Most people think paradoxes tend to be impossible occurrences. How could two futures be created from a single problematic event? Take for example the self-detonating automaton; At 3 pm it sends a message to itself two hours before, at 1 pm. The message tells it to self destruct at 2 pm. If it self destructs, it never sends the message back in time telling it to do so, and if it never receives the message it will survive long enough to send it. Another example is the tachyonic anti-telephone. Bill and Debby are in different parts of the universe. So far that light speed isn't fast enough to hold a "real time" conversation. It's a good thing they have special phones to compensate for the distance between them. The phones send messages to the past, meaning they'll receive replies before they initially send a message. This leads to a series of backwards conversations that seem to break causality. If Bill already knows what Debby is going to say in response to a question, for example, what's the purpose of asking? And if Debby never receives the question, she never answers, prompting Bill to ask. Both possible outcomes have observers, making both real in slightly different realities. They're not parallel realities, but perpendicular, intersecting at just the right moment to break causality. That is, of course, unless there's more than one universe to hold possible outcomes. There's a myriad of simultaneously existing universes, but we can only view one at a time. That's what makes a paradox so problematic. While they can't happen in a single universe, they can easily happen over a multitude of universes, because the quantum paths are different in all possible realities. If this is true, then a paradox is simply a crux of different observable outcomes that happen at the same time, in different universes.