What can animals' inventiveness teach us about their agency?

The philosophical ambition of this first stage is to propose a definition of agency that can fully embrace the active role animals play in defining their own behaviours, without making the potentially anthropomorphic assumption that animals act according to goals. Thus, we will be interested in the animals’ behaviours of which they themselves are the triggering cause (otherwise they could not be said to be the agents), and which are characterised by their originality, i.e., which are not part of the known behavioural repertoire. The combination of these two conditions suggests that the behaviours studied cannot be explained solely by genetic determinism or by environmental conditions and are thus truly created by the animals under study. In this condition, the animals are both the triggering and the structuring cause of these behaviours (Ramsey et al. 2022).
Our analysis of agency will imply conceiving a form of non-human inventiveness, phenomenally visible, without anthropomorphic presuppositions. We will unify several fields—evolutionary biology, behavioural ecology, psychology, and philosophy—that have historically studied inventiveness in animals separately. We will analyse the existing literature to elucidate the distinction, if any, between invention, innovation and problem-solving across these fields. We will also question the links between behavioural flexibility and inventiveness to identify whether there is a threshold at which a behaviour becomes an invention. It will also be necessary to allow for the individual character of inventiveness (an individual is the inventor) while taking into consideration the behavioural repertoire of the population. This clarification is required to provide a common framework for future research on the subject.
This examination of animal agency will also lead us to reconsider related concepts. In particular, the behavioural repertoire should be redefined. Given the importance of animal inventions and innovations, the repertoire can no longer be understood as a static set of patterns but must be conceived as a dynamic, changing complex. This will bring us to question the way in which inventions relate to existing behaviours. Similarly, the relationship of agents with their environment can no longer be seen as occurring through the filter of affordances constrained by the goals imposed by natural selection; rather, it should be characterised by its equivocity. Yet, while we may develop hypotheses regarding the constitution of the behavioural repertoire and the relationship between animals and their environment by the end of this stage, it is stages 2 and 3 that will provide further empirical enlightenment on these issues.