[boost] [thread] Question about MT API C++0X N2094

I'd like to know the rationale behind mutex concept requiring explicit locking signatures. Are users of the library "allowed" to access them directly? If yes why? Isn't this redundant to locks? Not requiring them, but letting them be implementation defined will allow the implementor more freedom. The only required concept should be: being able to take the address of the mutex. (Because this makes it unique.) E.g. the requirement for these N2094 mutex concepts would make implementation of my proposed POD mutex variant impossible. E.g. the current boost mutex also does not need to make these explicit functions. Sorry if I misunderstood the document. I would be glad if someone (H. Hinnant?) could explain to me. Thank you. Roland