From Hans Joas and W. Knöbl’s Twenty Lectures in Social Theory pg 254:
Luhmann thus conceives his functional-structural systems theory quite explicitly as a ‘systems-environment theory’ (ibid.), allowing him to extend his analysis of organizations beyond their internal mechanisms to include a broader context. This also enables him to drop one of the core hypotheses of traditional organization theory, which states that it is ultimately the organization’s internal goals or certain internal values which regulate what happens within it. Luhmann was to show that everything is far more complicated and that the many ties binding systems and subsystems to the wider environment rule out such a simple assumption
As they later put it, he was fascinated by “the procedures developed by formal organizations in order to assert themselves within an environment and set themselves apart from it, and in order to function in highly routinized fashion.” If I understand correctly his point is that a focus on goals/values obscures how the challenges of stabilising itself as a system within an environment is what really drives organisational dynamics. The articulation of values may be an element in that sometimes but taking it as our starting point fundamentally misrepresents the organisation. As they summarise, “The sociology of organizations must pay heed to this and must no longer assume that consistency and total stability are absolute system imperatives; rather, it must accept that systems need inconsistencies if they are to exist in an environment which can never be entirely controlled”.
Mark Carrigan 