All systems of social control require choices about the balance to be struck between encouraging or incentivizing good behaviour and sanctioning bad behaviour. The control systems in science are no different. In general, encouraging good behaviour is to be preferred. This generally tends to be more effective. In this particular instance, it also reflects the particular problems of controlling innovation, where the innovator necessarily knows more than the regulator. A virtue-based approach to ethics and integrity seeks to stimulate self-reflection and self-restraint such as to minimize the need for regulatory interventions. Where successful, the innovator anticipates and addresses challenges in advance so that problems are avoided
It would, though, be naïve to suppose that virtue alone would be sufficient to overcome the competing incentives offered by the reward systems in science, for individuals and organizations, to be less than scrupulous in the conduct of research and the treatment of human subjects. The control system must, then, also have an element of deterrence and sanctioning. While this exists for the protection of wider social interests in preserving confidence in the reported outcomes of research, and in respect for human life and dignity, it will only command confidence if it also meets accepted standards of natural justice for those accused of misconduct.
In the design and development of a fair and acceptable control system, it is likely that the following issues will need to be dealt with: