In this book, I have sought to provide a comprehensive, empirically grounded account of the collaborative creation of scientific knowledge in research groups. I hope to have contributed to a detailed understanding of the collective character of science, an understanding that reflects actual scientific practice and the perspectives of practicing scientists.
In Chap. 2, I outlined what I meant by “research groups." Having outlined the subject of the book, I then described my methodological approach to this subject in Chap. 3, reflecting on the role that qualitative empirical data can play in philosophical theorizing and in detailing the process of data collection and analysis upon which the empirical elements build. In Chaps. 4 and 5, I portrayed the two research groups that I studied empirically. In Chap. 6, I analyzed the ways in which these two groups divided scientific labor among their group members, and I argued why their division of labor is not well accounted for by existing community- focused approaches.
This book’s most important conceptual contributions are in Chaps. 7, 8 and 9, where I explored how detailed qualitative data can contribute to ongoing epistemological debates about dependence, trust and collective
© The Editor(s) (if applicable) and The Author(s) 2016 S. Wagenknecht, A Social Epistemology of Research Groups, New Directions in the Philosophy of Science,
knowledge. In Chap. 7, I provided two analytic distinctions—between first and second-order reasons, and between opaque and translucent epis- temic dependence. First-order reasons are immediate, evidential reasons to accept a proposition p. Second-order reasons, instead, concern the trustworthiness of a speaker who testifies that p is the case. When a person has only second-order reasons at her disposal, her epistemic dependence upon the testifier is opaque. When she is in a position to acquire first- order reasons, then her epistemic dependence is translucent. In Chap. 8, I showed how epistemically dependent scientists gauge the trustworthiness of collaborators, that is, by which strategies they acquire and corroborate second-order reasons, and how they seek to minimize their reliance upon such reasons. In Chap. 9, I argued that mutual epistemic dependence allows only groups of scientists to provide a scientific justification for collaboratively formulated knowledge, rendering some scientific knowledge irreducibly collective.
When writing this book, something, that I at first perceived as a peculiar tension, was underpinning my reflections on research group collaboration—the relation between an ethos of individual freedom and collaborative commitment, between the pursuit of individual research interests and the necessities of collaboration. I perceived this relation as a tension because I looked upon it alternately through the lens of epistemic individualism and the lens of its critique (as presented, e.g., in Fricker, 2006; Grasswick, 2004; Hardwig, 1985, 1991; Kusch, 2002; Nelson, 1995).
Eventually, however, I came to see that collaborative scientific practice intertwines the individual and the collective. Professional autonomy, understood as the volitional autonomy to pursue individual research interests, often requires scientists to make themselves epistemically dependent upon collaborators. For example, in my interview with Claire, senior scientist in the molecular biology laboratory, she emphasizes that she feels “very independent" in the sense that she “can do what she wants." At the same time, she describes herself as highly dependent upon collaborators in daily laboratory practice (Claire, interview, group2). Adam, senior scientist in the planetary science group, expresses a similar view when he underlines that all of his fellow group members have different research interests, but that they “[...] would miss out on a lot of research" if they worked
- 10 Concluding Remarks
alone (Adam, interview, groupl). For group members to collaborate and depend upon one another, they need to make themselves, their labor and their expertise available to one another—becoming a “knowledge base" for fellow group members (Laura, interview, groupl).
Which conclusions should epistemology draw from such observations? Do they undermine epistemic individualism, or do they vindicate it? Epistemic individualism, in its pure form, upholds the ideal of the autonomous, self-sufficient, individual knower. Its critics maintain that the ideal of individual epistemic self-sufficiency should not be realized, arguing that “[i]f rational at all, she [the autonomous knower] would not be ideal, but rather a paranoid sceptic about others’ intentions and capacities" (Fricker, 2006, p. 243; cf. Foley, 1994). The inter-individual account ofresearch group collaboration that I have developed in this book seeks to navigate past the ideal of the self-sufficient individual knower while avoiding the devaluation of individual knowing. For, what would science be without its competent, reflective practitioners? Do they not deserve to be called genuine “knowers"? Don’t they have a responsibility to “know what they are doing"? I think so. Yet at the same time, my account of research group collaboration recognizes that the kind of scientific knowledge that research groups produce collaboratively often cannot be had by group members individually—not as scientific knowledge, that is, justified first-hand according to scientific standards. However, when its scientific justification cannot be had by any single scientist, scientific knowledge is irreducibly collective knowledge.
The question as to which conclusions philosophy should draw from empirical observations touches upon a more general issue: What is the role that descriptive accuracy can have for philosophical theorizing, normative efforts that typically aim at analyzing what “ought to be (done)"? After all, what “is," however accurately described, does not tell us what “ought to be." Still, descriptive accuracy is of value to philosophy, particularly to social epistemology and philosophy of science, which seeks to assist practicing scientists and help explicate what scientific practice actually is and what it means to practice science well.
Epistemology’s core concern is for the acquisition of knowledge: How can we know? On the one hand, this concern is tied to conceptual challenges in defining what “knowledge" is (or in defining different types of knowledge and their respective characteristics). On the other hand, this concern is tied to questions that require prescriptive answers: What should we do to obtain knowledge? And, for the case of scientific knowledge, what ought good science be like?
Addressing some of social epistemology’s conceptual challenges, I have suggested ways to differentiate, contextualize and modify some of the field’s existing terminology in, for example, Chaps. 7, 8 and 9. And while I have refrained from formulating an outright prescriptive account of scientific practice, I do think that the book contributes to the question of what good science ought to be like, because any reasonable, practically relevant, philosophical account of what scientific practice “ought to be" requires a terminology appropriate to the subject matter—and it has to rely upon an understanding of actual scientific practice which is descriptively accurate and tailored to philosophical concerns.
How precisely prescriptive accounts relate to description is a difficult issue. At this point, let it suffice for us to focus on the relation between prescription and the description of capacity, a relation that can be phrased as “ought implies can." Whether or not ought does imply can is much debated, particularly among ethicists (for an overview, see Vranas, 2007). But it stands to reason that for the kind of “ought" that social epistemology and philosophy of science are interested in, “ought implies can" should hold true. If concerned with prescriptive questions, social epistemologists and philosophers of science typically investigate how the creation or acquisition of scientific knowledge can be ensured. For as this kind of agency-concerned “ought to ensure" (as distinguished from “ought to be"), Streumer (2003) shows, “ought" does indeed imply “can." In addition, more arguments have been made to support a relation between prescription and description. For example, Schleidgen and collaborators distinguish between abstract prescriptive principles and rules of practice that convey a “practical ought," arguing that such rules of practice need to take into account the various constraints—cognitive, financial, etc.—that human agents face in given situations (while abstract principles need not). Empirical insight in such constraints, the authors hold, can “help adapting basic principles to the capabilities of human agents" (Schleidgen, Jungert, & Bauer, 2010, p. 67). Empirical insight, the authors continue, also
help to evaluate whether practice is in accordance with prescriptive rules. But for empirical insights to fulfill this function, descriptive accuracy is necessary. Intuition, imagination or eclectic first-hand experience won’t do here; comprehensive data about actual scientific practice are necessary.
Therefore, I have collected qualitative empirical data through participant observation and interviewing. These data provide an account of what scientists actually do in day-to-day collaborations, and what they think they can and what they themselves believe they should do. The methodological approach I have chosen is, I believe, a step in avoiding prescriptive accounts of science that are too demanding, too idealized or too “embellished" (Soler et al., 2014, p. 14) to gain traction with actual scientific practice. It is also, I hope, a step toward avoiding an epistemology that is stylized as “science of science" and adopts “a theoretical position ‘outside’ and ‘above’ scientific practices" (Rouse, 2002, p. 180). Instead, I have sought to formulate an epistemology of science that acknowledges the experience and the professional reflection of practicing scientists.
Foley, R. (1994). Egoism in epistemology. In F. F. Schmitt (Ed.), Socializing epistemology: The social dimensions of knowledge (pp. 53—73). Lanham: Rowman and Littlefield.
Fricker, E. (2006). Testimony and epistemic autonomy. In J. Lackey & E. Sosa (Eds.), The epistemology of testimony (pp. 225—250). Oxford: Clarendon.
Grasswick, H. E. (2004). Individuals-in-communities: The search for a feminist model of epistemic subjects. Hypatia, 19(3), 85—120.
Hardwig, J. (1985). Epistemic dependence. The Journal of Philosophy, 82(7), 335— 349.
Hardwig, J. (1991). The role of trust in knowledge. The Journal ofPhilosophy, 88(12), 693-708.
Kusch, M. (2002). Knowledge by agreement: The programme of communitarian epistemology. Oxford: Oxford University Press.
Nelson, L. H. (1995). A feminist naturalized philosophy of science. Synthese, 104(3), 399-421.
Rouse, J. (2002). How scientific practices matter. Chicago: University of Chicago Press.
Schleidgen, S., Jungert, M. C., & Bauer, R. H. (2010). Mission: Impossible? on empirical-normative collaboration in ethical reasoning. Ethical Theory and Moral Practice, 13(1), 59—71.
Soler, L., Zwart, S., Lynch, M., & Israel-Jost, V. (Eds.). (2014). Science after the practice turn in the philosophy, history and social studies of science. New York: Routledge.
Streumer, B. (2003). Does “'ought” conversationally implicate “'can”? European Journal of Philosophy, 11(2), 219—228.
Vranas, P. B. (2007). I ought, therefore I can. Philosophical Studies, 136(2), 167— 216.