Some people rely on scientific data to decide whether a particular physical condition calls for a medical intervention.

Others rely on scientific data to decide what they should and shouldn’t eat. Still others attend to data from social psychology, economics, or climate science to deter­mine which social policies they should support, and which they should resist.

Fortunately, many scientists are committed to getting things right, and to telling the truth as they see it. And this should make outright fraud and fabrication the exception rather than the rule.Yet, questionable research practices that fall short of fraud and fabrication persist across the sciences (Fanelli 2009; John et al. 2012). By dropping data points on the basis of gut feelings, refusing to publish data that contradicts their previous research, or citing papers that they take to be flawed or problematic, scientists can shape our understanding of the world in ethically problematic ways (Martinson et al. 2005). It is difficult to draw a precise boundary between these questionable research practices and cases of outright fraud. But at minimum, a scientist who commits fraud must knowingly enter what they take to be falsehoods into the scien­tific record (Bright 2017: 291). This can take the form of fabricating data; claiming to have carried out an analysis without actually doing so; or claiming to have support for a hypothesis after knowingly omitting data that defeat this support. In many cases, fraudulent practices are carried out by scientists who know that they are acting against scientific norms (cf. Fallis 2009). But individual intentionality becomes less salient when responsibility becomes more diffuse and distributed.

Research in social ontology and the philosophy of science has started to explore the questions about who should be held responsible for harms that emerge as a result of the structure of scientific communities. Such approaches shift attention away from questions about individual responsibility, and toward the role of scientific communities in shaping the collection, analysis, and dissemination of scientific data. In this chapter, we build on this trend by examining some of the effects that institutional factors and collaborative practices have on moral responsibility in scientific communities. We begin by summarizing an influential strand of thought in the sociological literature, according to which scientific fraud is likely to emerge as an effect of participating in the scientific credit economy under stress. We then turn to a recent account of the different kinds of collective responsibility that are at play in scientific communities. And with these preliminaries on the table, we turn to two contexts where the factors that are likely to foster fraud emerge within scientific communities: (1) large-scale col­laborative research; and (2) ghostwritten and ghost-managed research. Finally, in light of these cases, we turn to questions about who should be held accountable for the bad outcomes of questionable or fraudulent research, and we argue that the scientific community as a whole should be held responsible for seeing to it that there is less scientific fraud, and that the harmful effects of fraud are mitigated.

25.1