To be more clear, this flaw happened because way back when we thought UCS-2 was sufficient. Then UTF-16 happened and all these systems designed for UCS-2 had to be hacked up for UTF-16 instead.
To be even more clear, this happened because early versions of Unicode consisted of fewer than 2^16 code points, and this was thought to be enough. In 1996, Unicode 2.0 was published and allowed characters whose encoding required more than 16 bits. This necessitated UTF-16 which allows expansion up to 32 bits to represent characters in higher planes (and is compatible with most text that had been encoded in UCS-2).
It doesn't allow expansion up to 32 bits. How would that even be possible? Some of the bits are needed to indicate to use the second two bytes, so there's a hole in the range and it's less than 32 bits.
