It's almost the physical equivalent of Gödel's incompleteness theorem.
Consider a finite collection of atoms consisting of a human and an environment. For simplicity, assume these atoms move classically. The human is a subsystem, and since the entire system involves deterministically, there is only a limited number of configurations that the human subsystem can realize within the environment. If a "human" is what is necessary to comprehend reality, then there are aspects of the system that cannot be realized by the subsystem, since there are far more states available to the total system.
Then again, I could just be rambling.
Think of a fractal. http://en.wikipedia.org/wiki/Mandelbrot_set
This negation is just as hard to prove as the multiverse hypothesis.
Lack of evidence of the existence of other universes doesn't prove their non-existence, nor does it rule out their existence.
However, note that the multiverse hypothesis is simpler in its content. The statement that this universe exists, while all the other ones do not, goes one step beyond the statement that they all exist.
Is there no evidence that there are no other universes?
Mathematics shows us that there are: it shows us that entities exist which are not located anywhere in this universe.
For instance, the number pi. We accept pi as real, but where in this universe is it? It is not computable, so that even if we use all the particles in this universe as the beads of an abacus, it will not represent pi.
If this universe is all there is, then there is no pi; we must stop talking in terms such as "there exists a real number that gives the ratio of the circumference to the diameter of a circle".
(Of course, the existence of this one is shared between the multiverse hypothesis and its negation.)
* Well, various "physical" representations/encodings of this concept must exist in different people's brains.
Concepts can be reshaped. If pi is merely a concept, why can't we conceive it to be something else, like 4.0?
Pi is an idealised approximation of a certain kind of curvature in a certain kind of spacetime. Pi appears to have infinite precision only if you assume the curvature and the spacetime are perfectly smooth and have infinite resolution.
Given there's a Planck limit, it's unlikely the universe really is perfectly smooth. So Pi in the real universe is not very likely to have infinite precision.
That's where the concept of Pi as a Platonic example of mathematical perfection trips over the reality of physics as a messy thing where traditional math structures can only ever provide approximate descriptions.
Beyond a certain level of detail Other Stuff is happening, and you can't assume smoothness or infinite precision.
We don't know a whole lot about the Other Stuff yet. One reason is that math was built to work with smooth functions with infinite precision, and there's no equivalent math that describes the lumps of whatever spacetime is made of, or the relationships between the lumps.
pi doesn't help you find the distance around a circle in some curved space-time, but pi is what it is.
I think that the hypothesis will be quasi-proven when science hits the ultimate limit beyond which there is no explanation. This will show that the world just proceeds by arbitrary rules that are traceable to axioms. Once you have that, you have shown that it's indistinguishable from any other such a formal system. That is to say, that it exists in the same way: the world is no more or less real than, say, the plane of the complex numbers. After that, although you still don't have empirical evidence of other worlds, there is no short supply of imaginable worlds (such as in mathematics) which are just as real.
In other words, once it becomes obvious that the world is just "math all the way down", it will be as foolish to deny other existences as to deny the square root of -1.
That's another way of saying Godel's theorems didn't end Mathematics, mathematicians just redefined what Mathematics was, presumably so they could keep their jobs.
Only true if the box is not everything.
"Science is prediction, not explanation." - Fred Hoyle.