When Help Becomes Control

Expertise can justify advice. It does not automatically create authority.

There is a strange problem built into almost every institution designed to help people.

The better the helper becomes at helping, the easier it becomes for the helper to believe they should decide.

A teacher understands the subject better than the student. A physician understands medicine better than the patient. An organizer understands labor strategy better than a worker who has never organized. A social worker understands institutional systems better than a client trying to navigate them. A manager may understand regulatory obligations better than the employee performing the work.

None of those knowledge differences are inherently problematic. In many cases, they are exactly why mediation is useful.

The problem begins with a subtle inference:

I know more about this problem than you do.

therefore

I can probably make a better decision than you can.

therefore

I should be the one making the decision.

The first statement concerns knowledge.

The second concerns judgment.

The third concerns authority.

Those are not the same thing.

Confusing them is one of the ways help becomes control.

The basic structure of mediation

At its simplest, mediation can be represented as:

Actor₀ → Mediator → Actor₁

The actor begins in one state. The mediator introduces something the actor does not currently possess: information, coordination, resources, expertise, access, training, protection, representation, or infrastructure. The actor then emerges in a different state.

The intuitive ideal is:

Capability(Actor₁) > Capability(Actor₀)

The person is more capable after receiving help than before.

But capability alone cannot tell us whether a transformation was legitimate.

A person can emerge healthier, wealthier, safer, or better educated from a process they never authorized. An institution can produce a desirable outcome while still exercising authority it never legitimately possessed.

Better outcome ≠ legitimate transformation

That distinction matters. If mediation is judged only by outcomes, almost any sufficiently benevolent form of control can justify itself after the fact.

The harder question is not merely:

Did the intervention help?

It is:

What happened to the agency of the person being helped?

Paternalism is a particular kind of control

Philosophers disagree about the exact boundaries of paternalism, but a common core is recognizable: one actor interferes with another person's liberty or autonomy without that person's consent because the intervening actor believes the intervention will benefit that person. The Stanford Encyclopedia of Philosophy surveys this structure and the long-running distinctions between hard and soft, strong and weak, and pure and impure forms of paternalism.

John Stuart Mill's On Liberty remains one of the classic objections to adult paternalism. His harm principle draws a sharp distinction between preventing harm to others and coercing a competent adult merely for that adult's own supposed good. Later theorists such as Gerald Dworkin and Joel Feinberg complicated the picture by examining capacity, voluntariness, information, and the conditions under which limited intervention may be justified.

That distinction is important.

Stopping someone who is about to act on a serious factual misunderstanding is not structurally identical to overriding a competent person who understands the relevant consequences and still chooses differently.

Children can require protection. Unconscious patients can require emergency treatment. People can temporarily lose decision-making capacity. One person's action can threaten others. Communities can legitimately delegate authority to institutions.

The relevant question therefore cannot simply be:

Was someone's choice restricted?

It has to include:

Was the restriction sufficiently authorized,
necessary, proportionate, bounded, and accountable?

There is another useful distinction. Not every illegitimate form of control is paternalistic. Someone can control another person for money, status, obedience, efficiency, ideology, or power. Paternalism is narrower: it is control defended by reference to the welfare or good of the person being controlled.

Paternalism ⊂ Control

That is precisely why paternalism can be difficult to recognize. Its justification is often benevolent.

Agency is more than having choices

One weakness in simplistic accounts of autonomy is the assumption that freedom means merely having more options.

Meaningful agency requires more than a menu.

The capability approach, associated especially with Amartya Sen and Martha Nussbaum, distinguishes formal freedom from substantive opportunity. A person may nominally possess a right while lacking the information, health, resources, access, or social conditions necessary to exercise it.

That gives us two different distinctions:

Choice ≠ Capability

Capability ≠ Authority

A person may need substantial assistance in order to become capable of exercising meaningful freedom. That assistance is not necessarily a threat to autonomy. It may be what makes autonomy possible.

Self-Determination Theory reaches a related conclusion from psychology. The theory identifies autonomy, competence, and relatedness as basic psychological needs associated with healthy functioning. Its research program distinguishes autonomy-supportive environments from controlling ones and examines how those environments affect motivation and behavior. The theory's own overview is available through the Self-Determination Theory research site.

The important point is not that people should operate in isolation.

Human agency develops through relationships.

Parents teach children. Teachers teach students. Mentors guide apprentices. Experts advise nonexperts. Communities coordinate people who could accomplish very little individually.

The alternative to paternalism is therefore not radical independence.

It is supported agency.

The same structural question appears across very different fields

What makes this problem interesting is how often distinct traditions encounter the same structural tension under different vocabulary.

Labor organizing distinguishes worker self-organization from models where professional staff primarily solve problems for members.

Social work explicitly recognizes client self-determination. The National Association of Social Workers Code of Ethics instructs social workers to respect and promote clients' self-determination while also recognizing narrow circumstances in which serious, foreseeable, and imminent risk can justify limitation.

Medicine speaks of patient autonomy and shared decision-making.

Disability-rights movements condensed a similar demand into the phrase Nothing About Us Without Us, insisting that people affected by policies and institutions participate meaningfully in the decisions that shape their lives. The broader legal shift toward supported rather than substituted decision-making is reflected in the Convention on the Rights of Persons with Disabilities.

Development economics speaks of capabilities.

Political theory speaks of participation and subsidiarity.

Elinor Ostrom's work on commons governance documented how communities can create durable institutions for collective resource management rather than fitting every problem into a simple centralized-control-versus-private-property binary. Her work was recognized by the Nobel Prize in Economic Sciences.

These traditions are not interchangeable and should not be treated as though they derive from one intellectual source.

But they repeatedly encounter a similar question:

Does the institution enable people to act,
or does it increasingly act instead of them?

Labor organizing: representation or substitution?

Labor organizing gives a particularly clean version of the problem.

One distinction in union practice contrasts a service model with an organizing model. In simplified form, the service model treats the organization as a provider that handles problems for members. The organizing model attempts to develop member participation and collective capacity so workers themselves become active agents in solving workplace problems.

Worker₀
  ↓
resources + knowledge + coordination
  ↓
Worker₁

The worker remains the principal actor.

But expertise accumulates. The organizer has seen more campaigns. The representative understands the law. Staff understand institutional procedure.

And then a dangerous inference becomes possible:

Knowledge(Mediator) > Knowledge(Principal)

DOES NOT IMPLY

Authority(Mediator) > Authority(Principal)

An organizer can know more about organizing without thereby acquiring ownership of the workers' goals.

Education has the same problem

Paulo Freire's critique of the banking model of education is useful here. In Pedagogy of the Oppressed, he describes a relationship in which knowledge belongs to the teacher while students are treated primarily as recipients. His alternative emphasizes dialogue and the development of learners as participants in the production of understanding.

Whatever one thinks of Freire's broader political commitments, the structural distinction is valuable:

Teacher → Answer → Student

versus

Teacher → Capability → Student

The purpose of teaching cannot simply be to make a learner capable of reproducing the teacher's conclusions.

A successful teacher should be able to produce a student capable of disagreeing with the teacher intelligently.

Otherwise the teacher may have transmitted compliance rather than judgment.

Medicine exposes the difference between means and ends

Medicine may provide the clearest example because the knowledge asymmetry can be enormous.

A physician may know vastly more about pathology, pharmacology, prognosis, treatment risk, and surgical technique than the patient.

But medical expertise does not automatically determine what the patient values.

A physician may know which intervention maximizes expected survival. The patient may care about mobility, pain, cognition, fertility, independence, time at home, or quality of life in ways that cannot be derived from medical knowledge alone.

That produces one of the most important distinctions in the framework:

Expertise about means ≠ authority over ends

The expert may possess extraordinary knowledge about how to pursue a goal without possessing the authority to determine which goal another person must pursue.

Modern informed-consent and shared-decision-making models developed in part through recognition of this boundary. The clinician contributes expertise. The patient contributes values and preferences. The decision is not reducible to either one alone when the patient has decision-making capacity.

"Nothing About Us Without Us"

Disability-rights movements make the principle even more explicit.

Nothing About Us Without Us rejects the assumption that professionals, guardians, institutions, or policymakers may define the good of disabled people while excluding disabled people from meaningful participation in the decisions themselves.

The problem is not expertise.

Expertise is often necessary.

The problem is assuming that expertise itself confers authority to substitute one person's judgment for another person's life.

This is the structural difference between substituted and supported decision-making:

Substituted decision-making:
What should I decide for this person?

Supported decision-making:
What does this person need in order to make
or communicate their own decision?

The mediator remains.

The direction of mediation changes.

Subsidiarity: help without absorption

A related formulation appears in the principle of subsidiarity. Catholic social teaching describes higher-order communities as having a role in supporting and coordinating lower-order communities without unnecessarily depriving them of functions they can perform themselves. The principle is summarized in the Catechism of the Catholic Church, including its treatment of subsidiarity around paragraph 1883.

Structurally:

Higher capacity
  ↓
support
  ↓
Lower capacity strengthened

rather than:

Higher capacity
  ↓
absorption of lower authority

This does not imply that every problem should be solved locally. Some problems genuinely require larger institutions.

The narrower principle is that greater capacity alone is not sufficient reason to remove competent agency from a lower-order actor.

Participation is not the same as power

Sherry Arnstein's 1969 essay A Ladder of Citizen Participation gives us another useful distinction.

Arnstein separated forms of nominal participation from arrangements in which citizens possess meaningful decision-making power. Informing and consultation may be valuable, but they are not identical to partnership, delegated power, or citizen control.

That reveals an uncomfortable truth:

Voice ≠ Decision Rights

An institution can listen to people without allowing those people to govern anything.

A survey is not authority. A listening session is not authority. A comment box is not authority.

Whenever an institution claims that people "participated" in a decision, the useful follow-up is:

What could their participation actually change?

The critical transition: from expertise to authority

Across these domains, the same transition repeatedly appears.

The mediator begins by saying:

I can help you understand the problem.

Then:

I can help you evaluate the available options.

Then:

I think one option is clearly better.

None of those statements necessarily threatens agency.

The critical transition is:

Because I know which option is better,
I should determine which option you choose.

The transformation is:

Expertise → Recommendation → Claimed Authority

The first transition is ordinary mediation.

The second requires justification.

The central mistake is treating epistemic superiority as though it automatically creates normative authority.

The Agency Substitution Test

That suggests a practical diagnostic.

Call it the Agency Substitution Test:

Has the mediator's judgment begun replacing
judgment that properly belongs to the principal?

The test becomes especially important when the original actor is competent, sufficiently informed, acting voluntarily, and capable of understanding the relevant consequences.

If those conditions hold, replacing the actor's judgment requires an independent justification.

Emergency may provide one. Temporary incapacity may provide one. Protection of third parties may provide one. Explicit delegation, a previously agreed rule, or legitimate institutional authority may provide one.

But this does not:

I know better.

A compact working formulation is:

Paternalistic Substitution
=
Judgment Replacement
+ Welfare Justification
- Sufficient Authorization

That is not offered as a complete philosophical definition of paternalism. It is a diagnostic for recognizing the transformation.

The Goal Ownership Test

There is another question that may be even more revealing:

Who owns the goal?

Consider this structure:

Principal → Goal
Mediator → Means

That is ordinary expertise.

But if the relationship quietly becomes:

Mediator → Goal
Principal → Compliance

the structure has changed.

The Goal Ownership Test asks:

Is the mediator helping the person pursue the person's goals,
or redefining those goals on the person's behalf?

This does not mean every requested goal must be accommodated. Institutions have legitimate purposes and boundaries. A surgeon is not required to perform a medically inappropriate procedure because a patient requests it. An employer does not surrender responsibility for workplace safety because an employee prefers a dangerous method. A government does not lose the ability to restrain conduct that threatens other people.

Those are independent limitations.

They should not be confused with the proposition that expertise itself grants ownership of another person's good.

The Mediator Dependency Test

A tempting principle is that a good mediator should eventually become unnecessary.

That is too strong.

Some mediation remains permanently useful because coordination itself is valuable. Courts remain useful. Teachers remain useful. Doctors remain useful. Infrastructure remains useful. Unions remain useful.

The better question is:

Has the mediator made itself more necessary
than the underlying function actually requires?

Call this the Mediator Dependency Test.

A healthy institution can remain permanently useful without making the people it serves permanently incapable.

The danger is not continuing mediation.

It is manufactured dependence.

Historical warning: benevolence can legitimize domination

History provides extreme cases where institutions described domination in the language of education, civilization, protection, or improvement.

The federal Indian boarding-school system in the United States is one of the clearest examples. Native children were separated from families and communities and subjected to assimilation policies intended to replace Indigenous language, culture, and identity with the dominant society's preferred model of life. The U.S. Department of the Interior's modern investigation of the Federal Indian Boarding School Initiative documents the scope and consequences of that system.

This is an extreme case and should not be casually equated with ordinary education, workplace policy, medical advice, or social services.

But extreme cases make one structural point unmistakable:

Benevolent intention ≠ legitimate authority

A person or institution can sincerely believe it is helping while exercising authority it has no right to possess.

Good intentions are not enough

This is why paternalism cannot be diagnosed simply by asking whether the mediator cares.

The parent may love the child. The doctor may care about the patient. The organizer may care deeply about workers. The social worker may genuinely want the client to thrive. The institution may genuinely believe its policies protect people.

Intent matters morally.

But structurally:

Benevolent Intent ≠ Legitimate Authority

A legitimate system therefore needs more than goodwill.

It needs boundaries that can be inspected.

A practical boundary test

No single criterion settles every case. But together the following questions reveal the direction of the relationship.

1. Authorization

Did the affected person authorize the mediator to perform this function?

2. Scope

Is the mediator staying within the authority actually delegated?

3. Competence

Can the principal reasonably make the relevant decision?

4. Information

Does the principal understand the important options and consequences?

5. Voluntariness

Can the principal meaningfully disagree, refuse, or choose another course?

6. Necessity

Is intervention actually required, or is substitution merely easier for the mediator?

7. Proportionality

Is the intervention broader or more restrictive than the problem requires?

8. Goal ownership

Who determines the desired end state?

9. Contestability

Can the mediator's judgment be challenged, reviewed, or appealed?

10. Transparency

Can the affected person understand the relevant rule, reason, and decision process?

11. Capability transfer

Is the intervention increasing the principal's future ability to understand, decide, or act?

12. Dependency

Is the mediator becoming more necessary than the underlying function requires?

13. Reversibility

Can transferred authority return when the justification for that transfer ends?

The point is not to turn moral life into a checklist. The point is to make authority visible.

The deeper principle

The research ultimately forces a refinement of the simple claim:

Good mediation increases agency.

That is useful, but incomplete.

A system can increase someone's objective capability while still exercising authority illegitimately.

A stronger formulation is:

Good Mediation
=
Capability Expansion
+ Legitimate Authority
+ Principal Authorship

The mediator may transform information, resources, coordination, access, skills, options, protection, and means.

But absent an independently justified transfer of authority, the mediator should not silently transform:

Principal's Goal → Mediator's Goal

or:

Principal's Judgment → Mediator's Judgment

while continuing to describe the principal as the actor.

That is the structural deformation.

The person has not merely been helped.

The person has been replaced.

So when does help become control?

The answer is not simply when someone loses a choice.

It is not whenever an institution exercises authority.

It is not whenever a mediator remains necessary.

The boundary is crossed when the mediator's role changes from enabling another actor's legitimate action to appropriating decision-making that properly belongs to that actor, without sufficient authorization, necessity, proportionality, or accountability.

"I can help you act."

becomes

"I can act better than you."

becomes

"Therefore I may act instead of you."

The first proposition concerns capability.

The second concerns judgment.

The third claims authority.

That final transformation is the one worth learning to notice.

Because the most durable forms of paternalism rarely introduce themselves as control.

They introduce themselves as help.

References and further reading

1. John Stuart Mill, On Liberty (1859).
2. Gerald Dworkin, "Paternalism," The Monist 56, no. 1 (1972).
3. Joel Feinberg, Harm to Self, Volume III of The Moral Limits of the Criminal Law (Oxford University Press, 1986).
4. Stanford Encyclopedia of Philosophy, "Paternalism".
5. Edward L. Deci and Richard M. Ryan, "The 'What' and 'Why' of Goal Pursuits: Human Needs and the Self-Determination of Behavior," Psychological Inquiry 11, no. 4 (2000).
6. Self-Determination Theory: Theory Overview.
7. Amartya Sen, Development as Freedom (Oxford University Press, 1999).
8. Martha C. Nussbaum, Creating Capabilities: The Human Development Approach (Harvard University Press, 2011).
9. Stanford Encyclopedia of Philosophy, "The Capability Approach".
10. Paulo Freire, Pedagogy of the Oppressed (English edition, 1970).
11. Sherry R. Arnstein, "A Ladder of Citizen Participation", Journal of the American Institute of Planners 35, no. 4 (1969): 216–224.
12. Elinor Ostrom, Governing the Commons: The Evolution of Institutions for Collective Action (Cambridge University Press, 1990).
13. James I. Charlton, Nothing About Us Without Us: Disability Oppression and Empowerment (University of California Press, 1998).
14. United Nations, Convention on the Rights of Persons with Disabilities.
15. National Association of Social Workers, Code of Ethics, Standard 1.02: Self-Determination.
16. Catechism of the Catholic Church, treatment of subsidiarity including paragraph 1883.
17. U.S. Department of the Interior, Federal Indian Boarding School Initiative.