In studies on childhoods in the Global South, which are influenced by conceptual considerations on the social geography of children’s worlds (see Holt, 2011; Kallio and Hakli, 2015; Kraftl and Horton, 2018)13, and under the influence of feminist and other relational action theories, the key concept of agency is (self-)critically discussed by childhood researchers. In an anthology on this subject (Esser et al, 2016), the editors explain the ‘almost dogmatic insistence on agency and its constitutive importance for Childhood Studies’ (Esser et al, 2016: 2) as a critical response to the adult-centricity of almost all previous research on children and childhood. This research had seen children primarily as a result of socialization processes and as an appendix to a family. In contrast, agency has now been discovered as a specific property of children, which allows their emancipation. Their voice should be heard and their subordinate and marginalized position overcome. The ‘new social movements’ (see, for example, Laraña et al, 1995) and the Ariès’ (1962) perspective on childhood as an invention of modern bourgeois society, or, in the words of Shulamith Firestone (1970), as an ‘oppressed class’, had a great influence on this new orientation.
Today, the agency concept is once again under criticism. The previously dominant concept of childhood as a development stage, with an adult imagined as perfect and the child imagined as vulnerable by nature, has been replaced by an essentialist version of agency. This has been hypostatized, on the one hand, as an anthropological fact opposite to the idea of the vulnerable, developing child, and, on the other hand, as the most advanced expression of ‘modernity’ (see Esser et al, 2016). By imagining the child in an absolute manner as an actor in itself, the link to the biological basis (body) and the social conditions of the life of children were lost or not sufficiently observed (see Prout, 2000). The relation to the generational order, another key concept of childhood research, has also been lost where the actions of children can have reproductive as well as transformative functions (see Närvänen and Näsman, 2007). This is now connected with the basic question of how children are individually and collectively positioned in different social contexts, making it necessary to introduce not just one, but different childhoods.
Another critique comes from feminist-oriented care concepts, which had already conceptualized interdependent relationships against the conceptions of an autonomous subject (see Wihstutz, 2016; Cheney, 2018). A theory of agency cannot simply assign the fictions of autonomy attributed to the (male) adult to children and thus negate their dependence on other people who care about them. In this context, relational social theories, in particular the Actor-Network-Theory founded by Bruno Latour (1993; 2005), have also had an influence on the re-conceptualization of agency (see Oswell, 2016; Raithelhuber, 2016; Spyrou, 2018). They are based on the assumption that agency is not an inherent personal property, but is always inherent in and interwoven with social relationships. Instead of hypostatizing agency as a quasi-natural property, it must be seen as part of a complex network of different human and non-human actors.14 This could also refer to the concept of ‘multidimensional agency’, which Daniel Stoecklin and Tobia Fattore (2018) formulated with reference to the Capability Approach (Sen, 1999; Nussbaum, 2011). According to them, ‘agency is constituted intersubjectively, within parameters set and enabled through structures represented in social space. Children’s agency is thus always constituted through constraints and opportunities, whether they are accepted, negotiated or resisted’ (Stoecklin and Fattore, 2018: 15; see also Stoecklin, 2013; Larkins, 2019).
The new theoretical reflection on the concept of agency is remarkably related to non-European contexts. In the past few years, various concepts of agency have been formulated in order to analyse, in accordance with the concrete conditions of life and in a culture-sensitive manner, the ability of children in the Global South to act. These concepts are to be critically appreciated here and linked with my own thoughts, which also form the basis of the following chapters.
Most attention so far has been paid to a concept developed by Natascha Klocker (2007) in research with children from rural areas of Tanzania working as domestic workers in homes of more or less well-off families. Klocker (2007: 85) distinguishes between ‘thick’ and ‘thin’ agency:
‘Thin’ agency refers to decisions and everyday actions that are carried out within highly restrictive contexts, characterized by few viable alternatives. ‘Thick’ agency is having the latitude to act within abroad range of options. It is possible for a person’s agency to be ‘thickened’ or ‘thinned’ over time and space, and across their various relationships. Structures, contexts, and relationships as ‘thinners’ or ‘thickeners’ of individuals’ agency, by constraining or expanding their range of viable choices. Between ‘thick’ and ‘thin’ agency there is a continuum along which all people (including rural young people) are placed as actors with varying and dynamic capacities for voluntary and willed actions.
The author justifies the concept by the fact that it was difficult for her to ignore the pressure on the girls, which is caused by poverty and various sociocultural factors. Above all, the girls were affected by ‘powerful hierarchical age-structures’ which largely restricted their options for action (Klocker, 2007: 85). Nevertheless, according to the author, among the child domestic workers, ‘all of the girls replied unequivocally that they had decided for themselves’ (Klocker, 2007: 91; italics in original). This obvious contradiction, into which the author does not go further, points out that the distinction between ‘thick’ and ‘thin’ agency she has made does not meet the complexity of the contexts of action (Esser, 2016).
It is also apparent from other research in Africa and Middle East, as well as Latin America, that the material and sociocultural framework conditions do not necessarily lead to restrictions on the capacity to act, but can also become a kind of action provocation, which causes children and youths to take on new and independent actions. For example, studies on children in Ghana (Mizen and Ofosu-Kusi, 2010; Ofosu-Kusi and Mizen, 2012), Kenya (Omolo 2015) or Peru (Aufseeser, 2014a; 2014b) who live on the street show, in the words of Alderson and Yoshida (2016: 77), ‘how children’s self-reliant agency similarly keeps knocking against very hard contexts and discrimination against children’. A study carried out in a rural region of Mexico (Carpena-Méndez, 2007: 45) concludes that boys and girls between the ages of 12 and