In psychiatry, the level of research has been increasingly evaluated based on the hierarchy model which classifies evidence into different levels in recent years. Even in clinical research, systematic reviews and meta-analyses receive the highest regard, classified as Level 1, and case reports the lowest, classified as Level 5. This evidence-centered sense of value naturally grants priority to evidence-based clinical practice, which in turn can lead to under-evaluation of clinical experiences accumulated by our predecessors. With the value assigned to case reports lowered, practicing clinicians in academia can also find clinical experiences with real patients less interesting, which is concerning. Clinicians used to see individual patients, their pathology and the meaning behind it, which constitutes the essence of descriptive psychiatry, which provides basis for the traditional psychiatry; this attitude is increasingly rare to find in psychiatrists today. This climate in psychiatry has been reinforced by DSM-based diagnosis. The initial objective of DSM was to abandon the concept of disease and adopt operational diagnosis instead, and its clear goal was to provide improved reliability and thereby to provide practical utility in prevention and treatment, even at the expense of finding meaning in the sense of traditional psychiatry, i.e. validity. In those days many of the psychiatrists chose to use appropriate instrument depending on the purpose, including DSM, which was supported by post-modernism. However, once DSM, which was originally used mainly for research, walked in to the realm of clinical settings, demanding more focus on evidence, the concept of DSM started to be diffusive: this is most notable in DSM asking its users to find a meaning in the DSM-based diagnosis, which is operationalized, and further asking to see the clinical phenomenon in a categorical manner. I propose that we reconsider the basis of psychiatric diagnosis with an emphasis on gaining and accumulating knowledge from individual cases.
 <Author's abstract>