
From linear promises of Nature-based solutions to relational practices of Nature-based assemblages
Nature-based Solutions (NBS) promise numerous social, environmental, and economic benefits inspired and supported by nature. However, as the literature on NBS has evolved, more critical perspectives have emerged, emphasizing that naive assumptions about universal co-benefits must be abandoned. Nonetheless, the potential of NBS remains strong in many policy initiatives. For instance, the recent EU Nature Restoration Regulation mandates the widespread implementation of ecological restoration projects across member countries. Ecological restoration is a key component of NBS. Proponents of the Nature Restoration Regulation (NRR) highlight its potential benefits for biodiversity and climate change, along with future projections of ecosystem services that are expected to offset the losses associated with land-use restrictions (such as forestry) and the costs of implementing restoration projects. However, there is a significant discrepancy between the optimistic visions and the potential challenges associated with the NRR.
This disparity can be traced back to the differing interests and values of various stakeholders, as well as the tension between viewing NBS as a promise versus as a practice. When treated as a promise, NBS tends to suggest a definite future with specific benefits, based on a linear assumption that well-planned interventions—like restoration, conservation, and the modification or creation of new ecosystems—will yield guaranteed co-benefits. Such promises often alienate both proponents and critics of NBS because the future remains uncertain and unpredictable. This uncertainty can lead to unmet expectations, as the social-ecological dynamics resulting from NBS interventions are often complex and surprising. Consequently, the promises surrounding NBS frequently go unrealized, leading to failures in delivering on those commitments.
We suggest that the promises associated with NBS should be complemented by approaches that embrace a relational perspective on the emergent and evolving nature of practices that arise from NBS interventions (figure 1). The new concept of Nature-based assemblages seeks to accomplish this by recognizing that the promises related to NBS are always subject to change due to ecological and social dynamics, rather than being fixed by engineering solutions. While NBS as a promise sets certain expectations for diverse stakeholders, the inherent unpredictability of ecological and social systems means that these expectations may not be fully met, resulting in disappointment and perceptions of NBS as failures.

The concept of NBS suggests a linear and instrumental causal relationship between the use of a specific ecosystem service for a human-defined purpose. In principle, the COEVOLVERS framework approaches NBS as a form of technology. However, in practice, NBS interventions cannot exert the same level of control over nature as one would with a grey infrastructure project. Nature possesses many degrees of freedom that cannot be fully managed by humans. The COEVOLVERS initiative has recognized this “fact of nature”, arguing that in light of rapid and unpredictable environmental changes, human design must not only accept these degrees of freedom but intentionally harness them to create solutions for an uncertain future. Moreover, from a normative standpoint, there is widespread agreement that NBS should also benefit nature, not just human interests. Therefore, respecting and recognizing the autonomy of nature is an ethical imperative in the design of NBS.
In recent work on a deliverable forthcoming soon, WP2 has turned to assemblage theory to suggest a way out of this conceptual impasse. Assemblage theory was systematically developed in Deleuze and Guattari’s classic “A Thousand Plateaus” and has been made more accessible by Manuel DeLanda, particularly in his 2016 book “Assemblage Theory.” This theory provides a different ontological framework for understanding NBS and their interventions. The key idea in the COEVOLVERS applications is that an assemblage is not a “system” as the term is used in “ecosystem.” Instead, assemblages are temporally and spatially contingent combinations of elements that retain their autonomous causal powers while simultaneously manifesting newly emerging causal powers in the assemblage. This concept is better captured by the original French term “agencement.” Unlike “systems,” which have clearly defined boundaries and follow an internal logic, like an engine, assemblages are open, evolving, and emergent.
Therefore, we propose conceptualizing an NBS intervention as arranging an assemblage of various elements in a specific location to trigger coevolutionary dynamics that produce certain desired causal effects. However, these causal effects are not directly controlled by human intervention; they are intentionally left to the coevolutionary dynamics. In other words, the “solution” is not designed by humans but discovered by the NBA. “Solution” is not a term of engineering but a term of appreciation.
For conceptual clarity, this approach proposes a new understanding of NBS as a “thing-idea.” The NBS is an idea in the sense that humans envision a certain causal chain between nature and a solution. It is also a thing, metaphorically likened to an egg, where the human idea fertilizes a “thing” in the sense of a material entity in the world (The “corps sans organes” in Deleuze and Guattari’s terminology). After human intervention, we no longer refer to this as NBS but as Nature-Based Assemblage NBA.
The NBA represents a coevolutionary process in which humans and non-humans co-create certain causal outcomes. As we have argued in previous publications and deliverables, this co-creation means that human interventions intentionally create evolutionary potential without planning or knowing the outcomes, although they may form scientifically grounded expectations. However, humans do not view deviations from the original idea as failures in “engineering” the NbS, but rather as desired manifestations of nature’s creativity, with humans being a part of nature equal to all other beings.
To make this general view practically relevant, we adopt DeLanda’s concept of “assemblage of assemblages” and apply assemblage theory across all ontological levels. We eventually progress to a flat ontology where agency and causal powers are not confined to fixed entities but are associated with assemblages that are seen as the constituents of the NBA. For instance, coevolutionary stakeholder analysis no longer assumes that certain actors are predefined but instead examines emerging stakeholder assemblages as the primary units of analysis. For example, under specific conditions, we can consider sheep, dogs, and shepherds as an assemblage, thereby recognizing them as stakeholders. Similarly, in a healing garden, patients and plants can also be viewed as an assemblage. We define criteria to determine when such combinations differ from mere conjunctions of members, with a key focus on whether the relationships among the members of an assemblage are embodied—grounded in emotional connections, corporeal interactions, and empathy, as well as mutual attention and care.
Practical consequences in research methodology are numerous. For example, in COEVOLVERS, one prominent method involves various forms of mapping a local NBA through counter-mapping procedures that go beyond conventional cartographic representations. In recent decades, practical tools have been developed across the social sciences, including emotion mapping, participatory mapping, and behavioral mapping. The NBA perspective allows for the systematization of these existing methods in NBS research, design, and implementation, all of which adhere to a coevolutionary logic grounded in assemblage theory. Countermapping not only helps identify stakeholder groups but also serves as a way to nurture these groups, ensuring the sustainability of the NBA over time. For instance, storytelling methods can evoke emotional connections between people and places, which can then be incorporated into participatory community maps. These community maps enhance the sense of place and foster relationships of identity and care among residents.

We summarise this new approach by conceptualising a dialectic between NBS as a "thing-idea" and NBA as practices of co-evolutionary world-making (see figure 2). Originally, the NBS is a human concept that combines with an intended materialization, referred to as the “thing-idea.” Through intervention, an NBA comes into being, initiating an autonomous co-evolutionary dynamic. This process, indicated by the size of the circles, represents the creative and emergent powers of the NBA driving the growth of coevolutionary potential and the growth of knowledge.
The outcome is a spiralling evolution, as humans learn from the realisation of their intervention. This learning manifests in the form of transdisciplinarity, which means that experts and non-experts, as members of the NBA, co-create a new version of the original thing-idea. Based on these insights, we transition from a mode of intervention to a mode of caring for and tending to the NBA. This shift further enhances its emergent powers, allowing for the shared flourishing of human and non-human members of the NBA. The spiralling causal dynamics unfold over time, provided that appropriate conditions are met.
Carsten Herrmann-Pillath and Simo Sarkki
Cover photo: Kevin Young on Unsplash