Blokland and Soenen use ethnographic research of tramcars in Antwerp to discuss how incidental contacts in public transport relate to safety and trust in the public space. They argue that anonymity, although often blamed, is not the culprit. Blaming certain categorically labeled groups as 'problematic' is not a fruitful approach either. Through an analysis of social realms (public, private and parochial, as in Lofland) that customers create in interaction with each other, the authors show that public transport incorporates potentials both for 'thin community' and for conflict and anxiety. Whether people experience one or the other depends on the social trust. The key to understand how such trust can grow and decrease cannot be found in crime statistics, individual attitudes or categories of 'problematic groups', but in the constructions of the social realms through which people opt out of the public realm or, in groups, appropriate the public space at the expense of others. Describing the various services of 'Lijnspotters', drivers and controllers, the authors discuss what types of social control in public transport is most likely to enhance social trust.

To live in a safe neighbourhood is to live in a neighbourhood where one knows what to expect and can navigate the public space. This article argues that issues of public safety in urban disadvantaged neighbourhoods are often understood as depending on crime and social control; but it may well be that whether or not people feel safe depends just as much on the degree of public familiarity of the context in which they live – and consequently on their abilities to know who to trust and distrust – as it depends on the usual suspects of crime rates, social cohesion, broken windows and collective efficacy. Using empirical data from four neighbourhoods in Rotterdam, the Netherlands, the article shows that data may substantiate this claim, formulates some policy recommendations and proposes an agenda for further research.