One way to look at it (and I am not sure if this is correct, but it may be what the essay author meant) is to not treat the `unreachable` as affecting the presence of the decision, but only the result of the decision. If `unreachable` was replaced by a normal statement, we'd have:

    if (argc <= 2)
        do_something();
    else
        return printf("%s: we see %s", argv[0], argv[1]);

So the `return printf` is executed when `argc` is greater than 2. If we remove just the body of the first branch:

    if (argc <= 2)
        ;
    else
        return printf("%s: we see %s", argv[0], argv[1]);

the same thing holds. And additionally when `argc <= 2`, control will move past the `if`.

Under this view, if the `unreachable` won't cause the entire removal of the `if`, the compiler will produce the equivalent of:

    if (argc > 2)
        return printf("%s: we see %s", argv[0], argv[1]);

    return puts("this should never be reached")

Again, I don't say this is the correct interpretation, but it is one possibility, that would have to be ruled out by other parts of the standard.