This dissertation examines the enregisterment of a particular way of speaking, referred to by the preadolescent pupils in this study as e.g. gangster, slang, or perker language. This register is enregistered in opposition to a more standard Copenhagen register and associated with indexical values such as masculinity, toughness, academic non-achievement, and ethnic minority status. In order to explore the semiotics of this register I combine analyses of situated discourse, language ideologies, and macro discourses. My study focuses on a group of preadolescent pupils and combines ethnographic fieldwork with two sociolinguistic experiments focusing on how this way of speaking is perceived by a group of preadolescent pupils for whom this register is a vital part of their linguistic everyday.The aim of my study is to examine, who uses this gangster register, how, and why, and what stereotypical indexical values they associate with it. I also explore how they utilize this way of speaking in their expressions of identity and belonging, e.g. by using it both to signal in-group solidarity and to exacerbate (potentially physical) conflicts. With my combination of linguistic ethnography and social psychology I have amassed a wide range of data. The ethnographic data include: ethnographic observations, inter actional data (including Facebook data), retrospective data (peer group interviews and a class session focusing on a comedy sketch featuring the gangster register), and teacher interviews. The experimental data consist of two experiments: a guise experiment including 352 respondents and studying how the contemporary urban vernacular is evaluated in comparison with another Copenhagen register and a supplementary experiment studying to what degree the preadolescent pupils from my ethnographic fieldwork perceive the same guises as speaking gangster. The analyses of the different data point to the importance of a range of concepts in the enregisterment of the gangster register, chief among them ethnicity, authenticity, gender, and a fluctuation between cohesion and dissension. The experimental data show that this register has a high degree of recognizability, and that the more gangster the speakers are perceived to be, the less intelligent, wealthy, nice, and attractive they are perceived to be. The micro-analyses of the ethnographic fieldwork indicate that that while this register does not inherently index ethnicity and while it can be used to signal in-group solidarity, the pupils in this study repeatedly make ethnicity relevant in their everyday interaction and besides using this register in friendly, jocular exchanges, they also use it to escalate conflicts and amplify aggression.