The emergence of objectivity: Fleck, Foucault, Kuhn and Hacking

Over the past century, scholars have put forward different analytical notions that aim to shed light on the historical development of science. It was in his paper ‘Some Specific Features of the Medical Way of Thinking’ (1986 [1927] pp. 39–46) and then in his book Genesis and Development of a Scientific Fact (1935) that Fleck introduced the notion of thought style, the distinctive mode of thinking of a community in which knowledge is produced (a ‘thought collective’). Kuhn, on his part, defined a paradigm in The Structure of Scientific Revolutions (1962) as a general framework with components such as exemplars, ontological assumptions and values. The notion of episteme essentially highlights the structures common to the practices of a historical period and has been proposed in The Order of Things (1966), a book in which Foucault wanted ‘to reveal a positive unconscious of knowledge: a level that eludes the consciousness of the scientist and yet is part of scientific discourse’ (Foucault, 1994 [1966], p. xi).

More recently, in two papers published in 1982 and 1992 (Hacking, 2002, pp. 159–199), Hacking outlined ‘the project of scientific styles of reasoning’ (Hacking, 2012). The label ‘style of reasoning’ came from the historian Alistair Crombie (Crombie, 1994), who had specified six methods of scientific enquiry central to the history of Western thought: the style of geometry, the experimental style, the style of hypothetical modelling, the taxonomic style, the statistical style and the historico-genetic style. Hacking noted that each of these styles involves new types of objects, standards of evidence, laws and true-or-false sentences. He also suggested that styles have sharp beginnings and are self-authenticating, i.e. they do not answer to any higher standard of truth. In his later writings (e.g. Hacking, 2009), he insisted that styles are not only ways of thinking but also ways of doing: humans are embodied creatures that use their minds and bodies to think and act in the world (Hacking, 2012, p. 2).

Although Fleck, Kuhn, Foucault and Hacking belonged to a stream of epistemologists who reflected on the process of acquiring knowledge in close relation with the analysis of historical cases, their philosophical contexts were different. Fleck is one of the thinkers who, in the 1930s, rejected the ‘outside time and history’ investigation of scientific activities conducted by the Vienna Circle (Fleck, 1979 [1935], p. 50) and claimed that scientists' practices do not follow a timeless rule. As Rheinberger (2010, p. 19) made evident, his thought has ‘a surprising number of points in common’ with Gaston Bachelard's, traditionally considered, together with Georges Canguilhem, the acme of what is known as ‘historical epistemology’. The latter is a philosophical tradition that fully emerged in France at the beginning of the twentieth century, when different thinkers assumed that knowledge could be fully understood only by investigating the historical conditions under which it is produced.¹

Fleck was probably unfamiliar with many works of French historical epistemology, including Bachelard's. On the other hand, Foucault was a student of Canguilhem and belonged to the genealogy of French historical epistemologists. The relationship between Kuhn and French philosophy of science is not perfectly clear. The Structure of Scientific Revolutions is widely presented as a work that has substantive continuities with French historical epistemology. Indeed, Gutting (2003) compared Kuhn with Bachelard regarding their discontinuist reading of the history of science. However, Simons (2017, p. 46) argued that Kuhn has been influenced by Alexandre Koyré and Hélène Metzger and that ‘the relation to the better known brands of French epistemology remains unclear’.

When we turn to Hacking, we need to remember that the expression ‘historical epistemology’ has been recently ‘reinvented’. Indeed, Daston (1994, p. 282) characterised historical epistemology as ‘the history of the categories that structure our thought, pattern our arguments and proofs, and certify our standards of explanation’ and cited Hacking and Arnold Davidson as its leading practitioners. However, instead of ‘historical epistemology’, Hacking preferred the expression ‘historical meta-epistemology’ to mark a distinction with the previous use: ‘where Bachelard insisted that historical considerations are essential for the practice of epistemology, the historical meta-epistemologist examines the trajectories of the objects that play certain roles in thinking about knowledge and belief’ (2002, p. 9). Hacking described ‘historical meta-epistemology’ as the study of organising concepts: epistemological concepts such as probability, evidence, reason and objectivity that shape our ideas and practices of knowledge (Hacking, 1999a, p. 53) (see also Sciortino, 2017). He added that historical meta-epistemology, falls under the concept of ‘historical ontology’, the study of the historical trajectories of ‘not only “material” objects but also classes, kinds of people and ideas’ (2002, p. 2).

Despite the difficulties in pinpointing the common beliefs of these thinkers, their notions have been compared with each other since at least 1968, when Piaget (2015, p. 132) observed that ‘Foucault's epistemes are reminiscent of Kuhn's paradigms' and discussed their differences from a structuralist perspective. In more recent times, Möβner (2011) pointed out important differences between the concept of thought style and that of paradigm; Elwick (2012) depicted Hacking's styles of reasoning as conditions of possibility and likened them to Foucault's epistemes; and Winther (2012) investigated the simultaneous application of the notions of paradigm and style of reasoning to the history of science. The fact that these four notions are often compared to each other conceals the admission that they belong to the same genus. However, the general physiognomy of this genus has not been described yet: are these notions tools for achieving overlapping objectives? How do they reflect the different perspectives of their proponents?

Clarifying these questions will help me to combine the ideas of Fleck, Kuhn, Foucault and Hacking in a way that enables their works to be used as part of the same project. Indeed, the aim of this paper is to make instrumental use of their insights to form a new combined framework that illuminates the process of production of scientific knowledge. Section 2 lays the foundations for a reasoned answer to the questions above: I shall argue that we can view Fleck's, Kuhn's, Foucault's and Hacking's notions as frameworks that offer explanatory narratives of how objectivity is possible. The respective attempts of these thinkers to explain the possibility of objectivity will be viewed as part of their effort to historicise Kant. It would have been possible also to look at how Ernst Cassirer has had similar ambitions and to discuss how these four thinkers have been influenced by an important strand of Neo-Kantianism. However, given that Fleck, Kuhn, Foucault and Hacking have been more central in recent history and philosophy of science, I will confine my attention to their ideas. In Section 3, I shall point out some differences between the four frameworks. Finally, in section 4, I shall argue that they can be used together as a toolkit for understanding how concepts, propositions and other epistemological items emerge. Afterwards, I shall propose a layered diagram that helps to highlight the similarities and differences between the frameworks of these authors and combine their insights. I shall put particular stress on Hacking's project because I want to show that his framework can play a crucial role in revealing fundamental aspects of the process by which objectivity becomes possible.