[tt] [s-t] An Exceptionally Simple Theory of Everything

Wed Nov 28 07:59:30 UTC 2007

----- Forwarded message from "Joshua XXXXXXX" <XXX at post.harvard.edu> -----

From: "Joshua XXXXXX" <XXX at post.harvard.edu>
Date: Wed, 21 Nov 2007 16:41:05 -0600
Subject: Re: [s-t] An Exceptionally Simple Theory of Everything

<XXXXXXXXXXXX at verizon.net> wrote:

> A paper by that title, claiming to unify quantum theory and
> general relatively without string theory, found via
> Slashdot by a cow-orker who is wondering what it
> might mean:
>
> http://arxiv.org/abs/0711.0770

Short version:  Joshua XXXXXX, whose PhD is from a slightly better
U of C than the paper's author, is pretty sure it's kookery.

Postwar particle physics in a nutshell, beginning with Shelter Island
and Dyson's QED paper (for which Schwinger, Tomonaga and Feynman were
rightly awarded the Nobel, and thereby hangs a tale).  This gives us
a working relativistic quantum theory of one field (the photon) and
one particle (the electron/positron), with which we can work around
apparently divergent calculations through a mathematical trick known
as renormalization.  This is masterfully described for the layman in
Feynman's own little book, "QED."  A fair litmus test:  if you think
this is the single most beautiful piece of physics since Einstein's GR, 
you have the soul of a physicist.  If you think it's hackery, try math.

1950s and 60s:  the particle zoo.  New fundamental physics concerning
parity, which was unexpected but could be accommodated within the QED
paradigm.  "Who ordered that?" particles by the score, with hints of
order in their spectrum reminiscent of the hydrogen spectrum before QM,
or the periodic table before the Bohr atom.  Deep despair about the
prospects of a perturbative QED-like approach, since the new forces
were neither weak nor renormalizable.  Crazed ideas like S-matrix
analyticity, nuclear democracy and bootstrapping, hoping to get true
answers without quantum field theory, since the sums over partial answers
looked untameable.

1970s:  the pendulum swings back.  't Hooft and Veltman prove that 
Yang-Mills theories, which are fancy QED with many fields, are 
renormalizable and hence candidates for old-fashioned QFTs.  Gross
and Wilczek show that the strong interaction gets weak at very high
energies/short distances ("asymptotic freedom"), which means that
the complexity of the particle zoo is irrelevant to the fundamental
physics underneath.  Weinberg and Salam exploit unfashionable work on
spontaneous symmetry breaking to show that the electromagnetic and
weak force may be described by a Yang-Mills theory, even though the
weak force alone cannot be.  But Coleman (peace upon him; he died
this Sunday, which saddens me greatly) and Manjula proved that there
is no way to mix spacetime symmetry up with Yang-Mills symmetry,
using a deep general argument from S-matrix theory.  So now we have
some hope of a complete description of strong, weak AND e-m physics on
terms that Feynman would have accepted in 1947.  But not gravity.

1980s:  the Standard Model triumphant.  The i's are dotted, t's are
crossed, and Z's are discovered in the Weinberg-Salam program laid out
above.  Easy generalizations to unify strong with electroweak by
Georgi and Glashow, Pati and Salam etc. all turn out to be ruled out
by a failure to find proton decay, but something along those lines is
clearly going to work eventually.  Wess and Zumino have found a 
fascinating hole in Coleman's no-go theorem, called supersymmetry, 
which can mix up spacetime with Yang-Mills by mixing up bosons (which
are gregarious and good for making lasers) with fermions (which are
aloof and good for making lab benches that lasers don't fall through).
Extensions of the supersymmetry idea called supergravity offer some
faint hope of a renormalizable theory of everything.

But then strings burst on the scene.  Accidental fallout from the
crazed groping of the bootstrapping era yields a series of purely
mathematical miracles, not fully understood to this day.  It becomes
possible to create a quantum theory of strings that is actually finite
and needs no renormalization, *and Einstein gravity falls out of it
inevitably and for free*.  Except, only in 26 dimensions, or 10 with
supersymmetry.  And, over the last twenty years, string theory has 
sucked most of the life out of the subject, attracting top minds as 
a flame attracts moths, or like the broken advanced alien artifact 
that it probably is.

One way to stay clear of strings is to generalize the standard model
in cautious ways and avoid the big questions.  We're in a golden age
of cosmology, and we can probably afford one more generation of big
accelerators; in fifty or a hundred years there may be enough new data
to build a revolution upon, but for now phenomenology is just botany.
Another way is to perversely attack the gravity problem as if strings
had never been dreamt of.  The loop quantum gravity approach of
Ashtekar and others is the most successful feint in this direction;
they throw out the elegant Einstein paradigm of curved spacetime, and
consider flat spacetime with origami folds (angle deficits, which are
measured around loops and quantized).  None of the perturbative
machinery of Yang-Mills carries over; so far the LQG people have their
hands full trying to reproduce empty space as a quantum theory, never
mind particles.

So where does this E8 paper fit into the bestiary?  There seem to be
basic errors involving the mixing of boson and fermion fields without
supersymmetry; I'm not completely convinced the paper is even coherent
at the elementary level.  But lazily spotting the author that much, he
has written down a very big model that uses a Yang-Mills theory to
mix up things that are experimentally known with very high precision
not to mix (strong-electroweak with flavor, for starters) and also to
mix the giant Yang-Mills symmetry with gravity, as forbidden by that
Coleman result.  He does this in a LQG picture of gravity where the
renormalization theorem that makes Yang-Mills interesting does not seem
to work, and he assumes someone else will eventually discover why it's
renormalizable or finite.  And that's about it, so far as I can tell.
E8, the largest exceptional group, is an enormous symmetry with room
for all sorts of things in it, and (1) often emerges from the top down
in trying to make some contact between string theory and 4D reality, as
well as being (2) often considered from the bottom up as a tempting 
extension of the SO(10) or E6 symmetries that arise naturally in the
"grand unification" of strong-electroweak.

I once read a devastating summary of Benoit Mandelbrot's contribution
to his eponymous set.  He didn't discover it; that was Julia.  He
didn't prove [various interesting results] about it; that was [other
people].  He _did_ prove that it is not simply connected; however, the
proof was wrong, and it _is_ simply connected.  And he wasn't the
first to use a computer to draw a picture of it; that was [someone
else in turn].  HOWEVER, he got a lot of people interested in the set
and in drawing pictures of it, just at the time when drawing pictures of
it became technologically feasible, and for that he deserves credit.
(If anyone has the original text of this rant, please post it; my
vague half-memory does its invective no justice at all!)  Anyway, the
kind thing to say about this paper is that it is a stake in the ground,
which, in the event of unexpected timely success in solving some of the
deep underlying problems, might result in the author's name becoming
associated with a colorful visualization of the problem's superficial
face.  Except that even that won't work:  the group E8 has only real
representations, so his embedding of chiral fermions (particles with
parity) into it is either wrong, or requires some symmetry breaking
that cannot come from inside his model.  And the literature is littered
with E8 string phenomenology; if there's any mapping from his model to
sense, it must already be out there somewhere.

    Theoretical physicists are |==============================================
postulated to make the numbers | Joshua XXXXXXXXXXXXXXXXXXXXXXXX 
balance, not seen in the lab.  |==============================================

----- End forwarded message -----
-- 
Eugen* Leitl <a href="http://leitl.org">leitl</a> http://leitl.org
______________________________________________________________
ICBM: 48.07100, 11.36820 http://www.ativel.com http://postbiota.org
8B29F6BE: 099D 78BA 2FD3 B014 B08A  7779 75B0 2443 8B29 F6BE