Cornell realism is a view in meta-ethics, associated with the work of Richard Boyd, Nicholas Sturgeon, and David Brink, who took his Ph.D. at Cornell University but never taught there. There is no recognized and official statement of Cornell realism (Brink's Moral Realism and the Foundation of Ethics comes close), but several theses are associated with the view.
Contents
Moral realism
There are suitably mind-independent and therefore objective moral facts that moral judgments are in the business of describing. This combines a cognitivist view about moral judgments (they are belief-like mental states in the business of describing the way the world is), a view about the existence of moral facts (they do in fact exist), and a view about the nature of moral facts (they are objective: independent of our cognizing them, or our stance towards them, etc.). This contrasts with expressivist theories of moral judgment (e.g., Stevenson, Hare, Blackburn, Gibbard), error-theoretic/fictionalist denials of the existence of moral facts (e.g., Mackie, Richard Joyce, and Kalderon), and constructivist or relativist theories of the nature of moral facts (e.g., Firth, Rawls, Korsgaard, Harman).
Naturalistic non-reductionism about metaphysics
Moral facts are natural facts. They fall within the province of the natural and social sciences. But while they are not supernatural (as in divine command theory) and they are not non-natural (as in Moore's Principia Ethica or Mackie's picture of a realist world), they cannot be reduced to non-moral natural facts. That is, while moral facts are natural facts and supervene on non-moral natural facts, they cannot be identified with non-moral natural facts (see, e.g., Miller's An Introduction to Contemporary Metaethics).
Non-reductionism about semantics
There is no reductive connection between moral terms and concepts and natural terms and concepts. This gives Cornell realists a simple response to the charge that you cannot have naturalism without naturalistic fallacy: namely, that metaphysical reduction doesn't imply semantic reduction. This usually goes with a Kripke-Putnam semantic story: moral terms and concepts pick out certain natural properties in virtue of those properties standing in an appropriate causal (social-historical) relation to our tokenings of the terms and concepts.