The LEGIT project is an academic research project funded through a European Research Council Starting Grant (ERC-StG). The project is hosted at Wageningen University (the Netherlands).

Models facilitate anticipatory governance      
Human beings live in constant interaction with their environment, both challenging it and being challenged by it. Earth and environmental researchers aim to understand this environment and the expected changes, often with the use of computer models. The understanding through computer models can feed into decisions on actions. Prominent contemporary examples are the use of climate models to support far-reaching policy decisions such as the 2015 Paris Agreement, and the use of epidemiological models to support policy decisions on Covid measures.

But models are not neutral
Models are simplified representations of reality. An incomplete or simplified understanding of natural and social processes can result in a range of possible model outcome. Both in simplifying the real world and in dealing with incomplete understanding of relevant processes, epistemic, political, and methodological choices have to be made to configure the model.

Value-laden models have real-world consequences
The implication is that certain perspectives and values are favoured by the model and its results. This subsequently steers the agenda for the decision-making process. This can steer focus towards one narrative, while other options are completely overlooked. Participatory modelling, as a way of co-production, is suggested to create support for model-based decision making and to decrease the risk of tunnel vision – but this comes with its own challenges of representation and power.

Yet, models have the authority to support policy decisions
Despite their inherent value-ladenness, whether originating from a participatory approach or not, models are frequently used tools to support decision making. This means that these models have acquired legitimacy: they have gained the authority to support policy decisions. Therefore, model use comes with a high responsibility, both for their developers and their end-users

Therefore, LEGIT aims to uncover the factors and processes through which computer models and their outcomes acquire legitimacy for decision support.

Questions we will answer are why a model structure was selected in the first place, which questions were asked and who formulated them, which considerations were made in setting up the model, how uncertainty was treated, which negotiations took place along modellers and between modellers and the decision makers before the final decision-support model was established, and how stakeholders perceive the model. All these factors not only influence the model results, but also what the model can do, for what purpose, and for whom.