Professional socialization and training, informal peer influence, and professional norms and sanctions were once regarded as effective tools in the social control of medicine. However, these intraprofessional controls over medicine have come to be seen as ineffective and new extraprofessional controls have been considered requisite for several decades. In this paper, I focus on the historical development of governmental controls over biomedical research in Japan. A general characteristic of Japanese regulations is the existence of numerous subdivided guidelines and laws. This situation was prompted by the governmental response to bioethical issues on a case-by-case basis. While this response has advantages in speed and flexibility, it also has disadvantages in that it lacks consistency, and frequent revisions lead to complicated provisions that cannot be understood by researchers. Another feature of the Japanese situation is excessive reliance on non-binding guidelines published by various administrative agencies. One reason for the government to rely on these guidelines is that they are formally non-binding but regarded as “effective” measures because regulated parties perceive them as having binding power. Indeed, some guidelines have their own sanctions against researchers or institutions, which include public announcement of the names of violators. However, if this measure is actually taken without basis in the law, it can be illegal. In this case, Japanese experiences show that excessive reliance on “soft” regulations can lead to paradoxical situations concerning social control of biomedical research.