Plausibly Deniable Harm: How Everyday Engineering Practices Produce Hidden Risk

Harm in engineering and infrastructure organisations is commonly attributed to failure, misconduct, or technical breakdown. This paper argues that harm can also arise through ordinary, rule-following practices that appear responsible on paper. Drawing on science and technology studies scholarship, it introduces the concept of plausibly deniable harm to describe situations in which operational and experiential knowledge is systematically excluded through formally legitimate organisational processes, producing foreseeable adverse outcomes that remain administratively unclaimable. Through conceptual analysis informed by 25+ years of commissioning practice, the paper identifies five mechanisms through which this exclusion occurs: representational abstraction (knowledge translation losses), procedural compliance (substituting process for engagement), temporal deferral (normalising delay), credibility asymmetries (privileging formal over experiential authority), and interactional ambiguity (softening dissent through professional courtesy). These mechanisms interact to produce harm that is formally compliant yet substantively consequential. By reframing harm as an emergent feature of compliance-oriented governance rather than exceptional failure, the paper extends science and technology studies (STS) debates on responsibility, accountability, and care in sociotechnical systems. It offers an analytical framework for understanding how organisations maintain procedural accountability while remaining substantively unresponsive, with implications for infrastructure governance, organisational design, and future empirical research on epistemic inequality in high-stakes environments.