A quite interesting paper on the topic is McMonigal & al. [1], it is a bit technical but they discuss the Alcubierre warp drive quite nicely.
To elaborate on the second paragraph about special relativity, one way to define Lorentz transformations is, that they are the transformations which preserve the causal structure of Minkowsky space. On the other hand, the speed limit is a direct consequence of the Lorentz transformations. So the causal structure and the speed limit are directly related. To look again at the argument that a velocity larger than the speed of light implies causality violation: The moment you argue that a speed faster than the speed of light is possible, you implicitly change the meaning of causality. Then you use the Lorentz transformations to conclude that in on frame of reference event A precedes B and that event B precedes A in another. So you first deny the special status of the Lorentz transformation, just to go on and use Lorentz transformations. And my take on this is a strong maybe, the moment you start talking about local Lorentz invariance violation, anything goes. Perhaps there is causality violation, perhaps not, but this depends on your specific meaning of 'causality' and 'faster than light' and not very much on special relativity.
The other thing is the "sort of" parenthesis, the Lorentz transformation is directly linked to the causal structure of special relativity. A statement like "if* you could go faster than the speed of light, then you could violate causality" is somewhat nonesensical, since if you could go faster than the speed of light, then you would break local Lorentz invariance and therefore you break your very definition of causality.