An Inquiry into Model Validity When Addressing Complex Sustainability Challenges
Table 2
Philosophical foundations of complexity-compatible modelling.
(1) Foundation 1: an ontological lens that includes unpredictability and uncertainty
The nature of the world is characterised by unpredictability and uncertainty, a reality that is not knowable in its entirety [19]. When it comes to an emergent property of a complex system, such as sustainability in society, many causal relationships of that property may be hidden within a multilayered reality [33]. Modelling of sustainability necessarily requires consideration of known factors (“known knowns”), factors known but not easily quantifiable (“known unknowns”), and factors not yet identified that even if they were to be discovered would likely not lend themselves readily to quantification (“unknown unknowns”) [34].
(2) Foundation 2: an epistemic stance that acknowledges the occurrence and significance of interactions between values and science, objectivity, and subjectivity
Both subjective and objective analyses are critical to advancing scientific understanding [35]. It is important to acknowledge the influential role of the observer in interpreting the observed system and engage more directly with the users of models at different stages of the modelling process. Many sustainability challenges need to be viewed from multiple stakeholder perspectives and require solutions that achieve shared mutual understanding and consensus [36]. Understanding interactions between human values and science and between objectivity and subjectivity supports modelling that can capture the social, cultural, and structural influences critical in sustainability.
(3) Foundation 3: a methodological angle that portrays the particular characteristics of complex systems such as interdependency and emergence
Complex systems can include characteristics such as emergence, self-organisation, nondeterministic behaviours, adaptation to environment, and hierarchies of agents [37]. To understand and model complex systems requires capturing their dynamic, systemic characteristics, including interdependencies between subsystems (or discrete components). Complex systems need to be understood in their integrity and interactivity [38], and in their intrinsic openness, in the sense of being subject to potential connectivity, learning, evolution, and adaptation [39]. Modelling methods may need to account for feedbacks to achieve this understanding [40].
