The police, in particular the riot police, can be a rather inaccessible object of investigation, whose reservations towards research are analysed with reference to five barriers: 1) police control of access to the field, 2) the doubly asymmetric research relationship, 3) attempts by the police to steer the process, 4) the sceptical attitude of (potential) interviewees, and 5) the restrained discussion behaviour. However, what appears as a hurdle from a researcher's perspective allows structures

more »

... of the object itself to be reconstructed. These include a prevalence of narratives of police "innocence" and "powerlessness" with which resistance against external aspirations for control is buttressed. The police view themselves as constantly being under public scrutiny and being unjustly publicly criticised. In this manner the predominant attitude towards research is reserved if not hostile. The police definitional power in its fields of action is thus partially transferred to research on the police. However, police interference has its limits, and counterstrategies will be set forth. Most data used are from a grounded theory methodology (GTM) project on video surveillance and countersurveillance of demonstrations, based primarily on group discussions and expert interviews with riot police.