Introduction

Where this article fits: This is a Life-demarcation clarification. It asks whether intelligence should be identified with life itself. It does not attempt to provide a complete theory of intelligence, cognition, consciousness, or artificial intelligence, and it does not infer APS explanatory superiority from the distinction.

Intelligence has become an increasingly important explanatory concept across biology, cognitive science, artificial intelligence, and philosophy. Systems may be described as intelligent when they learn, solve problems, adapt flexibly, integrate information, anticipate possibilities, or coordinate behaviour across changing conditions. Living systems can exhibit many such capacities, and advances in artificial intelligence have made the relationship between intelligence and biological organisation increasingly difficult to ignore. Yet the presence of intelligence, or of behaviour reasonably described as intelligence-like, should not be confused with a definition of life.

APS defines life as viability-oriented, constraint-closed organisation. This definition does not require strategic reasoning, abstraction, symbolic representation, planning, or the other capacities commonly associated with intelligence. Conversely, systems may exhibit sophisticated problem-solving or adaptive performance without thereby satisfying the APS definition of life. The central claim of this article is therefore one of non-identity: life is not defined by intelligence, and intelligence-like capacity alone does not establish life.

This distinction is deliberately limited. It does not show that theories of intelligence are explanatorily inadequate, nor does it establish that APS offers a better account of the phenomena those theories seek to explain. Rather, it separates two questions that are easily conflated: what makes a system alive, and what makes a system intelligent. Once those questions are distinguished, the relationships among life, biological agency, cognition, intelligence, and artificial systems can be examined without assuming that one of these concepts defines the others.

Intelligence Within the APS Architecture

APS distinguishes life, biological agency, Biological Evaluation, biological significance, Integration, cognition, Mind, intelligence, Selfhood, Reflective Agency, and Meaning rather than treating them as interchangeable descriptions of increasing biological complexity. For the purposes of the present argument, only a small part of this architecture is required.

Life is viability-oriented, constraint-closed organisation, while biological agency is viability-oriented organisational activity. Biological Evaluation is the process through which agency generates significance, and biological significance is a relationally real product of biological evaluation. Integration concerns the organisation of what matters into functional unity, while cognition is the organisation of biological significance across time. Intelligence can then be used for a further family of capacities that may include flexible problem-solving, abstraction, planning, learning, strategic coordination, and other forms of context-sensitive cognitive performance.

These distinctions establish an explanatory architecture rather than a hierarchy of increasingly important biological properties. Nor should their ordering automatically be read as a universal causal, temporal, or evolutionary sequence. APS uses them to distinguish explanatory questions that might otherwise be collapsed. In particular, the present article does not require the stronger proposition that every form of intelligence must arise through precisely the same sequence of organisational transitions.

The Life-demarcation point is narrower. Living organisation need not display intelligence in any strong sense, while intelligence-like performance need not establish viability-oriented, constraint-closed organisation. Life and intelligence may therefore be related in biological systems without being identical explanatory categories.

Why Intelligence Became Explanatorily Central

The association between life and intelligence is understandable because living systems display many capacities that overlap with contemporary conceptions of intelligent behaviour. They respond adaptively to changing conditions, coordinate activity flexibly, regulate their interactions with their surroundings, and in some cases learn from prior interactions or alter behaviour in ways sensitive to future possibilities. At greater degrees of cognitive elaboration, organisms may plan, communicate, solve unfamiliar problems, or coordinate behaviour socially. It is therefore unsurprising that intelligence has often seemed to capture something important about the distinctive capacities of living systems.

The growth of artificial intelligence has made this association more complicated. Artificial systems can now solve problems, optimise performance, recognise patterns, generate complex outputs, and adapt their responses to changing inputs without resembling conventional organisms in their material organisation. Behaviour that once seemed to provide strong evidence of intelligence can therefore be produced by systems whose biological status is not thereby established.

APS recognises the explanatory importance of these developments but asks a different question: what distinguishes the organisation of a living system from capacities that may be exhibited by living or non-living systems? This directs attention away from behavioural sophistication alone and toward the organisation in which that behaviour is realised. Optimisation, prediction, learning, and problem-solving may all be explanatorily important, but none by itself determines whether the system exhibiting those capacities is alive.

Biological Intelligence Occurs Within Living Organisation

Biological intelligence, where it occurs, is realised within living systems. The intelligent capacities of an animal, for example, are materially realised through cells, tissues, nervous systems, bodies, developmental histories, and organism–environment relations that depend upon the continuing organisation of the organism. In such cases, intelligence does not occur independently of biological organisation but through material processes maintained within it.

This claim should not, however, be universalised into the proposition that intelligence as such presupposes life. Once intelligence is characterised in functional terms such as problem-solving, learning, prediction, planning, or adaptive coordination, comparable capacities may be realised in artificial and computational systems that are not biologically alive. The appropriate distinction is therefore between biological intelligence, which is realised within living organisation, and intelligence or intelligence-like capacity considered more generally, which does not by itself establish biological status.

This distinction also prevents a material-dependence claim from becoming an assertion of explanatory superiority. The fact that biological intelligence is realised within living organisation does not show that every theory of intelligence must begin with the APS definition of life, or that biological organisation provides the best explanation of every phenomenon associated with intelligence. It establishes only that intelligence and life should not be treated as identical categories.

Life Does Not Require Intelligence

The APS definition of life does not require intelligence. A living system must maintain and re-establish the conditions of its own persistence through viability-oriented, constraint-closed organisation, but nothing in that definition requires strategic reasoning, abstraction, symbolic planning, explicit representation, or flexible problem-solving of the kinds ordinarily associated with intelligence.

This point is especially important because biological organisation extends far beyond organisms conventionally described as intelligent. Living systems can regulate metabolism, repair damage, maintain boundaries, respond to changing conditions, and coordinate viability-relevant activity without possessing the elaborated cognitive capacities ordinarily invoked when intelligence is discussed. Intelligence is therefore not necessary as a defining criterion of life.

This conclusion should not be confused with the separate and contested question of how broadly cognition should be attributed across living systems. APS distinguishes biological agency from cognition and cognition from intelligence. The claim that life does not require intelligence consequently does not decide whether particular organisms, cells, or other biological systems should be regarded as cognitive. It establishes only that a system may satisfy the APS definition of life without satisfying a demanding criterion of intelligence.

Intelligence-Like Capacity Does Not Establish Life

The converse distinction is equally important. Contemporary artificial and computational systems can exhibit capacities associated with intelligence, including problem-solving, optimisation, adaptive response, prediction, complex pattern generation, and flexible coordination of outputs. Such performances may be scientifically and technologically significant, but their presence does not by itself establish that the system producing them is biologically alive.

Many current artificial systems depend extensively upon external infrastructures for energy, maintenance, fabrication, repair, training, and continued operation. This observation is relevant to their organisation, but it should not be converted into the stronger assertion that artificial systems could never satisfy a biological or APS criterion of life. Whether some future artificial system could instantiate viability-oriented, constraint-closed organisation is a separate question and should not be decided merely from its artificial origin.

The bounded conclusion is therefore that intelligence-like performance is not sufficient by itself to establish life. Determining whether any particular artificial system satisfies the APS definition of life requires analysis of its organisation, not simply an assessment of the sophistication of its behaviour. Behavioural intelligence and biological status remain distinct explanatory questions.

Biological Agency Does Not Require Intelligence

APS defines biological agency as viability-oriented organisational activity. Agency, in this sense, does not require deliberation, symbolic reasoning, self-awareness, strategic planning, or intelligence. It is the present-tense activity through which viability-oriented living organisation is enacted and sustained.

This conception prevents biological agency from being treated as a specialised cognitive capacity. A living system does not first have to become intelligent in order to act in ways organised relative to its own viability. The modulation of activity through which conditions of persistence are maintained can occur without the cognitive elaboration ordinarily associated with intelligence.

At the same time, APS does not infer from this relationship that agency is “more fundamental” than intelligence in some universal hierarchy of explanation. The relevant claim is more precise: biological agency can occur without intelligence. Agency and intelligence are therefore non-identical explanatory concepts, and the presence of one should not automatically be inferred from the presence of the other.

Cognition Is Not Intelligence

APS also distinguishes cognition from intelligence. Cognition is defined as the organisation of biological significance across time. On the stronger boundary currently used to distinguish cognitive activity from immediate or fixed regulation, biological agency becomes cognitive when integrated biological significance modulates activity across a temporal field of viability-relevant possibilities in ways not exhausted by immediate or fixed regulation.

Intelligence occupies a different explanatory role. It commonly refers to comparatively elaborated capacities such as flexible problem-solving, abstraction, planning, strategic coordination, learning across changing contexts, and other forms of cognitive performance. Although particular accounts of intelligence will draw the boundary differently, the important point for APS is that cognition and intelligence should not simply be treated as synonyms.

A living system may therefore exhibit biological agency without cognition, while a cognitive system need not exhibit intelligence in a demanding sense. Conversely, intelligence-like performance considered functionally does not by itself establish biological agency. These boundaries require substantive assessment rather than inference from behavioural complexity alone. Their purpose is not to arrange organisms along a ladder of cognitive sophistication, but to distinguish the organisational claims involved when biological activity is described as agential, cognitive, or intelligent.

Constraint Closure and the APS Definition of Life

Constraint closure belongs to the APS definition of living organisation. In constraint-closed organisation, relevant constraints are maintained through the activity of the system whose continued organisation they help make possible. The concept therefore contributes to specifying what APS means when it characterises life as an organisation capable of actively maintaining and re-establishing the conditions of its own persistence.

Intelligence specifies something different. A capacity for problem-solving, learning, optimisation, prediction, or flexible behaviour does not by itself tell us whether the organisation realising that capacity is viability-oriented and constraint-closed. The distinction is consequently not between a weak theory of intelligence and a stronger theory of life, but between different explanatory targets.

Constraint closure can therefore help determine whether a system satisfies the APS definition of life without thereby demonstrating that APS explains intelligence better than cognitive, computational, artificial-intelligence, predictive, or other approaches. The distinction is definitional and organisational before it becomes comparative. Any stronger claim of explanatory gain would require a specified shared explanandum and evidence that the relevant difference in organisation actually matters to its explanation.

Intelligence, Significance, and Meaning

APS does not treat intelligence as the source of biological significance. Biological Evaluation is the process through which agency generates significance, and biological significance is a relationally real product of that evaluation. Integration concerns the organisation of what matters into functional unity, while cognition concerns the organisation of biological significance across time.

Meaning belongs to a later development in the APS pathway and should not be treated as interchangeable with biological significance, semiosis, cognition, or intelligence. This distinction matters because biological relations can be significant for the viability of an organism without requiring the reflective or interpretative capacities often associated with human meaning. It also prevents intelligence from being made a prerequisite for every form of biological significance.

The present article does not require a complete reconstruction of the pathway from life through cognition toward Mind, Selfhood, Reflective Agency, and Meaning. Its relevant conclusion is simply that living systems need not be intelligent in order for their organisation to exhibit viability-relative significance, while intelligence-like behaviour does not by itself establish those biological relations.

This preserves the distinction between intelligence and the wider APS architecture without treating intelligence as universally caused by, chronologically subsequent to, or explanatorily subordinate to every concept that precedes it in the conceptual pathway.

Intelligence and Predictive Processing

Predictive-processing and Active-Inference approaches provide important resources for explaining prediction, learning, action, regulation, and related cognitive phenomena. Some forms of intelligent behaviour may therefore be fruitfully modelled through prediction, optimisation, inference, model updating, or closely related processes.

Nothing in the APS distinction between life and intelligence requires rejecting such explanations. The fact that prediction or inference can contribute to the explanation of intelligent behaviour does not establish that prediction or inference defines life. Conversely, the APS definition of life does not establish that predictive or inferential explanations are subordinate to APS.

The relationship becomes a genuinely comparative explanatory question only where predictive, cognitive, and APS approaches address the same biological explanandum. At that point, explanatory contribution must be assessed by examining the dependencies each approach specifies, the evidence supporting them, and whether the differences between them produce explanatory gain, complementarity, redundancy, or comparator advantage.

The Life-demarcation claim made here is much narrower: neither prediction nor intelligence is, by itself, the APS definition of life.

Intelligence Across Scale and Time

Intelligence, where it exists, can be organised across multiple spatial and temporal extents. Intelligent activity may integrate memory with present conditions, anticipate future possibilities, coordinate behaviour across different parts of an organism, exploit communication among individuals, or depend upon continuing interactions with features of the environment. Learning, planning, social coordination, and other intelligent capacities therefore cannot always be understood by examining an isolated event or component.

In biological systems, such capacities are materially embedded in ongoing organismal, developmental, physiological, ecological, and behavioural processes. In artificial systems, comparable functional capacities may be realised through different technical architectures and may depend upon extensive external infrastructures. These differences can matter when the organisation realising intelligence becomes part of the explanandum.

APS treats Scale as an analytic projection concerned with where persistence is organised across spatial and temporal extents. Applying that perspective to intelligence does not imply that there is a privileged biological level at which intelligence must be explained. It asks instead how the processes relevant to a particular intelligent capacity are organised across the spatial and temporal extents that materially realise it.

The resulting analysis is therefore contextual rather than hierarchical. Intelligence may be realised differently in biological and artificial systems, and those differences may sometimes matter explanatorily. They should not, however, be converted into the universal claim that intelligence as such is biologically constituted.

Intelligence Within the APS Explanatory Grammar

APS situates intelligence among a wider set of biological and cognitive concepts without treating those concepts as interchangeable. For the purposes of this article, three distinctions are decisive: life is viability-oriented, constraint-closed organisation; biological agency is viability-oriented organisational activity; and cognition and intelligence are not synonymous.

Together, these distinctions establish why intelligence should not be used as the defining criterion of life. They do not establish that theories of intelligence cannot explain biological phenomena, that APS contains a complete theory of intelligence, or that APS has demonstrated comparative explanatory gain over cognitive, computational, predictive, or artificial-intelligence approaches.

This distinction is central to the comparative methodology now used across APS. Conceptual difference does not itself constitute explanatory gain. Where APS and another approach address different explananda, both may be legitimate without being direct competitors. Where they address a genuinely matched explanandum, comparison must concern the dependencies each specifies and the evidence supporting them rather than the vocabulary in which those dependencies are described.

The Life–intelligence distinction therefore performs a clarificatory role before it performs a comparative one. It prevents a capacity associated with some living systems from being mistaken for the organisation by which living systems are defined, while leaving open the substantive question of how intelligence should best be explained.

Why This Clarification Matters

Distinguishing life from intelligence prevents several explanatory questions from being collapsed. Biological status is not established by behavioural sophistication alone, biological agency does not require intelligence, and cognition and intelligence remain distinct within APS. At the same time, intelligence-like capacities can be investigated in biological or artificial systems without assuming in advance that they either constitute life or derive their explanatory significance from the APS account of living organisation.

This clarification is especially important as the behavioural boundary between biological and artificial systems becomes less intuitively obvious. Increasingly sophisticated artificial performance makes behavioural criteria alone a poor basis for deciding questions about biological organisation. Conversely, the extraordinary regulatory and adaptive capacities of organisms should not automatically be redescribed as intelligence merely because they contribute effectively to persistence.

The distinction is useful precisely because it is limited. APS does not need to deny the explanatory importance of intelligence in order to reject intelligence as its definition of life. Nor does it need to classify every adaptive biological activity as intelligent in order to recognise the organised activity through which living systems persist.

Conclusion

Intelligence describes an important family of capacities that may include learning, flexible problem-solving, abstraction, planning, prediction, and adaptive coordination. Such capacities occur in living systems, sometimes in highly elaborated forms, but intelligence-like capacities can also be exhibited by artificial and computational systems. Their presence therefore does not provide a sufficient criterion for distinguishing the living from the non-living.

APS approaches that distinction through organisation. Life is viability-oriented, constraint-closed organisation, while biological agency is viability-oriented organisational activity and cognition is the organisation of biological significance across time. These concepts identify different explanatory features of biological organisation and should not be collapsed into a single scale of increasing sophistication.

The resulting conclusion is deliberately bounded: life is not intelligence, and intelligence-like capacity alone does not establish life. This conclusion does not establish that APS is explanatorily superior to theories of cognition or intelligence. Comparative explanatory questions arise only when APS and another approach address the same explanandum and can be assessed in relation to the dependencies and evidence each provides.

The importance of the distinction lies precisely there. Intelligence can be biologically important without defining life, just as living organisation can be explained without treating intelligence as its organising principle.

Key Point

Intelligence is neither the APS definition of life nor by itself sufficient to establish biological status. Biological intelligence is materially realised within living organisation, while intelligence-like capacities may also occur in non-living systems. The non-identity of life and intelligence does not by itself establish comparative explanatory superiority.

Explanatory Architecture

Central Question

The central question is whether intelligence should be identified with life or distinguished from the organisation through which living systems maintain and re-establish the conditions of their persistence. APS adopts the latter position while leaving open the broader scientific question of how particular forms of intelligence are best explained.

Architectural Role

This article establishes a Life-demarcation claim. Its purpose is to distinguish life from intelligence without claiming that APS provides a complete theory of intelligence or that cognitive, computational, predictive, or artificial-intelligence approaches are explanatorily subordinate to APS. The article therefore establishes non-identity while withholding claims of comparative explanatory gain.

Preceding Explanatory Dependencies

Within the APS corpus, the clarification draws upon the accounts developed in What Is Life?, Biological Agency, Biological Evaluation, Biological Significance, Integration, Cognition — Where Does It Belong in Biology?, Organised Persistence, Constraint Closure, and APS and Contemporary Explanatory Approaches. These are relations within the APS explanatory architecture rather than claims that the concepts involved form a universal causal, temporal, evolutionary, hierarchical, or comparative ordering.

Subsequent Explanatory Developments

The distinctions established here provide a basis for more specialised questions concerning minimal cognition, artificial systems, biological and artificial agency, Mind and Selfhood, and comparative approaches to predictive, computational, and cognitive explanation. Those later investigations can ask whether particular forms of intelligence require specific organisational dependencies without reopening the narrower Life-demarcation result established here.

The article consequently opens several questions without attempting to settle them. These include what distinguishes cognition from intelligence, which forms of biological intelligence require cognition in the APS sense, whether artificial systems could instantiate viability-oriented organisation, when intelligence-like behaviour warrants attribution of cognition, and which proposed relations among agency, significance, cognition, intelligence, and meaning correspond to genuine biological dependencies rather than conceptual ordering. A further comparative question arises whenever APS and a theory of intelligence address a genuinely matched explanandum: whether their differing explanatory resources produce gain, complementarity, redundancy, or comparator advantage.

Position Within APS

This article belongs to the Life-demarcation family. Its function is to prevent behavioural or cognitive sophistication from being treated as the defining basis of life while preserving intelligence as a legitimate and important explanatory target in its own right. It therefore clarifies the boundary between life and intelligence without adjudicating the general comparative adequacy of theories of intelligence.