How to Characterise Pure and Applied Science

ABSTRACT

Regarding the dichotomy between applied science and pure science, there are two apparently paradoxical facts. First, they are distinguishable. Second, the outcomes of pure sciences (e.g. scientific theories and models) are applicable to producing the outcomes of applied sciences (e.g. technological artefacts) and vice versa. Addressing the functional roles of applied and pure science, i.e. to produce design representation and science representation, respectively, I propose a new characterisation of the dichotomy that explains these two facts.

Acknowledgements

Special thanks to the anonymous referees of this journal for helpful comments.

Notes

1 Related to this distinction is the dichotomy of basic/applied research. Although my analysis has some implications for the latter distinction, characterising it according to my analysis needs more effort, particularly since the concept of research as a scientific activity realised institutionally has not yet been philosophically analysed, though the issue has been tackled from the social-political point of view (Pielke 2012; Godin and Schauz 2016).

2 Science policy studies practitioners since the late 1950s have been dealing with these issues surrounding the dichotomy. Among time-honoured analyses are Nelson (1959), Arrow (1962), Polanyi (1962), and Toulmin (1964, 1966). For new studies investigating the division from an economic point of view, see Rosenberg (1990); Pavitt (1991); David, Mowery, and Steinmueller (1992); Partha and David (1994); Salter and Martin (2001); and Nelson (2006). Roll-Hansen (2017) has recently explored the issue with an emphasis on education policy.

3 Simon (1968) and Kroes (2002) have developed a similar account on which professions, e.g. engineering, medicine, architecture, and law, are separated from natural sciences because they are centrally concerned with design (how things out to be), whereas the natural sciences with knowledge (how things are) (Simon 1968, 114). In section 6, I will illustrate the differences between my account and theirs. Also, Niiniluoto (1993, 2014) has proposed a similar account.

4 ‘Practical science’ popularised in the early nineteenth century was ‘used to describe the practice that complemented theoretical knowledge’. ‘Science applied to art’ imported from the educational culture of the early nineteenth-century France represented ‘the desire to promote industry through scientific education’ (Bud 2012, 541–542).

5 For instance, the Kaiser-Wilhelm-Gesellschaft was established in Germany in 1911 to overcome the limitations imposed by the ‘university-centered organization of research’ (Schauz 2014, 284). In Britain, the Department of Scientific and Industrial Research was founded in 1916 ‘to encourage some sectors of British industry to spend more on research’ (Clarke 2010, 289).

6 As Popper (1959, 149–152) argued, if science is nothing but an instrument, then there is no difference between pure and applied science.

7 Bush uses ‘pure science’, ‘pure research’, and ‘basic research’ interchangeably to denote what deals with ‘the discovery of fundamental new knowledge and basic scientific principles’ (Bush 1945, 78).

8 ‘We do not wish to say that facts do not exist nor that there is no such thing as reality. In this simple sense, our position is not relativist. Our point is that “out-there-ness” is the consequence of scientific work rather than its cause’ (Latour and Woolgar 2013, 180–182).

9 ‘The sciences do not speak of the world but, rather, construct representations that seem always to push it away, but also to bring it closer’ (Latour 1999, 30).

10 (Pure) science as an activity has both cognitive and non-cognitive features. Cognitive ones are those which result from the representational nature of its outcomes. For instance, that particle physics is dependent largely upon computer simulation is a cognitive feature of the discipline, since differential equations involved in (which represent a relation in the world) need to be solved numerically. On the other hand, that its experimental researches are supported financially by governments is a non-cognitive feature. A parallel distinction between functional and non-functional features holds for engineering disciplines involved in design. In sections 4 and 5, I will characterise these types of disciplines according to their outcomes.

11 Kitcher (2004) has taken up a normative stance and claimed that the distinction should be thrown away for political reasons. For a criticism, see Roll-Hansen (2017).

12 Appealing to the internal–external division, which has been allegedly transcended by new methodologies of historiography, e.g. microhistory, cultural and contextual history of science, seems likely to be old-fashioned. However, the point here is that approaches dealing with the dichotomy are all not in history and social studies of science, but partly in philosophy of science which its methods tilt toward internalism. Furthermore, as Shapin (1992) has argued, the transcendence is not at all incontestable. Investigating the debate philosophically, Nanay (2017) has explained why we need both internal and external history. According to Nanay, internal history is indispensable in explaining external history and external history in explaining internal history. Metaphorically, for him, the relation of internal to external history in history is as theory to data in science.

13 Any account appealing to the concept of application should say how scientific knowledge is applied. According to a prevailing view named ‘covering law’, it is applied by implementing the boundary conditions of a concrete phenomenon which may be pertained to a specific artifact. For details of this view and its shortcomings see Boon (2006), 34–39.

14 For the history of the challenge, see Kline (1995); Clarke (2010); Bud (2012); Lucier (2012); Pielke (2012); Kroes and Bakker (2013); Bud (2014); Schauz (2014); and Roll-Hansen (2017). Some have shown that technology as the production of applied science evolves independent of scientific knowledge (Kohlmeyer and Herum 1961; Hughes 1976).

15 For Niiniluoto, applied sciences, e.g. engineering sciences, agriculture, medical sciences, and practical social sciences, fall between basic science and technology. For him, while basic and applied sciences produce knowledge, technology produces artifacts. The difference between basic and applied science lies in the functional role of knowledge that each one produces. While basic sciences produce knowledge having ‘epistemic utilities’, applied sciences produce knowledge having both epistemic utilities and ‘instrumental value for the associated human activity’ (Niiniluoto 1993, 3–6). The problem of this account, in particular, is that any kind of knowledge due to its dependence upon truth is instrumentally valuable. If so, the supposed borderline between the two kinds of knowledge becomes hard to be drawn.

16 Scientific realists claim that, given a mature scientific theory, not only what it says about observables, but what it does about unobservables is (approximately) true. But, scientific empiricists accept just the first claim (Duhem [1906] 1991, 107; van Fraassen 2008, 87). In other words, all agree that a mature scientific theory has a semantic content corresponding to the world, though empiricists just admit the correspondence between the semantic content of the theory about observables and the world.

17 In the next sections, we will see how the productions of applied sciences affect the functioning of pure sciences.

18 The agents perform also other actions, e.g. interpreting the productions. For our purposes, I ignore these.

19 Similarly to Galle’s conception of design representation, for Kroes ‘a design is not a technical artefact in itself but merely a representation thereof’ (Kroes 2009, 513). The action-theoretical account of design of Houkes and co-workers explores systematically how the design of artefacts, as a phase of the plan-design cycle, is developed according to its own action-theoretical structure (Houkes et al. 2002, 309).

20 See section 6.

21 Incidentally, Simon (1968, 12–13) warned us that in order to know the properties of an artefact it should be allowed to work, since its internal structure would be known through its interaction with environment.

Additional information

Funding

Funding for this research was provided by Shahid Beheshti University (grant number sad/600/1096).