When conservators strip centuries of grime, discolored varnish, and earlier retouching from a Renaissance fresco, they present the result as a recovery: the artist's original palette, we are told, has been returned to view. Yet the rhetoric of recovery obscures a genuine interpretive act. A fresco is not a fixed object awaiting excavation but a palimpsest, its present surface the cumulative record of the painter's hand, the slow chemistry of pigment and plaster, and the interventions of every prior restorer. To remove one stratum is necessarily to privilege another, and the decision about where to stop is not dictated by the object itself. The dominant school of modern conservation, which I will call the purist position, holds that the restorer's task is to arrest deterioration and eliminate all accretions foreign to the artist's execution, thereby exposing the authentic surface. Purists point, reasonably, to cases in which darkened varnish had been mistaken for deliberate tonal unity; once cleaned, such works reveal a chromatic brilliance that earlier critics, theorizing from the grime, had wrongly attributed to the painter's intent. The lesson, purists argue, is that our aesthetic judgments must be disciplined by material fact. But the purist conflates two distinct claims. That a given layer is not original is a claim about history; that its removal restores the artist's vision is a claim about aesthetics, and the second does not follow from the first. Consider the painter who, anticipating the mellowing of his colors, keyed them deliberately high, expecting time to temper them. For such a work, the freshly cleaned surface displays not the intended effect but a garish anticipation of it: the pigments as mixed, not as meant to be seen. Here fidelity to the material substance betrays fidelity to the intention, and the purist has no principled way to choose between them. I do not conclude that cleaning is illegitimate. The skeptic who would leave every surface untouched merely freezes one arbitrary moment - the grime of a particular century - and grants it a spurious authority. My claim is narrower: that no restoration is a neutral return to origins, and that the conservator, in deciding which history to erase and which to preserve, unavoidably authors an interpretation. The honest restorer therefore owes the public not the fiction of recovery but an account of the choices made and the evidence for them. This reframing has a practical consequence. If restoration is interpretation, then its results should be defeasible in the way interpretations are: documented, reversible where technique allows, and open to the revision that new evidence about a painter's methods may compel. The language of recovery, by contrast, presents each intervention as final, foreclosing the very scrutiny that the uncertainty of the enterprise demands. What conservation loses in the aura of objectivity it would gain in the candor appropriate to a discipline that, at its best, reasons under irreducible doubt.