Persons adapt to systems until such time as they choose not to, and then the systems fall apart

From When Rome's system falls apart: the repersonalisation by Michael G. Heller.

In the Rome introductions I proposed some plausible claims about the society of ancient Rome. In contrast to most other historical societies Rome’s governance was organisationally differentiated by categories of participation. Participation operated as a system of structured interactions in the absence of any single power centre. System interactions were organised through relationships between the power zones of senate, magistrates, and assemblies. The operative system interactions were individualistic and diffusively clientelistic.

Systems are inherently impersonal. As explained in my theoretical introductions, whole governance systems, which are rare, are always programmed as instructions to give preference to an impersonal rule over a discretionary choice. In practical terms, systems operate regardless of personal preferences and personal discretion. Rather, persons adapt to systems until such time as they choose not to, and then the systems fall apart. Patron-client exchanges, which are highly particularistic and personalised in operational terms, can, in the impersonal aggregate, be a long-lasting generalised system of interactions based on known or intuited rules about how one must conduct personal relationships. Sharing as they might the same programming code (a binary choice when an immutable need to preserve society takes precedence over immediate individual need) individualistic-clientelistic systems can work in tandem with systems of organisations. This is what we find in Rome where both intra-elite and aristocratic-plebeian clientelist relations coordinated decision making about shared and core governance issues affecting relations between the subsystems (i.e. the competing zones of organisational power) constituted as perpetual participation mechanisms.