Manson and O'Neill (2007) reject the use of autonomy as the underlying principle justifying the requirement for informed consent for biomedical ethics. Instead, they advocate a more practical theory of informed consent as a waiver of expected behavioural norms, and advocate a focus on the communication transactions that are required for informed consent to be effective. These normative expectations are legitimate expectations a person might have, such as an expectation that someone else will not suddenly start stabbing them with a knife. For surgery, however, this expectation would need to be waived by the person, because without the waiver of the expectation there is no consent to the surgical procedure. In order to make the consent informed, there needs to be successful communication of information and feedback between the parties involved. This idea is encapsulated in the concept of communicative transactions. The idea isn't without its criticisms, which I will address later, but it is an interesting and different approach that I believe is a lot more realistically achievable and easier to fit into current IT practice as a normative theory than the more aspirational autonomy-based theories of Faden and Beauchamp and Appelbaum.
Manson and O'Neill outline their reasons for rejecting autonomy as a justification for informed consent in bioethics. They claim that there are many problems with some current ways of thinking about consent, with the largest the disparity between the aspirational goal of autonomy that informed consent claims to attempt to achieve and the real life situations that show that this goal is essentially (although unfortunately) impossible. In the following sections I outline and expand upon their arguments, looking at how their theory relates to information technology situations (which are in some ways quite similar to the biotechnology situations they cover).