Agency as the Defining Activity of Life

Where This Article Fits.

The paper “Agency as the Defining Activity of Life: A Viability-Oriented Framework Integrating Process and Scale,” published in Biological Theory, was an important step in the development of APS. It proposed that biological agency—the viability-oriented activity through which living systems contribute to maintaining the conditions of their own continued existence—is fundamental to understanding Life. It also brought Agency, Process and Scale together within a single framework and helped establish the conceptual direction of subsequent APS research.

APS has continued to develop since that paper was written. Some of its central claims have been retained, while others have been clarified, narrowed or repositioned. In particular, subsequent work has distinguished more carefully between Life and Agency, between living organisation and organised persistence, and among Agency, Process and Scale. It has also placed tighter limits on what APS can claim merely from providing a common conceptual framework.

This article explains that development. It first states the argument of the Biological Theory paper as clearly as possible, without rewriting it in terms of later APS conclusions. It then identifies what subsequent research has retained, what has changed, and why those changes matter. The result is not a repudiation of the original Agency thesis, but a more precise account of its place within APS.

For a detailed account of Agency within living systems, see Biological Agency as the Activity of Self-Maintenance, which examines self-maintenance, viability regulation, organised persistence, function, normativity and constraint closure. For the wider theoretical context, see APS and Contemporary Theories, which considers mostly organism-centred evolution, autonomy theory, developmental agency and process biology, and situates APS among contemporary approaches to biological Agency.

The three articles therefore now have distinct roles:

• Agency as the Defining Activity of Life — The original Agency thesis and how subsequent APS research has refined it.
• Biological Agency as the Activity of Self-Maintenance — The detailed account of viability-oriented Agency in living systems.
• Agency in Contemporary Biology — The wider theoretical context in which APS approaches Agency.

1. The Original Claim - Agency as the Defining Activity of Life

The starting point of the published paper was a simple question: what distinguishes a living system from a nonliving one?

Many familiar answers begin with what living things contain or with the properties they display: metabolism, reproduction, heredity, cells, genetic material, homeostasis, or the capacity to evolve. The paper took a different approach. It asked what living systems do in order to remain living.

Its answer was biological Agency.

Here, Agency did not mean consciousness, intelligence, deliberate choice, or human-like intention. It meant the capacity of a living system to act in ways that contribute to maintaining its own viability. A cell regulates its internal conditions, maintains a boundary, repairs damage, responds to stress, and reorganises its activity when circumstances change. These activities differ enormously across organisms, but the paper proposed that they share something important: they contribute to maintaining or re-establishing the conditions under which the system can continue to exist.

The paper described this as “the viability-oriented activity through which systems sustain and reconstitute their own organization across time and scale.” Life was therefore presented as an ongoing organisational process in which living systems actively regulate the conditions of their own persistence. In its strongest formulation, the paper put the point very directly: a system is alive insofar as it acts to maintain and reconstitute the conditions of its own persistence (Spencer 2026).

That was the sense in which Agency was proposed as the defining activity of Life.

The emphasis on activity matters. The proposal was not that living systems possess a mysterious additional property called Agency, nor was Agency intended as another causal force added to metabolism, gene regulation, development, physiology, or behaviour. Rather, the paper used Agency to redescribe established biological processes in terms of the organisational conditions under which they occur (Spencer 2026).

This also explains the role of viability. Not every change made by a living system counts as Agency simply because the system caused it. The relevant activities are organised in relation to the system’s continued ability to exist: some conditions support continued living organisation, while others undermine it. Agency concerns activity understood within that difference.

The paper also tried to make this proposal empirically usable rather than leaving Agency as an abstract philosophical idea. It proposed six biosignatures—homeostasis, adaptive plasticity, selective boundary regulation, goal-directed behaviour, recursive self-repair, and cross-scale coordination—as observable ways in which viability-oriented activity might be investigated. These were presented as empirically tractable indicators, while the paper also acknowledged unresolved questions about thresholds, borderline cases, and the limits of nonliving self-organisation (Spencer 2026).

Importantly, the paper did not present APS as a replacement for ordinary biological science. It explicitly said that the framework did not replace standard biological explanations, but sought to make explicit an organisational condition that those explanations typically presuppose without themselves making it their primary explanatory target (Spencer 2026).

The original claim can therefore be stated quite simply: living systems do not merely persist. They actively contribute to maintaining the conditions that make their continued existence possible. The published paper called this viability-oriented activity biological Agency and proposed it as the defining activity of Life.

That is the claim from which the subsequent development of APS begins.

2. Why the Paper Began with Agency

The paper began with Agency because it wanted to shift attention from what living systems have to what living systems continually have to do.

A living cell is not alive simply because a particular collection of molecules is present at one moment. Its components are continually being produced, transformed, transported, damaged, repaired, and replaced. Its internal conditions change, as do its surroundings, yet through this continuing activity the cell can remain a recognisably continuous living system.

The same general problem appears at larger scales. Organisms grow and develop, respond to injury, shortages, and environmental disturbance, alter their physiology and behaviour, and coordinate activities among different tissues and organs. In each case, persistence occurs through change, not by avoiding it.

The published paper treated this active maintenance of organisation as fundamental. It described living systems as initiating, regulating, and coordinating their organisation in response to changing conditions. Process captured the temporal organisation through which such activity is enacted and renewed, while Scale recognised that biological organisation extends across different spatial and temporal domains. Agency became the paper’s main explanatory entry point because it brought a particular biological fact into focus: living organisation has to be (Spencer 2026).

This was also why viability mattered so much. A rock may persist for a very long time, but it does not continually regulate itself so that it remains within conditions compatible with being a rock. Living systems are different: their continued existence depends upon organised activity that maintains, restores, or alters their own conditions.

The paper expressed this by saying that persistence in living systems is not merely a passive outcome but an ongoing achievement (Spencer 2026). That formulation explains why the paper was dissatisfied with accounts of Life that simply listed characteristic traits. Metabolism, boundaries, repair, regulation, and responsiveness mattered not merely because living systems happened to possess them, but because these processes participated in maintaining living organisation.

Agency therefore connected two questions that can otherwise appear separate:

What is a living system?

and

What does a living system do to remain living?

The paper described Agency as the “operational nexus” between those questions. Its formal definition accordingly identified Life with the capacity actively to coordinate and maintain organisation across interacting spatial and temporal domains in ways that promote continued viability and adaptive potential (Spencer 2026).

The published paper really did give Agency a very strong role, and that claim should not be retrospectively weakened simply because subsequent APS research has made finer distinctions. At the same time, the paper was not entirely committed to a simple equation of Life = Agency. It repeatedly spoke of viability-oriented organisation, constraint closure, and organised persistence, while its conclusion treated constraint closure, Agency, Scale, and evolution as analytically distinguishable aspects of an integrated organisation (Spencer 2026).

There was therefore already a tension within the paper. One formulation made Agency the defining feature or capacity of Life; another increasingly located Agency within a broader account of viability-oriented organisation.

Subsequent APS research would eventually make that distinction explicit. Before examining what changed, however, it is important to identify what the later work retained.

3. What Subsequent Work Has Retained

The central insight of the published paper has survived: living systems are not passive structures but actively participate in maintaining the conditions under which their organisation can continue. That remains central to APS.

Agency remains viability-oriented activity

Subsequent work has retained the idea that biological Agency concerns activity organised in relation to viability. Agency is therefore not simply another word for causation. A biochemical reaction can cause another reaction without thereby constituting biological Agency. The more specific question is how the activity of a living system bears upon maintaining or re-establishing conditions compatible with its continued organisation.

This remains close to the published paper’s original formulation, which described Agency in terms of activity through which systems sustain and reconstitute their organisation and explicitly distinguished this usage from intention and cognition (Spencer 2026).

Agency still does not mean consciousness or intelligence

That distinction has also survived. A bacterium need not understand its circumstances in order to regulate its internal conditions, and a cell need not consciously intend to repair damage. Biological Agency, in the APS sense, therefore begins much earlier than psychological Agency.

This matters because the word agency can otherwise make the theory sound more anthropomorphic than it is. The original paper explicitly rejected that interpretation, and subsequent APS work has retained the distinction.

Living organisation still has to be actively maintained

The later work has also retained the paper’s emphasis on active maintenance. Living systems undergo continuous material and organisational change, so maintaining Life cannot mean keeping everything unchanged. What persists is an organisation capable of continuing through turnover, perturbation, and reorganisation.

The original paper captured this with the idea that persistence is an ongoing achievement. Subsequent APS research retains that phenomenon while becoming more careful about where it sits within the explanatory architecture.

Constraint closure remains important

Constraint closure has likewise not been abandoned. The published paper described living systems as organisations in which the constraints regulating activity are themselves generated and maintained through the system’s activity, and it used viability-oriented, constraint-closed organisation when discussing how Agency might be diagnosed empirically (Spencer 2026).

Later APS work has made this organisational dimension more important, not less.

The relationship between Agency, Process and Scale remains important

Nor has subsequent work discarded the three APS perspectives. Agency still asks about viability-oriented activity; Process brings temporal organisation and transformation into view; and Scale asks how relevant organisation and causal relations extend across spatial, temporal, and organisational extents.

What has changed is the relationship among them, not the need for them. The published paper itself was already moving towards this position when it described apparently different features as “analytically distinct aspects of a single, integrated organization.” Subsequent work makes that implication more explicit.

APS still seeks an organisational account of Life

Finally, subsequent work retains the paper’s most general ambition: to understand Life in organisational rather than merely compositional terms. The question is not simply which molecules, structures, or traits are present, but how those components participate in an organisation capable of continuing as a living system.

What has changed is the precision with which the pieces are arranged. The original paper made Agency carry several jobs at once: it helped define Life, described viability-oriented activity, connected organisation with persistence, linked Process and Scale, and supported broader claims about biological explanation. Later work has separated those jobs.

The original Agency insight therefore remains, but it now leads to a more exact question: if Agency is the viability-oriented activity characteristic of living systems, does Agency itself define Life, or is it one aspect of a more fundamental description of living organisation?

That is where subsequent APS research begins to modify the original claim.

4. What Subsequent Work Has Changed

The central claim of the published paper has survived, but its place within APS has changed.

The original paper sometimes asked Agency to perform several theoretical functions at once. Agency was the activity through which living systems maintained viability, but it was also presented as the defining feature of Life, the link between what Life is and what Life does, and an organising concept connecting Process and Scale. At the same time, another formulation was already developing within the paper: living systems were increasingly described in terms of viability-oriented organisation, constraint closure, and organised persistence, and the conclusion treated Agency as one analytically distinguishable aspect of an integrated organisation.

Subsequent APS research has resolved this tension in favour of the organisational formulation. The change can be stated simply: Agency remains central to APS, but it no longer has to provide the whole definition of Life. This is a clarification rather than a rejection of the original insight.

The current position distinguishes more carefully between the living organisation and what that organisation does. Life concerns the kind of organisation that exists and persists as a living system; Agency concerns the viability-oriented activity through which that organisation responds, regulates, maintains, and, where possible, restores conditions compatible with its continuation. Giving the concepts different jobs avoids several problems.

First, it reduces the risk of circularity. If Agency is defined as what living systems do and Life is then defined simply as whatever exhibits Agency, the concepts become difficult to distinguish. Giving Life an organisational formulation and Agency an activity-oriented formulation makes their relationship clearer.

Second, Agency no longer risks becoming an invisible explanatory ingredient added to ordinary biological mechanisms. It is not another force operating alongside biochemical reactions, physiological regulation, or developmental processes. Rather, it is an analytic way of asking how activities of an organised living system relate to viability.

Third, the distinction clarifies what is being claimed when a phenomenon is described as agential. APS need not imply that Agency independently caused the phenomenon. It can instead examine particular biological activities in relation to their contribution to maintaining or re-establishing viable organisation. The original paper had already guarded against treating Agency as an additional causal entity by saying that the formulation introduced no new causal forces but redescribed established biological processes in terms of their organisational conditions (Spencer 2026). Subsequent work makes that restraint more systematic.

A second development concerns the explanatory ambitions of APS. The published paper sometimes presented APS as a unified framework for defining, detecting, and explaining living systems (Spencer 2026). Subsequent work requires a sharper distinction between conceptual integration and explanatory gain.

A common vocabulary for describing relationships among viability, organisation, Agency, Process, and Scale can be useful, but it does not by itself demonstrate that APS explains a particular biological phenomenon better than physiology, developmental biology, evolutionary theory, molecular biology, ecology, or another established approach. That stronger claim has to be demonstrated against an appropriate alternative.

The result is a more restrained APS, but also a more testable one. Instead of assuming that a general framework must be explanatorily superior because it encompasses more phenomena, APS can ask a more demanding question:

What does an APS analysis explain in this particular case that the relevant alternative explanation does not?

Subsequent work therefore narrows some of the original paper’s ambitions while strengthening the standards by which those ambitions can be assessed. Agency remains important, but the amount of theoretical work it is expected to perform has become more precisely defined.

5. Life, Agency, and Organised Persistence

The revised position becomes clearer if three ideas that were closely intertwined in the original paper are separated: Life, Agency, and organised persistence. They concern the same biological reality, but they are not interchangeable.

Life

The current APS formulation treats Life as viability-oriented, constraint-closed organisation.

In plain English, a living system is organised in such a way that its continued existence depends upon activities that help maintain the conditions under which that organisation can continue. The system does not merely occupy a stable state; its organisation is continually enacted, maintained, and reconstructed through the processes occurring within and through it.

This formulation is not foreign to the published paper, which repeatedly used viability-oriented, constraint-closed organisation when characterising living systems and applying its proposed biosignatures (Spencer 2026). What has changed is that this organisational formulation now carries more of the burden of answering what Life is.

This definition is neither circular nor tautological. Viability does not mean simply “being alive” or “remaining alive.” It identifies the asymmetry among conditions under which a biological organisation can continue functioning as that organisation and conditions under which its organisation is impaired or lost. Constraint closure identifies a different feature: the organisation of relations through which relevant constraints are themselves generated, maintained, or re-established through the activities they help regulate. Neither concept therefore defines Life by presupposing Life. The APS definition instead proposes that what distinguishes living organisation is the conjunction of viability orientation and constraint closure: an organisation whose activities occur relative to conditions of its continued functioning and whose relevant constraints participate in mutually maintaining that organisation.

Agency

Agency answers a different question:

What does that living organisation do in relation to its viability?

Agency concerns activities through which a living system contributes to maintaining, regulating, or re-establishing conditions compatible with its continued organisation. A cell repairing damage, physiological regulation within an organism, or context-sensitive activity that alters the relationship between an organism and its environment can all be analysed agentially.

Calling such activity agential does not introduce a new force behind the molecular or physiological processes involved, nor does it imply that a cell consciously wants to survive. It means examining those processes in relation to the viability of the organised system of which they form part. The original paper already insisted that Agency need not imply intention or cognition; that distinction remains unchanged (Spencer 2026).

What has changed is the stronger claim that Agency itself exhausts the definition of Life. A useful shorthand is therefore:

Life identifies the organisation. Agency identifies viability-oriented activity of that organisation.

This is an analytic distinction rather than a separation into independent things. There is no free-floating Agency apart from an organised system whose activity can be assessed in relation to viability, while living organisation itself is not static: its continuation depends upon activity.

Organised persistence

Organised persistence brings a third question into view. The published paper was right to emphasise that living persistence is not passive: living systems continually re-establish the conditions under which they can continue to exist (Spencer 2026). APS can now distinguish more carefully, however, between what needs explaining and the resources used to explain it.

Organised persistence is principally the phenomenon that needs explanation.

The puzzle is straightforward. Living systems change continuously. Their components turn over. Their internal states fluctuate. They develop, repair, respond, and reorganise. Yet some form of organised continuity can persist through those changes.

How?

Agency contributes to the answer because living systems perform viability-oriented activities. Process contributes because persistence occurs temporally, through ongoing change rather than outside it. Scale matters because the relevant organisation and causal relations may extend across different spatial and temporal extents, while constraint closure concerns relations in which relevant constraints are themselves maintained through the organisation they help regulate.

Organised persistence should therefore not be treated as something produced by Agency alone. It is the central phenomenon, while Life names the viability-oriented, constraint-closed organisation exhibiting that persistence and Agency, Process, and Scale allow different aspects of its organisation and continuation to be analysed.

This arrangement makes the APS architecture less linear and clarifies why subsequent work does not repudiate the published paper. The original phenomenon remains the important one: living systems somehow maintain organised continuity through continual change. Later APS work has mainly become more careful about which concepts address which parts of that problem.

6. Agency, Process, and Scale in the Subsequent APS Framework

The same clarification changes the relationship among the three terms that give APS its name.

The easiest mistake would be to imagine a sequence:

Agency → Process → Scale

On that picture, Agency would be foundational, Process would describe how Agency unfolds, and Scale would describe where Agency operates. That is close to the architecture sometimes suggested by the earlier presentation, but it is no longer the best way to understand APS (Spencer 2026).

Agency, Process, and Scale are better treated as different analytic projections on the same living organisation. Each asks a different question.

The term analytic projection is important. APS does not propose that living organisation is divided into three corresponding components called Agency, Process, and Scale. The biological organisation is one integrated reality, while explanation can interrogate that organisation under different questions. An analytic projection selectively brings particular relations of that organisation into explanatory focus: Agency foregrounds viability-oriented activity; Process foregrounds continuity and transformation through time; Scale foregrounds the spatial and temporal extents across which the relevant organisation and relations are realised.

The distinction is therefore analytic without being arbitrary. Agency, Process, and Scale are distinguished by the explanatory questions being asked, but those questions concern materially realised features and relations of the same biological organisation. “Analytic projection” consequently avoids two opposite errors: reification, in which Agency, Process, and Scale are treated as independent things or causal domains; and explanatory collapse, in which their different explanatory questions become interchangeable simply because they concern the same organisation.

Agency: what is the system doing relative to viability?

The Agency projection directs attention to activity. It asks how activities of the system contribute to maintaining, restoring, or altering conditions relevant to continued viable organisation. This may involve regulation, repair, behavioural change, boundary maintenance, or many other biological activities; the important point is not the particular mechanism but the question being asked of it: how does this activity bear upon viability?

Process: how does the organisation continue through change?

Process directs attention to time, transformation, and continuity. Living systems are never simply finished objects: metabolism proceeds, components turn over, development changes organisation, damage occurs and may be repaired, environmental circumstances alter, and evolution changes lineages across generations.

The Process projection therefore asks how organised continuity is enacted through change. It is not merely “Agency over time.” Many biological processes can be studied without being classified primarily in terms of Agency, because temporal organisation and transformation pose explanatory questions in their own right.

Scale: across what extents is the relevant organisation realised?

Scale directs attention to spatial and temporal extent and to relations among domains of organisation. A biological phenomenon may depend upon molecular interactions, cellular organisation, tissue relations, whole-organism physiology, environmental conditions, or combinations of these. Relevant timescales may likewise range from rapid molecular events through development and ecological change to evolutionary history.

Scale therefore asks: at what spatial and temporal extents does the organisation relevant to this explanation occur, and how are those extents related? It is not merely a place in which Agency happens, but a distinct explanatory projection.

Three questions, one living organisation

The resulting architecture is therefore not hierarchical:

• Agency: What is the living system doing in relation to viability?
• Process: How is its organisation enacted and transformed through time?
• Scale: Across what spatial and temporal extents are the relevant organisation and relations realised?

Their complementarity therefore arises not because three independent dimensions combine to constitute Life, but because one living organisation can present different explanatory dependencies according to whether the inquiry concerns viability-oriented activity, continuity through transformation, or the extents across which the relevant organisation is realised.

These questions can overlap, and often should, but none is automatically prior to the others. This develops something already visible near the end of the published paper, where apparently different features of the APS account were described as “analytically distinct aspects of a single, integrated organization.” Subsequent work makes that implication explicit and extends it to the architecture of APS itself (Spencer 2026).

An APS analysis therefore does not require every biological question to be translated into Agency. The important explanatory issue may concern temporal organisation, cross-scale relations, or viability-oriented activity; in other cases an established biological explanation may already answer the relevant question adequately. Agency, Process, and Scale provide different ways of interrogating living organisation rather than three hidden substances, causal forces, or stages through which Life must pass.

This also changes the sense in which APS can be called integrative. Its strongest defensible integrative claim is that Agency, Process, and Scale provide a common analytic architecture within which otherwise different biological questions can be related. Whether that architecture produces additional explanation in any particular case is a further question.

That distinction marks one of the largest methodological developments since the original paper. APS need not demonstrate its value by claiming that established biology has failed to explain Life. Instead, it can identify a particular explanatory target, establish the strongest existing explanation of that target, and ask whether the APS analysis contributes something additional.

This makes fewer automatic claims for APS, but it also makes the framework clearer. Agency identifies the viability-oriented activity of living systems; Process addresses continuity and transformation through time; Scale identifies the spatial and temporal extents relevant to an explanation. Life brings these questions together not because Agency, Process, and Scale manufacture Life in sequence, but because each examines a different aspect of the same viability-oriented, constraint-closed living organisation.

That is the principal conceptual change. It does not overturn the insight that gave the published paper its title; it tells us more precisely what that insight is an insight about.

7. What APS Now Claims—and What It No Longer Claims

The development described in the previous sections makes it possible to state the present APS position more precisely. Some claims made in the original paper remain central, while others have been narrowed or repositioned. This does not require treating the earlier paper as a mistake; it requires distinguishing between an initial theoretical proposal and what can reasonably be claimed after that proposal has been developed further.

APS still claims that living systems actively contribute to their own continuation

The central Agency insight remains: living systems do not persist in the same way that an inert object can persist. They regulate internal conditions, maintain boundaries, repair damage, reorganise activity, and respond to changes that bear upon their continued viability. APS uses Agency to examine this viability-oriented activity.

Nothing in subsequent work requires abandoning that idea or reducing Agency to consciousness, intention, intelligence, or behaviour. The original paper explicitly distinguished biological Agency from those stronger psychological meanings. What has changed is not the importance of Agency, but the claim that Agency by itself supplies the complete definition of Life.

APS now distinguishes more clearly between Life and Agency

The original paper sometimes moved directly from viability-oriented Agency to a definition of Life, saying that a system is alive insofar as it acts to maintain and reconstitute the conditions of its own persistence (Spencer 2026). The current position makes a finer distinction:

Life is viability-oriented, constraint-closed organisation. Agency concerns the viability-oriented activity of that organisation.

The concepts remain closely related because living organisation is active and Agency cannot be understood independently of an organised system whose viability is at stake. They nevertheless perform different theoretical jobs.

The title Agency as the Defining Activity of Life can therefore still be read literally: Agency can still be described as the defining activity of living organisation without being required to provide the complete definition of Life.

APS still treats organised persistence as a central problem

The original paper emphasised that living persistence is an ongoing achievement rather than passive endurance, and that insight survives. Living systems remain recognisably continuous despite material turnover, developmental change, environmental disturbance, and continual internal activity (Spencer 2026). Explaining that continuity-through-change is a genuine biological problem.

What has changed is the place of organised persistence within APS. Rather than treating it as something simply generated by Agency, APS now treats it primarily as a phenomenon requiring explanation. Agency helps analyse the viability-oriented activity involved in that persistence; Process addresses its temporal organisation; Scale identifies the spatial and temporal extents across which relevant relations are realised; and constraint closure contributes to explaining how living organisation maintains the conditions of its own continuation.

This clearer division of explanatory labour prevents organised persistence from being collapsed into Agency while preserving the contribution Agency can make to its analysis.

APS still uses Agency, Process and Scale—but not as a sequence

Agency, Process, and Scale remain central, but APS no longer treats them as a progression in which Agency comes first, Process describes the unfolding of Agency, and Scale describes where Agency occurs. Instead, they are complementary analytic projections on living organisation:

Agency asks what the system is doing in relation to viability; Process asks how its organisation is enacted and transformed through time; and Scale asks across what spatial and temporal extents the relevant organisation and relations are realised.

The original paper was already moving towards this position when it described apparently different features of living organisation as analytically distinct aspects of a single integrated organisation (Spencer 2026). Subsequent APS work makes that relationship explicit.

APS still proposes a general framework—but generality is not explanatory superiority

This is perhaps the most important methodological change.

The original paper was ambitious about the explanatory reach of APS. It argued that the framework could connect questions about mechanisms, development, ecology, and evolution within a common account of living organisation, while also stating that APS was not intended to replace standard biological explanations (Spencer 2026).

Subsequent work has made that second commitment more demanding. APS may legitimately show that different biological phenomena can be described or compared using common concepts such as viability, organisation, Agency, Process, and Scale, but a common vocabulary is not the same thing as an explanatory advance, and a broader framework is not automatically a deeper explanation. Nor does asking a question that another theory does not ask show that the other theory has failed.

If APS claims explanatory gain, the comparison therefore has to be made on a matched target: what exactly is being explained, what is the strongest existing explanation of that phenomenon, and what does the APS account add?

The answer may sometimes be substantial; in other cases APS may principally provide conceptual organisation, or an established biological explanation may already do the required work. The current framework leaves those possibilities open.

APS does not claim that established biology becomes meaningful only when placed inside APS

This follows directly from the previous point. Molecular mechanisms do not need APS in order to become genuine biological explanations, evolutionary explanations do not require an APS foundation before they can explain evolutionary change, and developmental, physiological, ecological, and systems explanations retain whatever explanatory achievements they can independently demonstrate.

APS may identify relationships among these explanations or ask additional organisational questions, but it should not assume that it grounds, completes, subsumes, or makes intelligible the rest of biology merely because it offers a general account of living organisation.

This is a real narrowing of some of the stronger implications surrounding the original framework, but it is also a methodological gain. A theory becomes more informative when it is clear not only about what it claims, but also about what would have to be shown before a stronger claim was justified.

APS still makes substantive theoretical proposals

This restraint does not make APS merely descriptive. The proposal that Life should be understood in terms of viability-oriented, constraint-closed organisation is substantive, as are the proposal that Agency identifies viability-oriented activity, the treatment of organised persistence as a central explanatory problem, and the claim that Agency, Process, and Scale provide analytically distinguishable ways of investigating the same living organisation.

These proposals can be criticised, compared with alternatives, and tested for explanatory usefulness. The difference is that APS no longer needs to build its case by assuming that the existence of a unified framework already demonstrates its explanatory superiority. Its claims can instead stand or fall more locally.

That makes the framework narrower in some respects, but also more defensible.

8. Conclusion: Refinement Rather Than Repudiation

The paper Agency as the Defining Activity of Life … began with a strong proposal: living systems are distinguished by viability-oriented activity through which they contribute to maintaining and re-establishing the conditions of their own continued existence. That proposal remains important to APS.

Subsequent research has not overturned this central Agency insight. It has not made Agency synonymous with cognition or intention, abandoned viability or constraint closure, discarded the problem of organised persistence, or removed Agency, Process, and Scale from the framework. What has changed is the relationship among these ideas.

The original paper asked Agency to perform several theoretical jobs simultaneously: it characterised what living systems do, helped define Life, connected viability with persistence, provided an entry point into Process and Scale, and supported broader claims about the explanatory integration APS might provide across biology. Subsequent development has separated these roles.

The current developed and clarified position can therefore be stated concisely:

Life is viability-oriented, constraint-closed organisation. Agency is the viability-oriented activity of that living organisation. Organised persistence is the continuity-through-change phenomenon requiring explanation. Agency, Process, and Scale are complementary analytic projections on the same living organisation.

They are analytic projections because they distinguish explanatory questions directed at one integrated biological organisation, not because Life consists of three corresponding components or levels.

These distinctions make the framework less dependent upon any single concept. Agency remains central without having to carry the entire ontology of Life, which is why the development is better understood as refinement rather than repudiation.

Indeed, the published paper already contained signs of the distinction that later work would make explicit. Although its strongest formulations defined Life through Agency, its conclusion also described Life in terms of viability-oriented, constraint-closed organisation and treated several features of the framework as analytically distinct aspects of one integrated organisation (Spencer 2026). The later development resolves that internal tension rather than replacing the original proposal with an unrelated one.

It also resolves a second tension. The paper sought both to make an ambitious general claim about living organisation and to insist that APS did not replace ordinary biological explanations. Current APS can hold those commitments together by distinguishing conceptual integration from demonstrated explanatory gain. It can offer a general theoretical proposal about Life and a common analytic architecture without assuming that either automatically supersedes molecular, physiological, developmental, ecological, or evolutionary explanation.

This allows APS to coexist with explanatory pluralism. Different biological questions may require different explanatory resources, and APS has to demonstrate its contribution at the point where it claims one.

The original Agency thesis can therefore be read more precisely. Its strongest surviving claim is not that Life is Agency, nor that all biological phenomena are ultimately explanations of Agency. It is that living organisation is distinctive in part because it is active in relation to the conditions of its own continued viability, and Agency is the APS concept for examining that activity.

That formulation is narrower than some readings of the original proposal, but clearer about what Agency identifies, what it does not replace, and what further work is required to demonstrate explanatory value. The title of the original paper can therefore remain meaningful: Agency can still be described as the defining activity of Life without treating Agency as the complete definition of Life itself.

The central intuition has survived, while the architecture around it has become more precise. APS is consequently in a better position to say not only what it claims about Life, Agency, Process, Scale, and organised persistence, but also what evidence would be needed to show that those concepts add something to the explanations biology already possesses.

Explanatory Architecture

Central Question

How has the APS claim that Agency is the defining activity of Life changed as the framework has developed?

Architectural Role

This article examines the development of one of the earliest central claims of APS. The published Biological Theory paper proposed biological Agency—the viability-oriented activity through which living systems contribute to maintaining the conditions of their own continued existence—as fundamental to understanding Life.

Subsequent APS research has retained that central insight while clarifying its theoretical role. APS now distinguishes Life, understood as viability-oriented, constraint-closed organisation, from Agency, understood as the viability-oriented activity of that organisation. Organised persistence is treated principally as a phenomenon requiring explanation, while Agency, Process, and Scale are treated as complementary analytic projections on living organisation rather than as a foundational sequence.

The role of this article is therefore not to establish Agency as the foundation of APS, but to explain what the original Agency thesis claimed, what subsequent work has retained, what has changed, and why those changes matter.

Relevant Conceptual Foundations

The following concepts and articles provide the current APS context needed to understand the development examined in this article:

• What Is APS?
• What Is Life? A Biological Question Revisited
• Viability
• Biological Organisation
• Constraint Closure
• Organised Persistence

These relationships identify conceptual connections within the current APS framework. They should not be read as a chronological sequence, causal hierarchy, or series of levels of organisation.

Related Explanatory Developments

The following articles develop or apply concepts discussed here in greater detail:

• Biological Agency as the Activity of Self-Maintenance
• Viability: The Organising Principle of Biological Persistence
• Biological Explanation: What Needs to Be Explained
• Process: The Dynamics of Living Biological Organisation
• Scale: How Life Is Coordinated Across Space and Time
• Agency in Contemporary Biology
• Agency and Organised Persistence
• APS and the Free Energy Principle: Complementary Explanatory Frameworks in Theoretical Biology

These articles should not be understood as deriving from an Agency-first architecture established by the present article. They investigate related aspects of living organisation or compare APS with other explanatory approaches.

Related Explanatory Questions

• In what sense can Agency still be described as the defining activity of Life?
• What is the difference between defining Life and identifying its characteristic activity?
• How does viability orient biological Agency?
• What is the relationship between Life, Agency, organisation, and organised persistence?
• Why are Agency, Process, and Scale treated as complementary analytic projections rather than as a sequence?
• When does an APS analysis provide explanatory gain rather than conceptual redescription?
• How should APS relate its account of Agency to established biological explanations?

Position Within APS

The original Biological Theory paper was important in establishing Agency as a central APS concept and in bringing Agency, Process, and Scale together within a common framework.

This article has a different role. It documents how that original proposal has subsequently been refined. It preserves the claim that living systems actively contribute to maintaining conditions compatible with their continued viability, while distinguishing Agency more carefully from the organisational definition of Life and from the phenomenon of organised persistence.

It also reflects a broader methodological development within APS: conceptual integration does not by itself establish explanatory gain. Claims that APS adds to existing biological explanations must be assessed against the relevant explanatory alternatives.

Position Within the APS Corpus

This article is a conceptual clarification and development article rather than the foundational Anchor or methodological parent of the APS comparative corpus.

Its distinctive role is to connect an important published stage in the development of APS with the framework’s current formulation. It allows readers to see which claims of the original Agency thesis remain, which have been narrowed or repositioned, and how subsequent research has changed the architecture of APS without retrospectively rewriting the published paper.

For the current operational account of biological Agency, readers should proceed to Biological Agency as the Activity of Self-Maintenance. For the broader theoretical context of Agency in contemporary biology, see Agency in Contemporary Biology.