Research misconduct - does self-policing work?

The world of academic research has been described as The Republic of Science. This has various characteristics, one being that it operates on the basis of self-policing. This self-policing is believed to be effective in ensuring that academic misconduct is rare, generally low-level and selfcorrecting. Any serious misconduct, it is assumed, will be quickly detected by peer review and stopped. The risks of being caught and the severity of the sanctions that follow are presumed to be so great that few will be tempted to stray down this route. While plagiarism is increasing among students, such misconduct has not generally been seen as a major problem among academics. However, the case described here should force us to reconsider our preconceptions about the efficacy of self-policing.