What Is Free Will?

You stand before two doors.

You choose the left one.

Could you really have chosen the right?

Were you:

the author of the action?

Or was the decision fixed by:

• genes,
• brain state,
• prior causes?

This is the problem of free will.

One Phrase, Several Questions

“Free will” can mean different things.

At least four questions are often mixed together:

• Could I have done otherwise?
• Did the action come from me?
• Was I controlled or coerced?
• Am I morally responsible?

A theory may answer these differently.

Freedom from Coercion

One everyday sense of freedom is simple:

you acted without external force.

If someone holds a gun to your head, your freedom is reduced.

This is political and practical freedom.

Metaphysical Freedom

A deeper question asks whether your action was determined by prior states of the universe.

This is metaphysical freedom.

It concerns causation.

Agency

Agency is the capacity to:

• form goals,
• choose actions,
• influence outcomes.

Agency does not automatically imply metaphysically undetermined choice.

A deterministic system can still be an agent in a functional sense.

Voluntary Action

A voluntary action typically involves:

• intention,
• control,
• absence of compulsion.

This is already meaningful even before solving metaphysics.

Authorship

We often care whether an action expresses:

• my values,
• my reasoning,
• my character.

This is the idea of authorship.

Free will may concern ownership of action more than causal uncausedness.

Alternative Possibilities

A traditional intuition says freedom requires:

I could have done otherwise.

This is the principle of alternative possibilities.

It has been heavily debated.

What Does “Could Have” Mean?

There are several interpretations.

Conditional

If I had wanted differently, I would have acted differently.

Categorical

Holding the entire past fixed, I still could have acted differently.

Compatibilists and incompatibilists divide sharply here.

Determinism

Determinism says that given:

• the complete state of the world,
• laws of nature,

only one future is physically possible.

This does not mean humans know that future.

Determinism ≠ Predictability

Chaotic systems can be deterministic yet unpredictable.

Human behavior can be causally determined without being easy to forecast.

This distinction is essential.

Fatalism

Determinism is not fatalism.

Fatalism says outcomes occur regardless of what we do.

Determinism says what we do is part of the causal chain producing outcomes.

Your actions still matter causally.

Example

If you study, you may pass.

If you do not, you may fail.

Under determinism, your studying itself has causes.

But it remains causally relevant.

Randomness

Suppose determinism is false.

Quantum events introduce randomness.

Does randomness create freedom?

Not obviously.

A random coin flip is not a choice.

Control

Freedom seems to require:

control.

Pure randomness reduces control.

Therefore:

indeterminism alone ≠ free will.

The Core Triangle

Free-will theories often balance three ideas:

• causation,
• alternatives,
• control.

Different theories prioritize different combinations.

Hard Determinism

Hard determinists say:

determinism is true

and:

free will is incompatible with determinism.

Therefore:

no free will.

Hard Incompatibilism

A broader view says:

whether determinism or indeterminism is true, the kind of free will required for ultimate moral responsibility does not exist.

This is hard incompatibilism.

Compatibilism

Compatibilists say:

free will is compatible with determinism.

Freedom concerns:

• acting from one’s reasons,
• absence of coercion,
• appropriate control.

Causal determination does not destroy agency.

Libertarian Free Will

Libertarianism about free will says:

we have genuine freedom

and:

that freedom is incompatible with determinism.

Therefore some human choices must be indeterministic in the relevant way.

Not Political Libertarianism

This philosophical use of libertarianism concerns free will.

It is unrelated to political ideology.

The terms share a word, not a theory.

Sourcehood

Some incompatibilists focus less on alternatives and more on sourcehood.

To be free, you must be the ultimate source of your action.

But this creates a regress.

The Ultimate Source Problem

You act from your character.

But did you choose your character?

Your character came from:

• genes,
• upbringing,
• prior choices.

Those prior choices came from an earlier character.

Where does ultimate authorship begin?

Galen Strawson’s Basic Argument

Galen Strawson argues, roughly:

to be ultimately responsible for what you do, you would need to be ultimately responsible for the way you are.

But you cannot create yourself from nothing.

Therefore ultimate responsibility is impossible.

Self-Creation

No one chooses:

• birth,
• initial temperament,
• early environment.

This weakens the idea of absolute self-authorship.

But Ordinary Responsibility May Survive

We do not usually require people to create themselves ex nihilo.

Responsibility may depend on:

• reasons responsiveness,
• control,
• understanding.

This supports compatibilist approaches.

Reasons-Responsive Control

A person is free when their decision mechanism responds appropriately to reasons.

If new evidence would change the choice, the system exhibits rational control.

This is a sophisticated compatibilist criterion.

Manipulation

Suppose someone secretly rewires your brain to desire something.

You then act willingly.

Was the action free?

Manipulation cases challenge simple:

“do what you want”

theories.

Deep Self

Some theories distinguish:

• surface desire,
• higher-order values.

Freedom may require acting from desires you identify with.

This connects to Harry Frankfurt.

Frankfurt’s Hierarchy

Frankfurt distinguished:

first-order desires:

want to do X.

and:

second-order desires:

want to want X.

A person can reflect on which motivations they endorse.

Addict Example

An unwilling addict may:

want the drug

but:

wish not to want it.

The conflict reveals different levels of agency.

Identification

Freedom may involve acting from motivations the person identifies with.

This shifts focus from alternative possibilities to internal structure.

Frankfurt Cases

Frankfurt constructed thought experiments in which a person appears morally responsible even though they could not have done otherwise.

These challenge the principle of alternative possibilities.

Basic Structure

Suppose a hidden controller would force Jones to choose A if Jones showed signs of choosing B.

But Jones chooses A entirely on his own.

The controller never intervenes.

Was Jones responsible?

Many think yes.

Why Frankfurt Cases Matter

If responsibility survives despite no genuine alternative, then:

could have done otherwise

may not be necessary.

What matters may be:

how the actual decision arose.

Coercion

Coercion reduces freedom because the action no longer reflects normal preferences and reasons.

This supports control-based theories.

Compulsion

Addiction, obsession, and certain disorders can reduce control internally.

Freedom is not only about external force.

Agency depends on psychological organization.

Ignorance

A person may act voluntarily but without relevant knowledge.

Responsibility can be reduced.

Freedom and responsibility are related but distinct.

Moral Luck

Outcomes partly depend on factors beyond control.

Two reckless drivers behave identically.

Only one happens to hit someone.

Do they deserve different blame?

This is the problem of moral luck.

Consequence vs Choice

Responsibility can attach to:

• decision process,
• outcome.

Legal and moral systems often treat them differently.

Neuroscience

Every decision has neural correlates.

Does discovering them undermine freedom?

Only if free will requires decisions to occur outside the brain.

A physicalist can say:

you are the brain process.

Brain Causes vs My Causes

If my brain causes an action, did I cause it?

The question assumes:

I

and:

my brain

are separate agents.

A physicalist rejects that split.

Conscious Intention

We often experience:

deliberation, decision, action.

But introspection may not reveal the full causal chain.

Neuroscience can study the timing.

This leads to Libet experiments later.

Consciousness and Agency

Does free will require consciousness?

Some actions are automatic yet voluntary.

Others involve explicit deliberation.

Consciousness may increase flexible control without being necessary for every free action.

Habit

A skilled driver performs many actions automatically.

We still attribute agency to the driver.

Control can be distributed across conscious and unconscious processes.

Decision as Process

A decision may not occur at one instant.

It can emerge gradually from:

• evidence accumulation,
• competing goals.

The idea of a tiny inner chooser may be misleading.

Homunculus Again

If free will requires a little self inside the brain making decisions, we have explained nothing.

The chooser would need its own chooser.

Agency must emerge from the whole system.

Free Will as Emergent Control

One possibility is that free will is a higher-level capacity.

A person can:

• model options,
• predict consequences,
• regulate impulses.

These are real abilities even if implemented physically.

Degrees of Freedom

Freedom may come in degrees.

A child, an addicted person, a reflective adult

may possess different levels of self-control.

This fits ordinary practice better than an all-or-nothing metaphysical switch.

Learning Increases Freedom

A person who learns:

• emotional regulation,
• critical thinking

can respond to more reasons.

In this sense, freedom can be cultivated.

Social Conditions

Freedom also depends on:

• education,
• poverty,
• coercive institutions.

Metaphysical free will should not erase social constraints.

Legal Responsibility

Law usually cares about capacities such as:

• intent,
• knowledge,
• control.

It does not require proving metaphysical indeterminism.

Practical responsibility is already compatibilist in many respects.

Ultimate Responsibility

Some philosophers demand something stronger:

ultimate desert.

The person deserves blame or praise in a deep sense independent of prior causes.

This is much harder to defend.

Desert vs Consequential Responsibility

Punishment can be justified because someone:

deserves it

or because it:

• deters,
• rehabilitates,
• protects.

Free-will debates affect these justifications differently.

Free Will and Physics

Newtonian determinism once seemed to threaten freedom directly.

Quantum mechanics removed strict classical determinism.

But randomness does not automatically produce agency.

Physics alone does not settle the philosophical issue.

Free Will and Computation

A deterministic algorithm can still evaluate reasons and choose based on goals.

Would we call that free?

For machines, the same conceptual questions reappear.

Machine Agency

An artificial agent can:

• represent alternatives,
• plan,
• select.

This is functional agency.

Whether it deserves moral responsibility is another matter.

The Master Question

Free will is not one metaphysical switch.

It is a family of questions about:

• alternatives,
• control,
• sourcehood,
• responsibility.

Clarity begins by separating them.

The Beginning of Part XVII

Part XVII asks:

Are our choices genuinely ours?

We will examine:

• determinism,
• compatibilism,
• libertarianism,
• neuroscience,
• responsibility,
• purpose.

The Philosophical Lesson

Free will becomes less mysterious once we stop asking only:

“Was the action uncaused?”

A useful concept of freedom may instead ask:

Did the action flow from a system capable of:

• understanding reasons,
• regulating itself,
• acting without coercion?

Whether this is enough is the central dispute.

The Next Question

Before evaluating the theories, we must understand the strongest apparent threat.

If every event follows from prior conditions and laws, is the future already fixed?

The next essay is:

Determinism.