I'm the man in the box...
Society is boxes.
At least, that's what Talcott Parsons claimed. His concept of functionalism had society continually striving towards an equilibrium. He sought to identify the necessary parts to a functioning society. So far, so good.
Except that he wanted to preserve those bits of society in place. Men as the head of the household, for example, because that was the way it was done. That's the first of the big criticisms of Parsons. The other two are that his concept was a tautology, and that he gave no mechanism for the generation of either the structures or the deviant procedures.
The first is Parsons's mistake between confusing the goal and the mechanism to reach the goal. Parsons made a good argument that multiple roles and needs must be met in a family unit. His mistake is in presuming that there is a real reason that only certain individuals can take certain roles. There is no allowance that the role a confessor priest plays can be substituted by a therapist, for example. The goals he describes and the needs that are modeled - those do seem to be pretty valid. How those concepts and needs are filled... well, that's a different story.
And that brings us to the second criticism. Parsons' concept is a tautology, because it's "just a model". It has a great deal of usefulness, but it cannot be confused with reality. Models are by definition simpler than the reality they try to describe. Horribly useful - but you've got to know when to hold them, and know when to run.
The final bit might be solved by the concept of memetics - a field of study that ascribes the attributes of genes to ideas. Ironically - because Parsons was originally a biology wonk - the mechanisms of evolution serve well to describe where deviance and the institutions that fill a society's roles come from. Random chance keeps trying until something sticks.
Perhaps the reason this hasn't been more widely appreciated is this: Self-described functionalists tend to not like change, and also like the idea that actions have intent. The notion that human society is not an intended innovation, but rather a hodgepodge collection of lucky accidents and winnowed-out problems seems repulsive.
Yet it explains so much more. Natural selection only - only - removes attributes that decrease an organism's ability to continue the race. That's why the appendix might be vestigal, but isn't gone. It's not a big enough drain on our individual resources to matter. Likewise, any part of a society that doesn't actively disrupt it can persist indefinitely... even if it's not beneficial.
These tweaks to Parsons's concepts can rescue the model of modern functionalism into something useful - but in doing so, undermine the often-unspoken values of functionalists from Comte onward.
(Can you tell I'm reading theory again?)