ABSTRACT
Responding to the so-called reproducibility crisis, various disciplines have proposed – and some have implemented – changes in research practices and policies. These changes have been aligned with a restricted and rather uniform conceptualization of what science is, and knowledge is made. However, knowledge-making is not a uniform affair. Here, we reflect on a salient fault line running through Wissenschaft (the whole of academic knowledge making, spanning the sciences and humanities), grounded in the relationship between the acts of research and writing, separating research as reporting from research as writing. We do so to demonstrate that replication and replicability cannot be treated as uniformly applicable and that assessment and improvement of research quality invites various tools and strategies. Among those, replication is important, but not omnipresent. Considering these other tools and strategies in context allows us to situate the value of replication for knowledge making as a whole.
Disclosure statement
No potential conflict of interest was reported by the authors.
Notes
1. Research on research is hardly new and finds its roots in the philosophy, sociology, history, and anthropology of science (De Solla Price Citation1965). Similarly, the pursuit of openness in scientific conduct and reporting can be traced back to, for instance, public witnessing of experiments (Shapin and Schaffer Citation1985). Finally, discussions on how to assess whether a research finding holds up are as old as research itself. Repetition, in its many forms, has always been an important part of that discussion.
2. Many frameworks for plurality in science have been proposed, ranging from paradigms and thought styles (Fleck Citation[1935] 1980; Kuhn Citation[1962] 1970) to epistemic cultures (Knorr-Cetina Citation1999, Citation1991). For an overview of (the history of) plurality in science, see Chang (Citation2012).
3. See e.g., http://nanopub.org/wordpress/?page_id=65 (accessed 25 November 2019).
4. Some would argue plagiarism (we thank Matthias von Herrath, who reviewed this paper, for this point).