The Emergence of Agency III — From Possibility to Action

An organism lives among possibilities.

It can move toward something or away from it. It can grow in one direction rather than another. It can open, close, absorb, expel, pursue or retreat.

But a possibility is not yet an action.

What, then, turns one into the other?

We might be tempted to say: choice.

An agent considers alternatives and selects one. Action therefore begins with decision.

But this places agency rather late in the story.

A plant's root does not decide where to grow. A bacterium does not deliberate about a chemical gradient. A muscle does not choose to contract in the way a person chooses to raise a hand.

Yet these systems do not merely undergo events.

They are organised so that some conditions produce changes that alter what happens next.

The crucial transition may therefore be simpler than choice:

a possibility becomes an action when the system is organised to make that possibility happen.

Consider a seed.

Light, water, gravity and temperature establish a field of possible futures. But the seed does not passively wait for one of them to occur. Its internal organisation selectively transforms its circumstances: roots extend, shoots emerge, resources are mobilised, barriers are penetrated.

The organism converts possibilities into events.

This is not because it possesses a representation of the future. It is because its organisation is oriented toward maintaining itself through particular transformations.

That orientation can be extremely primitive.

A bacterium moving up a chemical gradient does not need to know where it is going. Its movement nevertheless changes the probability of reaching conditions under which it can continue to function.

Action, in this sense, is not defined by intention.

It is defined by organised intervention in the field of possibilities.

This gives us a useful distinction between reaction and action.

A stone rolls downhill because forces act upon it. The stone's organisation does not regulate its movement toward some condition required for its continued existence.

An organism, by contrast, can alter its relation to its surroundings in ways that contribute to its own persistence.

The distinction is not absolute. Biology is full of processes that look like ordinary physical reactions. What changes is the organisation in which those reactions participate.

A contraction of muscle is a physical event.

Within an organism, it can become part of a movement.

Movement can become part of behaviour.

Behaviour can alter the organism's circumstances.

And those altered circumstances can change the organism's subsequent possibilities.

The scale changes, but the underlying structure remains:

possibility → organised intervention → altered possibility.

This is why agency need not be imagined as a mysterious force added to physical processes.

Agency can emerge when physical processes become organised into a system capable of altering the conditions of its own continuation.

The important word is capable.

An organism need not successfully achieve what it is organised to achieve. Action can fail. A plant can grow toward light and die. An animal can flee and still be caught.

Failure is significant because it reveals that action is not simply what happened.

There was a distinction between possible outcomes, and the organism's activity participated in determining which outcome occurred.

That gives action a temporal structure.

The organism begins in a situation with possibilities.

It acts.

The situation changes.

A new set of possibilities appears.

Action therefore does not merely select from a pre-existing menu of futures.

It changes the menu.

This is where agency begins to become genuinely relational.

The organism is not an isolated source of causation imposed upon a passive environment. Nor is it merely the endpoint of causes originating outside itself.

It occupies a loop in which its own activity becomes part of the conditions that determine what it can do next.

And the loop can become increasingly elaborate.

The bacterium changes its position.

The plant changes its environment.

The animal changes its behaviour according to what happens.

The human can change the conditions under which future choices themselves become possible.

Agency grows by extending the loop.

Perhaps, then, the history of agency is not the history of the moment when matter somehow acquired the power to choose.

It is the history of possibilities becoming organised into interventions, and interventions becoming part of the production of new possibilities.

Choice is one extraordinarily sophisticated form of this process.

It is not necessarily its beginning.

The beginning may be much older:

something acts when its organisation makes a difference to which of its possible futures becomes actual.

The Emergence of Agency II — The World an Organism Can Afford

An organism does not live in the world as we describe it.

It lives in the world that its organisation makes possible.

A bacterium encounters chemicals, but only some chemical differences are relevant to what it can do. A plant encounters sunlight, gravity, water and countless other physical conditions, but these do not all enter equally into its life. A spider encounters a space in which vibrations, surfaces and gaps have consequences that would be invisible to us.

The organism therefore does not simply inhabit an environment.

It inhabits a field of possibilities.

This does not mean that the organism constructs an imaginary world. The external conditions are real. What changes is the relation between those conditions and the organism's capacities.

Water affords something to a fish that it does not afford to a terrestrial mammal.

A vertical surface affords something to a spider that it does not afford to a tree.

A particular wavelength of light can afford photosynthesis to a plant without being perceived as light at all.

An affordance, in this broad sense, is not a property of the object alone. Nor is it a projection of the organism alone.

It is a relation between what exists and what the organism can do.

This gives us another way of understanding the emergence of agency.

An organism's possibilities are not unlimited. Its structure selects among the possibilities offered by its environment.

A membrane permits some exchanges and prevents others. A digestive system makes some substances usable and others irrelevant or harmful. Muscles, limbs, roots, wings and sensory systems each open some possibilities while closing others.

To be an organism is therefore already to inhabit a structured space of possible action.

And this space is not fixed.

An organism changes as it acts.

A plant's roots alter the soil around them. A beaver changes a river. A spider changes its surroundings by constructing a web. An animal's repeated movements can turn a previously irrelevant feature of its environment into a usable resource.

The organism does not merely respond to its world.

It helps to make the world it can subsequently respond to.

This creates a reciprocal relation:

the environment constrains what the organism can do, while the organism changes the conditions under which it can act.

Agency therefore cannot be located entirely inside the organism.

Nor can it be located entirely outside it.

It arises in the relation between the two.

This also clarifies the connection between mattering and possibility that we began with.

Something matters to an organism partly because it changes what the organism can afford to do.

Food matters because it can sustain future activity. Shelter matters because it changes the possibilities of survival. A threat matters because it constrains the organism's future.

Mattering is therefore not simply a reaction to the present.

It is oriented, however primitively, toward what can happen next.

An organism lives in a world of affordances because it lives in a world of possible futures.

And those futures are not abstract possibilities. They are possibilities for this organism.

The distinction is important.

There is the possibility of climbing a tree.

And there is the possibility of climbing a tree for a squirrel.

The first is a statement about the world.

The second describes a relation between the world and an organism's capacities.

This may be the beginning of perspective in a very elementary sense. Not consciousness, not representation, not a private point of view—but a world organised according to what is possible here, for this system.

Agency begins to look less like a power added to life and more like a consequence of being alive.

To live is to be constrained.

To be constrained is to have some possibilities and not others.

And to have possibilities that differ in their consequences is already to have something at stake.

The organism inhabits, therefore, not simply a world of things, but a world of what it can afford to do.

And that world is continually changing because the organism is changing it.

Agency may begin precisely in that circularity:

the organism acts upon a world of possibilities, and in acting, changes the possibilities upon which it can act.

The Emergence of Agency I — When Does Anything Have a Stake?

We usually begin talking about agency when something appears to do something.

An animal hunts. A person decides. A government acts. A machine executes a command. Agency seems to belong to things that can intervene in the world.

But perhaps this starts too late.

Before anything can act on a possibility, something has to make a difference between possibilities. And before a difference can guide action, some possibilities must matter more than others.

This suggests a simpler question:

When does anything have a stake?

A stone can fall. A river can erode its banks. A crystal can grow. None of these processes, by themselves, give us reason to say that anything is at stake. The difference between what happens and what might have happened belongs to our description, not necessarily to the system itself.

An organism is different.

A bacterium can encounter a concentration of sugar and move towards it. It can encounter a toxin and move away. The relevant distinction is not merely that the environment contains two different chemical conditions. The distinction is that the organism's continuing existence is differently related to them.

The environment contains differences.

For the organism, some of those differences matter.

This does not require consciousness. The bacterium need not experience sugar as desirable or toxin as unpleasant. Nothing needs to feel like anything. Mattering can be entirely biological before it becomes experiential.

This gives us a useful distinction between difference and mattering.

A difference is simply a difference between states.

Mattering begins when a difference is incorporated into the organisation of a system such that different states have different consequences for its continuation.

The distinction is subtle but important. A change in temperature is a physical difference. It becomes biologically significant when an organism's capacity to continue depends upon remaining within some range of temperatures.

The organism therefore does not merely occupy a world of differences. It participates in a world of consequences.

And consequences introduce possibility.

For a stone, there may be countless things that could happen next, but none of them is intrinsically preferable to another. For an organism, by contrast, some possible futures are compatible with continued organisation and others are not.

The future has acquired a structure.

This may be one of the earliest beginnings of agency—not because the organism chooses among futures, but because its organisation distinguishes futures according to what they mean for its continuation.

We should therefore be cautious about locating agency in decision-making.

Decision may be a very late form of agency.

At its origin, agency may be much simpler: a system whose own organisation makes a difference to what happens next.

The organism does not merely undergo events. Its state alters the possibilities available to it, while those possibilities feed back into its continued existence.

That creates a rudimentary loop:

what happens → matters to the organism → changes what the organism can do → changes what happens next.

Agency begins to appear not as a thing possessed by an organism, but as a relation between organisation and possibility.

This also changes how we should think about value.

Value need not begin as an idea, a preference, or a judgment. It can begin as a difference in consequence.

For something alive, there can be better and worse conditions before there is anyone capable of saying better or worse.

Life therefore introduces something remarkable into the physical world:

a future that matters to the system that is living it.

That may be the smallest possible stake.

And perhaps agency begins there—not when something first chooses what to do, but when what happens next becomes a matter of consequence for what it is.