Ahhh, well if you're taking aim at Haskell somehow being "right" then I'm not going to say a thing. :)

I think as far as monads go I quite like your distinction drawn elsewhere between abstractions for mathematical sake and abstractions for human cognitive sake. I think ultimately it has a lot to do with someone's goals as a programmer and person as to which abstractions they should hold and realize.

I think, personally, monads have a very high power-to-weight ratio here, even outside of Haskell, but I've never been one to argue that one cannot live without them. Or even shouldn't.