C++ has a broken definition of NULL as 0, which is a consequence of breaking the semantics of void * by not allowing implicit conversions from it to any other pointer.

C doesn't have that problem. And in fact, POSIX requires that NULL be defined as "the value 0 cast to type void *", which is followed by all C implementations that I know.

Only the most horrible C implementations would define NULL to 0. And I don't know any of them.

The rationale for the addition of nullptr was rather an unjustified fear of the C standard committee (WG14) of annoying some proprietary C implementation that was stubborn and noisy enough to complain about it, combined with the feeling of C++ programmers that are members of WG14 that C should adapt to C++ to make it easier for them to reuse C code.

You seem to know way more about this than I do. Go ahead and edit this answer (or just post another one)

Makes sense, I will write one. :)

alx‭ C is still flawed. There's the definition of a null pointer constant in C, which is clear. And then there's an ambiguous part regarding the NULL macro saying "...NULL which expands to an implementation-defined null pointer constant;". I think the intention was to say it is implementation-defined whether 0 or (void*) was used (or 0ULL etc). But the maintainers of Death Station 9000 may read that as "ok we are free to do #define NULL 1". This ambiguous state of things is what I think Stroustrup's main criticism against C was about back in the days when he defined null pointers in C++. And not the 0 vs (void*)0 part. Related: What's the difference between null pointers and NULL?