The rationale for this chapter is closely linked to the context of foreign language learning in English secondary schools, a context in which motivation to learn languages is recognised as being in need of development (Nuffield Languages Inquiry 2000; King 2003). Research shows that languages are increasingly becoming an elite subject, with fewer children from poorer families choosing to continue to learn them (Centre for Information on Language Teaching and Research, Association for Language Learning and University Council of Modern Languages 2003). Problems of motivation can lead to disappointing standards of achievement or, at worst, disruptive behaviour in the classroom, truancy and exclusion from school (Social Exclusion Unit 1998; Department for Education and Skills 2003, 2004). Faced with such problems, many teachers believe it is important to 'tighten up' classroom discipline in an attempt to control the pupils' behaviour. In this chapter, I will suggest, however, that teachers' attempts to increase control may be counterproductive and themselves lead to higher levels of disaffection. Drawing on research into learners' constructions of language learning, I explore why there is an urgent need for teachers to consider the links between learner autonomy and motivation theory in order to find ways of transferring control to the learners. I take the position that locating the problem of poor motivation in learners themselves is socially unjust. Blaming the learners or their families for underachievement or lack of motivation is problematic especially given the differential levels of achievement and engagement between children of different socio-economic and ethnic backgrounds (Gillborn & Mirza 2000; Department for Education and Skills 2002). Rejection of such deficit theories leads to a more critical perspective, in which the education system itself (curriculum, structures, etc.) is construed as the problem, clearly failing particular sections of the population. Drawing on urban education theory and research from a critical theory perspective, I have previously reconceptualised disaffection as a search for a voice in a context of disenfranchisement (Lamb 2000). Here I develop this reconceptualisation further by reporting on a broader ethnographic research project designed to privilege young learners' voices, and position them as experts in their own learning. These expert voices belong to children aged 14 to 15 (i.e. in Year 9 of compulsory education), learning French or German in a secondary school in a northern English city, where several years earlier a number of language teachers had introduced flexible learning, a system of classroom organisation which involved learners in making choices of learning activities from a bank of resources according to their own individual needs (Lamb 2003). I accessed the voices of these teenagers by means of focused group conversations (FGCs) (Lamb 2005), specifically designed as a tool which would offer young learners an inclusive and supportive framework for constructing and articulating difficult constructs. Four groups of six learners were involved in the research, with the groups organised according to levels of achievement and motivation (A1: low achievers, motivated; A2: low achievers, less motivated; B1: high achievers, motivated; B2: high achievers, less motivated). I transcribed the focus group conversations as accurately as possible in the local dialect, though for the purpose of comprehensibility by an international audience I am reporting them here in a slightly amended form. My main intention in this research was not to make generalised statements about the ways in which motivated and unmotivated learners construe learning (though interestingly there were broad differences between the groups in many of the conversations). Rather, I wanted to offer an environment in which young learners could feel comfortable about expressing their thoughts about language learning, suspending as far as possible the usual power relationships between adult and children. This was achieved by means of carefully devised FGC protocols influenced by a range of questioning techniques (Holstein and Gubrium's  'active interviewing', Tomlinson's  'hierarchical focusing', and Roy's  work on cognitive interviewing); varied activities such as 'concept mapping' (Powney & Watts 1987: 30), projective techniques (LeCompte & Preissle 1993: 164), drawing and self-rating scales; as well as critical consideration of my persona and role in the group, and attention to environment and atmosphere (Lamb 2005: 184-212). In the rest of this chapter, I will focus on issues of control which emerged in the focus group conversations and which are specifically related to motivational beliefs. Here I will make particular connections to general motivation theory rather than to the social-psychological approaches of Gardner and Lambert (e.g. 1972) which are specific to language learning. In doing so, I am building on the paradigm shift in motivation research in language learning triggered by Crookes and Schmidt's (1991) encouragement to re-examine the direction which had been taken by research in this field. An important feature of this shift is that it encouraged a greater focus on the practical implications of motivation research for the classroom (Dörnyei 1994, 1996; Oxford & Shearin 1994, 1996; Williams & Burden 1997). It also ushered in aspects of general motivation theory which relate to autonomy, such as in the work of Dickinson (1995), Ushioda (1994, 1996) and Dörnyei (1998). Of course, motivation is contextualised (Dörnyei 1994, 2001b; Vallerand 1997), and learner autonomy is similarly 'situated' (Murphey 2003), which means that my arguments in this chapter need to be understood within the specific context in which the research was carried out. Nevertheless, I hope that it is possible to reflect on the implications for any context, be it teaching English or other languages to children or adults. © 2009 by Hong Kong University Press, HKU. All rights reserved.