Support That Preserves Agency

A manager watches an employee struggle with a decision.

The manager has more experience. She can see a likely mistake forming. She could explain the tradeoffs, offer context, point out something the employee has overlooked.

Or she could simply make the decision.

The second option is often faster.

It can also defeat the point of having someone else responsible for the work.

Management research offers another approach: autonomy-supportive leadership.

Autonomy-supportive managers provide useful information, acknowledge an employee's perspective, invite choice and input where possible, explain the reasons behind constraints, and encourage self-initiation rather than relying primarily on pressure and control.

That does not remove standards, deadlines, or accountability. The manager still contributes experience and judgment. But the employee retains room to understand the situation, weigh the tradeoffs, and act.

This matters because delegation produces more than a completed task. It can also produce someone more capable of exercising judgment the next time.

There is a recurring problem whenever one person has expertise, authority, or capability that another person lacks: how do you help without quietly taking ownership away from the person you are helping?

When the right to decide is at stake

People management is one place where preserving ownership matters. Disability advocacy confronts the question with far higher legal and personal stakes: not simply who leads a task, but who holds the right to decide.

For adults who need help making consequential decisions, the traditional legal response has often been guardianship: another person receives authority to make some or all decisions on their behalf.

An alternative called supported decision-making starts from a different premise. Someone may need help understanding a lease, comparing medical options, managing money, or communicating a choice without needing another person to make that choice for them.

The U.S. Administration for Community Living describes supported decision-making as an arrangement in which people retain the right to make their own decisions while receiving help from people they trust. The support can vary with the decision. Someone might need substantial help with finances but little with transportation or housing.

The important distinction is not between independence and dependence.

It is between supporting a decision and substituting for the decision-maker.

Parents encounter the same problem years earlier

Parenting research uses the term autonomy support for something that can easily be mistaken for permissiveness.

It is not the absence of limits.

Self-determination research describes autonomy-supportive parenting as giving children meaningful room for choice and participation while still providing structure, explanations, and boundaries. A parent can insist that homework gets done or that a child cannot run into the street without requiring obedience to become the child's only reason for acting.

The long-term objective is different from immediate compliance.

A parent who solves every difficult problem can produce the desired outcome today. The homework gets finished. The backpack is packed correctly. The disagreement with a friend gets handled.

But each successful rescue can also remove one opportunity for the child to practice judgment.

The support problem therefore contains a tension. Help too little and the person may fail for reasons that better guidance could have prevented. Help too much and the immediate success may conceal that the person never gained the capability to succeed without you.

Expertise creates a particularly difficult version

Medicine makes the tension harder because the expertise gap can be enormous.

A physician may understand probabilities, treatment effects, complications, and disease progression that a patient cannot reasonably be expected to understand unaided.

But expertise about medicine does not answer every part of a medical decision.

One treatment may offer greater expected longevity with severe side effects. Another may preserve day-to-day quality of life while accepting additional risk. The physician has expertise about the medical consequences. The patient has information the physician cannot supply: which consequences matter most in that person's life.

That is why medicine developed shared decision-making.

The Agency for Healthcare Research and Quality describes it as a process in which clinicians help patients understand the benefits, harms, and risks of available options while incorporating the patient's own values and preferences into the decision.

It does not pretend the patient possesses the physician's expertise.

It also does not treat the physician's expertise as authority over the patient's goals.

Support works because the two parties contribute different things to the decision.

Software has to decide who is in control too

The same problem appears in a much less human-looking place: interface design.

Good software increasingly predicts what people want, fills things in for them, recommends actions, catches errors, and automates work. Each capability can help, and each can also take control.

One of the oldest usability principles is user control and freedom. Good interfaces make it possible to back out, cancel, or undo an action rather than forcing the user to proceed along the path the software selected. The ability to reverse an action does more than repair mistakes. It makes experimentation safer because users can explore without surrendering control of the outcome.

That becomes more consequential as software grows more capable.

A system that warns that an action looks risky still leaves the choice visible.

A system that silently prevents the action has made a different decision about who should be trusted.

Sometimes that is appropriate. Airplanes should not depend on a passenger's preference about whether to deploy the landing gear.

But many systems move from assistance to control without making the transition obvious.

The temptation to take over

The helper may actually know better.

If the only objective is producing the best immediate outcome, taking over can be rational.

But immediate outcome is not always the only thing being produced.

The interaction may also be developing the person's judgment, preserving responsibility for the outcome, or affecting the ability to act without assistance next time.

The most useful help may not be the help that most completely prevents mistakes. It may be the help that makes the consequences understandable, makes dangerous errors harder to make, and leaves enough of the decision intact that it still belongs to the person making it.

There is no universal boundary. A toddler, a new employee, a patient under anesthesia, and an experienced software user obviously require different kinds of support.

But the distinction remains useful.

When someone is struggling, the question is not only how much help do they need?

It is also which part of the decision still needs to remain theirs?

Curated and edited by John Pile Jr.

Sources and further reading

• U.S. Administration for Community Living, “Alternatives to Guardianship.”
• Self-Determination Theory, “Parenting and Family.”
• Gavin R. Slemp et al., “Leader autonomy support in the workplace: A meta-analytic review,” Motivation and Emotion 42 (2018).
• Agency for Healthcare Research and Quality, “About Shared Decision Making.”
• Nielsen Norman Group, “User Control and Freedom (Usability Heuristic #3).”

Featured image: Photo by Christina Morillo on Pexels.