Argyris Double-Loop Learning L2: Question governing values, norms and objectives (double-loop learning)

What does L2 require?

When errors recur, when problems are non-routine or threatening, or when stated objectives and actual practice diverge, the organization examines and where needed changes the governing values, norms, policies and objectives themselves and then the actions, rather than only trying harder within them. Argyris's Product X case shows people correcting production errors (single-loop) long after some knew the question was whether the product should be made at all (double-loop). Double-loop learning always needs an opposing set of ideas against which to measure existing assumptions, and incongruities between what the organization espouses and what it does are open to challenge. Where it is absent, change arrives only through crisis, revolution from within or without, or a crisis manufactured by management, which comes late, teaches alert people that their loyalty is not valued, exhausts the organization and reinforces the conditions that blocked learning.