Abstract

Computational descriptions are often useful in biology, but APS argues that computation cannot define what life is. Living systems are not algorithms executing predefined rules, but viability-oriented, constraint-closed organisations that actively maintain the conditions of their own persistence. Computation may model aspects of biological activity, but it presupposes the organised persistence of living systems rather than explaining it.

Why Life Is Not Computation

Computational descriptions are now widespread throughout biology. Genes are said to execute programs, cells are described as information-processing systems, and brains are often characterised as computational machines. Because living systems transform inputs into outputs in structured and adaptive ways, it is tempting to conclude that life itself is fundamentally computational.

APS rejects this conclusion.

Computational models can successfully describe many aspects of biological activity, but computation does not explain what makes a system alive. Living systems are not defined by the execution of algorithms, but by viability-oriented, constraint-closed organisation through which they actively maintain the conditions required for their own persistence.

The distinction is fundamental. Computation may occur within living systems, but life is not computation.

APS Box — Realization Is Organisational, Not Merely Functional

In classical functionalist accounts, two systems may count as equivalent if they perform sufficiently similar causal or computational roles. APS argues that this criterion is insufficient for biological explanation because it abstracts away from the organisational conditions that make living activity possible.

In APS, realization is not understood primarily as abstract input-output equivalence. Rather, realization concerns participation within viability-oriented, constraint-closed organisation.

Biological systems are not merely collections of mechanisms performing isolated functions. They are temporally extended organisations whose processes contribute to sustaining the conditions of their own persistence. Functions therefore derive explanatory significance from their role within organised self-maintaining systems.

For this reason, materially distinct systems may realise comparable biological capacities only insofar as they participate in sufficiently similar forms of persistence-maintaining organisation. What unifies realizations is not identical material structure, nor abstract computation alone, but organisational participation within living systems.

APS therefore rejects both strict substrate essentialism and unrestricted substrate neutrality.

A system does not become biologically equivalent merely because it reproduces selected behavioural outputs or formal causal patterns. A computational simulation of metabolism, for example, does not thereby become metabolically alive, and a system that reproduces selected behavioural features associated with cognition does not necessarily instantiate biological cognition unless those activities contribute to the persistence-maintaining organisation of the system itself.

APS consequently treats realization as organisationally constrained. Biological capacities may be implemented through diverse mechanisms and structures, but those realizations remain governed by the requirements of viability, persistence, and constraint closure.

The Computational Temptation

Computational language is attractive because biological systems often exhibit organised regulation, responsiveness, coordination, and adaptation. Organisms process signals, regulate internal states, and modify behaviour in ways that can frequently be modelled computationally.

These descriptions are scientifically valuable. Computational methods are indispensable in neuroscience, systems biology, bioinformatics, developmental modelling, and artificial intelligence research.

APS does not reject computational modelling. It rejects the stronger claim that computational organisation is sufficient to explain life itself.

The problem is not computation as a tool, but computation as an ontology.

APS Box — How APS Relates to Other Theoretical Frameworks

Contemporary frameworks such as control theory and nonlinear dynamical systems theory provide powerful and indispensable accounts of living systems. They explain how systems regulate their behaviour, maintain stability, and exhibit complex dynamics across time.

APS does not replace these approaches, nor does it reduce them to a single underlying model. Instead, it addresses a different question: what kind of organisation makes a system a living system in the first place.

Multiple theoretical frameworks can therefore apply to the same system, but they do so in different ways and address different explanatory questions. A dynamical systems description may characterise the system’s trajectories, and a control-theoretic model may describe its regulatory structure. These accounts are often non-reducible to one another, and each can be independently valid.

However, such accounts do not by themselves distinguish living systems from non-living systems. Many non-living systems also exhibit regulation, feedback, and complex dynamics.

APS contributes by specifying the organisational conditions under which a system must actively maintain its own viability. A system is viability-oriented when its processes are organised such that their effects contribute, directly or indirectly, to maintaining the conditions required for the system’s continued existence.

Importantly, in APS, viability is not defined by reference to life itself. It is specified in terms of the persistence of an organised system within a bounded region of possible states. A living system is therefore not defined in a circular sense as “that which maintains viability”, but as a system whose organisation is structured around maintaining these conditions.

In this sense, APS provides a way of identifying what makes a system a living system, rather than a general account of everything such a system does.

This does not mean that APS provides a complete or self-sufficient account of living systems. Describing how such systems behave, regulate, and evolve requires additional theoretical frameworks. APS specifies conditions that are necessary for identifying living systems, but it does not by itself exhaust the explanations required to understand them.

A complete scientific understanding of living systems therefore requires multiple forms of explanation. Dynamical and control-theoretic accounts explain how systems behave and regulate. APS clarifies the conditions under which such systems count as living systems at all.

Key Point:
Explaining regulation, dynamics, or behaviour is not the same as explaining what makes a system a living system. APS addresses this latter question by specifying the organisational conditions under which viability is maintained.

Computation and Information Processing

APS distinguishes computation from information processing more broadly.

Information processing refers to the transformation and coordination of differences or signals within organised systems. Computation is a more specific concept involving formally specified operations, rule-governed state transitions, or algorithmic procedures.

Living systems undoubtedly process information in important ways. Computational models may successfully capture some of these dynamics. But neither information processing nor computation explains why a living system exists as a persistence-maintaining organisation in the first place.

APS therefore treats computational descriptions as secondary to biological organisation rather than constitutive of it.

What Computation Explains Well

Computational approaches are highly effective for describing:

pattern recognition
signal coordination
optimisation and search
control and feedback regulation
decision procedures
learning dynamics

Where viable biological systems already exist, computational models can illuminate how particular regulatory or behavioural processes unfold.

For example, neural networks may model sensory classification, regulatory systems may be described using control-theoretic frameworks, and bacterial chemotaxis may be analysed through information-processing dynamics.

APS fully accepts the scientific importance of these approaches.

What computation does not explain is why the organised system performing these activities must continuously sustain itself in order for any processing to occur at all.

Computation Presupposes Organised Persistence

Computation requires a system whose identity and operation are already sufficiently stable for formal operations to be defined.

Computational systems presuppose:

identifiable system boundaries
stable operational conditions
specified inputs and outputs
state-transition rules or algorithms
externally defined success conditions

These assumptions are not generated by computation itself.

APS therefore asks a prior biological question:

What makes there be a system whose continued existence matters?

Living systems must continuously:

maintain their own boundaries
regenerate the constraints enabling their activity
repair damage and reorganise under perturbation
preserve viability despite changing conditions

If these activities fail, the system ceases to exist as a living system.

Computation therefore operates within organised persistence rather than explaining it.

Algorithms Do Not Ground Biological Normativity

Computational systems can succeed or fail relative to predefined specifications. An output is correct or incorrect according to externally established criteria.

Biological normativity is fundamentally different.

In living systems, processes are persistence-relevant for the system itself. A failing heart is not malfunctioning because it violates an externally specified program. It is malfunctioning because its activity no longer contributes adequately to maintaining the organism’s continued viability.

Likewise, starvation, injury, or regulatory collapse are not merely computational errors. They are existential failures affecting the persistence of the system itself.

APS therefore locates biological normativity in viability-oriented organisation rather than in algorithmic correctness.

Computation Without Life Is Possible

Computation can occur in systems that are clearly not alive:

digital computers
distributed software systems
machine-learning architectures
simulation platforms
autonomous algorithms

Such systems may be adaptive, self-modifying, and highly complex. They may even exhibit forms of self-regulation.

But their continued existence does not matter to themselves in the biological sense. Their goals, evaluation criteria, and success conditions remain externally specified. This is the distinction APS develops in more detail in Why AI Is Not Biological Agency. Contemporary AI systems may optimise, learn, and respond adaptively, but they remain externally maintained optimisation systems rather than endogenously viability-oriented organisations.

Failure in such systems is typically technical or functional relative to external purposes. Failure in living systems is existential because the system’s own persistence is at stake.

This demonstrates that computation is not sufficient for life.

Living Systems Do Not Run Programs

Computational metaphors often imply that organisms execute stored instructions or predefined algorithms.

APS rejects this framing.

Living systems do not merely run programs; they sustain organisation.

Their activity consists in the continuous maintenance, repair, and reorganisation of the constraints that make their own continued activity possible. When conditions change, organisms do not simply select new computational states. They reorganise themselves in ways that preserve viability.

This may involve:

altering regulatory dynamics
reallocating energetic resources
modifying developmental trajectories
reorganising behavioural strategies
restructuring interactions with the environment

Living organisation is therefore not reducible to computation over predefined states. It is an ongoing process of persistence-maintaining transformation.

Constraint Closure and Multi-Scale Organisation

APS grounds life in constraint closure: the reciprocal organisation through which the processes maintaining biological constraints are themselves constrained and sustained by the organisation they generate.

This organisation unfolds across multiple interacting scales.

Molecular processes regulate cellular activity, cells contribute to organismal persistence, and organisms modify ecological conditions that in turn shape their own viability. Biological organisation therefore emerges through coordinated activity distributed across spatial and temporal domains.

Computational descriptions often abstract away from this multi-scale organisation in order to model specific operations or information flows. Such abstractions may be scientifically useful, but they do not capture the full organisational conditions required for living persistence.

Life is not defined by isolated computations, but by the integrated organisation sustaining the conditions under which any computation can occur.

The APS Position

APS does not reject computation. It repositions it.

The issue is therefore not whether computational descriptions are scientifically useful, but whether computation alone explains what makes living systems biological.

APS Box — What APS Changes — and What It Does Not

APS changes what biology treats as explanatorily fundamental.

Traditional biological explanation often begins with:

components,
mechanisms,
traits,
or historical outcomes.

APS instead begins with organised viability-maintenance: the ongoing activity through which living systems sustain the conditions of their own persistence.

This shifts the explanatory centre of biology from:

components → mechanisms → outcomes

toward:

organisation → activity → stabilised features.

Importantly, APS does not propose:

a special life substance,
separate physical laws,
non-natural causal forces,
or a distinct “realm of life.”

Living systems remain fully continuous with chemistry and physics. Agency, normativity, and purpose are understood as organisational features of living systems rather than as external metaphysical additions.

The shift is therefore not a rejection of scientific explanation, but a reorganisation of biological explanation itself. APS changes what biology treats as explanatorily primary: not isolated components or historical success alone, but the organised activity through which living systems maintain themselves across time.

In this sense, APS represents an explanatory and ontological re-centring within biology rather than the introduction of a separate realm of existence.

Within APS:

computation is a descriptive framework for certain biological processes
computational models presuppose viable organised systems
agency grounds persistence, function, and normativity
computation becomes biologically meaningful only within viability-oriented organisation

Computation is therefore instrumental rather than constitutive.

It describes aspects of what living systems do without explaining what living systems are.

Summary

Computational approaches are indispensable throughout contemporary biology. They successfully model coordination, regulation, learning, signalling, and control across many domains of life.

But life is not computation.

Living systems are viability-oriented, constraint-closed organisations whose continued existence depends on their own ongoing activity. They actively maintain the conditions required for their persistence across changing conditions and interacting scales.

Computation may describe aspects of this activity, but it does not explain why organised persistence exists or why failure is existential rather than merely technical.

In APS, computation is meaningful only because life already exists.

Why Life Is Not Computation — An APS Clarification

Key Points