Razor and the Multiverse

I was wondering if anyone here had any opinions about Occam's Razor and a multiverse scenario. I have seen several arguments that Occam's Razor is pro-multiverse. A piece from Victor Stenger (http://www.colorado.edu/philosophy/vstenger/Cosmo/ant_encyc.pdf):

"

Several commentators have argued that a multiverse cosmology violates Occam's razor.

24

This is wrong. The entities that Occam's law of parsimony forbids us from "multiplying beyond

necessity" are theoretical hypotheses, not universes. Although the atomic theory of matter

multiplied the number of bodies we must consider in solving a thermodynamic problem by 10

24

or so per gram, it did not violate Occam's razor. Instead, it provided for a simpler, more

powerful, more economic exposition of the rules that were obeyed by thermodynamic systems."

Also, I've seen Max Tegmark argue on similar lines before. Any thoughts?

Occam's razor is a rule of thumb. Its logical basis can be argued thus. There are very many complicated theories, and for one of these to explain a set of empirical facts is only weak evidence that it is correct (the agreement is more likely to be by chance). There are few simple theories so that if one of these is consistent with the facts it is stronger evidence that it is correct.

[This argument is my independent view. Without reading anything about Occam's razor, I can confidently state that someone else deserves credit for thinking of it decades earlier].

Thank you for your view, Elroch. However, I feel (and many do) that it is a safe bet to rely on Occam's Razor fairly far out. A classic example would be Bombieri and the Riemann Hypothesis. Don't have a link for this, sorry!

my opinion is that the multiverse theory is ridiculous.  What is the evidence?

I am not here to argue for a multiverse. Furthermore, one can argue for either Cosmological Darwinism, Parallel Universes, or Braneworld Scenarios. Look up Max Tegmark's arguments if you want as well as Lee Smolin.

Haaaaaaaaaaaaaaaaaaaaaaaaaaaaaa!

..

i think [positively] about "..a multiverse scenario.. "... but:

[a snippet from wiki article re Occam's razor:]

"...other things being equal...the simplest hypothesis proposed as an explanation of phenomena is more likely to be the true one than is any other available hypothesis, that its predictions are more likely to be true than those of any other available hypothesis, and that it is an ultimate a priori epistemic principle that simplicity is evidence for truth. [~ R Swinburne 1997.]"

strangequark and other devotees of Occam's razor, the history of science shows that Occam's razor often misleads. There are many examples of simpler theories which adequately explained the facts turning out to be approximations to more complex theories which were needed to explain subtle deviations. Unless you are happy with theories that work only approximately, this proves that Occam's razor can mislead.

A good example is the replacement of Newtonian gravitation with General relativity. Newton's theory is indisputably simpler, and adequately explained the experimental data until the precession of the orbit of mercury was observed, but it turned out to be a low speed, short distance, low mass approximation to Einstein's theory.

I think I understand your position, Elroch, for me this is more of "what looks best or simple right now", not necessarily always or not to be supplanted by future theories...I feel like my statements here are being thought of as stronger than they actually are, although this can be reasonable to assume because I did not say everything at once.

A common mistake in applying Occam's Razor is assuming that the simpliest explanation is the one with the fewest apparent components. This is not true.

Take Newton's theory and Einstien's theory.

Einstien's theory has more compents than Newton's theory, but it is not more complex because it is simply applied to a wider range of scenarios (cosmic and atomic level gravitation) than Newton's theory, which can only be applied to those scenarios if one were to make a whole number of assumptions, which themselves are components of the theory.

Ultimately Einstien's theory is the simpiler of the two, because it can be applied to more scenarios without modification or assumed behaviour.

I've heard the "gravitational argument" from Elroch before. I'll I can say now is yes, maybe general relativity is truly discriminated by Occam's Razor. That is to say, we know it is incomplete because of quantum gravity. One may have a more simple theory or quantum gravity with less components: for example, so many important equations pop out of Riemann's Metric Tensor with extended dimensions in string theory. I'm not so sure that the above argument works.

A highly relevant point is that the status of the empirical facts influences what is the simplest theory. Newton's theory did not need a replacement until two pieces of empirical data were found which disagreed with it, one of which had nothing to do with gravitation at all! Apologies to the large number of people to whom this story is a bit of a bore - do skip to the next thread.

The first was that the speed of light was a constant, independent of the relative motion of the observer. This refuted Galilean invariance, which no-one had doubted before. In hindsight this is surprising, as Galilean invariance also meant that Maxwell's beautiful and complete theory of classical electromagnetism was was only true in one particular stationary frame, which seems silly now.

The simplest symmetry theory of space time consistent with a constant speed of light is of course Lorentz-Fitzgerald invariance, i.e. special relativity.

Then the last empirical fact was of course, the precession of the orbit or mercury. Tough cookie that one, for believers in Newtonian gravity.

And perhaps space-time being a manifold which is locally the same as flat special relativistic space-time is indeed the simplest theory possible that explains the final empirical fact, in some sense.

If one is talking about the status of empirical facts clearly one needs a unified theory for quantum gravity...if string theory (or some other beautiful mathematical theory) does so, then it appears to me that nothing is being accomplished by saying Newton's theory of gravity is more simple=correct, using your point.

Physicists generally seem to be of the opinion that gravity should be quantised, but I am not sure this is required by the empirical facts, at least not in a direct sense (like the above empirical reasons why GR was needed). The way in which I would be inclined to combine QM and GR is to merely change the theatre of QM from flat space time to the manifold of GR [perhaps like indicated in this article http://en.wikipedia.org/wiki/Quantum_field_theory_in_curved_spacetime which is all Hawking needed to apply QM to black holes ]

So is there actually empirical evidence that precludes gravity and everything else being intrinsically different, with GR providing the canvas on which other physics occurs? I suppose the simple observation is that if gravity is caused by the mass and energy of individual particles and the individual particles are quantised, you can't get around needing a quantum theory for gravity. Is this the essential (indirect) reason that everyone believes gravity is quantised (including me)?

What do you mean here (you may just be over my head, can't tell):

"So is there actually empirical evidence that precludes gravity and everything else being intrinsically different...."

Could you please rephrase that for me? Sorry, I need it "dumbed down" .

Probably what you need is it "intelligented up".

My point was this. For other physics, there is loads of direct evidence that behaviour is not classical (eg the photoelectic effect, twin slit experiments of various types, magnetic moment of the electron), but we have no direct evidence that gravity is not classical (we are still struggling to show gravity waves exist as predicted by Einstein, and as far as I know there is no even plausible experiment to show that gravity is quantised.

However, my picture of quantum physics taking place in a curved space-time of general relativity is ok until you think of what causes the curvature. It's particles which obey quantum theory, and if these are obeying quantum laws, how can the gravitational field be entirely classical? For example, if you have a superimposition of two states where there either is or is not a particle in a particular region, presumably you need a superimposition of two states of the gravitational field as well.

As far as I know, we have no way to detect the subtle differences in gravity due to such quantum effects - on a large scale the uncertainty is usually insignificant. [ This is what my unclear question was about ]. So is the above common sense argument the reason most people are sure gravity is a quantum theory as well?

I'm fairly sure that a widely accepted reason to accept gravity as quantum is the fairly simple basis of the graviton which must be quantized...Elroch, have you heard of LISA?

But what experiment makes the existence of the graviton necessary, strangequark? It is easy to describe experiments that show that light is quantised and electricity is quantised (and there are beautiful experiments explicitly showing the quantisation of the magnetic field). But for gravity, I don't think anyone has even described a plausible experiment that could do likewise.

For when I don't understand everyone's arguments, here are some reasons ppl have given why gravity should be quantized:

http://golem.ph.utexas.edu/~distler/blog/archives/000639.html

http://www.physicsforums.com/showthread.php?p=856677#post856677

http://tigger.uic.edu/~huggett/Nick/My%20Work_files/why.pdf

All Wikipedia says is: "Motivation for quantizing gravity comes from the remarkable success of the quantum theories of the other three fundamental interactions. Although some quantum gravity theories such as string theory and other so-called theories of everything also attempt to unify gravity with the other fundamental forces, others such as loop quantum gravity make no such attempt at unification, they simply quantize the gravitational field while keeping it separate from other force fields." (http://en.wikipedia.org/wiki/Quantum_gravity)

Excellent links, strangequark. My first reactions are

(1) wow - to the one from the string theory lecturer, which is like music to my eyes, but so far beyond my knowledge that I can get little more.

(2) stingray and garrett in physicsforums agree on exactly the same simple to understand reason that I gave why gravity cannot be classical

(3) is at a first glance intriguing as it says, contrary to what I believed, that there may be experiments that could detect quantum gravity effects in the near future, using the large scale universe as apparatus.

[If I had heard of LISA, I think it must have slipped my mind. I followed LIGO with interest for several years, and it looks as though I haven't missed a huge amount there!]