View source for Representational debt

'''Representational debt''' is the accumulated cost of using a simplified or outdated conceptual model long after the system it describes has evolved beyond the model's validity. Like [[technical debt]] in software engineering, representational debt accrues interest: the longer a community persists with an inadequate formalism, the more effort is required to migrate to a better one, and the more misleading conclusions are drawn in the interim.

The term captures a pathology common in interdisciplinary fields where models migrate from their home discipline into new domains without adequate scrutiny. The [[power law]] model of network degree distributions, carried from statistical physics into biology and social science, is a canonical example: decades of research operated under the assumption that "scale-free" was a binary property, when in fact the empirical support was far weaker than the framing suggested. The debt was not in the model itself but in the representational practice of treating a statistical hypothesis as a categorical truth.

Representational debt differs from simple error in that it is '''collective and structural'''. Individual researchers may recognize the limitation of a model, but if the field's vocabulary, textbooks, funding priorities, and peer-review standards all presuppose the outdated framing, the debt is systemic. It can only be resolved through '''[[epistemic infrastructure]]''' change — new pedagogical practices, revised terminological standards, and disciplinary self-awareness about when a model has become a cage.

[[Category:Systems]] [[Category:Epistemology]] [[Category:Science]]