You Do Not Need to Be Uncaused to Be a Cause

Determinism, agency, responsibility, and the Struggler inside causality

Argument note · physicalism, fatalism, mental causation, metacognition, moral desert, causal participation, and Berserk

Andraž Đurič

Version 1 · July 2026

Abstract

This note targets two nihilistic reactions sometimes built on determinism. The first is paralysis: if every action is determined, effort and deliberation are pointless. The second is license: if every action is determined, nothing is genuinely wrong and one may as well do anything.

Neither conclusion follows from determinism. Determinism is a thesis about how events arise. It does not remove deliberation, discipline, resistance, affection, or action from the causal process. Those processes may be determined and remain among the determinants of what happens next. Fatalism says that an outcome occurs regardless of what an agent does. Determinism says that what the agent does is part of the history through which the outcome occurs.

Two corrections precede the argument. Physicalism does not logically entail determinism: a wholly physical world may contain deterministic or stochastic dynamics. Indeterminism would not by itself restore libertarian freedom, because randomness is not authorship. The argument therefore proceeds conditionally: suppose determinism is true.

The central result is deliberately narrower than a solution to the free-will problem. A caused deliberative process can remain causally effective. This remains true whether one adopts compatibilism, doubts basic-desert responsibility, reduces deliberation to neural and bodily organization, or leaves the strongest ownership question unresolved. To deny the result requires more than determinism. It requires an additional thesis under which the relevant control processes are causally idle, unreal, or unable to make a difference.

The note also develops a recursive consequence. Beliefs about determinism are themselves causes. A fatalistic model can help produce withdrawal. A more accurate model can enter memory, attention, inhibition, and planning. Metacognition is treated, cautiously, as a physically realized monitoring-and-control relation through which an agent represents part of its own activity and feeds that representation back into what happens next. A text that teaches causal participation may therefore become one of the causes through which participation is strengthened.

Berserk supplies the artistic centre. Its supernatural causality is not identical to causal determinism in contemporary metaphysics, and Guts is not evidence in a philosophical proof. He is an embodied model of constrained agency: a being formed by forces he did not choose whose formed capacities become forces in turn. The Struggler does not escape causality. He struggles from within it.

1. The target

A person encounters determinism and reacts in one of two directions.

Everything is predetermined. I have no free will. Nothing I do matters.

Or:

Everything is predetermined. I am not ultimately responsible. I may as well do whatever I want.

The reactions look opposed. One produces resignation. The other produces liberation or recklessness.

Structurally, they make the same mistake.

They move from a descriptive claim about causal provenance to a practical or evaluative conclusion without supplying the bridge.

Determinism, in its standard causal form, says that the complete state of the world at one time, together with the laws of nature, fixes what follows. It does not by definition say that:

agent-level processes are unreal;
deliberation has no effects;
all trajectories are equally good;
harm ceases to be harm;
reasons cease to distinguish actions;
or no response to dangerous conduct is justified.

Each additional conclusion requires an argument of its own.

2. Map of the position

The whole note can be placed in front of the reader before the details begin.

Physicalism is the standing ontology, not a proved premise. Persons, thoughts, experiences, values, and decisions are treated as physically realized.
Determinism is considered conditionally. The note does not establish that our universe is deterministic.
Determinism is not fatalism. Determinism includes an agent's activity in the route to an outcome; fatalism bypasses it.
A caused process can remain a cause. Causal provenance does not erase causal efficacy.
Human agency is treated as organized causal control. Perception, memory, modeling, valuation, planning, inhibition, learning, and action can be physically implemented capacities.
Represented reasons can affect behaviour. This does not settle what objective normative reasons are or guarantee that reported reasons are accurate.
Indeterminism does not automatically create authorship. Randomness is not equivalent to control.
Determinism does not entail nihilism. Facts about causation do not by themselves negate meaning, value, or practical reasons.
Determinism does not erase harm. A determined injury still injures.
Ultimate self-authorship is doubtful. No agent selected the original materials from which its character and decision system developed.
Basic-desert responsibility is therefore contestable. The note does not assume that anyone deserves suffering merely for having ultimately created themselves as they are.
Forward-looking practices remain available in principle. Protection, explanation, restitution, rehabilitation, and behavioural change are not eliminated by determinism, although their moral justification requires further constraints.
Compatibilists and free-will skeptics share substantial ground. They can disagree about desert while agreeing that deliberation and practical intervention remain real.
Higher-level explanation is treated as tracking physically realized organization. The core argument does not require deliberation to be ontologically irreducible.
Berserk illustrates constrained agency. Guts is an artistic model, not empirical evidence or a metaphysical proof.
Central thesis: you do not need to be uncaused in order to be a cause.

This is less a defense of one complete school than an attempt to identify a point at which several schools converge.

3. Keep the theses separate

Four claims are routinely collapsed.

Physicalism

Whatever persons, thoughts, values, decisions, institutions, and experiences are, they are physically instantiated or physically realized.

Determinism

Given a complete physical state and the laws, only one later history is physically possible.

Fatalism

Some outcome will occur regardless of what an agent does, or agents are powerless to shape what occurs.

Nihilism

No value, meaning, practical reason, or norm exists in the relevant sense.

These are not synonyms.

Physicalism concerns what reality contains. Determinism concerns how one state constrains another. Fatalism concerns whether action is bypassed. Nihilism concerns value or meaning. Moving between them requires bridges.

The note does not attempt to prove physicalism or determinism. It asks what follows if a physicalist grants determinism for argument.

4. Physicalism is not determinism

A physical world need not be deterministic merely because it is physical.

Physicalism is an ontological thesis. It says that reality contains no independently operating nonphysical substance or domain of the relevant kind. It does not settle whether physical dynamics are deterministic, stochastic, branching, or otherwise structured.

Quantum theory makes the logical separation concrete. Everett's relative-state formulation and Bohmian mechanics supply deterministic physical frameworks. Objective-collapse theories such as the Ghirardi–Rimini–Weber model introduce stochastic dynamics. The success or failure of those interpretations is not decided here. Their coexistence is enough to show that physicalism by itself does not select determinism.

Physicalism removes the supernatural exemption. It does not automatically write one unique future into the laws.

The reverse correction matters as much. Indeterminism does not automatically create freedom.

If a decision contains an uncontrolled random element, that element is not thereby authored by the agent. A coin flip inside a nervous system would introduce unpredictability, not ultimate self-creation. This is why hard incompatibilists treat determination and luck as different threats to the same strong conception of authorship.

The argument below therefore does not depend on whether our universe is in fact deterministic. It is conditional:

Suppose determinism is true. Does fatalism or nihilism follow?

5. Determinism is not fatalism

The most important distinction is simple.

Fatalism says:

The outcome occurs whether or not I act.

Determinism says:

My action, including my deliberation and effort, arises from prior conditions and helps produce the outcome.

These claims are not equivalent.

Under determinism, studying can cause passing. Neglect can cause failure. Training can alter performance. Information can alter a decision. Kindness can reduce suffering. Recklessness can injure another person. A warning can prevent an accident. An institution can reshape the incentives and dispositions that generate future conduct.

All of those causes may themselves have causes.

That does not make them absent.

A caused cause is still a cause.

The metaphor of a finished film can mislead here. Determinism is not the claim that the future sits elsewhere already completed while present agents watch it arrive. It is a relation among states and laws. The later state occurs through the processes that generate it, including whatever agents actually perceive, infer, decide, and do.

The future does not happen instead of the causal route.

The route is how the future happens.

6. Chrysippus and the Lazy Argument

The error is ancient.

Cicero records the so-called Lazy or Idle Argument:

If it is fated that I recover, I will recover whether I call a doctor or not.
If it is fated that I do not recover, calling a doctor will not help.
Either recovery or non-recovery is fated.
Therefore, calling a doctor is pointless.

Chrysippus's reply is that some outcomes and actions are co-fated. Recovery may be fated together with calling the doctor, receiving treatment, and the treatment's physiological effects. The argument becomes invalid when it deletes those conditions while pretending to preserve the same outcome.

The doctor is not an optional ornament attached to the outcome. The doctor may be one of the causes through which the outcome occurs.

The modern nihilistic version has the same structure:

If success is determined, I will succeed without effort.
If failure is determined, effort cannot help.
Therefore, effort is pointless.

The error is identical. It removes the causal pathway from the history it is supposedly describing.

Under determinism, success can be determined through effort and failure through its absence. The effort is not an optional decoration added after the causal story is complete. It is inside the story.

7. The core result: provenance does not erase efficacy

The narrowest argument can be stated directly.

Determinism says that an agent's deliberation and action have sufficient antecedent causes.
Having antecedent causes does not entail lacking downstream effects.
An agent's deliberation and action can therefore be both determined and causally efficacious.
So the inference from determined to causally irrelevant is invalid.

The distinction is between where a process came from and what the process does.

A river is caused by topography, precipitation, gravity, and prior hydrological conditions. It still erodes stone.

A fire is caused by fuel, oxygen, heat, and ignition. It still burns.

A nervous system is caused by development, inheritance, learning, injury, culture, and the present environment. It still integrates information, inhibits some responses, strengthens others, predicts, plans, and acts.

To explain the causes of agency is not to subtract agency from the explanation. It is to explain how agency exists.

The conclusion is modest but hard to dislodge. A critic may reject the vocabulary of agency, deny basic desert, or insist that every high-level description reduces to lower-level physics. The critic must still answer the causal question: does the physically realized control process make a difference to what happens?

If it does, determinism has not produced fatalism.

If it does not, an additional thesis is doing the work.

8. What counts as an agent here

The word agent carries a large unresolved literature. This note should not hide that fact.

For present purposes, the human case is enough. I use agent for a sufficiently integrated, physically realized control system capable of:

receiving information from an environment and its own internal state;
retaining state through memory or other persistent organization;
tracking possible outcomes;
evaluating those outcomes relative to operative ends or norms;
selecting and executing actions;
and modifying later behaviour through learning, reflection, or reorganization.

This is an operational profile, not a universal set of necessary and sufficient conditions.

Barandiaran, Di Paolo, and Rohde propose a more demanding account centered on individuality, interactional asymmetry, and normativity. Bratman's planning theory emphasizes temporally extended intentions and partial plans. Other traditions use thinner behavioural, computational, biological, or social criteria.

Agency-boundary firewall. This note does not determine whether thermostats, simple organisms, octopuses, reinforcement-learning systems, language models, corporations, or other collectives count as agents, nor whether agency is binary or graded. It fixes the ordinary adult human case and argues from capacities plainly present there.

The central result does not require a final universal definition. It requires only that some physically realized human control processes affect behaviour. Denying that would be a much stronger claim than determinism.

9. Reasons, representations, and causes

The sentence reasons affect behaviour can name several different claims.

Represented or motivating reasons

Considerations encoded or otherwise carried within the agent and involved in producing action: the bridge is unstable, this person will be harmed, I promised, this plan conflicts with another commitment.

Normative reasons

Considerations that genuinely count in favour of an action, whether or not the agent recognizes them.

Reported reasons

What the agent later says motivated the act. Such reports may be accurate, partial, reconstructed, or confabulated.

The anti-fatalist argument needs only the first claim.

Donald Davidson defended the view that a reason can rationalize an action by causally explaining it. Fischer and Ravizza distinguish recognition of reasons from reaction to them within a reasons-responsive mechanism. Cognitive science describes goal representations and control signals that alter which mappings from inputs and internal states to outputs are selected.

Reasons do not need to float outside physics to alter behaviour. They need to be represented in a system whose behaviour is sensitive to them.

Reasons firewall. The claim concerns physically realized representations and reasons-responsive control. It does not assume that introspection is transparent, that every stated reason caused the act, that post-hoc rationalization is rare, or that the metaethical status of objective normative reasons has been settled.

A determined system can be sensitive to evidence. It can also be insensitive, biased, compulsive, manipulated, or mistaken. Determination settles none of those distinctions.

10. Deliberation and changing the conditions

Someone may object:

But if the exact past and laws were the same, I could not have deliberated or acted differently.

Granted, under determinism.

It still does not follow that deliberation performs no function.

Two claims must be kept apart:

With every cause held exactly identical, a different result would not occur.
Changing relevant causes cannot change the result.

The first belongs to determinism.

The second contradicts the ordinary meaning of causation.

A person who learns that a bridge is unstable is not in the same informational state as a person who does not know. A person who has trained inhibition for a year is not the same control system as before the training. A person who has integrated a convincing argument carries a different memory, distinction, and expectation into the next decision.

You do not change an outcome by somehow acting differently under identical total conditions. You change future outcomes by becoming, encountering, or constructing different conditions.

Determinism does not make intervention pointless. It explains why intervention must alter causes.

Education, therapy, practice, sleep, medication, institutional design, friendship, warnings, incentives, and argument are all attempts to modify the conditions from which later behaviour arises.

They may fail.

They may have unintended effects.

They do not become causally idle because they too have causes.

11. Metacognition and discipline

The same structure appears inside the agent.

A simple trajectory may look like:

anger → insult

A more reflective trajectory may look like:

anger → recognition of anger → evaluation → inhibition → another response

No supernatural interruption has entered the sequence.

Flavell's early account of metacognition, the monitoring-and-control framework developed by Nelson and Narens, and later experimental work all distinguish processes that track cognitive activity from processes that regulate it. Cognitive-control theories likewise describe maintained goal representations biasing the flow of neural activity toward task-relevant mappings.

Metacognition is not the mind stepping outside causality. It is a causal system representing part of its own operation and feeding that representation back into what happens next.

This is a functional interpretation, not a claim that introspection reveals the whole mechanism. Human self-models are incomplete. Monitoring can be inaccurate. A feeling of confidence may follow effort rather than accurately assess competence. Rationalizations can be constructed after the act.

But imperfect monitoring can still matter.

Noticing that one is escalating, rehearsing an if–then plan, changing the environment, or strengthening a competing habit can alter the later process. Gollwitzer's work on implementation intentions is an especially concrete example: a representation linking a future cue to a selected response can reorganize how action is initiated when the cue occurs.

Discipline, on this picture, is not a metaphysical power to originate oneself from nothing.

It is the construction and stabilization of causal structures that make some later actions more likely and others less likely.

12. “That reaction was determined too”

Another reply says:

My nihilism, resignation, or recklessness is determined too.

Under the hypothesis, yes.

But the sentence changes nothing about whether the reaction is true, coherent, useful, harmful, or corrigible.

A delusion can be determined.

A mathematical error can be determined.

A correction can be determined.

An accurate model can be determined.

A self-destructive impulse and the intervention that interrupts it can both be determined.

Causal determination is not epistemic endorsement.

The fact that a belief has causes does not establish its truth. The fact that a reaction has causes does not establish that it realizes the agent's ends. Saying this response was determined supplies an explanation of its occurrence. It does not supply a reason to preserve it.

The same applies reflexively to this note. It too has causes. If its argument is sound, those causes do not make it unsound. If it changes a reader's model, that effect does not become unreal by being determined.

13. Two nihilistic reactions

Paralysis

The first response is:

If the future is determined, effort cannot change it.

This is the fatalist error. Effort may be one of the causes on which the future depends.

A determined agent can remain sensitive to outcomes, update models, invest effort, and alter local conditions. The causal history does not happen around the agent while leaving the agent inert. It runs through the agent's capacities.

License

The second response is:

If both discipline and recklessness are determined, there is no reason to prefer either.

This introduces another missing premise:

Trajectories with the same deterministic status are normatively equivalent.

Determinism does not establish that premise.

A trajectory containing severe suffering differs from one without it.

A trajectory in which an agent destroys a valued relationship differs from one in which the relationship is protected.

A trajectory containing health, understanding, competence, affection, and rich experience differs from one containing preventable degradation, confusion, isolation, and injury.

The histories may both be determined. They are not therefore identical in content, consequence, or relation to the values of the agents inside them.

Equal causal provenance does not imply equal value.

14. The value firewall

A nihilist can answer:

You are assuming that value exists.

The objection is correct against any claim that this note has proved objective value.

It has not.

The present argument is defensive and conditional:

Determinism supplies no bridge from “actions are caused” to “nothing has value.”

A critic may independently defend moral error theory, global evaluative nihilism, or the absence of stance-independent value. That argument must stand on its own. Determinism does not complete it.

The conditional point survives across theories of value.

If pain is bad for a conscious subject, determinism does not make it painless.
If an agent values a project, determinism does not make that valuation absent.
If a norm is justified, determinism does not invalidate it merely by explaining its causal history.
If no value exists in any sense, that conclusion requires more than determinism.

Normative firewall. This note does not derive categorical value from physics, causal efficacy, or agency. It blocks one inference: determination does not itself entail valuelessness or normative equivalence.

15. Harm, responsibility, and desert

The sentence I do no wrong because I could not have done otherwise can negate several different relations.

Causal attribution: did the act arise through this person's system?
Harm: did it injure, deprive, deceive, or endanger someone?
Norm violation: did it violate a justified rule, claim, or obligation?
Agential ownership: did it issue through capacities appropriately attributable to the person rather than through direct seizure, coercive bypass, or severe impairment?
Answerability: may the person appropriately be asked to explain, acknowledge, repair, or revise?
Accountability: are blame, sanction, restraint, or other reactive responses fitting?
Basic desert: does the person deserve blame or suffering simply because of ultimate authorship, independently of future protection, repair, deterrence, or formation?

Determinism does not automatically erase the first three.

It leaves the fourth through sixth deeply contested.

It places serious pressure on the seventh.

A determined assault can still cause pain.

A determined lie can still corrupt another person's map.

A determined betrayal can still destroy trust.

A determined dangerous disposition can still predict future danger.

Those facts can support protection, restraint, treatment, rehabilitation, restitution, environmental reform, or the removal of access to victims. Each response still requires proportionality, evidence, rights constraints, and attention to side effects.

Determinism can weaken vengeance without weakening intervention.

Galen Strawson's Basic Argument presses the self-authorship problem: to be ultimately responsible for what one does, one would need to be ultimately responsible for how one is; but any attempt to make oneself depends on an earlier self not yet made by that act.

Removing ultimate self-creation does not remove the person as the organized process through which the action occurred. It changes what kind of responsibility can be defended and why a response is justified.

16. What makes a decision mine?

This is the deepest unresolved question near the centre of the paper.

Mine can mean:

the event occurred within my organism;
it was generated by my integrated decision machinery;
it reflected beliefs, values, intentions, or commitments with which I identify;
it issued from a reasons-responsive mechanism;
it was not produced through coercive override, seizure, manipulation, compulsion, or incapacitating disorder;
or I ultimately authored the character and mechanism from which it arose.

These are not equivalent.

Frankfurt locates person-level freedom partly in the structure of first- and second-order desires. Watson distinguishes an agent's motivational system from the evaluative standpoint with which the agent identifies. Fischer and Ravizza add reasons-responsiveness and a history through which the agent takes responsibility for the mechanism. Manipulation arguments and free-will skepticism challenge whether those forms of ownership reach ultimate desert.

The causal question and the ownership question must not be collapsed. A process can cause an outcome before we have settled whether it grounds ultimate authorship, moral desert, or the strongest form of personal ownership.

The anti-fatalist result requires less than a final ownership theory.

Even if a decision is not mine in the causa-sui sense, the process occurring through this organism may still alter the future. Causal efficacy can survive while the deeper ownership relation remains disputed.

Responsibility cannot be completed so cheaply. Where explanation becomes excuse, how impairment changes answerability, and which history makes a mechanism genuinely one's own remain open.

17. The compatibilist route

Compatibilists argue that the freedom worth wanting does not require exemption from causality.

Frankfurt attacked the principle that responsibility always requires the ability to have done otherwise. Fischer and Ravizza developed guidance control: action issuing from the agent's own moderately reasons-responsive mechanism. Dennett emphasized evolved capacities such as anticipation, learning, avoidance, self-control, and reflective revision. Bratman explains how intentions and partial plans organize action across time.

On this family of views, a decision may be free in the relevant sense when it issues through the agent's functioning capacities rather than through coercion, compulsion, manipulation, or bypassing.

The exact boundary remains disputed.

How much responsiveness is enough?

What makes a mechanism genuinely the agent's own?

Can a determined history satisfy the historical conditions of ownership?

Does the theory capture moral responsibility or only competent control?

This note does not settle those questions.

The anti-nihilistic result needs less than full compatibilism: it needs only the fact that organized, information-sensitive processes can be causally effective inside a determined world.

18. The skeptical route

Free-will skeptics reject or doubt the control required for basic-desert responsibility.

That need not lead to nihilism.

Derk Pereboom explicitly argues that rational deliberation, agency, relationships, criminal protection, achievement, and meaning can survive skepticism about basic desert. Gregg Caruso develops a non-retributive public-health approach to criminal justice. The wider responsibility literature distinguishes attributability, answerability, accountability, causal responsibility, and basic desert rather than treating responsibility as one indivisible property.

A skeptic can still say:

the act issued through this agent's dispositions and decision system;
the act caused harm;
the agent can answer questions about reasons;
the agent or environment can be modified;
victims may protect themselves;
future conduct can be deterred or prevented;
repair and reconciliation can be pursued;
and some trajectories remain better for conscious beings than others.

What the skeptic denies is that anyone ultimately deserves suffering merely for being the end product of a causal history they did not create.

One can reject metaphysical vengeance without rejecting morality, agency, or consequence.

19. Convergence without settlement

The compatibilist and the skeptic disagree about whether determined agents possess the control required for moral responsibility in the fullest relevant sense.

They can nevertheless converge on the present target.

Determinism does not entail fatalism.
Deliberation can be determined and causally effective.
Represented reasons can alter behaviour through physically realized mechanisms.
Different determined trajectories can differ in value and harm.
Dangerous conduct can justify forward-looking intervention.
Learning, therapy, institutions, education, incentives, and argument remain causal technologies.
Nihilism does not follow merely from the absence of ultimate self-authorship.

This is the strategic advantage of the argument.

We do not need to solve the entire free-will problem in order to defeat the slide from determinism to practical nihilism.

20. Higher levels and the reductionist objection

The note describes persons, reasons, and deliberation at a higher level than neurons, molecules, or fields.

A reductionist may reply:

Deliberation is only a useful shorthand. The lower-level physical process does all the causal work.

The objection opens the mental-causation and causal-exclusion problem. Jaegwon Kim argues that a physically sufficient cause can appear to exclude a distinct higher-level mental cause. Fodor, Davidson, and later nonreductive physicalists defend different forms of autonomy for higher-level explanation. List and Menzies use a difference-making account to argue that higher-level properties can remain causally relevant under conditions where their physical realizers do not screen them off in the required way.

This paper does not solve that dispute.

It does not need to.

The core argument does not depend on deliberation being ontologically irreducible. A reductionist may treat “deliberation” as a compact name for an organized neural and bodily process. So long as that process affects behaviour, causal efficacy remains.

To defeat the core result, the critic needs more than the claim that one description reduces to another.

The critic needs the relevant physically realized control process to be causally idle, nonexistent, or systematically disconnected from behaviour.

That is an epiphenomenalist or eliminativist commitment, not a consequence of determinism itself.

Higher-level realism still matters for the wider Epistemic Forge architecture. Human-level descriptions identify stable organization, counterfactual dependencies, and intervention-relevant regularities that an unstructured microphysical inventory does not transparently display. But the anti-fatalist proof is constructed to survive even if that stronger view remains contested.

21. Belief effects: a caution

It is plausible that some people react to anti-free-will messages with resignation or moral license. That observation should not be inflated into a settled population-level psychological law.

Influential experiments reported that deterministic prompts increased cheating. Later preregistered replications largely failed to reproduce the effect. A meta-analysis of 145 experiments found that anti-free-will manipulations could shift stated beliefs toward determinism but found no reliable downstream behavioural consequences.

The empirical result is therefore restrained:

Deterministic nihilism is a recognizable interpretive response, not an established universal behavioural consequence of learning about determinism.

The recursive claim developed below is equally bounded. Ideas can enter causal histories. It does not follow that every argument reliably changes behaviour, that persuasion is simple, or that this essay will transform every reader.

Bandura's self-efficacy theory and work on implementation intentions support a narrower practical point: beliefs about efficacy and explicitly encoded cue–response plans can influence initiation, effort, persistence, and goal-directed action. The magnitude and direction depend on the person, message, environment, and task.

22. The Struggler inside causality

Berserk gives the abstract structure a body.

Guts does not begin as a sovereign self authoring itself from nowhere.

He is born into conditions he did not choose.

Violence shapes his nervous system.

Training shapes his body.

Betrayal, attachment, grief, rage, responsibility, and repeated survival alter what he can perceive and what he can do.

His capacities are historical products.

They become causes in turn.

Guts does not have to stand outside causality in order to struggle against the forces bearing down on him. His resistance is itself a causal force. His body, rage, learned skill, attachments, decisions, and refusal to collapse all alter the trajectory.

The fact that these features have antecedent causes does not make them absent.

It explains how this particular force came to exist.

Guts is not powerful because he is untouched by circumstance. Much of his power is circumstance metabolized into capacity: injury converted into technique, terror into vigilance, attachment into protection, rage into motion, and memory into direction.

Not every conversion is healthy. Berserk does not romanticize struggle as pure improvement. The same history also produces distortion, isolation, cruelty, obsession, bodily damage, and the danger of becoming continuous with what one resists.

That makes the example philosophically stronger.

Constrained agency is not magic. A system can redirect some of the forces passing through it while being damaged by others. It can gain local control without becoming causa sui. It can struggle without becoming omnipotent.

The Struggler is not an exception to causality. He is causality organized into resistance.

23. What Berserk contributes

The work contributes more than an illustration of persistence.

It repeatedly places agency under pressure from structures larger than the individual: war, hierarchy, ambition, trauma, prophecy, supernatural manipulation, social dependence, bodily limitation, and the accumulated consequences of earlier choices.

Yet the answer is not that the individual therefore disappears.

Guts remains locally consequential.

He cannot guarantee victory.

He cannot erase the past.

He cannot choose the initial conditions of his character.

He cannot simply overpower every structure that constrains him.

He can still protect, fail, learn, injure, repair, persist, and redirect what happens nearby.

Agency need not mean command over the whole field. It can mean becoming a consequential process inside a field one did not choose.

This is good philosophy because it refuses two fantasies at once.

It refuses the fantasy of total sovereignty.

It also refuses the fantasy that anything short of total sovereignty is nothing.

The title the Struggler then carries a precise philosophical reading. A Struggler does not refute causality by leaving it. A Struggler is a constrained structure whose resistance becomes one of the forces determining what follows.

24. The Berserk firewall

Artistic-scope firewall. The supernatural “causality” of Berserk is not identical to causal determinism in contemporary metaphysics. It includes fate-like currents, prophecy, astral structure, manipulation, and narrative uncertainty. Guts is not empirical or logical evidence that determinism is compatible with agency.

The manga is used as an artistic model of constrained agency. The philosophical argument stands or falls on the distinctions among determinism, fatalism, causal efficacy, value, and desert. The artistic section succeeds if it renders that structure intelligible and does justice to the work; it fails if it is asked to prove the metaphysics.

25. Making Strugglers

The argument now turns reflexively.

A belief such as nothing I do matters because everything is determined is not merely a passive description. It can become part of the machinery producing withdrawal.

A different model can enter the same machinery:

My effort has causes, and my effort is one of the causes through which the future occurs.

If the reader integrates that distinction, the reader is not in the same condition as before.

There is now another representation available during despair, temptation, anger, or fatigue.

It may become:

a reason to pause;
a correction to fatalistic self-talk;
a memory available during temptation;
a model of discipline;
an instruction to alter an environment;
or a premise from which a new plan is built.

The text cannot guarantee any of those effects.

But it can participate in their production.

A philosophy of causal agency is not causally neutral. A fatalistic model can help produce resignation. A more accurate model can become part of reflection, inhibition, discipline, and change. The argument that ideas are causes may itself become one.

This is not a paradox.

Under determinism, the reader who encountered the argument and the reader who did not are not in identical total states. The argument becomes part of memory and control. If later conduct differs, it differs because the causes differ.

People also become parts of one another's causal environments.

A parent enters a child's developmental history.

A teacher becomes part of a student's conceptual apparatus.

A friend can become part of another person's emotional regulation.

An author can enter decisions made years after writing.

To speak to another person is sometimes to place a new possible cause inside the environment from which their future actions will emerge.

This claim needs an ethical constraint. Propaganda, coercion, addiction design, and manipulation are causal interventions too. The aim cannot be influence as such.

The ideal is narrower:

Supply truthful distinctions that enlarge another agent's capacity to understand, evaluate, and regulate the processes acting through them.

The purpose is not to tell the reader that they stand outside causality.

It is to place within their causal history a model through which causality may become more reflective, disciplined, and capable of redirecting itself.

In that limited sense, a text can help make a Struggler.

26. Objections and boundaries

Objection: If everything is fixed, genuine alternatives do not exist

Under determinism, alternatives with the exact same total past and laws do not exist. Compatibilists dispute whether that strong alternative possibility is required for agency or responsibility. This note does not settle the dispute. Its narrower claim is that the actual deliberative process can remain causally operative even if only one total history is possible.

Objection: Conscious deliberation may be epiphenomenal

Determinism alone does not entail epiphenomenalism. A physicalist may identify deliberation with, or realize it through, causally active neural and bodily processes. Should phenomenal consciousness itself prove causally inert, ordinary claims about conscious authorship would require revision. The broader control process could still affect behaviour, but the phenomenology-to-action relation would remain an unresolved debt.

Objection: Reasons are only post-hoc stories

Some reasons are confabulated. It does not follow that every information-bearing or goal-representing state is causally idle. Behavioural sensitivity to evidence, incentives, instructions, commitments, and argument is compatible with poor introspective access to the full mechanism.

Objection: Calling determined behaviour “agency” merely changes the definition

That is possible if agency was stipulated in advance to require uncaused self-creation. The cost is that the stipulated property may be incoherent or disconnected from the control capacities actually used in explanation, prediction, medicine, law, learning, and coordination. The dispute then concerns which property deserves the name, not whether the organized capacities exist.

Objection: A determined intervention could not have been omitted

Given the exact total antecedent state, perhaps not. It still follows that the intervention is one of the conditions upon which the result depends. Counterfactual and interventionist causal analysis does not require an event to be metaphysically uncaused in order to count as a difference-maker.

Objection: “Making Strugglers” is manipulation

Any communication can influence. The relevant distinction is between bypassing an agent's evaluative capacities and supplying considerations that the agent can inspect, reject, revise, and integrate. The essay can fail this standard. Its defense is transparency: premises, scope, debts, and kill conditions are placed in view.

27. What remains unresolved

The debts should be stated without euphemism.

This note does not settle:

the universal necessary and sufficient conditions for agency;
whether agency is categorical, graded, biological, computational, social, or some combination;
whether simple machines, artificial systems, or collectives qualify as agents;
the metaphysical status of normative reasons;
whether objective or stance-independent value exists;
the precise relationship between physical description and higher-level ontology;
the causal-exclusion problem for mental properties;
whether phenomenal consciousness itself is causally efficacious;
what constitutes genuine ownership of a decision;
where causal explanation becomes excuse;
how coercion, manipulation, compulsion, ignorance, development, and mental disorder alter responsibility;
which conditions ground attributability, answerability, or accountability;
whether any form of basic-desert responsibility survives;
or which forward-looking interventions are morally proportionate and justified.

These are real philosophical debts. They are not silently solved by the central argument. But neither are they required to establish its narrowest result: caused control processes can remain causes, and determinism alone does not entail fatalism or nihilism.

28. What follows and what does not

The argument supports the following claims:

physicalism does not logically entail determinism;
indeterminism does not automatically create authorship or control;
determinism is distinct from fatalism;
a determined deliberative process can remain causally efficacious;
changing information, training, or environment changes relevant antecedent conditions;
the fact that a response is determined does not validate it;
equally determined trajectories need not be equally valuable;
determinism does not by itself erase harm, norms, or practical reasons;
determinism may place pressure on ultimate self-authorship and basic-desert responsibility;
compatibilists and free-will skeptics can converge against practical nihilism;
the core argument survives a reductionist treatment of deliberation if the realized process remains causally active;
metacognitive monitoring and control can be understood as physically implemented feedback;
and arguments can enter the causal histories through which later action occurs.

The argument does not establish:

that the universe is deterministic;
that physicalism is certainly true;
that compatibilism is the final theory of free will;
that objective value exists;
that every human action is rational, voluntary, owned, or answerable;
that consciousness is causally irreducible or causally efficacious;
that punishment is generally justified;
that every forward-looking intervention is humane or effective;
that reading this essay will reliably improve behaviour;
or that Berserk proves any metaphysical thesis.

29. Placement in the literature

The anti-fatalist core is ancient. Chrysippus's answer to the Lazy Argument already identifies actions and outcomes as co-fated rather than causally independent.

The claim that reasons can be causes belongs near Davidson. The account of planning and temporally organized action belongs near Bratman. Compatibilist ownership and reasons-responsiveness belong near Frankfurt, Watson, Fischer, Ravizza, and Dennett. The challenge from ultimate self-authorship belongs near Nietzsche, Galen Strawson, Pereboom, and contemporary free-will skepticism.

The higher-level issue lies inside the mental-causation and causal-exclusion debate. Kim supplies the central adversarial pressure. Fodor, Davidson, List, and Menzies supply different defenses of higher-level explanatory or causal relevance.

The metacognitive extension belongs near Flavell, Nelson and Narens, Koriat and colleagues, and cognitive-control work such as Miller and Cohen. The claim is not that those literatures prove a complete metaphysics of reflexive agency. They support the monitoring, representation, and control architecture from which the philosophical interpretation is drawn.

The responsibility taxonomy belongs near Watson, Smith, Shoemaker, Scanlon, Fischer and Ravizza, Pereboom, and Caruso. No single taxonomy is treated as final.

The empirical caution about belief in free will belongs to the experimental sequence from Vohs and Schooler to later replication work and the Genschow et al. meta-analysis.

The components are therefore established or strongly prefigured.

The synthesis is personal: separate determinism from fatalism and nihilism; make the core result survive both compatibilism and skepticism; expose the unresolved ownership and mental-causation debts; extend the account through metacognitive self-modification; and embody it through the Struggler inside causality.

This note should not be presented as a proof of compatibilism, a new theory of agency, or an unprecedented discovery that caused causes can cause.

Its contribution is the architecture, compression, scope discipline, recursive extension, and artistic integration.

30. You do not need to be uncaused, compressed

Physicalism does not entail determinism.

Suppose determinism is true anyway.

Then your thoughts, values, habits, deliberations, and actions have causes.

That does not make them causally absent.

Fatalism says the result occurs regardless of what you do.

Determinism says what you do is part of how the result occurs.

You could not produce another result under an exactly identical total state.

You can produce different later results by encountering, creating, or becoming different conditions.

Education is a cause.

Practice is a cause.

Discipline is a cause.

A warning is a cause.

A represented reason is a cause when the system responds to it.

Metacognition does not escape causality. It feeds a partial model of the process back into the process.

Determinism does not prove value. It also does not erase it.

It can weaken ultimate desert without erasing harm, protection, repair, or change.

What makes a decision fully mine remains disputed.

Its causal efficacy does not wait for that dispute to be settled.

Guts did not choose the forces that formed him.

The formed person became a force in turn.

I do not stand outside causality. I am one of its organized forces. I do not need to be uncaused in order to be a cause. The Struggler does not escape the chain. He becomes one of the links through which the chain changes direction.

Standing of this essay

This is a bounded argument note, not a complete theory of free will, agency, mental causation, value, or moral responsibility.

Its strongest result is narrow: determinism states that deliberation and action have antecedent causes. It does not state that those processes lack effects. Fatalism and practical nihilism therefore require additional premises.

The note adopts physicalism as its standing ontology but treats determinism conditionally. It remains compatible with compatibilism and with skepticism about basic-desert responsibility.

The operational account of human agency is sufficient for the target and explicitly incomplete as a universal definition. The account of reasons is restricted to physically realized motivating representations and reasons-responsive control. The account of value blocks an inference and does not derive objective value.

The higher-level causal claim is deliberately non-load-bearing. If deliberation reduces to a lower-level physical process, the argument survives so long as that process affects behaviour. The causal status of phenomenal consciousness remains open.

The post should be corrected if one of the following occurs:

determinism is shown to entail that agent-level or control-level processes cannot be causes;
the distinction between determinism and fatalism collapses under a defensible account of either thesis;
the Lazy Argument can validly remove the action while preserving the same causally dependent outcome;
having antecedent causes is shown, by itself, to entail lacking downstream effects;
all physically realized deliberative and control processes are shown to be causally idle;
equal deterministic status is shown to entail normative equivalence;
determinism alone is shown to entail evaluative nihilism;
the distinction among harm, attribution, ownership, answerability, accountability, and basic desert proves unsustainable;
the metacognitive monitoring-and-control interpretation materially misstates the relevant cognitive science;
or the Berserk interpretation materially misrepresents the work's treatment of Guts, causality, limitation, and struggle.

The unresolved questions listed above are not hidden implementation details. They mark where this note stops.

References

Bandura, A. (1977). “Self-Efficacy: Toward a Unifying Theory of Behavioral Change.” Psychological Review, 84(2), 191–215. — efficacy expectations as influences on initiation, effort, and persistence; used to bound the claim that models of one's causal capacity can alter conduct. Verified · July 2026
Barandiaran, X. E., Di Paolo, E., & Rohde, M. (2009). “Defining Agency: Individuality, Normativity, Asymmetry, and Spatio-temporality in Action.” Adaptive Behavior, 17(5), 367–386. — a demanding naturalized account of agency built around individuality, interactional asymmetry, and normativity; cited to expose rather than conceal the agency-boundary problem. Verified · July 2026
Bohm, D. (1952). “A Suggested Interpretation of the Quantum Theory in Terms of ‘Hidden’ Variables. I.” Physical Review, 85, 166–179. — deterministic physical interpretation used only to demonstrate that physicalism does not logically select indeterminism. Verified · July 2026
Bratman, M. E. (1987). Intention, Plans, and Practical Reason. Harvard University Press / CSLI. — intentions and partial plans as structures organizing practical reasoning and action across time. Verified · July 2026
Caruso, G. D. (2021). Rejecting Retributivism: Free Will, Punishment, and Criminal Justice. Cambridge University Press. — non-retributive criminal-justice model centered on prevention, social conditions, rehabilitation, and public safety rather than basic desert. Verified · July 2026
Cicero. (44 BCE/1992). On Fate, especially §§28–30, in R. W. Sharples (ed. & trans.). Liverpool University Press. — primary surviving source for the Lazy Argument and Chrysippus's co-fated reply. Primary text verified · July 2026
Davidson, D. (1963). “Actions, Reasons, and Causes.” The Journal of Philosophy, 60(23), 685–700. — classic defense of rationalization as a species of causal explanation. Verified · July 2026
Dennett, D. C. (2003). Freedom Evolves. Viking/Penguin. — naturalistic compatibilism: anticipation, learning, avoidance, self-control, and reflective revision as evolved capacities inside the causal order. Verified · July 2026
Everett, H., III. (1957). “‘Relative State’ Formulation of Quantum Mechanics.” Reviews of Modern Physics, 29, 454–462. — deterministic no-collapse physical framework; paired with stochastic collapse work solely to separate physicalism from determinism. Verified · July 2026
Fischer, J. M., & Ravizza, M. (1998). Responsibility and Control: A Theory of Moral Responsibility. Cambridge University Press. — guidance control, moderate reasons-responsiveness, historical ownership, and taking responsibility for a mechanism. Verified · July 2026
Flavell, J. H. (1979). “Metacognition and Cognitive Monitoring: A New Area of Cognitive–Developmental Inquiry.” American Psychologist, 34(10), 906–911. — foundational account of metacognitive knowledge, experience, and monitoring. Verified · July 2026
Fodor, J. A. (1974). “Special Sciences (Or: The Disunity of Science as a Working Hypothesis).” Synthese, 28(2), 97–115. — classic defense of higher-level scientific generalizations against simple reduction. Verified · July 2026
Frankfurt, H. G. (1969). “Alternate Possibilities and Moral Responsibility.” The Journal of Philosophy, 66(23), 829–839. — challenges the claim that responsibility always requires the ability to have done otherwise. Verified · July 2026
Frankfurt, H. G. (1971). “Freedom of the Will and the Concept of a Person.” The Journal of Philosophy, 68(1), 5–20. — hierarchical structure of desires and identification; a major predecessor for one sense of a decision's being the agent's own. Verified · July 2026
Genschow, O., Cracco, E., Schneider, J., Protzko, J., Wisniewski, D., Brass, M., & Schooler, J. W. (2023). “Manipulating Belief in Free Will and Its Downstream Consequences: A Meta-Analysis.” Personality and Social Psychology Review, 27(1), 52–82. — 145 experiments; belief manipulations shifted reported beliefs but produced no reliable downstream behavioural effects. Verified · July 2026
Ghirardi, G. C., Rimini, A., & Weber, T. (1986). “Unified Dynamics for Microscopic and Macroscopic Systems.” Physical Review D, 34, 470–491. — stochastic objective-collapse dynamics; used only to illustrate that physical theories need not be deterministic. Verified · July 2026
Gollwitzer, P. M. (1999). “Implementation Intentions: Strong Effects of Simple Plans.” American Psychologist, 54(7), 493–503. — cue–response plans as physically and psychologically implemented structures that can improve translation of goals into action. Verified · July 2026
Kim, J. (1998). Mind in a Physical World: An Essay on the Mind–Body Problem and Mental Causation. MIT Press. — central causal-exclusion pressure against nonreductive mental causation; treated as a genuine unresolved adversarial boundary. Verified · July 2026
Koriat, A., Ma'ayan, H., & Nussinson, R. (2006). “The Intricate Relationships Between Monitoring and Control in Metacognition.” Journal of Experimental Psychology: General, 135(1), 36–69. — evidence for bidirectional monitoring–control relations and a caution against treating introspective experience as a simple transparent cause. Verified · July 2026
List, C., & Menzies, P. (2009). “Nonreductive Physicalism and the Limits of the Exclusion Principle.” The Journal of Philosophy, 106(9), 475–502. — difference-making account under which some higher-level properties remain causally relevant despite physical realization. Verified · July 2026
Miller, E. K., & Cohen, J. D. (2001). “An Integrative Theory of Prefrontal Cortex Function.” Annual Review of Neuroscience, 24, 167–202. — cognitive control through actively maintained representations of goals and means that bias later processing and action. Verified · July 2026
Miura, K. (1990–2021). Berserk, vols. 1–41. Hakusensha; English edition, Dark Horse Manga. — primary artistic source for Guts, causality, trauma, attachment, bodily limitation, and struggle. The philosophical use in this note is interpretive. Primary work · July 2026
Nelson, T. O., & Narens, L. (1990). “Metamemory: A Theoretical Framework and New Findings.” Psychology of Learning and Motivation, 26, 125–173. — influential monitoring-and-control architecture for metacognition. Verified · July 2026
Pereboom, D. (2014). Free Will, Agency, and Meaning in Life. Oxford University Press. — hard incompatibilism together with rational deliberation, forward-looking responsibility, agency, relationships, and meaning without basic desert. Verified · July 2026
Shoemaker, D. (2011). “Attributability, Answerability, and Accountability: Toward a Wider Theory of Moral Responsibility.” Ethics, 121(3), 602–632. — responsibility as multiple distinguishable relations rather than one undifferentiated property. Verified · July 2026
Strawson, G. (1994). “The Impossibility of Moral Responsibility.” Philosophical Studies, 75(1–2), 5–24. — the Basic Argument: ultimate responsibility would require responsibility for how one is, producing a regress of self-creation. Verified · July 2026
Vohs, K. D., & Schooler, J. W. (2008). “The Value of Believing in Free Will: Encouraging a Belief in Determinism Increases Cheating.” Psychological Science, 19(1), 49–54. — influential initial behavioural report, retained as historical context and bounded by later replication and meta-analytic evidence. Verified · July 2026
Watson, G. (1975). “Free Agency.” The Journal of Philosophy, 72(8), 205–220. — distinction between motivational and evaluative systems in an account of free agency and identification. Verified · July 2026

Verification key: “Verified · July 2026” means the citation and the specific claim attributed to it were checked against a publisher, journal, institutional repository, or primary work during the final literature pass.

Internal lineage

This argument note connects and extends several Epistemic Forge commitments:

The Nihilist's Missing Premise, for separating a descriptive fact from the evaluative bridge and practical reaction;
The Exclusivity Error, for the rule that physical realization does not automatically eliminate higher-level organization;
The Fidelity Program, for finite model-bearing agents acting through corrigible physical processes;
Name the Relata, for indexing freedom, responsibility, wrongness, causation, and value to the relation actually under discussion;
The Is/Ought Firewall and Where the Is–Ought Question Bottoms Out, for keeping descriptive causation distinct from practical conclusions;
Value as Persistence and Persistence Is Not Stasis, for agents as adaptive processes rather than static objects;
Manual for the End of the Universe, especially the principle that a rigged board does not make moves causally idle;
and the standing physicalist levels architecture of Epistemic Forge.

The new contribution is the explicit separation of determinism from fatalism and nihilism; the paired defeat of paralysis and moral license; the distinction among causal efficacy, ownership, answerability, accountability, and basic desert; the construction of a reductionism-resistant narrow core; the metacognitive and recursively causal extension; and the integration of Guts as the Struggler inside causality.

Version history

Version 1, 30 July 2026: Initial publication. Separates physicalism, determinism, fatalism, and nihilism; maps the core commitments before the argument; reconstructs the Lazy Argument and Chrysippus's co-fated reply; argues that causal provenance does not erase causal efficacy; defines the human agency profile while leaving the universal boundary open; distinguishes represented, normative, and reported reasons; develops deliberation as sensitivity to changed conditions and metacognition as monitoring-and-control feedback; separates paralysis from moral license and determinism from value nihilism; distinguishes harm, ownership, answerability, accountability, and basic desert; identifies convergence between compatibilism and free-will skepticism; makes the higher-level causal thesis non-load-bearing; cautiously reviews behavioural evidence; interprets Guts in Berserk as the Struggler inside causality; and develops the recursive possibility that a philosophical text can enter a reader's causal history and help make a Struggler.