Van Zundert, Antonijević, and Andrews begin by highlighting the important role coding plays in digital humanities research, how it is easily overlooked, as well as the detrimental and undetectable effects this can have on the research outcomes itself. Using Latour’s concept of the ‘black box’, and drawing parallels between codework and rhetorical arguments, the authors argue that code is a social construct that inadvertently embeds social and ideological beliefs upon research; furthermore, they highlight the current lack of monitoring, crediting and critiquing of codework. They use an analytical autoethnography method to examine the experiences of digital humanities scholars proficient in both humanities research and coding, grouping their observations into categories inspired by ‘the five canons of rhetoric’. Their findings illustrate that while code and codework increasingly shape research, they are rarely part of disciplinary discussions and the consequences include: limitations to the integration of digital scholarship into the humanities, softwares becoming reduced to merely user interfaces, and the loss of recognition for hybrid scholars who function as digital humanities programmers. They conclude that interdependence of code and text should become an established trajectory in the humanities as well as the development of methods for documenting, analysing and evaluating codework.