“Gosh, I wished Garfinkel had never used the term ‘experiments’!”—This, over and again, has been my line of chagrin expressed at the recurring, yet fleeting, all too narrow and often misdirected attention given to Studies in Ethnomethodology (1967). Where does the chagrin originate? The title and subtitle above hint at the answer developed in this postface, a postface which, ironically, will be closing this volume dedicated to Garfinkel's “experimental legacy.”

That being said, the “experiment” that I became most interested in is one that was mentioned, I believe, in Garfinkel's notes for a course called “normal environments.” For this experiment, Garfinkel apparently requested his students to go into a public toilet, close the door behind them and then bang on the door. The results must have been intriguing, although the “toilet experiment” wasn't mentioned again—if it ever existed. Mostly though, I deplore the rendition of Garfinkel's ethnomethodology in terms of “experiments” or, worse, “breaching experiments.” In the two or three years following the publication of Studies, this led to crass interventions at least in British sociology which, more than anything else, were often just pranks, nurturing reductive, superficial interpretations of “ethnomethodology” (e.g., as a methodology for “social psychology” or “applied social science”).

What is the alternative? Among the various ways of practicing ethnomethodological research, there are “conceptual investigations,” taking inspiration from Wittgenstein and Winch, as much as Garfinkel and Sacks (e.g., Hutchinson et al. 2008). What is to be said on Garfinkel's “experiments” from the stance of conceptual inquiry, for which this postface pleads? First, it is worth recalling that Garfinkel devised his experiments as “classroom demonstrations” to teach elementary aspects of sociological concepts (Garfinkel 1956). Against this backdrop, the recurring interest in sociological theory for “breaching experiments” has turned them into a hugely inflated topic, only rarely if at all connected to Garfinkel's pedagogical uses of them. Second, a general note of caution is in order. What is to be understood as an “experiment” in the first place? And, depending upon that understanding, what would be the precise nature of experimental results? What would these results, and the materials they draw upon and make available, allow us to claim?