Anthropology Southern Africa's code of conduct was revised in 2005. It included several important features, some of them unique. One was that promise of a regular return to the code to check its premises against the world we encounter as anthropologists. This was intended to generate an iterative conversation about ethics in the Association and an ongoing critical assessment of the values that underpin our work, along with a lively record of experiences that might help shape our thinking about training and research. Despite regular reminders, however, the anticipated return has not happened as frequently as it ought. Instead, the code – if it is consulted at all – has begun to appear as a static set of principles rather than the responsive, reflective, reflexive and engaged conversation that the notes to the 2005 Code anticipated. Not only is this a lost opportunity to think about the value(s), politics and ethics of ethics processes but it also means that the code is out of step with contemporary developments, including in relation to institutional ethics procedures. A second unique feature was the suggestion in the notes that the Code could draw on local relational ontologies, ubuntu, as a way to generate an ethical stance that - at least potentially - might have reparative value. A locally grounded concept of this kind could make the Code responsive to southern African social life and might facilitate a broader discussion of anthropology’s place given its colonial and apartheid entanglements. Calling on ubuntu in this way is not uncontroversial. It runs the risk of reifying, instrumentalising and romanticising ontologies, among other things. How might we think about the relation between grounded concepts and the precepts of ethics adjudication?