Deportation due to criminal activity is often viewed as a neutral administrative practice and has to date received little research attention. This study views the phenomenon as part of a broader field focused on regulating people’s mobility. It also looks at the balance between the state’s interest in deporting non-citizens who commit crime and the goal of protecting refugees. Deportation due to criminal activity is first discussed from five perspectives: as alien control, as punishment and the spatial separation of criminal “others”, as migratory movement and forcible repatriation, in relation to human rights and as a “second asylum hearing”, and as border practice. The study then examines deportation in district courts between 1973 and 2003, via a quantitative study of all convictions involving deportation. Deportation practice differs between persons who are and are not registered as residents. Residents are deported for more serious offences than others and increasingly often over time for crimes against the person and drug crime. The number of non-residents deported increases greatly from 1985, which cannot be explained by an increase in convictions or by legislative changes. The study finally examines the reasoning of courts on possible impediments to deportation when the person convicted had refugee or equivalent status. The court collected an opinion from the Swedish Immigration Board in 80 percent of such cases. The opinions are very brief, often identical for different individuals and seem to be based on general guidelines for different countries rather than the individual’s fear of persecution at sentencing time. In the other cases the court makes its own assessment of impediments to deportation, but the risks faced by those convicted are rarely discussed in the court judgements. In 17 cases, the individual was deported despite the Board’s opinion noting a risk of persecution.

During the 19th century the European states experienced a new kind of threat to their existence. The military threats from other countries were now accompanied by civilian threats that inspired mass protest, terrorism and other menaces to the established order. In Sweden, these threats were mainly seen as connected to the rising labor movement and to a growing number of foreign citizens. The aim of the dissertation is to examine surveillance for national security carried out by the Stockholm Criminal Investigation Department and its Police Bureau between 1885 and 1922. Apart from examining what specific surveillance methods that were used, the dissertation gives an answer to the question why the surveillance was carried out, and why it was carried out the way it was. It also discusses how differences and similarities between the surveillance in Sweden and other countries can be explained and how the surveillance between 1885 and 1922 relates to the corresponding activities during the latter part of the 20th century. The main conclusions are that there was a lack of formal rules regulating the surveillance, and that it therefore was based on the following of routines. The bureaucratization process that characterized the period influenced the surveillance, which came to be performed as a bureaucratic machine, characterized by a tendency of expansion. This meant that the surveillance activities were constantly expanded and became more and more extensive. The expansion is connected to the surveillance phenomenon, which could be said to have an unlimited scope. Furthermore, it is suggested that this specific historic legacy has affected the development of Swedish security police activity during the second half of the 20th century.