@Caledonian—as you say, there’s no a-priori reason to believe that a thing composed of predictable parts must itself be predictable.
That is not what I said, and it’s not what’s true. Something composed of only predictable parts is predictable itself, because there’s no place for unpredictability to enter in.
The problem is in determining whether something has only predictable parts.
The problem is in determining whether something has only predictable parts.