Cyclical Causation

In Monday's class, we briefly brought up the possibility of cyclical causation.  On first glance this possibility could seem like an alternative to Aristotle's 'unmoved mover', or 'uncaused cause'.  However, unless certain seemingly impossible circumstances are in fact possible, it cannot be so.  Even a cyclical event - such as the swinging of a perfectly balanced pendulum, in a location devoid of all gravity, friction, and any other forces which might slow and eventually halt its movement - has to have begun at some point.  While in the above example the pendulum moves itself, it could not have always been moving; something had to start it moving, even if it can then continue doing so infinitely.  The only way, I think, for cyclical causation to work is if we accept the existence of time travel, whereby something could travel back in time and become its own cause.  As most of us do not accept the possibility of time travel, cyclical causation is not a valid alternative to the 'uncaused cause' theory - which is not, of course, to say that there are not other alternatives.