the MPTC Dilemma (please solve)

[Ashley Yakeley]

Simon Peyton-Jones wrote:

> Of course -fallow-undecidable-instances still lifts all restrictions,
> and then all bets are off.

Is the behaviour of GHC with -fallow-undecidable-instances (and 
-fcontext-stack) well-understood and specifiable? It is a very useful 
option, as you can join (meet?) classes like this:

   class P a
   class Q a

   class (P a,Q a) => PQ a
   instance (P a,Q a) => PQ a

-- 
Ashley Yakeley