When Everything Is Finally Under Control

17 min read

The Relief of Perfect Alignment

You hired people for judgment, not merely execution. Then you built a system for preventing them from using that judgment without permission.

You called it alignment.

When they stopped acting, you called it an ownership problem.

That sequence is so common in software organizations that it passes for management. It is not management. It is a self-protective loop in which one person’s discomfort with uncertainty becomes everybody else’s loss of agency.

The loop often begins with reasonable words. Accountability. Consistency. Risk management. Visibility. Standardization. Governance. None of those words is the problem. The question is what the mechanism actually does.

Does it make reality easier for everyone to see, or does it make people easier for you to direct?

Does it contain business risk, or does it contain dissent?

Does it clarify who may decide, or ensure that the answer is always you?

If your people can produce evidence but cannot change your mind, you do not have governance. You have a court ritual.

The Oldest Promise in Management

A manager who adds an approval gate is not a slaveholder. A developer with a salary, legal rights, and the practical ability to resign is not enslaved. The racialized chattel slavery of the transatlantic trade was an atrocity built on ownership, hereditary status, organized violence, and the deliberate destruction of human freedom. Using it as office rhetoric would trivialize the people who endured it and excuse sloppy thinking about the workplace.

The boundary matters precisely because the underlying subject is serious.

Slavery did not begin with the transatlantic trade, and human domination has never depended on one legal design. Across history, people have used war captivity, debt bondage, serfdom, forced state labour, servile marriage, domestic servitude, and other arrangements to capture another person’s labour and choices. These systems were not identical in law, brutality, duration, or ideology. The United Nations’ 1956 Supplementary Convention named debt bondage, serfdom, and several forms of servile marriage as institutions or practices similar to slavery. The International Labour Organization distinguishes slavery, forced labour, and trafficking rather than collapsing them into one slogan.

Different systems. Different crimes. A recurring move.

Make exit impossible or ruinously expensive. Make dependence normal. Convert a human life into a source of labour, status, sex, care, revenue, or convenience for someone with more power. Then create a moral story in which the arrangement is natural, necessary, civilizing, contractual, divinely ordered, economically unavoidable, or somehow good for the person being controlled.

The modern workplace is not that. Its legitimate basis is mutual consent, limited authority, compensation, legal protection, and the ability of either party to end the relationship. Those distinctions are not polite details. They are the line.

But the human mind that likes turning people into means did not disappear when the law improved.

It learned better vocabulary.

Headcount. Capacity. Resources. Hands. Utilization. Compliance. The business needs. The transformation requires. The process says.

The person fades. The instrument remains.

Why Control Feels So Good

There is no single psychological organ called the desire to control. Several motives converge on the same behavior, and leaders are unusually well positioned to turn those motives into organizational machinery.

Order Feels Like Safety

You are responsible for an outcome you cannot produce alone. The board wants a date. Customers want reliability. Regulators want evidence. A critical digitalization program crosses business units, vendors, old systems, and policies nobody has written down. The software keeps revealing facts the plan did not contain.

You cannot make uncertainty leave. You can make people behave as if it has.

Demand estimates. Freeze scope. Require sign-off. Add reporting. Centralize decisions. Make every deviation visible and every doubt explainable. The anxiety drops—not because the situation became safer, but because the room now performs order for you.

Research on compensatory control shows that when people’s sense of personal control is threatened, they become more attracted to external systems that promise order. That research does not diagnose a particular CTO’s approval workflow. The organizational inference is still hard to miss: a system can be emotionally useful to the person imposing it while being operationally useless to everyone doing the work.

Your weekly status meeting may exist because uncertainty is expensive for the company.

It may also exist because uncertainty is unbearable for you.

Those are not the same problem.

Authority Should Mean Something

A title does more than allocate responsibility. It becomes a story about who you are.

The CEO decides. The CTO understands technology. The vice president sees the whole. The manager knows what the team should do. Then someone lower in the hierarchy knows more about the specific problem and says your preferred decision is wrong.

That should be useful information.

It may not feel like information. It may feel like a status injury.

If your self-respect depends on being the person with the answer, another person’s expertise does not help you. It reduces you. The easiest repair is to make expertise subordinate to rank: thank them for the input, restate the decision, and quietly remember that they were “difficult.”

Now the hierarchy has done emotional first aid for the leader.

The company gets the scar.

A Room That Knows Who Decides

Human status has at least two familiar routes. Dominance extracts deference through the ability to impose costs; prestige attracts it because other people value someone’s knowledge or skill.

Prestige is fragile. You have to remain worth listening to.

Dominance is wonderfully efficient. The calendar invite, budget decision, performance rating, contract renewal, and reorganization can produce agreement by Thursday.

Many leaders claim they want respect while building systems that manufacture deference. They do not notice the substitution because both look the same from the front of the room. People nod. The decision stands. Nobody openly challenges them.

Then they leave the meeting and the real conversation begins.

If the truth consistently appears after you leave, you are not leading the room. The room is managing you.

The Process Keeps It Professional

It is difficult to look at an experienced professional hired for judgment and say, “I need you to suppress that judgment so I can feel in charge.”

So nobody says it.

The person becomes a role. The instruction becomes policy. The threat becomes accountability. The punishment becomes a consequence. The decision becomes a governance outcome. Everybody contributed, which means nobody did it.

Albert Bandura called this kind of psychological machinery moral disengagement: moral justification, euphemistic labeling, advantageous comparison, displacement or diffusion of responsibility, minimizing consequences, and dehumanizing or blaming the person harmed.

Organizations are excellent at industrializing those moves.

“The steering committee decided.”

“Procurement requires it.”

“The framework says so.”

“We are just following the process.”

No single sentence is dramatic. That is the point. Domination becomes durable when nobody has to experience themselves as dominating.

The View Is Better From the Top

Power does not automatically corrupt everyone. It changes the psychological environment.

People with more power face fewer immediate constraints, receive more rewards, and can act with less fear of punishment. Research on the approach-inhibition theory of power connects power with greater attention to rewards, more automatic information processing, and more disinhibited action. Research on power and perspective-taking found that people placed in high-power conditions were less likely to account for what others knew, less accurate at reading emotion, and more anchored in their own vantage point.

That does not mean your title erased your empathy. It means your confidence in your empathy deserves less confidence than you probably give it.

Power also changes what other people show you. They edit bad news. Laugh sooner. Challenge later. Bring finished recommendations instead of uncertain observations. Translate reality into the vocabulary that survived the last disagreement.

You receive cleaner information and conclude that the organization is becoming clearer.

It is becoming quieter.

The silence then confirms your self-image: calm under pressure, decisive, able to align a room. What you cannot see is the cognitive work happening below you as people predict which truths you can tolerate.

Your organization may spend more time regulating your emotions than serving your customers.

Some People Really Do Need Direction

Douglas McGregor’s Theory X and Theory Y are often reduced to two management styles: strict versus nice. That misses the wound.

Theory X begins with assumptions: people dislike work, avoid responsibility, and must be directed, controlled, or threatened to produce effort. Theory Y begins elsewhere: people can exercise self-direction, accept responsibility, and apply imagination when conditions allow it.

McGregor’s enduring question was not, “Which type of employee did you receive?” It was, “What assumptions about human behavior are embedded in your management?” MIT’s account of his work also warns against turning Theory Y into another monolithic doctrine. The point is not cheerful faith in everyone. It is noticing how management beliefs create the evidence used to defend them.

Assume people cannot be trusted. Remove discretion. Require approval. Punish deviation. Reward compliance. People learn that initiative carries risk and waiting carries none. They stop deciding. They stop offering unfinished ideas. They do exactly what the work item says, including the foolish parts.

Then you look at them and see people who refuse responsibility.

Congratulations. Your theory has acquired a workforce.

This is why controlling leaders are so difficult to reach. The organization continuously manufactures confirming evidence. Every act of withdrawal proves people need direction. Every covert workaround proves they cannot be trusted. Every departure proves they lacked commitment. Every failed initiative proves autonomy went too far.

The one hypothesis that never enters the review is that people adapted intelligently to the leader.

Software Refuses to Stay Predictable

Software work creates a particularly offensive dependency for anyone who needs control.

The work is largely invisible. Progress can reverse when learning improves. The elegant plan can be technically impossible. A tiny requirement can expose a deep architectural problem. A younger developer can know more about the consequential detail than the executive paying everybody in the room. Tests, production behavior, and users can disprove the highest-paid opinion before lunch.

Worse, developers do not merely execute business decisions. By encoding rules, they discover contradictions in those decisions. They learn where policy is vague, where departments disagree, where a supposedly standard process survives through human improvisation, and where the transformation slide deck has mistaken a political compromise for a requirement.

Digitalization makes this even less comfortable. A company says it is automating a workflow or rolling out AI, not developing software. Yet somebody still has to decide behavior, exceptions, ownership, permissions, failure recovery, data meaning, and what happens when the vendor’s happy path meets the company’s actual history. The label changes. The dependence on human judgment does not.

A mature leader welcomes that dependence because it extends what the organization can know.

A controlling leader experiences it as loss of rank.

So the work gets made legible to authority instead of effective in reality:

• every problem is decomposed before the people doing it are allowed to learn
• estimates become promises and new information becomes failure
• ticket movement stands in for value reaching users
• approvals accumulate because no individual may be trusted with a consequential choice
• architecture decisions move upward, away from the code and toward title
• vendors are bought as interchangeable hands, then punished for withholding judgment nobody purchased
• activity surveillance expands because observing output is harder than counting motion
• AI is sold as a way to remove troublesome experts instead of increasing the reach of good ones

None of this needs a screaming tyrant. The most effective control systems are maintained by polite people with templates.

The result is not predictability. It is an organization trained to hide discovery until it can be presented as execution.

That is fatal to software delivery. Software flow depends on shortening the path from discovery to action. Control inserts a political translation layer into that path. Reality appears, gets softened, routed, scheduled, approved, and stripped of the person who understood it.

By the time the truth reaches the executive dashboard, it has learned table manners.

People Learn the System Quickly

People do not become inert because the org chart wishes it. They adapt.

One response is resistance. Psychological reactance is the motivation to restore freedom after it is threatened. Sometimes it is open disagreement. More often it is delay, argument over wording, selective interpretation, or doing the forbidden thing where authority cannot see it. The large body of research on psychological reactance exists because threatened autonomy reliably changes how people receive demands.

Another response is controlled motivation: doing the work to obtain a reward, avoid punishment, or reduce pressure. Self-determination theory distinguishes this from autonomous motivation and connects healthy functioning with the needs for autonomy, competence, and relatedness. Ryan and Deci’s foundational review opens with a useful fact: people can be proactive and engaged or passive and alienated largely as a function of their social conditions.

Your “unmotivated people” may be a photograph of those conditions.

The third response is concealment. People stop telling you what they think and start telling you what will get through. They estimate the acceptable number. They mark the ticket green. They move the risky conversation into a private chat. They create a spreadsheet because the official system makes adaptation punishable. They let the bad decision fail cleanly because saving it would require admitting they disobeyed.

This is not healthy. It is also not irrational.

When honest initiative is punished and ceremonial compliance is safe, intelligent people become excellent at ceremonial compliance.

The final response is exit. The people with the most alternatives leave first. Leadership then points to attrition as evidence that talent is disloyal, the market is difficult, or younger workers lack resilience.

The controller keeps the explanation.

The competitor gets the developer.

Of Course You Want Ownership

“We need more ownership” is one of the most dishonest sentences in corporate life.

Ownership means the person can make consequential decisions, including decisions you would not have made. It means their judgment can alter scope, sequence, design, and sometimes your own conclusion. It means they can be wrong without being stripped of agency at the first bruise. It means authority travels with responsibility.

What many leaders want is different.

They want employees to experience the emotional burden of ownership while the leader retains the decision rights. Take initiative, but not in a direction I dislike. Challenge assumptions, but not after I have attached my identity to them. Own the outcome, but do not change the plan. Speak truth, but make it pleasant. Move fast, but ask first.

That is not ownership.

It is obedience with liability transfer.

If you want the credit for the decision and somebody else to carry the shame when it fails, the organizational dysfunction is not below you.

It is you.

A Small Test of Leadership

Your self-description is useless here. Almost nobody says, “I like controlling people for my own benefit.” The motive arrives dressed as responsibility.

Use behavior instead:

1. When someone contradicts you in front of others, does your body register useful information or disrespect?
2. Who can say no to you without paying a later price in access, rating, budget, visibility, or warmth?
3. Have you ever called someone “not strategic” when they understood a reality that threatened your strategy?
4. Do your dashboards help teams make decisions, or mainly help you question them?
5. When a plan changes because the work revealed something new, do you reward the learning or investigate the deviation?
6. Can people act inside clear boundaries without pre-approval, or is every boundary merely a shorter leash?
7. Do you ask for estimates to understand uncertainty, or to obtain a number you can later use as evidence against the estimator?
8. When people follow your instruction exactly and it fails, do you accept authorship?
9. Which truths reach you only through unusually brave people?
10. What does the organization routinely hide from you because your reaction costs more than the problem?
11. Do you want a strong CTO, product leader, architect, or developer—or do you want a strong person whose strength never competes with yours?
12. If everyone obeyed you perfectly for a year, would the company become smarter or merely more like you?

If several questions sting, do not reach for the flattering explanation. Do not announce a listening initiative. Do not ask Human Resources to run a workshop on psychological safety.

Stay with the injury.

What are you afraid another person’s freedom will take from you?

Keep the Control That Actually Works

The desire for control will not disappear because you read the right book. You will feel it whenever responsibility outruns certainty, expertise below you threatens identity, or another person can affect an outcome you care about.

Your job is to keep that feeling from acquiring a workflow.

That requires harder discipline than micromanagement:

Name the fear before designing the control. What concrete failure are you preventing? What evidence says the proposed mechanism reduces it? If the honest answer is “I need to know,” decide whether the company needs that knowledge or you need reassurance.

Control conditions, not minds. Put boundaries around money, security, legal exposure, data, and production access. Make those boundaries explicit and testable. Inside them, let the people with relevant competence and responsibility decide. Governance should make reality visible; it should not make permission the only safe behavior.

Attach authority to responsibility. If a person owns an outcome, state which decisions they can make without you. If you reserve the decision, own the outcome. Stop delegating blame while centralizing power.

Make dissent cheaper than concealment. Reward the person who changes your mind with bad news. Ask for the strongest case against your preference before declaring alignment. Track which leaders discover problems early, not which teams keep every status green.

Use evidence that can disagree with rank. Working software, automated tests, production behavior, customer use, recovery time, defect escape, and lead time do not care who won the meeting. That is why controlling organizations prefer status narratives. Choose the evidence anyway.

Keep decisions reversible where possible. Small releases, short feedback loops, and bounded experiments reduce the amount of faith anyone must place in a single authority. You do not need obedience when reality can answer quickly.

Let expertise embarrass you. You hired people who should know things you do not. If that never becomes visible, you either hired badly or frightened them into silence. Respect starts where hierarchy stops pretending it owns insight.

Accept consequences without demanding submission. Autonomy does not mean absence of standards. A qualified person can own a bounded decision, be held to an agreed outcome, learn from failure, and remain trusted to exercise judgment. Accountability asks for an honest account and a responsible response. It does not demand humiliation as proof that authority still works.

This is not softness. It is refusing to use organizational power as anesthesia.

The Question Beneath Alignment

You will still need controls. Financial controls. Security controls. Access controls. Safety controls. Legal controls. Production controls. A company without boundaries is not humane; it is negligent.

But every control over a system can quietly become control over a person. Every request for visibility can become a demand for emotional submission. Every claim of accountability can become a way to make someone else carry your uncertainty while you keep the authority.

That is the line to watch.

Not because crossing it makes you a slaveholder. It does not.

Because crossing it makes you the kind of leader who needs capable people to suppress the judgment they were hired to contribute so you can feel large enough to lead.

The next time someone resists your preferred decision, do not begin with, “How do I get them aligned?”

Begin with the question you have been making the organization avoid:

Why do I need this person to obey me?

Do not justify it.

Answer it.

Sources and Further Reading

• Douglas McGregor, Theory X and Theory Y at MIT Sloan
• Richard M. Ryan and Edward L. Deci, Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being
• Aaron C. Kay and colleagues, God and the Government: Testing a Compensatory Control Mechanism for the Support of External Systems
• Adam D. Galinsky and colleagues, Power and Perspectives Not Taken
• Dacher Keltner, Deborah H. Gruenfeld, and Cameron Anderson, Power, Approach, and Inhibition
• Joseph Henrich and Francisco J. Gil-White, The Evolution of Prestige
• Albert Bandura, Moral Disengagement in the Perpetration of Inhumanities
• Christina Steindl and colleagues, Understanding Psychological Reactance
• United Nations, Supplementary Convention on the Abolition of Slavery, the Slave Trade, and Institutions and Practices Similar to Slavery
• International Labour Organization, What Is Forced Labour?