This thesis is based on a research project conducted by the Norwegian Armed Forces named CAMO (Cultural Awareness in Military Operations). It is an investigation into how a virtual world can be used to support training in cross-cultural communication and cultural awareness for soldiers who are to be deployed to take part in international military operations. The informants were cadets in training to become officers, stationed at The Norwegian Military Academy. The data for the thesis was collected in a one-day experiment. Second Life, a multi user virtual environment (MUVE), was used as a platform in order to conduct a scenario-based role-play simulation related to cross-cultural challenges. The scenario was structured based on a set of mini-scenarios with accompanying learning goals, containing a selection of critical instances and situations that the soldiers could encounter during a deployment to Afghanistan and therefore should know how to handle. An Afghan village and virtual avatars were modeled in Second Life, and used to stage the role-play. I have created the following research questions: 1. How can Second Life support social learning? 2. How can Second Life support cross-cultural communication and cultural awareness in military training? The thesis is structured as a qualitative case study, in which we used a mixed-methods approach to data collection and analysis. Data collection methods were questionnaire, interview and observation. They were chosen to be appropriate in order to investigate how the soldiers learned about cross-cultural communication in a social context and how they evaluated the use of Second Life as a platform for teaching and learning about cross-cultural communication and cultural awareness. The interviews and observations were analyzed partially by using a grounded theory approach during data categorization. A combination of social learning theories was used to create an initial set of categories for the analysis that was later refined in a bottom up iterative process of data categorization. I have triangulated the questionnaire data with the interaction material and the interviews in order to evaluate these sources in relation to one another. We have found that Second Life supports aspects of interaction and learning in a social context by employing techniques such as perspective taking, reflexive practice and establishing common ground in communication. We also found that Second Life works well as platform for training in cross-cultural communication and cultural awareness by role-play scenarios where virtual representations of cultural artifacts are modeled and incorporate in the scenarios. However, we did come across some constraints that hindered the flow of communication. Finally, at the end, I present some suggestions for improvement to address the constraints.