In this article we analyse the significance of silence in qualitative interviews with 36 individuals interviewed about high-risk, illegal activities. We describe how silence expresses a dynamic power relationship between interviewer and interviewee. In the analysis, we focus on two different types of silence: ‘silence of the interviewee’ and ‘silence of the interviewer’. We analyse how silence functions as an interviewee’s resistance against being categorized as ‘social deviant’, how an interviewer may use silence strategically, and how silence stemming from an interviewer’s perplexity constructs significant data. We conclude that silence constitutes possibilities for interviewees and interviewers to handle the complex power at play in qualitative interviewing either by maintaining or by losing control of the situation.