Article révisé par les pairs
Résumé : The analysis of the ways early career investigators from 32 countries frame the notion of ‘research impact’ in the Social Sciences and the Humanities (SSH) reveal some obvious commonalities. There is a convergent tendency to describe impact as an object of great concern that operates mostly through communication, while avoiding any absolute definition of the notion. Respondents associate impact to the concept of accountability and share the perception that impact does not currently belong to the dominant academic culture. Differentiations and even divergences in framing appear though in regard to the cognitive or social nature of impact as well as in the extent to which respondents associate the processes of conducting scientific research with or dissociate them from the creation of social impact. As such, five common argumentative frames of impact and six differentiated or divergent ones are brought to light. While commonalities can be related to shared socio-professional circumstances, the observed differentiations and divergences in framing are linked to the diversity of disciplinary, national and institutional contexts within which scholars are trained and conduct their research. Because of the coexistence of commonalities, differentiations, and divergences in early career investigators’ framing of impact, it is argued that research impact in the SSH is best conceptualized as a ‘boundary object’ (Star and Griesemer, 1989). We conclude that attempts to frame impact within a narrower perspective – whether in scholarly discourse or policy making – would finally reduce the diversity of contextualized opportunities for SSH scholars to engage in any valuable creation of impact.