The ontologies of deep disagreements about children
: Towards a non-representational methodology for social science

Student thesis: Doctoral ThesisDoctor of Philosophy


This thesis explores the ontologies involved for different sides in persisting disputes about children, specifically how different positions are based in different understandings of what children are. It opens with a survey of normative epistemological responses to disagreement, turning to an account of so-called deep disagreements that appear to be beyond resolution due to framework differences. Three ways to characterise these are explored in terms of epistemic injustice, conceptual metaphors and multiple ontologies and how these can overlap. The middle section develops and proposes social participatory realism as a methodology for studying deep disagreements, and for social science more generally, building on work in science and technology studies and a performative turn in mixed methods research. The methodology is illustrated in a critique of type-based attachment theory, and implications discussed for accounts of positionality and objectivity in research. The final section provides a meta-ethnographic synthesis of primary literature to create a map for how ontologies of childhood can differ. A Q methodology study uses this map to show that real differences in childhood ontologies do exist among different individuals. Finally, a content analysis of contributions to the debate over banning smacking in Scotland shows that different positions in this agreement align with sharply different views of what children are. Conclusions are drawn on the utility of a social participatory realist approach for research, and on the practice of ontological virtue as a means to deal with deep disagreements in practical contexts.
Date of Award2023
Original languageEnglish
SupervisorLiz Lakin (Supervisor) & Sarah Hulme (Supervisor)


  • Children
  • Multiple Ontologies
  • Methodology
  • Mixed Methods
  • Epistemology
  • Disagreements
  • Positionality

Cite this