There is broad agreement among research ethicists that investigators have a duty to obtain the informed consent of all subjects who participate in their research trials. On a common view, the duty to obtain this informed consent follows from the need to respect persons and their autonomous decisions. However, the nature of informed consent and the demands it places on investigators are open to dispute and recently have been challenged. Respect for persons, it has been claimed, does not require investigators to guarantee that the subjects enrolled in their trials comprehend the risk/benefit information disclosed to them or even that they appreciate the difference between research and therapy. According to this critique, the significance of defects in informed consent, like therapeutic misconception or unrealistic optimism, has been greatly exaggerated. This article reevaluates informed consent in clinical research in light of this critique. It not only rebuts the main points the critics raise, but also shows that other points they raise can be accepted by a doctrine of informed consent that resembles the common view in maintaining that autonomous authorization is central to informed consent.