Any demonstration I've seen of using FTL signals to violate causality seemed to involve an exchange of FTL signals. Would FTL travel be consistent with causality if it always involved motion in one general direction?
For example, call the direction toward the Andromeda Galaxy from here (and continuing beyond in the same direction, of course) "universal west." You could go as fast as you want to if you're going west, northwest, or southwest, but not an inch east.
I think this gives up "Lorentz invariance," but that's OK. I can't remember getting any Christmas cards or birthday gifts from Lorentz invariance in, like, ever.
EDIT: I'm not worried about things appearing out-of-order to an external observer, but really only worried about preventing FTL signals from being able to directly or (as in every case I've seen) indirectly send a signal into one's own past. The "universal west" direction is arbitrary but would be the same for all observers.