A recently blossoming historiographical literature recognises that physical anthropologists allied with scholars of diverse aspects of society and history to racially classify European peoples over a period of about 100 years. They created three successive race classification coalitions – ethnology, from around 1840, anthropology, from the 1850s, and interwar raciology, each of which successively disintegrated. The present genealogical study argues that representing these coalitions as ‘transdisciplinary’ can enrich our understanding of challenges to disciplinary specialisation. This is especially the case for the less well-studied nineteenth century, when disciplines and challenges to disciplinary specialisation were both gradually emerging. Like Marxism or structuralism, race classification was a holistic interpretive framework, which, at its most ambitious, aimed to structure the human sciences as a whole. It resisted the organisation of academia and knowledge into disciplines with separate organisational institutions and research practices. However, the ‘transdisciplinarity’ of this nationalistic project also bridged emerging borderlines between science and politics. I ascribe race classification’s simultaneous longevity and instability to its complex and intricately entwined processes of political and interdisciplinary coalition-building. Race classification’s politically useful conclusions helped secure public support for institutionalising the coalition’s component disciplines. Institutionalisation in turn stimulated disciplines to professionalise. They emphasised disciplinary boundaries and insisted on apolitical science, thus ultimately undermining the ‘transdisciplinary’ project.