view, that string
theory has "failed", as worthy of consideration. But to be useful this
claim must be backed up with good examples of such failures. So far we
have been arguing about whether or not the non-uniqueness of the vacuum
is really such a failure. To me it seems, and please correct me if this
is wrong, that all the objections that you have brought forth against
string theory rest on the single assumption that it needs to be
considered a failure that the theory does not predict where in moduli
space we are, i.e. which state the world is in:

> A prediction is not something that tells you what "could" be seen
> in coming experiments, it is something that tells you what "will"
> be seen in coming experiments.

That's true as soon as the state is known, which the experiment is
probing. Given some state, theory predicts its properties. Without
knowledge of the state being measured theory can only say what could be
seen, namely anything compatible with some state.

For instance, in classical celestial mechanics, where (bound) states are
Kepler ellipses, let E be any such state. Then the theory says that we
could find a planet in the solar system with trajectroy E. The theory
does not predict that beyond Neptun we will find Pluto (ignoring
planet-planet interactions). But as soon as position and velocity of
Pluto are determined the state is known and theory predicts all the
future trajectory of the object.

The analogue applies of course to string theory. As long as we do not
know the state the world is in (the "vacuum") we don't know the string
scale, the type of string (I,II, etc.), the compactification geometry,
for instance. As soon as we know these, string theory predicts for
instance amplitudes of string scattering events.

I don't think this is unusual or even diconcerting. As another example,
classical cosmology predicts that we could inhabit an open FRW universe.
It does not predict that we necessarily do, because there are other
possible states in the theory, like flat and closed FRW universes. But
as soon as the state we observe is known (the value of the curvature
parameter k in this example) the theory predicts all properties of this
state that can be measured.

Also, as with any other theory, some predictions should be possible
without a full knowledge of the state. That's why the author of the
hep-ph/0111298 paper, that I had cited, says (in the abstract):

Peter Woit

unread,
Apr 13, 2003, 10:30:21 PM4/13/03
to
Urs Schreiber wrote:

>To me it seems, and please correct me if this
>is wrong, that all the objections that you have brought forth against
>string theory rest on the single assumption that it needs to be
>considered a failure that the theory does not predict where in moduli
>space we are, i.e. which state the world is in:

My objections to string/M-theory are not just based on the
vacuum selection problem. The vacuum selection problem is
just a symptom of more serious underlying problems.
The minute one tries to calculate anything in
string/M-theory one runs into trouble. The trouble
with vacuum selection occurs first because that's the first
thing one needs to calculate to use the theory to do physics.
It's not that we don't know where in moduli
space we are, more to the point we don't know what the
moduli space is, or what the theory is that corresponds
to a generic point in this supposed moduli space.
The more serious problems of string/M-theory are:

1. There is no theory, just a few fragments of a theory
which some people have been hoping for twenty years to turn
into a real theory, with no success. If you don't believe
me that there is no theory, consult the following web page

http://www.physics.ucsb.edu/%7Egiddings/Mquest.html

which has a list of unanswered questions in M-theory,
including "What are the fundamental degrees of freedom
of M-theory?", "What are the dynamical laws governing
their interaction?" "What are the observables of M-theory?"
If you don't know what the fundamental degrees of freedom,
dynamical laws or observables of your theory are, how can
you claim to have a theory?

2. The fragments of a theory that exist are ugly and
mathematically unconvincing. All attempts to relate
string/M-theory to the real world involve
very complicated and ugly constructions (e.g. an
11-dim space, R^4 X a seven dimensional singular
space with a complicated and random-looking set of
singularities). The standard model is based on two
of the deepest notions in modern geometry (the Dirac
operator and the notion of a connection). An
aesthetically and mathematically convincing extension
of the standard model should involve mathematics at
this depth or greater, not random complicated geometrical
constructions obviously set up to get the answer one
wants.

3. The effort to get beyond the standard model using
string/M-theory has now occupied most of the smartest
people in the field for nearly twenty years, producing
somewhere between 10K-100K scientific papers. Most
importantly, this has used up about twenty years of intense
intellectual effort by Witten. The end result is
not only no progress towards the goal, but the goal
is getting further and further away, with the
Susskind/Douglas papers just the latest example of how
little string/M-theorists now think their theory can ever
explain. In 84-85, many were convinced that they would
be calculating standard model parameters within months. This
now seems impossibly far away. There has been real
progress in understanding what string/M-theory might be,
but all this new understanding has just made it more and
more clear that the original hopes for the theory don't
work.

If this isn't "failure", then nothing is.
String/M-theory is still being heavily promoted
these days (as in the TV program I saw last night), but
it is not being held either to a standard of
consistency or to a standard of making falsifiable
predictions that can be checked. From the point of
view of its promoters, the theory can never be said
to have failed.

Thomas Larsson

unread,
Apr 13, 2003, 10:30:22 PM4/13/03
to
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote in message
news:3E96A902...@uni-essen.de...

> In the present state it is certainly a matter of individual opinion on
> which approach to put more personal hopes, if on any. I am prepared to
> seriously consider your and Thomas Larsson's point of view, that string
> theory has "failed", as worthy of consideration. But to be useful this
> claim must be backed up with good examples of such failures. So far we
> have been arguing about whether or not the non-uniqueness of the vacuum
> is really such a failure. To me it seems, and please correct me if this
> is wrong, that all the objections that you have brought forth against
> string theory rest on the single assumption that it needs to be
> considered a failure that the theory does not predict where in moduli
> space we are, i.e. which state the world is in:

That string theory fails to make unique predictions is embarrassing,
but it has never been my main objection. LQG does not make terribly
many predictions either (besides, Smolin's claim that special
relativity may break down makes me confused - it does not sound
right).

My main point, as expressed in

http://www.arxiv.org/abs/math-ph/0103013

is the existence of new mathematics, which both generalizes important
structures appearing in perturbative string theory, and is closely
related to theories describing real physics. You can also see that
although the e-print is a political pamphlet intended to provoke, I do
emphasize the successes of stringy mathematics, both as pure
mathematics and in statistical physics. It has rather been (some)
string theorists here that have described perturbative string theory
as boring and irrelevant, when I have pointed out that its mathematics
is less distinguished than previously thought.

This is IMO a very strong point, because string theory evolved
historically using mathematical beauty and consistency as sole
guidelines. In particular, string theory's pioneers were obsessed with
*central* extensions, as witnessed by terms like Virasoro,
Neveu-Schwartz and Ramond algebras. If they had known that similar
(but non-central) extensions exist and admit representations for all
algebras of vector fields, string theory might have taken a different
route during its formative years.

As you and other regulars on this newsgroup know (but others may read
this too), I refer of course to two algebraic structures:

The higher-dimensional Virasoro algebra, which combines key aspects of
general relativity (diffeomorphisms) and quantum mechanics (projective
representations).

mb(3|8), which is an exceptional simple Lie superalgebra closely related
to the symmetries of the standard model.

I believe that my point is both original and constructive, and I am
sad that my position has been construed as mere negativist
string-hating. It seems to me that investigating these structures, and
transforming them into physics, would be a much better use for all the
talent in string theory, especially since recent anthropic ideas seem
to be the alternative. The worst thing that can happen is that this
turns out to be cool mathematics only.

Sadly, my marketing efforts have obviously failed, but I don't really
see what I could have done differently. The chances that people look
into these structures must be better if they dislike them than if they
have never heard about them at all.

Jeffery

unread,
Apr 14, 2003, 11:58:16 AM4/14/03
to
ent...@farviolet.com (Lawrence Foard) wrote in message news:<b6nugl$fr5>


1) We are where we are because we can't be anywhere else. In this case
> where refering not only to position is space, but of one of many possible
> universes.

Not that it would be impossible, but you can point to characteristics
of our universe that would make life more likely.

>
> 2) Any combination of flexible parameters will give rise to a universe with
> life of some sort. This seems not to be the case in general, but might be
> the case for atleast some variation.

You can certainly come up with hypothetical universes in which life
was impossible. What if all matter in the Universe was in a giant
black hole? What if we had only bosons with no fermions? What if the
proton had a tiny lifetime like the other hadrons? What if there were
one, or none, extended spatial dimensions?

However, you are right in that there are probably universes that
contain life totally different from us, totally different from
anything we could imagine?


>
> 3) The weirdest possibility, bordering on religion and requiring time
> travel :) That intelligent life somehow plays a role in the existance
> of the universe. I'm thinking of some sort of odd bootstrapping,
> where adjustments influencing the past made by intelligent life
> allow the universe to exist.

I don't think this makes any sense at all.

Jeffery Winkler

http://www.geocities.com/jefferywinkler

Elliot Tarabour

unread,
Apr 14, 2003, 3:34:01 PM4/14/03
to sci-physic...@moderators.isc.org

ent...@farviolet.com (Lawrence Foard) wrote in message news:<b6nugl$fr5$1...@farviolet.com>...

No it doesn't say that. It is a constraint on the fundamental
physical parameters by the evidence of our existence. That is why it
is troubling. Its statement is causally incorrect.


>
> 2) Any combination of flexible parameters will give rise to a universe with
> life of some sort. This seems not to be the case in general, but might be
> the case for atleast some variation.

No it doesn't say that either although the strong version requires
carbon based life to emerge from the particular set of parameters that
govern our universe.


>
> 3) The weirdest possibility, bordering on religion and requiring time
> travel :) That intelligent life somehow plays a role in the existance
> of the universe. I'm thinking of some sort of odd bootstrapping,
> where adjustments influencing the past made by intelligent life
> allow the universe to exist.

This is where you are getting caught in the causal cul de sac of #1.
There is no evidence that time is reversable. Now if you want to
consider some type of cyclical universe model (which requires a closed
universe) then you might argue that there might be a "reprocessing" of
the parameters in which life in the collapsing universe might
influence the next cycle. (see Tiplers "The Physics of Immortality and
get rid of all the infinities and base it on a quantum theory of
gravity and it might come close to what you suggest.


>
> The interesting thing is all of the possibilities I can think of have
> implications which run beyond whats currently observable. I believe the
> fact that we exist is infact an important piece of data, but one which
> is hard to interpret.

If you can't falsify it (even in principle), it's not science, its
religion.

e.

Kevin A. Scaldeferri

unread,
Apr 14, 2003, 3:37:02 PM4/14/03
to
In article <3E96A902...@uni-essen.de>,
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote:

>Peter Woit schrieb:
>
>> String/M-theory doesn't just not uniquely predict the standard
>> model. It doesn't predict anything.

>I can well understand dissatisfaction with the achievements of string
>theory. But annoyance about the habit of some string proponents to
>overemphasize the success of the theory should probably not justify
>copying this habit under reversed sign and deliberately underestimating
>the success to the point of claiming that "there is no theory" and that
>"it does not predict anything". To take the most banal counter example
>(and there are less banal ones as you know better than I do): As soon as
>accelerators see a total of 12 or more spacetime dimensions (yes, that's
>not likely to happen anywhere soon, but that's how it goes with quantum
>gravity) string theory is physically falsified.

Do you truly beleive that? 10 years ago, the same claim would have
been made, but the number was 11 or more.

Gordon D. Pusch

unread,
Apr 14, 2003, 3:50:57 PM4/14/03
to sci-physic...@moderators.isc.org, jeffery...@mail.com

jeffery...@mail.com (Jeffery) writes:
[...]

> Also, there is nothing wrong with the anthropic principle itself.
> A point in the Universe chosen at random would be very unlikely to be as
> close to a star as we are.

That depends on what probability measure you use to choose your point
"at random." Since matter is clumpy, and tends to get concentrated
in and around stars, drawing points "at random" in proportion to
mass density may not only land your "near" a star, but _in_ a star !!!


> How do you explain the apparent unlikelihood of us being so close to a
> star? You explain it with the enthropic principle, meaning life would be
> much more likely to arise near a star.

Not necessarily. There is growing evide nce that large, tidally-heated
icy moons of gas giant planets and "brown dwarves" will have liquid water
"oceans" under their ice crusts, as Europa, Ganymede, and perhaps Callisto
do, and that if one also considers Water/Ammonia mixtures, then Titan,
Triton, and even Pluto may have liquid "oceans" under their ice crusts.
Even the larger Kuiper Belt Objects may have liquid Water/Ammonia "oceans"
that are kept liquid and supplied with energy from the decay of radioactive
isotopes in their cores. If so, then the majority of potential abodes for
water-based carbon life in the galaxy may not be "near" stars at all, but
rather in modest-sized iceballs floating in the outer darkness. Hence,
humanity's existence on a hot, oversized rock far too close to its sun
starts looking very unlikely indeed, and one should start talking about the
"Water/Ammonia-based Cephalopodic Principle," not the "Anthropic principle,"
since the squid-populated iceballs will outnumber the earthlike worlds by
one to three orders of magnitude... :-T

(The above, BTW, is a prime example of why the "Anthropic Principle"
has virtually ZERO predictive power --- one can make it fit just about
any scenario one pleases...)


-- Gordon D. Pusch

perl -e '$_ = "gdpusch\@NO.xnet.SPAM.com\n"; s/NO\.//; s/SPAM\.//; print;'

Urs Schreiber

unread,
Apr 16, 2003, 5:29:45 PM4/16/03
to sci-physic...@moderators.isc.org

Thomas Larsson schrieb:

[...]


> The higher-dimensional Virasoro algebra, which combines key aspects of
> general relativity (diffeomorphisms) and quantum mechanics (projective
> representations).

What I don't understand is why the higher-dimensional Virasoro algebra
should be the constraint algebra of (quantum) GR: Isn't there more to GR
than diffeo invariance? When starting from the Einstein-Hilbert action
one gets the constraint algebra that is used in LQG to quantize. Which
action would give the constraint algebra that you are proposing? Or
would it follow from the same action (the EH action) but using a
different quantization scheme?

In your paper you are saying that string theory used the Virasoro
algebra at a point when no (few?) "deeper structures" were known. I
don't know about that. But I know that the Virasor algebra turns up in
string theory not because somebody considered it deep enough to be
worthy of a TOE, but because it comes up all by itself when quantizing
the Polyakov action, which is the simple generalization of the point
particle action. Actuallly, this is where all of string theory follows
from by consistency requirements. In this sense I always find it a
little odd to complain about the structures in string theory not being
"deep enough": They are not put in by hand but follow from dealing with
the Polyakov action, in the end. This is where all these concepts are
rooted in something that is arguably physics (even if not necessarily
the physics of our world). But what is the physics behind your concepts,
in this sense. From which physical assumptions do we arrive at the
multi-dimensional Virasoro algebra? What if tomorrow somebody else comes
along and promotes that not depth is the ultimate indicator of
algebraic profoundness but some other algebraic characterization. How do
I know who's right?

Can you say anything about the classical limit of states that satisfy
your constraints, can you recover classical GR in some sense or at least
speculate about how that would have to be approached? In other words:
The higher-dim Vir algebra is nice mathematics: What does it have to do
with physics, with general relativity in particular? It is an extension
of the diffeomorphism group (right?) but that alone does not give GR,
does it? You say it incorporates diff-invariance and projective
representations and thus must be GR and must be quantum. This doen't
sound implausible (though I cannot really judge this statement) but
gives me much less information than I usually get with a (new) theory of
physics. I am left to trust your intuition about how QG should look
like. It's like somebody writing down just E8 and then claiming that
this must be the TOE. I wouldn't see why he could even claim so. But
since the appearance of E8 can actually be derived from an assumption
about physics (strings instead of points), I can see why this claim is
at least not without some basis.


> mb(3|8), which is an exceptional simple Lie superalgebra closely related
> to the symmetries of the standard model.

I read in your paper that the 0-grade subalgeba is sl(3) x sl(2) x
gl(1), which you call the standard model algebra. But of course that is
really su(3) x su(2) x u(1). You say the former is the non-compact form
of the latter. Well possible that I am ignorant of a basic fact of gauge
theory: How do you get from sl(3) x sl(2) x gl(1) to the observed gauge
group?

John Baez

unread,
Apr 18, 2003, 2:15:58 AM4/18/03
to sci-physic...@moderators.isc.org

In article <3E93DBB3...@uni-essen.de>,
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote:

>Peter Woit schrieb:

>> Urs Schreiber wrote:

>> >I recall that in his recent paper (hep-th/0303185) Lee Smolin advertised
>> >the fact that loop quantum gravity can consistentently be formulated for
>> >many kinds of matter couplings, for arbitrary dimensions and for
>> >supersymmetric as well as for non-supersymmetric scenarios. Why should
>> >versatility be a virtue for LQG but be a flaw for strings?

>> I don't think the people who do LQG think of this versatility as a
>> virtue,

>Lee Smolin emphasizes it because it means that LQG would not be
>invalidated by possible future discovery of such phenomena. It's not
>regarded as a "problem" because for some reason (that still escapes me)
>people do not demand of LQG (or any other approach to QG) what they
>demand of strings: Namely (among other things) that it fixes the why and
>not just the how of the world that we perceive.

People place different demands on loop quantum gravity and string theory,
because the supporters of these theories have made quite different
claims about these theories.

String theory has often been touted as a "theory of everything" which
could explain all the forces and particles of nature, if only the details
could be worked out. It's often been suggested that there's a more or less
unique consistent version of the theory. Claims like this help string
theorists get lots of money. They also raise big expectations - and lead
to criticism when these expectations are not quickly met.

Loop quantum gravity has always been a more humble, low-budget affair.
Its goal is to find a background-free theory of gravity that takes
quantum mechanics into account. Its supporters never seem to claim
that there's a unique way to incorporate matter into this theory, or
that it's a "theory of everything". So it gets less funding, and
people demand less of it.

Of course, one reason for this is that string theory came from particle
physics, which got lots of money in the heyday of particle accelerators,
before the debacle of the superconducting supercollider. Loop quantum
gravity came from general relativity, which got very little money in the
days before LIGO.

These issues are almost more sociological than scientific, but the
sociology affects the progress of science. People would reduce
their expectations of string theory, become less critical of it,
give it less funding, but give it more time to work, if string
theorists lowered their claims for it. I think this would be a
good thing. It may be in the process of happening right now, thanks
to the work of Susskind and others.

I should emphasize that both in superstring theory and loop quantum
gravity, the strength of the claims made varies drastically from
researcher to researcher. In loop quantum gravity, Smolin probably
makes more enthusiastic claims than almost anyone else.


Iain

unread,
Apr 18, 2003, 2:16:37 AM4/18/03
to sci-physic...@moderators.isc.org

gdp...@NO.xnet.SPAM.com (Gordon D. Pusch) wrote in message news:<giof3aq...@pusch.xnet.com>...
> jeffery...@mail.com (Jeffery) writes:

<snip>


> > How do you explain the apparent unlikelihood of us being so close to a
> > star? You explain it with the enthropic principle, meaning life would be
> > much more likely to arise near a star.
>
> Not necessarily. There is growing evide nce that large, tidally-heated
> icy moons of gas giant planets and "brown dwarves" will have liquid water
> "oceans" under their ice crusts, as Europa, Ganymede, and perhaps Callisto
> do, and that if one also considers Water/Ammonia mixtures, then Titan,
> Triton, and even Pluto may have liquid "oceans" under their ice crusts.
> Even the larger Kuiper Belt Objects may have liquid Water/Ammonia "oceans"
> that are kept liquid and supplied with energy from the decay of radioactive
> isotopes in their cores. If so, then the majority of potential abodes for
> water-based carbon life in the galaxy may not be "near" stars at all, but
> rather in modest-sized iceballs floating in the outer darkness. Hence,
> humanity's existence on a hot, oversized rock far too close to its sun
> starts looking very unlikely indeed, and one should start talking about the
> "Water/Ammonia-based Cephalopodic Principle," not the "Anthropic principle,"
> since the squid-populated iceballs will outnumber the earthlike worlds by
> one to three orders of magnitude... :-T
>
> (The above, BTW, is a prime example of why the "Anthropic Principle"
> has virtually ZERO predictive power --- one can make it fit just about
> any scenario one pleases...)

All very interesting speculation. However, the Anthropic Principle
does make a prediction - given that we are observers, and that we
exist when and where we do, and given no information as to whether
observers exist elsewhere, then observers *will* tend to be located
close to a star. Of course, one may be wrong, and self-aware entities
may be found to exist orbiting Saturn, but of course you adapt your
reasoning to fit the facts. All the Anthropic Principle says is that
observers are typically found in conditions no more unusual than those
required for their existence. It doesn't say much, true, but it does
indicate where to place your bet.

-I

Thomas Larsson

unread,
Apr 18, 2003, 10:51:47 AM4/18/03
to
Urs Schreiber <Urs.Sc...@uni-essen.de> wrote in message news:<3E9A6EEF...@uni-essen.de>...

> Thomas Larsson schrieb:
>
> [...]
> > The higher-dimensional Virasoro algebra, which combines key aspects of
> > general relativity (diffeomorphisms) and quantum mechanics (projective
> > representations).
>
> What I don't understand is why the higher-dimensional Virasoro algebra
> should be the constraint algebra of (quantum) GR: Isn't there more to GR
> than diffeo invariance?

Sure there is, but diff invariance is a crucial aspect (and one that
string theory misses, lacking a background-independent formulation). In
fact, tensor calculus is secretly the classical representation theory of
the diff group - think primary fields in CFT. There is certainly more to
GR than tensor calculus, but it is hard to imagine GR without it.
Similarly, the Fock modules of Vir(n) can be considered as the quantum
analogue of tensor calculus. This is not QG by itself, but probably a
necessary piece of mathematical infrastructure, which does for QG what
tensor calculus does for GR.

One reason why I believe very strongly in this has to do with how it was
discovered. The QG argument came first, at a time when there even was a
no-go theorem stating that a multi-dimensional Virasoro algebra could
not exist: vect(n) has no central extension when n > 1. That a
generalization of the Virasoro algebra nevertheless exists and has a
natural representation theory (something that other extensions don't
seem to have) confirmed to me that my instincts were right.

> When starting from the Einstein-Hilbert action
> one gets the constraint algebra that is used in LQG to quantize. Which
> action would give the constraint algebra that you are proposing? Or
> would it follow from the same action (the EH action) but using a
> different quantization scheme?

The latter. Now when the theory of Fock modules seems essentially
complete, I constructed in http://www.arxiv.org/abs/math-ph/0210023
more complicated modules that depend on dynamics in the
form of Euler-Lagrange (EL) equations. The idea is as follows:

Phase space is the space of solutions to the EL equations modulo gauges;
one may coordinatize phase space by specifying the values on a spacelike
surface, but this is not intrinsic. I have read this somewhere, probably
in

http://www.arxiv.org/abs/gr-qc/9910079
Loop Quantum Gravity and the Meaning of Diffeomorphism Invariance
Marcus Gaul, Carlo Rovelli

but now I am unable to find the quote (the paper is worth reading
anyway). Anyway, I consider a "large phase space" where gauges have not
been eliminated, but where the gauge group instead acts on (the space of
functions over) this large phase space. Classically, and in the absense
of anomalies, there is no difference; you can always mod out gauges by
passing to the orbit space.

Since the large phase space is built from tensor fields, the Fock
construction applies, and one winds up with Fock modules of Vir(n) which
to me seem like a quantization of the dynamical system - the modules
depend on the EL equations and carry a representation of the
diffeomorphism algebra, where some brackets have quantum corrections.

If you look at things more closely there are some consistency conditions
which only seem to work out if spacetime has four dimensions and there
are two bosons for every three fermions with the naive counting of
degrees of freedom; this relation holds in the standard model coupled to
gravity. However, the same argument also requires new gauge symmetries,
including fermionic ones, perhaps indicating the need for some new physics.

> What if tomorrow somebody else comes
> along and promotes that not depth is the ultimate indicator of
> algebraic profoundness but some other algebraic characterization. How do
> I know who's right?

You don't! If you want fool-proof guarantees in advance, you shouldn't
be doing research. Mathematical beauty alone can never be the ultimate
arbiter, since beauty lies in the eyes of the beholder. It is eventually
the job of the experimentalist to select between candidate theories.

Besides, the coolest thing about maximal depth is that it happens to
lead to a symmetry group which is closely related to the standard model,
i.e. experiments.


> I read in your paper that the 0-grade subalgeba is sl(3) x sl(2) x
> gl(1), which you call the standard model algebra. But of course that is
> really su(3) x su(2) x u(1). You say the former is the non-compact form
> of the latter. Well possible that I am ignorant of a basic fact of gauge
> theory: How do you get from sl(3) x sl(2) x gl(1) to the observed gauge
> group?

You go from compact to non-compact by complexification, and the other
way by picking a compact real form. One must eventually work with the
real form, but working with the complexification seems like a natural
first step.

Urs Schreiber

unread,
Apr 18, 2003, 11:17:45 AM4/18/03
to
Peter Woit schrieb:


> 1. There is no theory, just a few fragments of a theory
> which some people have been hoping for twenty years to turn
> into a real theory, with no success. If you don't believe
> me that there is no theory, consult the following web page
>
> http://www.physics.ucsb.edu/%7Egiddings/Mquest.html
>
> which has a list of unanswered questions in M-theory,
> including "What are the fundamental degrees of freedom
> of M-theory?", "What are the dynamical laws governing
> their interaction?" "What are the observables of M-theory?"
> If you don't know what the fundamental degrees of freedom,
> dynamical laws or observables of your theory are, how can
> you claim to have a theory?

Since this applies to M-theory only, I propose the following summary of
this discussion:

1) Perturbative string theory exists and makes predictions.

2) M-theory hardly exists and hence hardly makes predictions.

3) Many popular overly enthusiastic claims about string/M-theory are
problematic.

[...]


> The standard model is based on two
> of the deepest notions in modern geometry (the Dirac
> operator and the notion of a connection). An
> aesthetically and mathematically convincing extension
> of the standard model should involve mathematics at
> this depth or greater,

Isn't it fascinating how much string theory has to say about Dirac
operators, gauge theory and modern geometry?

Urs Schreiber

unread,
Apr 20, 2003, 8:22:14 PM4/20/03
to
"Kevin A. Scaldeferri" schrieb:

> In article <3E96A902...@uni-essen.de>,
> Urs Schreiber <Urs.Sc...@uni-essen.de> wrote:

> >Peter Woit schrieb:

> >> String/M-theory doesn't just not uniquely predict the standard
> >> model. It doesn't predict anything.

[...]
> >As soon as

> >accelerators see a total of 12 or more spacetime dimensions (yes, that's
> >not likely to happen anywhere soon, but that's how it goes with quantum
> >gravity) string theory is physically falsified.
>
> Do you truly beleive that? 10 years ago, the same claim would have
> been made, but the number was 11 or more.

Though the dimension did not just go up by one. The prediction is that
up to ten dimensions you see stringy signatures, e.g. Regge behaviour
and Hagedorn statistics. As soon as you begin to see an 11th dimension
this would gradually be replaced by the corresponding signatures of
full membranes.

Superstrings as such live in ten dimensions, and that cannot change.
They are limits of supermembranes which necessarily live in 11
dimensions, and that cannot change either. I don't know if membranes
could in priciple possibly some day be found out to be limits of
something even higher dimensional, as you suggest. Actually there is
F-theory, which lives in 12 dimensions. But that's usually considered
as a mere computational tool to get 11d physics. (Quite similar to how
the Wess-Zumino term in a 2-d WZ model is naturally written as a
topological term integrated over a 3-fold.)

So I think there is reason to believe in my claim above, but you are
right that it should be made more precise. Let's see: As soon as
accelerators see 11 dimensions and still no deviation from
Regge/Hagedorn signatures string theory is physically falsified. If no
Regge/Hagedorn behaviour (stringy signatures) have been seen before
that string theory is already physically falsified. As soon as
accelerators see 12 dimensions and spacetime is still described by
higher-order corrected gravity the theory is physically falsified
(since no sugra possible in d>11). If no signs that higher-order
corrected sugra applies before that have been seen the theory is
already physically falsified.

But I also think that this dimension-focused question is too narrow.
Long before accelerators (or any other experiment) can probe 10 or 11
dimensions all kinds of generic stringy signatures need to be
observed, as discussed in the papers that I had cited.

Gordon D. Pusch

unread,
Apr 21, 2003, 6:21:12 PM4/21/03
to

iainm...@yahoo.com (Iain) writes:

> All very interesting speculation. However, the Anthropic Principle
> does make a prediction - given that we are observers, and that we
> exist when and where we do, and given no information as to whether
> observers exist elsewhere, then observers *will* tend to be located
> close to a star.

I'm sorry, but I don't consider "we're here, because we're here,
because we're here, because we're here" to be a very interesting
or useful "prediction;" in fact, I don't consider it to be a
"prediction" in any meaningful sense of the word --- at best,
it is a particularly banal and useless tautology.

> All the Anthropic Principle says is that observers are typically found
> in conditions no more unusual than those required for their existence.
> It doesn't say much, true, but it does indicate where to place your bet.

You are confusing the "Anthropic Principle" with the "Generalized
Copernican Principle:" The former is the belief that the Universe must
have exactly the properties it does or we would not be here to observe it;
the latter is the belief that we must live in the most mediocre of all
possible worlds. The only thing these two beliefs share is that they are
both tautological in nature and both have ZERO predictive power.

Peter Woit

unread,
Apr 22, 2003, 4:04:43 PM4/22/03
to
Urs Schreiber wrote:

>Since this applies to M-theory only, I propose the following summary of
>this discussion:
>
>1) Perturbative string theory exists and makes predictions.

Sorry, but I can't agree with either of these statements.
The string theory perturbation series is well-defined to
two loops and may or may not have finite, well-defined
terms for higher loops. Even if the terms are finite,
the series is almost certainly a divergent series. So
I still have a problem with the statement that the theory
"exists", or string theorist's claims that they have a
consistent, finite theory of quantum gravity.

If you just define the theory by ignoring everything
after two loops, you do have a well-defined framework
in which to calculate things. Unfortunately it has lots
of problems including:
1. It's not unitary (since you are not including higher loops)
2. The theory has unbroken space-time supersymmetry
3. Before you can calculate anything you have to
choose a background geometry. There are an infinite
number of ways to do this.

So, the only thing that I see that really exists is
a calculational scheme that is not really consistent
and predicts just about nothing. One of the few
things you could possibly argue is a prediction
(space-time supersymmetry) disagrees with
experiment.

>2) M-theory hardly exists and hence hardly makes predictions.
>
>3) Many popular overly enthusiastic claims about string/M-theory are
>problematic.

Can't disagree with either or these. But when will leading figures
in the string/M-theory community stop making these problematic
claims?

Peter


Peter Woit

unread,
Apr 23, 2003, 2:06:50 AM4/23/03
to sci-physic...@moderators.isc.org

Urs Schreiber wrote:

>Peter Woit schrieb:
>
>
>
>>String/M-theory doesn't just not uniquely predict the standard
>>model. It doesn't predict anything.
>>
>>

>As soon as
>accelerators see a total of 12 or more spacetime dimensions (yes, that's
>not likely to happen anywhere soon, but that's how it goes with quantum
>gravity) string theory is physically falsified.
>
>

I'm still waiting to hear a single prediction from string theorists
about how the
physical world will behave at any energy scale. A prediction for what won't
happen isn't at all the same thing.

An example: suppose I announce that I have a wonderful theory of the
universe
that says that it is built out of golf balls. It might be true that if
tennis
balls start coming out of the interaction region at the LHC, that would be
evidence towards falsifying my theory. But calling "no tennis
balls at the LHC" a prediction of the theory is stretching the notion
of what a scientific prediction is past the breaking point.

zirkus

unread,
Apr 25, 2003, 2:14:10 AM4/25/03
to sci-physic...@moderators.isc.org

Peter Woit <wo...@cpw.math.columbia.edu> wrote in message news:

> I'm still waiting to hear a single prediction from string theorists

> about how the
> physical world will behave at any energy scale.

Since you are a mathematician perhaps you have not read about the
history of physics - I don't know if you have or not. Physicists did
not study the theory of black holes before the advent of GTR, however,
they did study various quantum phenomena before the advent of the
full-fledged quantum theory in the 1920s.

E. Witten once said something ala string theory is like 21st century
mathematics that fell into the 20th century. Physicists are not trying
to be rigorous logicians and if they happen to get
mathematical-physical insights before getting any clear experimental
hints then that is what they have to work with.

You can be assured that there is no good current reason for rejecting
either string theory or loop QG because if there was then the
discoverer would not keep it a secret. Also, you should probably
ignore anthropic speculations and the like in string theory because
such ideas are most likely premature and probably meaningless if they
have no a priori basis.

You keep wanting to emphasize the importance of Dirac operators but
these are also used in string theory, e.g. Pierre Ramond has written
about the role of the algebraic Kostant-Dirac operator in SUGRA.

In summary, ugly theories would only be good if they were actually the
truth but it is too early to give up on string theory, especially
since string theory is still smarter than you, me or any other
individual because no one person has produced as many new ideas in
math or (potentially) physics as string theory has.

Jeffery

unread,
Apr 25, 2003, 8:10:56 PM4/25/03
to sci-physic...@moderators.isc.org

zir...@hotmail.com (zirkus) wrote in message news:<8c7d34cb.03042...@posting.google.com>...

> Peter Woit <wo...@cpw.math.columbia.edu> wrote in message news:
>
> > I'm still waiting to hear a single prediction from string theorists
> > about how the
> > physical world will behave at any energy scale.
>
> Since you are a mathematician perhaps you have not read about the
> history of physics - I don't know if you have or not. Physicists did
> not study the theory of black holes before the advent of GTR, however,
> they did study various quantum phenomena before the advent of the
> full-fledged quantum theory in the 1920s.
>

In 1784, the English geologist John Michell realized that it would be
theoretically possible for gravity to be so overwhelmingly strong that
nothing, not even light could escape. So yeah, theory of black holes
existed before the advent of GTR.

Jeffery Winkler

Peter Woit

unread,
Apr 29, 2003, 4:33:56 AM4/29/03