Introduction

This chapter differs from most others in this book as it is based on reflections that inform plans for my doctoral studies rather than analysis of data already collected. It also differs in taking two separate areas for consideration. The first of these is my own main interest in the development of identity during the mid-teenage years, both in terms of narrative identity (developing a sense of auto-biography) and identity-in-action (developing a sense of self through social relations and ‘performance’). The second area focuses more firmly on transitions and specifically what happens in the process of maturation and creating – or at least assuming – adult identities. There are relatively few biographical studies offering close up examination of young people and their involvements in crime, and, other than Giordano et al (2002; 2007) long-term qualitative studies starting in adolescence are surprisingly thin on the criminological ground. Nevertheless, there are valuable insights from UK-based research with young adults on Teesside (for example, MacDonald and Marsh, 2005; MacDonald et al, 2011)) and in Sheffield (Bottoms and Shapland, 2011; Shapland and Bottoms, 2011). More pertinently, Halsey and Deegan's (2015) recent longitudinal study of young offenders in South Australia follows their lives from the mid-teenage years onwards. I draw on each of these to consider the development of identities in transition and the salience of social context, specifically criminal justice involvements, in opening up or foreclosing possibilities for positive change.

Crime over the young life course

Both these areas, of course, can be related back to the classic pattern of the age-crime curve. It has been noted elsewhere that, with some variation, the familiar bell-shaped curve, showing an increase and then a tailing off of anti-social activities, is still apparent whether the graph is plotted by age, the number of young people offending (prevalence), or the quantity of offences committed by young people (incidence) (see, for example, McVie, 2004). Bottoms and Shapland (2011) also suggest that charting the frequency of offending over time for a given population of offenders tends to produce similar bell-curves to the more familiar cross-sectional analyses. And this is despite these longitudinal measures excluding young people for whom offending is one-off or of short-term duration.