Abstract: To write about clients is an established routine in countless institutional settings, regardless of the fact that clients themselves seldom feel that the produced texts mirror or summarize their experiences. But what, more specifically, is left unwritten when staff starts typing on the keyboard to insert a piece of daily life into the computer? This article draws on data on violent events in Swedish detention homes, covering, on the one hand, interview accounts collected by ethnographic researchers and, on the other hand, formal journal reports on the “same” event written by staff. The analysis of one case exemplifies what written versions of a violent ward drama omit or transform: staff members’ “separation work” of the fighting actors and their local manufacturing of accountability, the involved actors’ conflict explanations in terms of ethnicity, gang culture, and “the first blow”, young people’s way of linking their self-control to the institution’s privilege system, and moral emotions as well as the significance of crucial details in the depicted course of events. The argument is not that staff should merely improve their routines of documenting events to really cover these or other facets of social life that are left behind at a detention home. Rather, the article attempts to explore why and in what sense institutional writing is incompatible with more informal, personal, and local accounting procedures.
Keywords: casebook journals; detention home; ethnography; institutional texts; interview; producing clients; total institution; written records