Drawing teaching into scorable form: A socio‐material critique of lesson observation

Abstract Amid rising accountability and standardisation mandates, paraprofessional educators constitute a growing yet precarious segment of the global education workforce. While their roles are expanding, the instruments used to assess their practice remain critically underexamined, despite having significant consequences for professional identity and employment. This reality highlights a deeper problem within global education policy: the unexamined assumptions that underpin large‐scale observation systems, which operate under a representational logic that assumes teaching can be fully captured. Nowhere is this clearer than in Japan's Assistant Language Teacher (ALT) system, a massive programme reliant on a transient, largely non‐certified workforce. This commentary employs a socio‐material lens to interrogate the ontological foundations of lesson observation, treating evaluation forms not as passive records but as active agents in pedagogical events. We analyse two observation rubrics, drawing on Actor‐Network Theory, Barad's agential realism, and feminist accounts of situated accountability to trace how the rubrics' materiality enacts pedagogical realities. Three claims follow: that observation tools (1) individualise structural constraints, (2) displace relational labour with what is scorable, and (3) produce epistemic torsion as expansion narrows into reduction. Our analysis shows how these tools produce the professional order they claim only to record, stabilising a narrow, individualistic, technicist vision of teaching that obscures the collaborative and contingent nature of paraprofessional work. Written critically from inside the system the authors helped structure, this commentary extends debates on lesson observation by shifting attention from observers to their instruments and challenges the assumption that teaching can be mapped without remainder. Context and implications Rationale for this study: Lesson observation instruments remain underexamined ontologically and epistemologically. Rubrics are not records but active agents in producing the professional realities they claim only to capture. Why the new findings matter: Epistemic torsion identifies the harm that rubrics inflict on teaching's irreducible messiness, shifting focus from how observation captures teaching to what it does to it. Implications for practitioners, policymakers and researchers: This commentary offers a socio‐material lens for reassembling the evaluation of teaching. It clarifies why rubric‐based observation distorts the relational and contingent nature of assistantship, and why ‘interpretive drift’ from the rubric is often professional judgement rather than non‐compliance. For policymakers, it demonstrates the limitations of representational approaches and the need for deliberative, practitioner‐defined alternatives that can hold the messiness of real teaching. For researchers, it identifies epistemic torsion as an emergent analytic category connecting tool design, professional visibility, and precarious labour, opening directions for interrogating how evaluative instruments materially shape educational practice.