'Feedback is the transmission and return of information.'
Hmm... Apparently we're ruling out things that sound that technical.
How about...
Forrester (1961) said:
An information feedback system exists whenever the environment leads to a decision that results in action which affects the environment and thereby influences future decisions.
One can shorten that a little to:
'A feedback loop exists when decisions change the state of the system, changing the conditions and information that influence future decisions.'
Richardson and Pugh (1981) said:
A feedback loop is a closed sequence of causes and effects, a closed path of action and information.
Don't these suffice, so people can get on with using the notions of feedback and circular causality to improve thinking? I would hope the less time spent on defining the concept of feedback loop the better.
I'd really prefer a diagram to words. E.g., how would you define a 'dollar bill'? Really tough without pulling one out of your pocket and showing it to somebody. I'd say it's easier to define a feedback loop than a dollar bill, but still a picture would help a lot. Drawing the circular thinking involved the prejudice loop or filling a glass gets across the feedback idea very quickly.
It's very important to me to link the idea of feedback loops to the more fundamental idea of endogenous thinking. If we try to think about endogenous causal chains responsible for some set of changes over time, we're forced to think in loops. However you define them. So feedback loops are causal chains that somehow double back upon themselves. They result from thinking causally within a causally closed boundary. That endogenous thinking is why feedback loops are important -- no, crucial -- to thinking reliably and powerfully in complex systems.