Abstract Humanitarian governance increasingly relies on standardized technical systems to make violence legible, comparable, and actionable. Yet the gendered implications of these “expert objects” remain underexamined. This article analyzes the Gender-Based Violence Information Management System (GBVIMS), a central interagency database used to record and classify GBV incidents in humanitarian settings. Drawing on qualitative research with humanitarian practitioners and refugee communities in Jordan and Bangladesh, and informed by Actor–Network Theory and feminist science and technology studies, the article examines how GBVIMS co-produces knowledge about GBV through its categories, definitions, and data practices. This research finds that while GBVIMS is designed to improve coordination and evidence-based response, its reliance on standardized categories—necessary for quantification—inevitably abstracts violence from social context, producing partial and exclusionary understandings of GBV. Practitioners consistently emphasized that the system’s value lies less in the numbers it generates than in the shared language and conceptual frameworks it provides, which shape how GBV is recognized, discussed, and acted upon across humanitarian settings. At the same time, refugees experience GBVIMS as an extractive data regime: their narratives are translated into predefined typologies that circulate beyond their control, often without transparency regarding how the data are used. By foregrounding the interpretive labor embedded in ostensibly neutral technologies, this article advances feminist critiques of technocratic humanitarianism and introduces the concept of object-ivity to describe how authority is delegated to standardized objects. It shows how tools intended to make violence visible can simultaneously obscure lived experience and reshape whose suffering gets counted in humanitarian governance.
Sarah Nandi (Tue,) studied this question.