Is Scientific Modeling an Indirect Methodology?

Abstract

Imagine, quarks will be retained as a construct in all future physical theories. Do physicists need more than this kind of invariance to claim the "real existence" of quarks and believe in having a "direct representation" of them?

If we consider modeling not as a heap of contingent structures, but (where possible) as evolving coordinated systems of models, then we can reasonably explain as "direct representations" even some very complicated model-based cognitive situations. Scientific modeling is not as indirect as it may seem. "Direct theorizing" comes later, as the result of a successful model evolution.