Proving What? On the Structure of Proof of Concept Research

Abstract

In his Representing and Intervening, Hacking (1983) argues that, contrary to common belief, many phenomena which the natural sciences investigate do not exist as such in nature. Rather, they only exist in highly specific, man-made experimental setups, and are as a result created by us themselves. Kroes (2014) has nuanced this by making a distinction between weak and strong creation: The phenomena Hacking talks about are weakly created, because they are not constructed according to an intelligent design. Artefacts like screwdrivers, bikes and scientific instruments are strongly created, because they are based on an intelligent design. While this latter category of objects usually lies outside of the realm of scientific investigation, one can find examples of research in which a seemingly strongly created object is created and investigated. This begs the question of the epistemic value of this type of research: Does the creation of these objects have an effect on scientific theory, for example? If it does not, what is the point of creating such an object for science? In this master thesis, I attempt to answer these questions by taking “proof of concept research” as a model for this kind of research. Building on previous work on the subject by Kendig (2016) and Elliott (2021), I introduce a framework to identify the relevant entities involved, as well as a relation between them. Using this framework, I argue that the products of proof of concept research are both strongly and weakly created. The epistemic products I identify reflect this duality: On the one hand techniques are developed which are relevant for the creation of artefacts. On the other hand the created object can act as a genuine object of investigation, and theoretical knowledge can be generated accordingly.