Causality could be a generalization that breaks down in extreme cases, like Newtonian mechanics.

This is scary.

Why?

>I like to think that we're all time travellers, collectively collapsing the universe's wavefunction through will alone... And now I'm onto determinism so I'll shut up.

I like that. Though I would think it is not 'will' but 'life' that interacts with universal determinism, through physical expression of information over time. How that information will express itself (how gene-information will translate into physical realities in each moment) seems in the realm of the unknowable (like turbulent flow or other instances of Lorenz-y chaos in the Universe), and consciousness as one of those gene-expressions makes this even more loopy.

But I've wandered into Leary-esque fascination with faults of perception lately, which ought to stick to the human-describing world of psychology, so I should probably shut up too.

What's to say that we're not existing in a reality created by a time traveler?

A new-reality model of time travel only affects the viewpoint of the time traveler.

Novikov's self-consistency principle implies that we can change the past as long as we don't know it's been changed. (Well, maybe more accurately, the time traveller has "always" done whatever it was he did in the past.) So even if causality is a requirement, it isn't necessarily violated.

Likewise, lots of time travel in fiction actually has the hidden assumption of a second time dimension. If a second time dimension existed, causality could be maintained along that dimension while seemingly broken in the time dimension we know and love.

And there's always the possibility that causality isn't a given and is simply how we experience the universe. Just because it has always been observed and it makes intuitive sense doesn't mean it is always true. :)

We only have a very firm belief that causality is inviolate. I guess there isn't a way to prove that it is.

Who says your reality is the same as my reality already? :)