Friendships formed in the course of scientific research are common and should be foregrounded in discussions of how the sciences are done. Inspired by the Sociology of Scientific Knowledge, I propose a ‘symmetrical’ analysis of friendships in both the social and natural sciences as a way of comparing knowledge-making practices. The research question that derives from this approach is: How are friendships with and between subjects generative of new forms of scientific knowledge and new types of relating? I provide an answer based on my experience of befriending a group of dendroclimatologists to whom I referred metaphorically as ‘my chimps’ in an analogy with the primatologist Jane Goodall’s affectionate relation with her research subjects. In my case, befriending dendroclimatologists involved cultivating a curiosity about each other’s research and worlds through different means. As a result, my work also came to matter to them and we produced it collaboratively. The instrumentalisation of friendships for the purpose of achieving a certain control and agreement with subjects and beings is, I argue, a normal aspect of knowledge formation, and should not be seen as unethical. If anything, befriending subjects promotes better research ethics as it generates a form of mutuality based on partial relatedness, constructive dissent and playfulness, rather than hybridity, totalising consensus and domination. Overall, my argument about friendship as a method in science seeks to criticise the ideal that isolation and indifference are at the heart of the way scientific knowledge, both social and natural, is and should be made.