There does appear to be one problem with the perspective you’ve addressed even operating under the assumptions you’ve mentioned (such as single timeline and free will). Namely, that reality itself is subjective, and if enough people believe a paradox didn’t happen, it didn’t. Doesn’t this put a LOT of emphasis on a person’s individual beliefs and perspectives (almost to a Copenhagen Interpretation level)? Or am I simply not following your case (i.e. it could be that the paradox did happen but our minds construct a narrative to cover our perception of what cannot be? This sounds like an odd hybrid of Terry Pratchett and H.P. Lovecraft.)

Oh, and, Kant is relevant to everything, Mike. Deal.

Kinda of the bastard child of Schroedinger’s Cat and Werner Heisenberg.

Kant has nothing to do with it.

That said, what about the Back to the Future or Terminator idea? That is, time travel wherein the past is actually altered really creates a branched timeline. So, taking the former as an example, when Marty went back in time and accidentally broke up his parents, he didn’t vanish from existence immediately. He created a new timeline where he and his siblings were never born. He then subsequently created a third timeline where his parents got together in a far more romantic fashion and were entirely different people. This requires distinguishing between the timeline of the world and the timeline of the person, but that doesn’t seem obviously objectionable.

Another way to go is to argue, with Kant, that time is a form of presentation not an ontological feature, in which case time travel is only a discontinuity in the order of the spatiotemporal manifold of perception.