Philosophy of Biomimetics

The state of the art

There are many meeting points of current biology and engineering, ranging from biotechnology via synthetic biology to the development of bio-hybrid systems (Johnson 2011). Recently, terms such as “bio-inspiration” and “biological transformation” have been introduced as umbrella terms to advertise the technological and economic advantages of fields such as biotechnology, biointelligence, and biomimetics (Dieckhoff et al. 2018, Marzi et al. 2018). Importantly, the disciplinary status of these new fields, the epistemic processes involved, and the ontological categories presupposed in this transformation remain poorly researched.

To better understand this transformation, biomimetics (or “Bionik” in German) is a promising object of study because it bridges the disciplinary borders paradigmatically by combining biology and technology in order to solve engineering problems “through the abstraction, transfer, and application of knowledge gained from biological models” (VDI-6220 2012).

Classical examples of biomimetic success are the flight of birds inspiring the development of aeroplanes, or the invention of Velcro® tape inspired by the interaction of burrs with the fur of animals. In recent decades, knowledge about biological systems has increased significantly due to new analysis and simulation methods, both on the molecular and the mesoscopic (organismal) level. Moreover, advancing production methods in engineering have considerably boosted biomimetic output. Biomimetics is a growing field of research, with its own societies (such as Biokon, biokon.de, or the International Society of Bionic Engineering, isbe-online.org), handbooks (Bar-Cohen 2012, Jabbari et al. 2014), and journals. Lenau et al. (2018) report a steady increase in both publications and patents in the field of biomimetics since the 1990s. In addition, we can assume a wide field of work that is biomimetic without explicitly using that label, e.g., in robotics, swarm intelligence, and artificial neural networks. The relevance of the field is reflected by the coalition agreement of the current German government, which explicitly endorses support for biomimetic research (Koalitionsvertrag 2018:35 and 58). The transfer of knowledge from biology to technology is expected to gain increasing relevance in the future (Gebeshuber et al. 2009, von Gleich et al. 2010) because nature still outperforms conventional engineering solutions in many cases. This includes tough and multi-functional spider-silk, photosynthesis, and the cognitive abilities of brains.

Despite their importance and successes, the new developments at the crossroads of biology and technology still lack a theoretical foundation as a basis for the systematisation of the knowledge generated in these fields. For biomimetics, this lack has been explicitly pointed out in the literature (Schmidt 2002, Nachtigall 2010:149), and the situation is no different today. The bibliographic database philpapers.org, for example, returns only very few philosophical papers explicitly addressing biomimetics. For instance, Dicks (2016) deals with the very general philosophical features of biomimicry, a field focusing on sustainability that partly overlaps with biomimetics, and Zoglauer (1994) analyses the transfer of models from nature to technology. Some authors suggest questionable sub-classifications of biomimetics. For example, Schmidt (2002) distinguishes various kinds of mimesis of which function-mimesis would be one sub-area of biomimetics among others. This has been criticised by Nachtigall (2010) because all biomimetics is about functions. Biomimetics has also been described as a technoscience, which does not seem to be adequate as the term “technoscience” refers to research areas without a clear separation between representation and intervention (“Darstellung und Eingriff”). However, the description of biological systems and the intervention in form of a technical application are clearly distinguishable in biomimetics. There are also related discussions in the broader philosophical literature, ranging from Kapp’s famous claim that all technical products have a biological precursor model (Kapp 1877) to Janich’s scepticism regarding the technical potential of biomimetics (Janich 1989).

A formal theoretical framework for biomimetic research would also clear the obstacles for biomimetic research projects identified by von Gleich et al. (2010). While some of these obstacles are merely of organisational nature, e.g., bringing together the necessary expertise and the supporting infrastructure, others are more principled. In particular, the tension in mixed teams between the scientific wish to gain insight and the engineering need to build market-viable applications would be ameliorated when such a formal framework provides a way of thinking shared by all members in biomimetic research teams. Such a foundation requires that both the epistemology and the ontology of biomimetics be known: How does biomimetics acquire knowledge, and about what?

The epistemology of biomimetics

With the exception of Zoglauer’s (1994) study on model transfer from nature to technology, hardly any research has been done on the epistemology of biomimetics. This is surprising considering that the basic idea that artificial devices imitate, and improve on, nature is a standard topos already in ancient philosophy (Schiemann 2014). So far, methodological analyses of this transfer have been conducted from a pragmatic point of view to more easily arrive at technical applications. Only few attempts have been made at posing more conceptual questions and at developing a theoretical basis (e.g., Vogel 2000, Vincent et al. 2006, Speck and Speck 2008, Nachtigall 2010, Fayemi et al. 2017). Commonly used labels such as “technology pull” and “biology push” (DIN-ISO-18458 2015) can, unfortunately, often be applied post hoc only, and they are descriptive rather than explanatory and do not provide any insight into the rational structure of biomimetic research.

Due to the lack of a theoretical foundation, it is even controversial whether biomimetics should be seen as a unified discipline. This controversy is also fuelled by the ambiguous use of terms such as “biomimetics”, “bionics”, or “biomimicry” on the one hand (Drack and Gebeshuber 2013), and by lumping together different fields under umbrella terms such as “bio-transformation” on the other hand. Situated at the crossroads between biology and engineering, biomimetics seems to have no clearly delineated object of research, nor a unified method or objective. One characteristic of biomimetics is that it combines knowledge from natural sciences with that from engineering. This implies that it can adhere neither to standard biological thinking nor to standard engineering routines. For this reason, neither established theoretical considerations about biology (e.g., Mahner and Bunge 1997, Krohs and Toepfer 2005) nor theories of engineering (e.g., Kroes and Meijers 2000, Kornwachs 2012, Kroes 2012, Hansson 2017) are sufficient to elucidate the status of biomimetics. It is an open question how these established accounts and other recent developments at the biology–engineering interface (which have also received little epistemological attention so far) can serve as starting points for investigations on biomimetics. The present proposal addresses this question by characterising the epistemological specifics of biomimetics.

Important preliminary work towards a theoretical basis of biomimetics has been conducted by Manfred Drack and colleagues in sub-project C02 of the Collaborative Research Center SFB TRR 141 “Biological Design and Integrative Structures” (2014–2019). Based on observations on the interdisciplinary workflow within biomimetic projects, Drack et al. (2018, 2020) developed a conceptual framework that is applicable to any biomimetic development, including those of material products, material processes and information processes. The main finding is that the engineering design approach of Pahl et al. (2007) can be successfully used to describe biomimetic transfer processes involving five levels of engineering design, namely (1) the overarching system (biological or engineering system, e.g., organism, machine), (2) the construction level (specifying the concrete parameters of the interacting entities), (3) the working principles (or mechanisms, i.e., abstract causal relations), and (4) the function in question, enabling the construction (or the concrete design) to perform (5) the task for which a machine, device or process is built. While the overarching system, the technical construction, and the task will differ from the biological role model, function and working principles remain the same – indicating that these are at the core of biomimetic knowledge transfer. Nachtigall (2010:83) has suggested using the term “abstraction” for the research process, which derives what Drack et al. (2018) identified as the function and working principle from the biological system. Nachtigall also introduced the term “concretisation” for the process of implementing function and working principle in the technical application. Combining these two ideas, the promising hypothesis emerges that the core of biomimetic knowledge generation involves the identification and implementation of functions and working principles, which are transferred together from biology to engineering. What remains unknown is whether a typical pattern of reasoning underlies these two types of processes.

The ontology of biomimetics

Today, biomedical ontology and its use for the structural representation of knowledge have a strong basis (e.g., Jansen and Smith 2008, Schulz and Jansen 2013). For biomimetics, however, this issue is still in its infancy. Several attempts have been made to structure and organise biomimetic knowledge in formal ontological structures from a computer-science perspective. Yim et al. (2008) develop a bio-inspired design ontology based on functions, strategies and design solutions. The approach of Kozaki and Mizoguchi (2014) aims at retrieving knowledge for technical applications, and Vincent (2014, 2017) discusses a systematization of biomimetic knowledge based on biological and technical trade-offs. While all of these approaches include interesting components, they also show severe shortcomings because they were designed to serve as practical tools for technical innovation (McInerney et al. 2018), and not to adequately represent the biomimetic domain of research. The latter remains an important desideratum. To further analyse the research objects of biomimetics and to indicate ways to systematise biomimetic knowledge, we can build on the following three results:

First, the discussion of the contrast between organisms and machines has a long history (Koutroufinis 2017). Within the SFB TRR 141, these core entities were comparatively analysed because of their presumed relevance for biomimetic knowledge transfer. In general, both “organism” and “machine” are true theoretical concepts in their original domains (Drack and Betz 2017), and, as it turned out, not straightforward to distinguish. Self-reproduction and self-maintenance are among the main features that, until today, are used to distinguish organisms from machines. However, many of the features that have previously been thought to mark the contrast, such as feedback or the open-system-in-flux equilibrium, are today found in both biological and engineered systems (Pouvreau and Drack 2007, Drack and Pouvreau 2015). For both machines and organisms, however, functions and working principles are relevant and serve as a common ground for biomimetic knowledge transfer (Drack et al. 2018).

Second, a general overview of terms and entities referred to in biomimetics has been developed by Drack et al. (“Glossary of architectural and constructional biomimetics”, to be published with Springer). The terms have been harvested from the relevant literature (norms such as DIN-ISO-18458 2015 and VDI-6220 2012, biomimetics books and websites) and from a survey among members of the SFB TRR 141. Among the roughly 2000 collected terms, basic terms were identified that are constitutive for the field of (architectural) biomimetics, including ‘function’, ‘working principle’ and ‘construction’. The vast majority of terms are technical but are nevertheless often used differently in the various disciplines (Nagel et al. 2013). Definitions for the terms were collected from architecture, engineering, biology and biomimetics reference sources, revealing commonalities and differences among the terminology of these fields.

Finally, our previous work suggests the centrality of functions, working principles and constructions. Relevant functions can be characterised by verb phrases (Pahl et al. 2007 suggest verb+noun combinations). Their theoretical analysis, however, is still a desideratum, as the word “function” itself is ambiguous and there are various competing theories of functions in both biology and artefacts. There is a rich literature on functions in the philosophy of biology and technology (e.g., Buller 1999, Krohs and Kroes 2009, Mossio et al. 2009, Houkes and Vermaas 2010). Standardly, causal-dispositional approaches are distinguished from aetiological approaches to functions, whereby the former consider the causal effects of functions, while the latter analyse functions in terms of their origins, be they a better adaptation to a given environment, or selective advantages in the phylogenetic past (Bedau 1998), or, in case of technical artefacts, the intentions of designers or users. In addition, the notion of normalcy or typicality is often held to be relevant (Wachbroit 1994, Schurz 2001, Krohs 2004). For the formal-ontological analysis of functions, we can build on our own work on how to integrate functions into a formal ontology (Röhl and Jansen 2014, Jansen 2018; cf. Spear et al. 2016).

Considerably less literature is available concerning the analysis of working principles and constructions as the causal principles that help fulfil a function but are distinct from it. On the one hand, however, there is a growing interest in mechanisms in the philosophy of biology (e.g., Machamer et al. 2000, Krickel 2018). On the other hand, it seems to be promising to model working principles as complexes of dispositions – i.e., in terms of systems of causal properties that can be realised by certain types of processes. For the latter, we can again build on own work (Jansen 2007, Röhl and Jansen 2011, Barton et al. 2017) to develop an account of working principles that is appropriate for the domain of biomimetics.

Objectives

The overall goal of the proposed project is to deepen our understanding of the new developments at the crossroads of biology and technology by a thorough analysis and consolidation of the theoretical foundations of biomimetics. The survey of the state of the art revealed the following open questions:

  • Can biomimetics be conceived as a coherent field of research? What is the nature of biomimetic knowledge, and how is it derived?
  • What is the specific object of biomimetic research and how is it to be analysed ontologically?

Answering these questions will shed new light not only on the exciting developments at the intersections of biology and technology, but also contribute to the ontological analysis of the underlying reality in domains that have long been considered as separate realms. Our leading hypothesis is that biomimetics is unified by epistemological features including the goal of studying the transfer of working principles for functions from biological to technical constructions. We will research how knowledge about functions, working principles and constructions can be gained and analyse ontologically what kind of entities they are and how they fit into the wider domain of biomimetics. This will contribute new insights to the debate about functions, adding an interdisciplinary perspective and an intertwined analysis of working principles and constructions. We expect that our results will also foster practical communication in biomimetics as well as help lay the basis to systematise existing knowledge and to generate new knowledge in the field.

Work programme

In order to accomplish the above-mentioned objectives, we need to

  • integrate the project within the wider biomimetics community and compose a corpus of research examples to be analysed in the project (work package WP 1),
  • analyse whether it is justified to view biomimetics as a unified field of research with its specific set of epistemological features (WP 2),
  • analyse reasoning patterns in biomimetic projects (WP 3),
  • review the ontological commitments of biomimetics to evaluate existing approaches (WP 4),
  • analyse the foundational categories of biomimetics (WP 5),
  • evaluate these results through presentation and discussion at an expert workshop (WP 6),
  • formalise the ontology (WP 7),
  • and evaluate it for consistency and adequacy (WP 8).