A Boundary Is Not a Line

How Living Systems Remain Distinct Without Sealing Themselves Off from the World

Life cannot continue by sealing itself away from the world. A living system must take in matter and energy, release waste and heat, detect changes, respond to some signals, resist some disturbances, tolerate others, and repair the material structures through which these relations occur. Its continued existence therefore depends on a basic condition:

A living system must remain causally distinct from its environment while continuing to interact with it.

This makes biological boundaries difficult to understand if we imagine them as simple geometric lines. A line can show where one region ends and another begins. But a living boundary concerns something deeper:

which causal relations distinguish the organization from its environment, and how those relations are materially implemented.

Connection Theory therefore separates three ideas that are often collapsed: Boundary
the empirically distinguishable causal differentiation between inside and outside; Interface
a materially implemented pathway through which causal exchange crosses that differentiation; Living Boundary
a specialized biological form of boundary in which selectively regulated coupling relations are integrated into the continuing causal organization of a living system.

1. Is Food Already Inside You?

Suppose you swallow a piece of food. It has crossed your lips. It is now enclosed by the external silhouette of your body. In an ordinary spatial sense, it appears to be inside you. But it has not yet entered your internal tissues.

The gastrointestinal tract forms a continuous passage from the mouth to the anus. Material in its lumen remains separated from internal tissues by layered barrier structures and processes, including mucus and epithelium (Johansson et al., 2008). A nutrient may therefore be:

• inside the body's external outline;
• still outside internal tissue;
• exposed to digestive enzymes;
• not yet absorbed;
• interacting with gut microorganisms;
• later transported into blood;
• eventually incorporated into the organism's material organization.

Which answer is correct? Several can be correct at once. They refer to different relations. This gives the first principle:

“Inside” and “outside” are not single biological facts. They are answers to specified causal and organizational questions.

2. A General Boundary Is Not Necessarily a Wall

The idea is broader than biology. For any organized system, a boundary appears when its regularized causal relations create an empirically distinguishable inside–outside pattern. That boundary may coincide with a physical surface. But it does not have to. A distributed organization can have members in several locations. A remote component can participate strongly in the internal causal organization. A nearby object can remain causally external. So a general boundary should not be defined by geometry alone.

A boundary is a pattern of causal differentiation, not merely a line in space. For a persistent organization, that differentiation must also be maintained through selective control of the flows that cross it.

The relevant question is:

Inside or outside with respect to which relation?

Boundary Strength Is About Selective Control, Not Closure

An open system can have a strong boundary.

People, matter, energy, information, signals, resources, and authority can cross a system boundary without erasing the system's individuality. What matters is whether the organization controls the terms of crossing strongly enough to preserve its internal relational pattern.

This suggests a useful general concept: organizational boundary sovereignty—the capacity of an organization to determine what may cross its boundary, under what conditions, and with what consequences for internal state, membership, authority, or access.

A company does not cease to be itself because employees enter and leave. It becomes organizationally vulnerable when outsiders can bypass its own processes and arbitrarily rewrite decision authority, assets, information pathways, or membership rules. A cell likewise depends not on perfect impermeability but on selective transport, signalling, recognition, and repair.

So:

Open system does not mean boundaryless system.

Boundary failure can therefore propagate inward. If cross-boundary flow exceeds or bypasses the system's regulatory capacity, controlled exchange can give way to loss of internal state regulation and then disruption of the relations that sustain the organization.

3. Boundary and Interface Are Different

A boundary describes differentiation. An interface implements exchange across that differentiation. A cell membrane is a material interface. But the boundary functions associated with it depend on many organized processes:

• channels;
• transporters;
• pumps;
• receptors;
• junctions;
• cytoskeletal relations;
• extracellular structures;
• repair mechanisms.

Likewise, one boundary function can be distributed across several physical interfaces. The intestinal barrier is not implemented by one membrane alone. It involves mucus, epithelial cells, junctions, immune processes, connective tissue, and vascular regulation. Therefore:

Boundary ≠ interface.

A boundary is the causal inside–outside distinction being studied. An interface is one of the concrete structures or processes through which that relation is enacted.

4. A Living Boundary Is a Specialized Boundary

Not every boundary is living. A dialysis membrane selectively separates substances. A reverse-osmosis membrane selectively permits some forms of transport. A valve controls flow. These can all implement real boundary-like relations. But they are not alive merely because they are selective. A living boundary belongs to a living causal organization. Its operation is integrated into the continuing dynamics of the same living system whose distinction from the environment it helps maintain. This makes the biological case more specific than the general concept. A working distinction is:

General boundary: an empirically distinguishable causal inside–outside differentiation.

Living boundary: a selectively regulated coupling relation, materially implemented through one or more interfaces and integrated into a continuing living causal organization.

The second is a specialized case of the first.

5. One Organism Can Have Several Boundary Dimensions

“The boundary of an organism” often hides several different questions. For example:

• Which substances can cross?
• Which external energy sources can be used?
• Which environmental changes can be detected?
• Which detected changes alter internal regulation or behavior?
• Which biological entities are tolerated or attacked?
• Which mechanical disruptions compromise integrity?
• Which damaged interfaces can still be repaired?

These questions need not have the same answer. It is therefore useful to speak of a family of boundary dimensions. Possible examples include:

Material coupling

Which substances cross, in what quantities, and through which interfaces?

Energetic coupling

Which external gradients or energy sources can be converted into processes usable by the organism?

Signalling coupling

Which external changes produce detectable internal state changes?

Regulatory coupling

Which detected changes alter resource allocation, regulation, or action?

Immune coupling

Which biological entities or molecular patterns are tolerated, contained, or attacked?

Integrity and repair coupling

Which disruptions are treated as damage, and how does the organism preserve or restore the relevant interface?

These are not asserted as a final canonical list. A boundary dimension should be retained only if it corresponds to an empirically distinguishable class of causal relations.

6. How Can a Boundary Dimension Be Identified?

If we allow several boundary dimensions, we need a way to prevent arbitrary proliferation. A proposed boundary dimension should satisfy empirical requirements.

1. A causal relation is specified

What kind of coupling is being studied?

Material transport? Electrical influence? Molecular signalling? Mechanical continuity? Immune recognition?

2. The relation is measurable

Different substances, signals, timings, directions, or states produce distinguishable effects.

3. The relation is state-dependent or regulated

The coupling can vary with the organism's current state, history, activity, or environmental conditions.

4. It can be perturbed relatively independently

Changing the proposed boundary relation can leave at least some other boundary functions temporarily intact.

5. Its disruption has a distinguishable consequence

Perturbing the relation changes the living organization in a characteristic way—for example by disturbing a gradient, transport process, signal relation, immune interaction, or mechanical integrity.

The observer chooses the system and the relation to examine. But once that frame is specified, the relevant coupling and perturbation effects are empirical.

The frame is selected; the causal relations within that frame are not created by the observer.

7. The Gut Shows Why Boundaries Do Not Coincide

The intestine makes this especially visible. A functional intestinal barrier must allow absorption while limiting uncontrolled passage of toxins, pathogens, and microorganisms. But this does not happen at one single line. The mucus layer affects which organisms and molecules can approach the epithelium. Epithelial cells and junctions regulate passage. Immune processes sample, tolerate, contain, or attack biological material. Additional vascular interfaces regulate access to circulation. These relations occupy nearby physical locations but answer different causal questions. A molecule may:

• cross mucus but not epithelium;
• enter an epithelial cell but not the bloodstream;
• influence host physiology without the producing microorganism entering tissue.

A commensal microorganism can be tolerated in the gut lumen but treated very differently if it enters blood. So:

Spatial enclosure does not settle biological membership or causal inside–outside status by itself.

8. The Microbiome Makes the Problem Even Clearer

Gut microorganisms participate in host physiology. They can contribute to host metabolism and immune development, among other functions (Bäckhed et al., 2004; Mazmanian et al., 2005). Yet they possess their own genomes and evolutionary histories. Are they part of the organism? There is no useful universal answer without specifying the relation. They can be:

• spatially enclosed;
• metabolically influential;
• immunologically tolerated;
• causally integrated into some host processes;

while still remaining distinct biological individuals under another individuation frame. Functional contribution does not automatically settle identity. This anticipates a later question:

What makes a changing living being remain the same individual through time?

Boundary and identity are related, but they are not the same problem.

9. Multiple Boundaries Are Not Peculiar to Animals

The same logic appears in organisms without guts, blood vessels, or adaptive immune systems.

Gram-negative bacteria

The bacterial envelope includes an outer membrane, a periplasmic region with peptidoglycan, and an inner cytoplasmic membrane. These layers contribute differently to:

• permeability;
• transport;
• mechanical integrity;
• resistance to external compounds.

Reducing entry and actively removing an internal compound can both alter intracellular concentration, but through distinct causal relations. The bacterium has one visible outline but multiple boundary functions.

Plant cells

The cell wall and plasma membrane also perform different roles. The wall contributes mechanical support. The plasma membrane performs selective transport and signalling. Plasmodesmata create regulated intercellular interfaces whose permeability can change in response to physiological conditions (Park et al., 2019). A plant cell can therefore preserve one boundary function while another is disrupted. Again:

One visible surface may implement several causal boundaries, while one boundary function may depend on several structures.

10. Detection and Regulation Are Not the Same Relation

An external event can affect a living system in several stages. A molecule contacts a surface. A receptor binds it. An internal state changes. A regulatory or behavioral response may follow. These stages should not be collapsed. A signalling relation exists when an external condition produces a detectable internal change. A regulatory relation exists when that change participates in altering subsequent regulation, resource allocation, or action. Not every detected signal changes system-level behavior. Some signals are ignored. Some produce only local effects. Some matter only under particular internal conditions. So:

Detection is not the same as regulation.

And, consistent with the broader framework:

Regulation is not the same as adaptation.

A living boundary can participate in all three without making them identical.

11. Boundary Functions Can Fail Separately

If all biological boundaries were one thing, they should fail together. They do not. A permeability defect is not identical to immune misclassification. Immune tolerance is not identical to mechanical integrity. Mechanical integrity is not identical to signal discrimination. Failures can propagate from one dimension into another, but propagation does not erase the distinction between the initiating causal relations. This gives the multiple-boundary proposal empirical content. Boundary dimensions correspond to distinguishable vulnerabilities, not merely different words for one wall.

12. Boundary and Flow Co-Maintain the Living System

Living organization depends on flows. But flows often depend on maintained differences:

• concentration gradients;
• electrical potentials;
• pressure differences;
• chemical potentials;
• uneven resource distributions.

Boundary processes help preserve these differences. The differences then support transport and energy conversion. Those flows provide material and energy for regulation and repair. Repair helps preserve the interfaces that sustain the differences. A mature living system can therefore contain a maintenance loop: Boundary Relations → Preserved Differences → Organized Flow → Regulation / Repair → Maintained Boundary Relations This is not a circular definition. It is a dynamical relation among mutually sustaining processes. A living boundary is not simply maintained by life. It participates in the causal organization through which living dynamics continue.

13. Boundaries Have History

A living boundary does not necessarily behave identically every time. Previous:

• exposure;
• inflammation;
• damage;
• repair;
• development;
• adaptation

can alter later permeability, signalling, recognition, or response. But not every historical effect should be called memory. A repaired membrane may behave differently because its current material state carries the consequences of earlier damage. Scarred tissue may respond differently because previous events changed present organization. This is broadly history dependence. A stronger form of structural memory exists when past causal history becomes retained in present organization in a way that systematically biases future dynamics. So:

History can change a boundary without historical change itself constituting a new boundary dimension.

Boundary describes an inside–outside causal relation. Structural memory describes how prior causal history remains present in the organization through time.

14. What Is Not Sufficient to Make a Living Boundary?

Several features are insufficient by themselves.

Spatial separation is insufficient

A wall can divide two spaces without belonging to a living organization.

Selective transfer is insufficient

A dialysis membrane can selectively separate molecules without being alive.

Detection is insufficient

A sensor can register an external signal without being part of a living boundary.

Regulation is insufficient

A fixed controller can regulate exchange without possessing living organization.

For a boundary to be specifically living, the relevant coupling must be integrated into the continuing causal organization of the living system. External support can assist this organization without automatically becoming a living component of the organism. A dialysis device, for example, can perform an important physiological support function while remaining an external technological system.

15. A Working Definition of a General Boundary

Connection Theory uses the following public working definition:

A boundary is an empirically distinguishable causal differentiation between an organized system and its environment with respect to a specified class of relations. In a persistent open system, that differentiation is maintained through selective regulation of the relevant cross-boundary flows.

This definition contains several points.

Empirically distinguishable

The boundary is not merely conceptual; the inside–outside difference should correspond to measurable causal organization.

Causal differentiation

The distinction concerns how parts of the system relate to one another and to the environment, not simply where they are located.

Specified class of relations

Information, matter, signalling, authority, mechanical continuity, and other relations can generate different boundary dimensions.

System-relative

A boundary is always stated with respect to a specified system and explanatory question.

16. A Working Definition of a Living Boundary

A living boundary is more specific:

A living boundary is a selectively regulated causal coupling relation, materially implemented through one or more interfaces and integrated into the continuing causal organization of a living system, such that changing the relation produces a distinguishable change in the system's living dynamics.

This distinguishes a living boundary from:

• a geometric outline;
• a passive wall;
• an externally imposed filter;
• a selective interface that does not belong to the living organization.

The definition is deliberately relational. A living boundary is not simply located somewhere. It is enacted through causal organization.

17. The Central Claim

A living system is not separated from its environment where interaction stops. Interaction does not stop. The system remains distinct because some causal relations are organized differently across the inside–outside distinction. So:

A line tells us where a geometric division has been drawn. A boundary tells us how an organized system is causally distinguished from its environment.

And for living systems:

A living boundary is a family of selectively regulated causal relations, materially realized through interfaces and integrated into the continuing organization of life.

These relations can:

• overlap spatially;
• differ functionally;
• change with state and history;
• be measured separately;
• fail separately;
• support one another.

The central shift is therefore: from geometry to causal organization.

18. The Next Question

If a living organism can replace matter, alter many internal relations, and change its boundary functions while continuing to exist, what exactly persists through time? It cannot simply be the same material. It cannot simply be the same shape. It cannot require every internal causal relation to remain unchanged. The next question is:

What makes a living being remain the same being while its causal organization continues to transform?

Part III — What Makes a Living Being the Same Being?

References

• Bäckhed, F., Ding, H., Wang, T., et al. (2004). The gut microbiota as an environmental factor that regulates fat storage. Proceedings of the National Academy of Sciences, 101(44), 15718–15723. DOI: 10.1073/pnas.0407076101.
• Johansson, M. E. V., Phillipson, M., Petersson, J., et al. (2008). The inner of the two Muc2 mucin-dependent mucus layers in colon is devoid of bacteria. Proceedings of the National Academy of Sciences, 105(39), 15064–15069. DOI: 10.1073/pnas.0803124105.
• Mazmanian, S. K., Liu, C. H., Tzianabos, A. O., & Kasper, D. L. (2005). An immunomodulatory molecule of symbiotic bacteria directs maturation of the host immune system. Cell, 122(1), 107–118. DOI: 10.1016/j.cell.2005.05.007.
• Park, K., Knoblauch, J., & Jensen, K. H. (2019). Controlling intercellular flow through mechanosensitive plasmodesmata nanopores. Nature Communications, 10, 3564. DOI: 10.1038/s41467-019-11201-0.