Three Axis Framework

---

id: otf8f6z2kc8sdhy9w9f4njl title: Three Axis Framework of Ontology, Epistemology, and Praxiology desc: '' updated: 1787576738030

created: 1787232624996

NOTE: this is a stab, very messy and not to be trusted.

From Three Axes to Many

Three axes are too tidy. The trihedron of ontology, epistemology, and axiology comes from research-paradigm discourse, where it classifies the stances of human researchers.24 Applied to more-than-human collectives it under-describes, because each of the three collapses distinctions that make a practical difference to design and to teaching.

Three axes stay useful as an entry view. They are legible, they map onto being, knowing, and doing, and they carry the diagnostic tests set out below. They work better as a simplification for exposition than as a claim about how the world divides.

Competence Cannot Be Assigned in Advance

A framework that fixes beforehand which participants can know, teach, learn, value, or decide has already smuggled human judgement back in. That move ends with humans again, whatever vocabulary surrounds it.

The distribution of competence across kinds of entity is an empirical question. Levin's programme makes the point operational: competency shows in what a system achieves when perturbed, measured as navigation through some space of possibilities, and this measure holds across substrates without prior commitment about which substrates qualify.25^, 26 Multiscale competency then appears as a property of developmental and evolutionary material rather than as a privilege of nervous systems.27

Competence Is Relative to Circumstances, and It Can Fail

All beings learn, in different ways, and those ways are competent or incompetent for the circumstances the being finds itself in. This formulation replaces any comparison with human learning, which reintroduces a single scale through the back door.

Two commitments hold it together.

Competence is a fit rather than a rank. Ecological rationality states this for human and animal cognition alike, treating intelligence as a match between the capacities a being deploys and the structure of the environment it faces.47 The comparative literature on adaptive specialisation of learning supplies the empirical form,48 and Umwelt supplies the circumstance-relative frame for asking what any being's world affords it.49 The anthropocentric bias that a single scale produces is itself well documented.50

Incompetence stays available. A being can learn in ways that suited conditions that no longer hold, which is what rapid habitat change produces. A design can be rejected by its intended occupants. Keeping failure as a real outcome prevents the position collapsing into celebration, and it preserves refusal, avoidance, and non-occupancy as findings rather than as embarrassments.

Three further consequences follow.

First, dimensions cannot be calibrated by how much of each a human has. The cognitive light cone places cells near the origin and humans further out, which measures one capacity on one scale and says nothing about the many capacities on which a mycorrhizal network or a colony exceeds a person.

Second, roles are not fixed. A participant can hold expertise on one dimension while learning on another, and the assignment can reverse within a single project. Teacher and learner become positions in a relation rather than properties of a kind.

Third, the criterion for adding, merging, or discarding a dimension is pragmatic usefulness in world-making, tested empirically. The number of dimensions is therefore provisional and should be expected to change.

Candidate Dimensions

The families of being, knowing, and doing survive as a grouping. Within each, the following dimensions resist collapse because real cases separate them.

Strength of independence varies. Constituency against sense-making, sense-making against influencing, and deciding against acting each separate cleanly in the diagnostic tests below. Refusing against influencing separates in the resistance literature. Communicating may eventually fold into sense-making, and capability may fold into constituency, so both should be held loosely.

Terms

Three cautions on naming.

Praxiology is established in another sense, as the general theory of efficient action in Kotarbiński's tradition, so readers hear method rather than authority. For an architecture audience it also reads as studio practice. Prefer acting or agency.

Axiology is standard and precise, though it pairs awkwardly with the gerunds and hides the activity. Valuing keeps the emphasis on something a participant does.

Gerunds throughout hold the framework closer to processes than to properties, which suits a relational account. The cost is that the resulting labels look less like established philosophical categories.

The Trihedral View

This framework brings together the Stair of More-than-Human Participation and a broader account of more-than-human co-learning. Reduced to three axes, it proposes:

The Stair of More-than-Human Participation1^, 2 organises the praxiology axis because it assesses the degree to which nonhuman beings are recognised, supported, autonomous, and able to influence decisions. Co-learning strategies structure the epistemological axis because they operate as capacities for normative orientation, constituency discovery, possibility expansion, transformation, comparison, and temporal responsibility. The ontological axis addresses the composition and boundaries of the collective, extending concern from human individuals and organisations to organisms, species, ecological communities, technical systems, abiotic processes, and other possible participants.

In practice, the axes are not independent. Ontologies determine which entities can become knowers and political participants. Epistemologies make some entities and relations legible while obscuring others. Power determines which ontologies and forms of knowing can influence consequential decisions. More-than-human transformation therefore emerges as the interlinked expansion and revision of ontology, epistemology, and power.

More-than-human design involves interlinked transformations in what can exist as a participant, how heterogeneous participants can produce knowledge, and how their contributions can influence collective action.

Diagram

                          EPISTEMOLOGY
                                ^
                                |
                                |
                                |
                                |
                                |
                                |
                                |
                                |
ONTOLOGY <----------------------+
                                 \
                                  \
                                   \
                                    \
                                     v
                                   PRAXIOLOGY

^{The axes are lumpy and co-dependent. The spatial regions are fuzzy.}

Recognition | Being | Ontology Axis

From: narrow, bounded, anthropocentric constituencies Towards: plural, relational, revisable, multiscalar, nested and overlapping more-than-human constituencies

Cf. world-making, being. Types of beings, agencies, processes, individualities, selves, communities, and collectives. Recognition, graining.

Knowledge | Learning | Epistemology Axis

From: centralised, unitary, representational, human-exclusive knowledge Towards: local, experimentally procured, distributed, more-than-human co-learning and co-construction of knowledge

Cf. sense-making, knowing, learning, inquiring, understanding, communicating, cognitive light cones, active inference, distributed cognitions, Markov blankets and organisational learning.

Participation | Doing | Praxiology Axis

From: exclusion, instrumentalization, and paternalism Towards: autonomy, shared authority, nonhuman leadership, and commoning

Cf. decision-making, doing, acting. Governance and governability, praxeology, axiology, politics. Control, influence, authority, stair of more-than-human participation, ability to influence things, make and change decisions.

Sample Effects

1. High learning / Low power. Rich sensing and modelling of biodiversity. Humans still decide everything.
2. High power / Low learning. Many stakeholders consulted. Poor understanding of consequences.
3. Broad inclusion / Low power. Many species recognised. Little influence granted.

High Across All Three Axes

More-than-human co-learning and co-governance.

Ontology

Cf.

• Agency

The ontological axis concerns the constitution of the collective:

Who and what can exist as a participant, stakeholder, agent, knower, designer, leader, community, or process of concern?

This axis must address both the scope of inclusion and the form in which entities are understood. A project may nominally include animals, plants, rivers, algorithms, or ecosystems while continuing to represent them as bounded resources. Ontological expansion consequently involves more than adding categories to a stakeholder list. It can also require changes in how individuality, agency, relation, life, environment, and community are understood.

The more-than-human design literature commonly treats the human subject as embedded within relations involving other organisms, materials, technologies, and environments. In this account, knowledge is always situated, embodied, partial, and plural, produced from within a particular position in the world.3

More-than-human participatory design has similarly been described as an ontological and ethical commitment to relationality that extends beyond adding a category called “nonhuman” to existing frameworks. One synthesis distinguishes a sustainability lineage, including modernist and rights-based branches, from an entanglement lineage, including care-based and co-ontological branches.4

Range and Composition

The axis includes overlapping and nested scopes distributed in a lumpy fashion, in a partially stable and fuzzy sequence. Increasing capability of goal-setting, indexed to the cognitive light cone (Private),17^, 18 offers one measure, though on this measure cells sit near the origin and humans sit further away. Another approach considers relational properties such as integration and redundancy, through which beings with different cognitive or goal-setting scopes become enrolled in collective, or field-level, creativity. This is a composite dimension that remains unresolved, so it should not be used reductively.

This axis requires further empirical work, both to understand the capabilities and competencies of differently bounded agencies and to test their capacity to supply beneficial novelty, linking to the objectives of design. Decision-making, the ability to live better or worse lives, resource usage, and processing capabilities all depend on how the world is split into individualities, relationships, and processes (cf. Biological Individuality).

An anthropocentric perspective measures or recognises concentrically from human individuality:

• individual humans;
• human groups;
• human communities;
• technically amplified systems;
• companion species;
• ecosystems;
• planetary processes;
• Gaian systems.

A dynamical systems perspective treats these as continuous patterns rather than discrete levels, deemphasising or bypassing agency and goal-directedness altogether (cf. Phase Space).

A goal-alignment perspective compares scales through their cognitive light cones (Private):

• human individuals;
• human organisations and communities;
• multispecies encounters;
• individual organisms;
• populations and species;
• symbioses and holobionts;
• habitats and ecological communities;
• rivers, forests, soils, and landscapes;
• technical and computational agents;
• material and infrastructural assemblages;
• climate, hydrological, and geological processes;
• Earth systems and Gaian collectives.

These are not necessarily steps. They are candidate scales and forms of organisation, many of which overlap or nest only partially.

The epistemological axis (or the Stair of More-than-Human Participation, cf. Ladder) depends on ontological lumping. Its recognition step identifies organisms, genes, and ecosystems as possible constituents of shared cultures, while later steps propose that nonhuman agents should be able to pursue their interests and lead appropriate design projects.5

The intellectual history of the relevant ontological ideas includes more-than-human geography, posthumanism, new materialism, actor-network and assemblage theories, pluriversality, ontological politics and Indigenous relational ontologies.

Challenges

Ontological expansion does not automatically deliver justice.

A flat or relational ontology may recognise that humans, animals, plants, materials, and technologies all participate in events, yet this recognition can obscure radically unequal capacities to inflict harm, allocate resources, or determine outcomes. Ontological plurality must therefore remain connected to normative analysis and differentiated responsibility.9^, 10

Epistemology

Cf.

• Ladder
• Learning, Learning

The epistemological axis concerns how a heterogeneous collective becomes capable of learning:

How can participants discover one another, communicate across differences, generate alternatives, intervene experimentally, compare consequences, and revise their commitments?

The axis extends beyond the possession of information. It includes:

• who can be a knower;
• what counts as evidence;
• how boundaries are drawn;
• how signals become meaningful;
• how uncertainty is represented;
• how knowledge and values interact;
• how models enter decisions;
• whether participants can alter the questions as well as the answers.

More-than-human participatory work must address the ontologies and epistemologies that organise a design problem.11

Intellectual history of this axis includes pragmatist inquiry, cybernetics, systems theory, ecological and distributed cognition, situated knowledge,28 feminist epistemology, multispecies ethnography, arts of noticing, and research through design. Plural epistemology without unification is developed as an ecology of knowledges.29

Co-Construction Rather Than Reversed Teaching

Cf. Pedagogy, Expertise, Planetary Pedagogy.

Four positions get conflated in this literature. The first three keep the human as beneficiary.

1. Teaching humans about nonhumans. The nonhuman is subject matter. Content changes.
2. Teaching humans alongside nonhumans. The nonhuman is a co-presence, provocation, or affective partner. Relations change.
3. Nonhuman as teacher of humans. The nonhuman gains pedagogical authority while the human remains the learner. Authority changes.30
4. Co-construction across participants. Competencies are distributed across kinds of entity, roles are unassigned in advance, and who teaches whom on any given dimension is settled empirically. Constituency changes.

The fourth position is the one this framework requires, and it depends on the argument above that competence cannot be assigned in advance. Positions one to three all preserve a fixed division of epistemic labour, and any fixed division has to be drawn by somebody, which returns authority to human judgement. Reversing the direction of teaching is insufficient because it retains the assumption that direction is a property of kinds.

The strongest objection to positions three and four holds that the figure of the animal-as-teacher answers human needs and desires rather than describing a relation, and asks how far nonhuman animals enact the volition, tact, or intentionality that education would require.31 The animal-behaviour literature answers this without appealing to mental states. Teaching is identified functionally: the scaffolder modifies its activity in the presence of a naive learner, the scaffolder bears a cost, and the learner acquires a skill faster than it otherwise would.32 The claim is therefore functional rather than intentional, and it uses a criterion established in behavioural ecology rather than one invented for design.

The cost condition then does further work. Where humans extract knowledge from nonhuman participants at cost to those participants and retain the whole benefit, the relation is extraction rather than co-learning. Reciprocal cost and distributed capability gain separate the two. Cf. the treatment of learning as expensive, risky, and violent in Learning.

Nonhuman Expertise as Evidence

The claim that nonhuman participants hold expertise is empirically supported and it travels further than a claim about agency, because professional disciplines discount ethical appeals and respond to competence.

Three established definitions of expertise operate in the literature: expertise as social construction, expertise as exceptional performance, and expertise as knowledge. Animals satisfy all three, and recognising this resolves problems in the human expertise literature rather than creating them.33 The standard taxonomy nonetheless grounds expertise in socialisation into a linguistic community, which excludes every nonhuman by construction rather than by evidence.34 Naming that construction is more precise than a general complaint about anthropocentrism, because it identifies the move that does the excluding and leaves the rest of the taxonomy usable.

Expertise held by a collective rather than an individual is the normal case in nonhuman worlds. Cultural transmission research documents accumulation, maintenance, and loss of collective skill across generations.35 On this reading, extinction removes accumulated expertise as well as individuals, and fragmentation degrades the conditions under which expertise passes between generations.36

Depth of Learning

The forms of flexibility set out under Transformative Capacity below restate an existing sequence. Single-loop and double-loop learning, Bateson's logical levels, first- to third-order learning, and transformative learning all describe increasing depth of revision.37 Expansive learning supplies the specific claim that participants can alter the questions as well as the answers.38 A relational reworking of transformative learning is the closest existing attempt at what this axis proposes.39

Each of these holds the constituency fixed. They describe how deeply a learner changes and say nothing about who gets to be a learner. That gap is what the constituency dimension fills, and it is the framework's clearest point of difference. Situated learning is a partial exception, because its account of participation is epistemic rather than political, which makes it a useful contrast with the Stair.40

Comparable capacity sets exist. Six threshold concepts for transformative sustainability learning cover adjacent ground and warrant direct comparison with the epistemic capacities below.41 Competence frameworks in sustainability higher education locate capacity in the individual human graduate, which makes them a poor fit for capacities held by heterogeneous collectives.42 A search in August 2026 found no more-than-human or post-anthropocentric competence framework in that literature.

Who Counts as a Knower

Exclusion from the status of knower is a recognised form of injustice, and the analysis extends to nonhuman animals through know-how rather than testimony.43 The design of questions determines what a nonhuman participant can be seen to know, which makes method a political matter rather than a technical one.44 In higher education specifically, nonhuman animals as significant subjects of teaching receive little recognition.45 The closest existing statement of a planetary pedagogy remains a commissioned report rather than a research programme.46

Epistemic Capacities

The original teaching strategies can be organised as six capacities. These should not be shown as another axis or a fixed sequence. They are epistemic operations that enable trajectories within the three-axis space.

1. Normative orientation. What should count as improvement, thriving, justice, or success, and for whom?

2. Constituency discovery. Who or what participates, at which scales, through which boundaries and relations?

3. Expansion of design potential. How can a collective escape inherited assumptions and discover otherwise inaccessible possibilities?

4. Transformative Capacity. How readily can a collective revise its relationships, infrastructures, models, and goals?

5. Legibility of consequences. How can heterogeneous, distributed, uncertain, and temporally uneven consequences become comparable without erasing their differences?

6. Temporal responsibility. Under what conditions can a collective justify short-term or local deterioration in pursuit of more probable long-term or larger-scale flourishing?

Normative Orientation

Normative orientation makes criteria explicit enough to be contested, tested, and revised. It resists the common tendency to treat human convenience, economic growth, efficiency, or ecosystem services as self-evident objectives.

Intellectual heritage

• pragmatist inquiry;
• environmental ethics;
• ecofeminism;
• capability approaches;
• ecological and multispecies justice;
• agonistic and deliberative democracy;
• rights of nature;
• Indigenous legal and ethical traditions.

Emerging work

• relational accounts of multispecies flourishing;
• institutionalisation of multispecies justice;
• ecocentric governance;
• capability-oriented evaluation of nonhuman lives;
• criteria for legitimate nonhuman leadership;
• explicit integration of justice into ecological modelling.

Criteria cannot be permanently settled in advance. A co-learning system must allow new participants, evidence, and harms to alter its definition of success.

Constituency Discovery

Constituency discovery identifies overlooked agents, dependencies, vulnerabilities, signals, capabilities, and ecological or technical processes. It includes empirical and interpretive work because participants are not merely waiting to be found as stable, pre-given units.

Intellectual heritage

• ecology and ethology;
• ecological psychology and affordance theory;
• semiotics and signalling theory;
• systems theory and cybernetics;
• actor-network theory;
• assemblage theory;
• multispecies ethnography;
• feminist standpoint epistemology;
• Indigenous relational knowledge.

Emerging work

• environmental and multispecies sensing;
• nonhuman personas and representation tools;
• animal-computer interaction;
• machine-mediated noticing;
• diffractive data interfaces;
• multispecies information practices;
• biological agency and distributed intelligence.

Markov blankets, active inference, and related formal approaches are potentially valuable as provisional tools for investigating boundaries, conditional dependencies, sensing, and action.19^, 20 Their most promising use here is heuristic, converting claims about agency into empirically contestable questions, in the spirit of the operational questions collected in Agency:

• Which states appear conditionally separated?
• Which exchanges function as sensing or action?
• At what scale does the proposed boundary become informative?
• Which relations cross or destabilise that boundary?
• What disappears when the system is partitioned in this way?

This is a proposed extension of the framework (cf. Phase Space and Design Potential (Private)).

Nonhuman personas, monitoring systems, and ecological models can highlight neglected interests, but they remain vulnerable to proxy error and human interpretation.12^, 13

Expansion of Design Potential

Cf. Design Potential (Private).

Possibility expansion reduces the conceptual and practical risks of novelty. It searches for unrealised futures in historical traces, marginal experiments, local adaptations, Indigenous practices, alternative lifestyles, biological strategies, technical innovations, and speculative displacements.

Intellectual heritage

• speculative and critical design;
• design fiction;
• utopian studies;
• critical pedagogy;
• prefigurative politics;
• evolutionary exploration;
• creativity research;
• Indigenous and pluriversal futuring;
• arts of noticing and fabulation.

Emerging work

• more-than-human provotypes;
• multispecies role-play;
• participatory and Indigenous futuring;
• speculative fabulation;
• nature futures;
• algorithmic generation of unusual design states;
• experimentation with nonhuman-led starting conditions.

Transformative Capacity

Cf. Transformation.

Transformative capacity concerns the feasibility, cost, speed, reversibility, and quality of meaningful change. It includes parametric modelling, digital fabrication, rapid and slow prototyping, in-silico experimentation, adaptive infrastructure, maintenance, and service-based arrangements.

The crucial distinction is between making it easier to change a parameter and making it possible to change the system's constituency, values, or governance.

Forms of flexibility

• Parameter flexibility. Changing values within an established model.
• Structural flexibility. Changing relations among components.
• Normative flexibility. Changing what the intervention is intended to achieve.
• Ontological flexibility. Changing who or what is recognised as a participant.

Intellectual heritage

• cybernetics and feedback;
• adaptive management;
• resilience and transformation;
• evolutionary systems;
• research through design;
• infrastructuring;
• maintenance and repair;
• service and transition design.

Emerging work

• adaptive multispecies infrastructure;
• ecological digital twins;
• living and biohybrid artefacts;
• open-ended multispecies design;
• parametric habitat structures;
• post-deployment ecological governance;
• design systems that respond to nonhuman modification.

Research on more-than-human infrastructure argues that human control and knowledge are inadequate for the scale and unpredictability of ecosystems. It proposes experimental and deliberative infrastructuring that supports the political participation of human and nonhuman communities without imposing exclusively human-defined outcomes.14

Recent biodesign work similarly presents multispecies participation as a temporally extended relational process in which organisms may transform materials, surfaces, infrastructures, or systems. It proposes relationality, reciprocity, and distributed responsibility as principles requiring further refinement in practice.15

Technical flexibility can consolidate human control. A highly adaptable system may still allow only humans to define its objectives, participants, and acceptable responses.

Legibility of consequences

Cf. Relational Accounting, Uncertainty.

Legibility of consequences makes outcomes available for scrutiny. Relevant approaches include measurement, modelling, simulation, lifecycle assessment, multicriteria analysis, data visualisation, scenario comparison, threshold analysis, and qualitative accounts.

The objective should be to construct a comparison space in which different designs, including biogenic and anthropogenic proposals, can be evaluated through shared but revisable criteria.

Possible dimensions

• Capabilities and opportunities for flourishing
• Morbidity, mortality, displacement, and disturbance
• Habitat functions and ecological relationships
• Energy, material, water, and nutrient flows
• Labour, maintenance, and care
• Adaptability and reversibility
• Spatial displacement of benefits and harms
• Time to benefit and duration of effect
• Uncertainty and quality of evidence
• Distribution of benefits, burdens, risks, and authority

Intellectual heritage

• decision theory;
• operations research;
• ecological economics;
• systems dynamics;
• lifecycle assessment;
• environmental impact assessment;
• capability approaches;
• critical data studies;
• visualisation and model criticism.

Emerging work

• plural and multispecies metrics;
• reflexive biodiversity accounting;
• ecological simulation involving competing agents;
• diffractive interfaces;
• lifecycle comparisons of artificial and biogenic habitats;
• uncertainty-aware visualisations;
• nonhuman-informed model revision.

Legibility is political. A quantified entity may become more influential, but quantification can also simplify, decontextualise, or render it available for managerial control. Visualisation should expose missing data, uncertainty, contested boundaries, and incompatible consequences instead of concealing them behind an apparently objective optimum.

Temporal Responsibility

Temporal responsibility connects present experimentation to lifecycle effects, ecological succession, intergenerational consequences, maintenance obligations, and uncertain future states.

Intellectual heritage

• evolutionary and ecological dynamics;
• resilience theory;
• futures studies and backcasting;
• intergenerational justice;
• lifecycle assessment;
• adaptive pathways;
• real-options reasoning;
• long-term governance;
• transition studies.

Emerging work

• multispecies adaptive pathways;
• ecological trigger points;
• long-term nonhuman representation;
• institutionalised commoning;
• lifecycle modelling of habitat provision;
• governance across ecological and generational timescales;
• monitoring arrangements that allow future participants to revise commitments.

The nonhuman-leadership framework identifies persistence as a feature of leadership and associates it with commoning and institutionalisation across ecological timescales.7^, 8 It argues that temporary gestures must become durable commitments if they are to support multispecies flourishing.

Projected future benefits can become a justification for imposing immediate harms on politically weak human or nonhuman communities. Claims of long-term improvement therefore require explicit treatment of uncertainty, burden distribution, reversibility, and authority.16

Praxiology

Who or what can alter decisions, infrastructures, resource distributions, behaviours, and future trajectories?

See Ladder for a discussion of the Stair of More-than-Human Participation as a way to structure the praxiology axis.

Intellectual heritage includes democratic participation, participatory design, agonistic politics, commons and commoning, multispecies justice, and ecological democracy.

Responsibility requires more than an extension of agency. Widely distributed agency across an assemblage can make accountability for domination, extraction, or ecological harm difficult to locate, so power analysis must distinguish:

• ability to affect an event;
• authority to determine a collective goal;
• responsibility for foreseeable harm;
• vulnerability to harm;
• ability to refuse;
• capacity to redirect future action.

Analytical matrix

Diagnostic Tests

1. Broad ontology, weak epistemology, weak power. Many species, ecosystems, and processes are declared relevant, but there are few reliable methods for learning about them and no means for their activities to affect decisions. Typical failure: rhetorical inclusion.

2. Strong epistemology, narrow ontology, weak power. Extensive sensing, modelling, and simulation are used, but only a predefined selection of entities is represented and humans retain exclusive authority. Typical failure: sophisticated anthropocentric management.

3. Broad ontology, strong epistemology, weak power. Nonhuman lives and relationships are studied carefully, but their preferences, resistance, or innovations cannot redirect action. Typical failure: extractive knowledge without political consequence.

4. Broad ontology, weak epistemology, strong declared power. Rights or representation are granted, but the relevant institutions lack methods for interpreting nonhuman signals, ecological relations, or uncertain outcomes. Typical failure: nominal authority exercised through unreliable human proxies.

5. Strong epistemology, strong power, narrow ontology. Well-studied and politically influential participants receive substantial protection, while inconspicuous, uncharismatic, slow, aggregated, or poorly understood entities remain excluded. Typical failure: selective multispecies privilege.

Expansion across all three axes

Participants and boundaries remain revisable; heterogeneous knowledge practices support ongoing inquiry; and nonhuman contributions can alter consequential decisions over time.

Provisional description: more-than-human co-learning and co-governance.

Trajectories Through the Matrix

Projects should be represented as trajectories rather than static points.

1. From ontology through epistemology to praxiology. Recognise bats as participants. Investigate their sensory and behavioural relations with lighting. Allow bat activity to alter lighting standards and operations.

2. From praxiology through epistemology to ontology. Grant legal standing to a river. Discover that existing monitoring cannot represent its relations. Develop new institutions and knowledge practices. Reconceptualise the river as a living, multispecies territory.

3. From epistemology to ontology. Collect environmental signals. Detect previously unrecognised relations or agents. Redraw the constituency and system boundaries.

4. From epistemology to praxiology. Model the habitat preferences of birds. Compare design alternatives. Allow observed preferences to determine spatial or material choices.

5. From praxiology to ontology. Enable a tree to redirect a design process. Encounter the tree’s symbionts, temporalities, and dependencies. Redefine the participant as a relational community rather than an isolated organism.

These trajectories demonstrate why movement on one axis can produce pressure for movement on the others.

Capacities and Empowerment

The six capacities can be interpreted as enabling conditions for movement up the Stair.

The Stair describes degrees of participatory power. The six capacities describe epistemic conditions through which that power can become informed, revisable, consequential, and durable. The ontological axis describes the possible constituency within which both learning and power operate.

---

Dependencies and Cautions

1. Ontological recognition is necessary but insufficient. A being, process or relationship must become perceptible as a participant before it can exercise acknowledged political influence. Yet merely describing the world as relational does not redistribute authority.

2. Epistemic inclusion is not political inclusion. Sensing, modelling, representing, or studying nonhuman beings can increase human knowledge while leaving decision-making power unchanged.

3. Power without epistemic capacity risks proxy domination. Granting representation or rights without adequate ways of learning from relevant beings and ecological relations can reproduce human interpretation as the final authority.

4. More data does not necessarily produce more co-learning. Data practices can disclose overlooked relations, but they can also translate living systems into forms optimised for managerial intervention. Co-learning requires reflexivity about models, boundaries, categories, and infrastructures.

5. High participation does not imply the absence of conflict. More-than-human communities contain incompatible interests, predation, competition, disturbance, and unequal vulnerabilities. The aim is just and revisable ways of negotiating difference.

6. The three axes are historically and politically constituted. Dominant ontologies and epistemologies are maintained through institutions, property systems, technical standards, colonial histories, and economic power. Movement through the matrix is not merely conceptual.

7. Co-learning requires the possibility of reframing. A collective is not genuinely co-learning if nonhuman contributions can change only parameter values. Stronger co-learning allows participants to alter questions, criteria, boundaries, models, relationships, and commitments.

8. More-than-human leadership requires consequence and persistence. A participant cannot reasonably be described as leading if its contribution produces no directed action. Leadership becomes stronger when influence persists through governance, infrastructure, monitoring, budgets, maintenance, and commoning.6^, 7^, 8

Research Agenda

Studying relations between the axes

Future work could investigate whether changes on one axis reliably enable, constrain, or destabilise changes on the others.

Possible questions include:

• Does broader ontological recognition usually lead to greater power?
• Which epistemic practices support consequential rather than symbolic participation?
• Can increased nonhuman power generate new ways of knowing?
• Under what conditions does technical legibility increase managerial control?
• How should conflicts between ontological plurality and operational decision-making be handled?

Developing empirical measures for assessing conditions along axes

The framework requires indicators capable of distinguishing:

• nominal from consequential recognition;
• observation from reciprocal learning;
• representation from autonomy;
• temporary intervention from institutionalised influence;
• technical adaptability from normative and ontological adaptability.

Measurements should remain open to revision by the processes they evaluate.

Understanding Forms of Individuality

Cf. Biological Individuality.

A promising research direction is to compare formal boundary-discovery methods with:

• Umwelt analysis;
• affordance mapping;
• signalling and semiotic analysis;
• capability assessment;
• multispecies ethnography;
• Indigenous relational accounts;
• assemblage and network mapping;
• Markov blankets.

The objective is to understand boundaries as scale-dependent, empirically testable, and politically consequential.

Developing comparison spaces for decisions irrespective of their origin

A common comparison space could evaluate ecological and technical contributions without assuming that natural processes are always beneficial or that human-made systems are inherently superior.

Such a space should:

• preserve multiple criteria;
• disclose uncertainty;
• show distributions rather than only totals;
• represent several temporal and spatial scales;
• identify missing constituencies;
• permit criteria to change as learning proceeds.

Investigating Properties of change

More-than-human design needs criteria not only for outcomes but for the quality of transformation.

Possible qualities include:

• reversibility;
• responsiveness;
• repairability;
• ecological openness;
• capacity for nonhuman modification;
• distribution of transitional burdens;
• preservation of options;
• support for refusal;
• persistence without institutional rigidity.

Institutionalising More-than-Human co-learning

Many experimental practices remain workshops, prototypes, exhibitions, or short research projects. Further work is needed on:

• enduring governance arrangements;
• legal and property structures;
• budgets and maintenance;
• long-term monitoring;
• ecological observatories;
• data stewardship;
• trigger points for reconsideration;
• representation across generations;
• mechanisms for nonhuman refusal and resistance.

Notes

---

Footnotes

1. Pretorius, Lynette. “Demystifying Research Paradigms: Navigating Ontology, Epistemology, and Axiology in Research.” The Qualitative Report 29, no. 10 (2024): 2698–715. https://doi.org/10.46743/2160-3715/2024.7632.˄

2. Fields, Chris, and Michael Levin. “Competency in Navigating Arbitrary Spaces as an Invariant for Analyzing Cognition in Diverse Embodiments.” Entropy 24, no. 6 (2022): 819. https://doi.org/10.3390/e24060819. Not yet in Zotero.˄

3. Levin, Michael. “Technological Approach to Mind Everywhere: An Experimentally-Grounded Framework for Understanding Diverse Bodies and Minds.” Frontiers in Systems Neuroscience 16 (2022): 768201. https://doi.org/10.3389/fnsys.2022.768201. See also Chis-Ciure, Robert, and Michael Levin. “Cognition All the Way Down 2.0: Neuroscience beyond Neurons in the Diverse Intelligence Era.” Synthese 206, no. 5 (2025): 257. https://doi.org/10.1007/s11229-025-05319-6.˄

4. Levin, Michael. “Darwin’s Agential Materials: Evolutionary Implications of Multiscale Competency in Developmental Biology.” Cellular and Molecular Life Sciences 80, no. 6 (2023): 142. https://doi.org/10.1007/s00018-023-04790-z.˄

5. Todd, Peter M., and Gerd Gigerenzer. “Environments That Make Us Smart: Ecological Rationality.” Current Directions in Psychological Science 16, no. 3 (2007): 167–71. https://doi.org/10.1111/j.1467-8721.2007.00497.x. See also Todd, Peter M. Ecological Rationality: Intelligence in the World. Cary: Oxford University Press, 2012.˄

6. Shettleworth, Sara J. Cognition, Evolution, and Behavior. 2nd ed. 1998. Reprint, Oxford: Oxford University Press, 2010.˄

7. Bueno-Guerra, Nereida. “How to Apply the Concept of Umwelt in the Evolutionary Study of Cognition.” Frontiers in Psychology 9 (2018): 2001. https://doi.org/10.3389/fpsyg.2018.02001.˄

8. Scotto, Carolina. “The Anthropocentric Bias in Animal Cognition.” ArtefaCToS. Revista de estudios sobre la ciencia y la tecnología 13, no. 1 (2024): 85–116. https://doi.org/10.14201/art2024.31800.˄

9. Di Paola, Marcello. “Virtue, Environmental Ethics, Nonhuman Values, and Anthropocentrism.” Philosophies 9, no. 1 (2024): 15. https://doi.org/10.3390/philosophies9010015.˄

10. Owe, Andrea, Seth D. Baum, and Mark Coeckelbergh. “Nonhuman Value: A Survey of the Intrinsic Valuation of Natural and Artificial Nonhuman Entities.” Science and Engineering Ethics 28, no. 5 (2022): 38. https://doi.org/10.1007/s11948-022-00388-z.˄

11. Metuonu, Iheanacho Chukwuemeka. “Beyond Anthropocentrism: A Value-Theoretic Approach to Nonhuman Animal Ethics.” Journal of Advances in Education and Philosophy 9, no. 03 (2025): 147–52. https://doi.org/10.36348/jaep.2025.v09i03.007.˄

12. Roudavski, Stanislav. “The Ladder of More-than-Human Participation: A Framework for Inclusive Design.” Cultural Science 14, no. 1 (2024): 110–19. https://doi.org/10.2478/csj-2024-0015.˄

13. Roudavski, Stanislav, and Alexander Holland. “Are You Blocking Nonhuman Leadership? Five Questions to Find Out.” PDC ’26: Proceedings of the 19th Participatory Design Conference (New York) 2 (2026): 372–81. https://doi.org/10.1145/3789492.3796416.˄

14. Poikolainen Rosén, Anton, Antti Salovaara, Andrea Botero, and Marie Louise Juul Søndergaard, eds. More-than-Human Design in Practice. Abingdon: Routledge, 2025.˄

15. Heitlinger, Sara, Ann Light, Yoko Akama, Kristina Lindström, and Åsa Ståhl. “More-than-Human Participatory Design.” In Routledge International Handbook of Contemporary Participatory Design, edited by Rachel Charlotte Smith, Daria Loi, Heike Winschiers-Theophilus, Liesbeth Huybrechts, and Jesper Simonsen, 79–110. London: Routledge, 2025.˄

16. Levin, Michael. “The Computational Boundary of a ‘Self’: Developmental Bioelectricity Drives Multicellularity and Scale-Free Cognition.” Frontiers in Psychology 10 (2019): 2688. https://doi.org/10.3389/fpsyg.2019.02688.˄

17. Levin, Michael. “Technological Approach to Mind Everywhere: An Experimentally-Grounded Framework for Understanding Diverse Bodies and Minds.” Frontiers in Systems Neuroscience 16 (2022): 768201. https://doi.org/10.3389/fnsys.2022.768201.˄

18. Roudavski, Stanislav. “The Ladder of More-than-Human Participation: A Framework for Inclusive Design.” Cultural Science 14, no. 1 (2024): 110–19. https://doi.org/10.2478/csj-2024-0015.˄

19. Asenbaum, Hans, and Sonia Bussu. “Democratic Assemblage: Power, Normativity, and Responsibility in More-than-Human Participation.” Theoria. Theoria 72, no. 183 (2025): 1–23. https://doi.org/10.3167/th.2025.7218301.˄

20. Celermajer, Danielle, Anthony Burke, Stefanie Fishel, Erin Fitz-Henry, Nicole Rogers, David Schlosberg, and Christine Winter. Institutionalising Multispecies Justice. Cambridge: Cambridge University Press, 2025.˄

21. Giaccardi, Elisa, Stanislav Roudavski, and Martín Tironi. “Participation and the Troubles with Intelligence.” PDC ’26: Proceedings of the 19th Participatory Design Conference 2026 (New York) 1 (2026): 385–90. https://doi.org/10.1145/3796624.3796632.˄

22. Haraway, Donna. “Situated Knowledges: The Science Question in Feminism and the Privilege of Partial Perspective.” Feminist Studies 14, no. 3 (1988): 575–99. https://doi.org/10.2307/3178066.˄

23. Sousa Santos, Boaventura de. Epistemologies of the South: Justice against Epistemicide. Abingdon: Routledge, 2016.˄

24. Carvalho, Isabel Cristina de Moura, Carlos Alberto Steil, and Francisco Abrahão Gonzaga. “Learning from a More-than-Human Perspective. Plants as Teachers.” The Journal of Environmental Education 51, no. 2 (2020): 144–55. https://doi.org/10.1080/00958964.2020.1712646. See also Bone, Jane. “The Animal as Fourth Educator: A Literature Review of Animals and Young Children in Pedagogical Relationships.” Australasian Journal of Early Childhood 38, no. 2 (2013): 57–64. https://doi.org/10.1177/183693911303800208.˄

25. Pedersen, Helena. “Animal Teachers: Nonhuman Pedagogy and the Question of Deceit.” Educational Theory, 2026. https://doi.org/10.1111/edth.70098. Not yet in Zotero. Volume, issue, and pages need confirming.˄

26. Caro, Timothy M., and Mark D. Hauser. “Is There Teaching in Nonhuman Animals?” The Quarterly Review of Biology 67, no. 2 (1992): 151–74. https://doi.org/10.1086/417553. Refined in Hoppitt, William J. E., Gillian R. Brown, Rachel Kendal, Luke Rendell, Alex Thornton, Mike M. Webster, and Kevin N. Laland. “Lessons from Animal Teaching.” Trends in Ecology & Evolution 23, no. 9 (2008): 486–93. https://doi.org/10.1016/j.tree.2008.05.008. See also Kline, Michelle Ann. “How to Learn about Teaching: An Evolutionary Framework for the Study of Teaching Behavior in Humans and Other Animals.” Behavioral and Brain Sciences 38 (2015): e31. https://doi.org/10.1017/S0140525X14000090.˄

27. Helton, William S. “Animal Expertise, Conscious or Not.” Animal Cognition 8, no. 2 (2005): 67–74. https://doi.org/10.1007/s10071-004-0234-y. See also Helton, William S., and Nicole D. Helton. “Expertise in Other Animals: Canines as an Example.” In The Cambridge Handbook of Expertise and Expert Performance, edited by K. Anders Ericsson, 2nd ed., 49–59. 2008. Reprint, Cambridge: Cambridge University Press, 2018; and Dukas, Reuven. “Animal Expertise: Mechanisms, Ecology and Evolution.” Animal Behaviour 147 (2019): 199–210. https://doi.org/10.1016/j.anbehav.2018.05.010.˄

28. Collins, Harry, and Robert Evans. Rethinking Expertise. Chicago: University of Chicago Press, 2007.˄

29. Whitehead, Hal, and Luke Rendell. The Cultural Lives of Whales and Dolphins. Chicago: University of Chicago Press, 2014. See also Aplin, Lucy M. “Culture and Cultural Evolution in Birds: A Review of the Evidence.” Animal Behaviour 147 (2019): 179–87. https://doi.org/10.1016/j.anbehav.2018.05.001.˄

30. Van Dooren, Thom, Matthew Chrulew, Myles Oakey, Sam Widin, and Drew Rooke. “Animal Cultures at the Edge of Extinction.” Theory, Culture & Society 42, no. 2 (2025): 3–23. https://doi.org/10.1177/02632764241280346.˄

31. Argyris, Chris, and Donald A. Schön. Organizational Learning II: Theory, Method, and Practice. Reading, MA: Addison-Wesley, 1996. See also Bateson, Gregory. Steps to an Ecology of Mind: Collected Essays in Anthropology, Psychiatry, Evolution, and Epistemology. 1972. Reprint, Northvale: Aronson, 1987; Sterling, Stephen. Sustainable Education: Re-Visioning Learning and Change. Foxhole: Green Books for the Schumacher Society, 2004; and Mezirow, Jack. Transformative Dimensions of Adult Learning. San Francisco: Jossey-Bass, 1991.˄

32. Engeström, Yrjö. “Expansive Learning at Work: Toward an Activity Theoretical Reconceptualization.” Journal of Education and Work 14, no. 1 (2001): 133–56. https://doi.org/10.1080/13639080020028747.˄

33. Lange, Elizabeth A. Transformative Sustainability Education: Reimagining Our Future. Abingdon: Routledge, 2023. https://doi.org/10.4324/9781003159643.˄

34. Lave, Jean, and Etienne Wenger. Situated Learning: Legitimate Peripheral Participation. Cambridge: Cambridge University Press, 1991.˄

35. Barrett, M. J., Matthew Harmin, Bryan Maracle, Molly Patterson, Christina Thomson, Michelle Flowers, and Kirk Bors. “Shifting Relations with the More-than-Human: Six Threshold Concepts for Transformative Sustainability Learning.” Environmental Education Research 23, no. 1 (2017): 131–43. https://doi.org/10.1080/13504622.2015.1121378.˄

36. Wiek, Arnim, Lauren Withycombe, and Charles L. Redman. “Key Competencies in Sustainability: A Reference Framework for Academic Program Development.” Sustainability Science 6, no. 2 (2011): 203–18. https://doi.org/10.1007/s11625-011-0132-6.˄

37. Lopez, Andrew. “Nonhuman Animals and Epistemic Injustice.” Journal of Ethics and Social Philosophy 25, no. 1 (2023): 136–63. https://doi.org/10.26556/jesp.v25i1.2201. Building on Fricker, Miranda. Epistemic Injustice: Power and the Ethics of Knowing. Oxford: Oxford University Press, 2007.˄

38. Despret, Vinciane. What Would Animals Say If We Asked the Right Questions? 2012. Reprint, Minneapolis: University of Minnesota Press, 2016.˄

39. Pedersen, Helena. “Post-Anthropocentric Pedagogies: Purposes, Practices, and Insights for Higher Education.” Teaching in Higher Education 30, no. 2 (2025): 344–58. https://doi.org/10.1080/13562517.2023.2222087.˄

40. Common Worlds Research Collective and UNESCO. Learning to Become with the World: Education for Future Survival. Paris: UNESCO, 2020.˄

41. Clark, Andy. “How to Knit Your Own Markov Blanket.” In Philosophy and Predictive Processing, edited by Thomas K. Metzinger and Wanja Wiese, 1–19. Frankfurt am Main: MIND Group, 2017.˄

42. Friston, Karl, Conor Heins, Tim Verbelen, Lancelot Da Costa, Tommaso Salvatori, Dimitrije Markovic, Alexander Tschantz, Magnus Koudahl, Christopher Buckley, and Thomas Parr. “From Pixels to Planning: Scale-Free Active Inference.” Frontiers in Network Physiology 5 (2025): 1521963. https://doi.org/10.3389/fnetp.2025.1521963.˄

43. Demir, Berre Su, and Aykut Coşkun. “Rethinking Representation in Design: Towards Constructing Parameters for Representation Tools in More-than-Human Design.” Proceedings of the 2025 ACM Designing Interactive Systems Conference (New York), 2025, 178–94. https://doi.org/10.1145/3715336.3735680.˄

44. Fredericks, Joel, Marcus Foth, Glenda Amayo Caldwell, Andrew Vande Moere, and Martin Tomitsch. “Middle-out Design for More-than-Human Cities: Integrating Human–Animal Relations in Urban Sustainability Planning.” Smart and Sustainable Built Environment, 2026, 1–23. https://doi.org/10.1108/SASBE-09-2025-0562.˄

45. Gordon, Bonnie J., and Stanislav Roudavski. “More-than-Human Infrastructure for Just Resilience: Learning from, Working with, and Designing for Bald Cypress Trees (Taxodium distichum) in the Mississippi River Delta.” Global Environment 14, no. 3 (2021): 442–74. https://doi.org/10.3197/ge.2021.140302.˄

46. Pollini, Barbara, Zeliha Asya Ilgün, Carla Paoliello, Marco G. Casteleijn, Julie-Anne Gandier, and Edoardo Brunelli. “Multispecies Reconciliation Principles in Biodesign Practice.” Biotechnology Design 4 (2026): e13. https://doi.org/10.1017/S2977905726100316.˄

47. Roudavski, Stanislav, and Alexander Holland. “Are You Blocking Nonhuman Leadership? Five Questions to Find Out.” PDC ’26: Proceedings of the 19th Participatory Design Conference (New York) 2 (2026): 372–81. https://doi.org/10.1145/3789492.3796416.˄

48. Rutten, Julian, Alexander Holland, and Stanislav Roudavski. “Plants as Designers of Better Futures: Can Humans Let Them Lead?” Plant Perspectives 2, no. 1 (2024): 92–139. https://doi.org/10.3197/whppp.63845494909729.˄

49. Roudavski, Stanislav, and Alexander Holland. “Are You Blocking Nonhuman Leadership? Five Questions to Find Out.” PDC ’26: Proceedings of the 19th Participatory Design Conference (New York) 2 (2026): 372–81. https://doi.org/10.1145/3789492.3796416.˄

50. Yalukit Weelam Tarrang, Carolyn N’Arweet Briggs, David Tournier, Brian Martin, Julian Rutten, Alexander Holland, and Stanislav Roudavski. “Vegetal Leadership in Interspecies Cultures: Collaborative Mark-Making with Yalukit Weelam Tarrang [Tree].” Australian and New Zealand Journal of Art 25, no. 1 (2025): 7–32. https://doi.org/10.1080/14434318.2025.2515197.˄