<p>Conservation and restoration use concepts that have both empirical and ethical content. For example, “ecosystem integrity” is thought to both empirically describe a system and to be a property that is morally valuable. However, the idea that ecosystems have integrity doesn’t seem literally true. In this paper I argue that ecosystem integrity should be thought of as a scientific idealization that has been useful to simplify complex systems to better theorize, gather data, and model the systems of interest. However, if ecosystem integrity is a scientific idealization, then what should we make of the normative content of the concept? I argue that ecosystem integrity can also be thought of as a Galilean normative idealization—an action guiding tool that was helpful to simplify the complexity of the moral value of ecosystems. I also argue that while this kind of normative idealization may have been useful in the past, its potential pitfalls mean that it is time to de-idealize ecosystem integrity and eliminate this concept from conservation and restoration.</p>