How Psychological Safety Became a Substitute for Leadership

A condition designed to protect candour has been elevated into a governing theory of leadership. In the process, organizations have become more attentive to how authority feels—and less certain about what authority must accomplish.

Organizations need employees who will speak when something is wrong.

They need nurses who question a physician, analysts who challenge an assumption, engineers who report a design flaw, managers who disclose a deteriorating result, and employees who tell senior leaders what those leaders would prefer not to hear. Silence can protect personal standing while exposing the organization to error, misconduct, operational failure, and unnecessary risk.

Psychological safety gave organizations a useful way to understand that problem. It explained why capable people sometimes withhold relevant information and why teams learn more effectively when members can ask questions, admit mistakes, request help, and challenge prevailing views without expecting humiliation or retaliation.

But the concept did not remain within those boundaries.

Psychological safety gradually moved from being a condition that enables people to contribute information into a standard by which leadership itself is judged. Leaders are now routinely assessed by whether people feel safe, heard, included, respected, reassured, and comfortable expressing themselves. Difficult conversations, performance decisions, organizational changes, and ordinary exercises of managerial authority are evaluated increasingly through the emotional experience they produce.

This shift appears humane. It has also created a serious institutional mistake.

The ability to speak without illegitimate fear is necessary. It is not leadership.

A team can feel free to speak and still be badly led. Employees can feel heard while decisions remain unresolved. Managers can invite dissent while avoiding judgment. Leaders can respond empathetically to every concern while failing to clarify responsibility, maintain standards, correct performance, protect responsible conduct, or accept the consequences of their decisions.

Psychological safety did not make leadership weaker. Its institutional misuse gave organizations a respectable way to substitute a managed interpersonal climate for the responsible exercise of authority.

The result is a leadership doctrine increasingly attentive to how people experience authority, but insufficiently exact about what authority owes them.

A Bounded Idea Became a Universal Answer

Amy Edmondson introduced team psychological safety as a shared belief that a team is safe for interpersonal risk-taking. The construct was developed to explain learning behaviour: whether people would ask questions, admit errors, seek feedback, offer ideas, and raise concerns when doing so could make them appear ignorant, incompetent, intrusive, or disruptive.

The evidence supporting its importance is substantial. A major meta-analysis covering 136 independent samples, more than 22,000 individuals, and nearly 5,000 groups found meaningful relationships between psychological safety and workplace outcomes, including task performance and organizational citizenship behaviour.

Google’s Project Aristotle further accelerated the concept’s influence. Its study of team effectiveness identified psychological safety as an important team dynamic. But Google did not conclude that psychological safety alone created an effective team. Its findings also identified dependability, structure and clarity, meaning, and impact. The original account expressly included reliable performance, clear expectations, effective decision processes, and the consequences of performance.

That larger finding has frequently been compressed into a simpler management message: psychological safety is the foundation of high-performing teams.

The compression matters. A multidimensional account of team effectiveness became a dominant leadership imperative. A specific construct concerning interpersonal risk expanded into a broader promise that leaders could improve performance by creating an environment in which people feel safe.

Once an idea enters the leadership industry, precision is difficult to preserve. Concepts must become teachable, measurable, repeatable, and commercially recognizable. Boundaries are reduced. Conditions become principles. Principles become universal prescriptions. Eventually, the concept becomes a test of whether a leader is enlightened, humane, inclusive, and worthy of trust.

Psychological safety was especially vulnerable to this expansion because its ordinary-language meaning appears broader than its scholarly definition. “Safety” can easily be heard as protection from discomfort, criticism, disagreement, pressure, negative evaluation, or unwelcome consequences. The American Psychological Association has cautioned that psychological safety is a specific and targeted concept rather than a general description of psychological well-being.

Edmondson has also repeatedly clarified that psychological safety does not mean niceness, comfort, lowered expectations, freedom from accountability, or an anything-goes workplace. Her own work calls for psychological safety, accountability, and high standards to exist together.

Those corrections are important. They also reveal what happened to the concept after it entered mainstream leadership practice.

The original theory concerned whether people could take interpersonal risks in service of learning and performance. The popularized version increasingly concerns whether people experience the exercise of authority as psychologically acceptable.

That is not a minor change in terminology.

It changes the governing question of leadership.

Instead of asking whether authority was exercised responsibly, organizations begin by asking whether authority produced the desired emotional experience. Instead of asking whether a decision was necessary, well reasoned, fairly applied, and properly communicated, they ask whether people felt sufficiently included in reaching it. Instead of asking whether a standard was legitimate and consistently enforced, they ask whether the person held to that standard experienced the conversation as supportive.

Psychological safety becomes more than a condition surrounding leadership.

It becomes a substitute for deciding whether leadership occurred.

The Leadership Substitution

The substitution begins when an enabling condition becomes the principal obligation.

Leaders should create conditions in which people can raise concerns without retaliation. They should listen seriously, acknowledge uncertainty, invite relevant challenge, and avoid using authority to humiliate or silence. These are legitimate requirements of responsible leadership.

But leadership cannot end with the successful expression of a concern.

Someone must determine what the concern means. Someone must evaluate the evidence, reconcile competing interests, decide which judgment will govern, state what is required, allocate responsibility, and accept accountability for the result.

Psychological safety can improve the information available to that person.

It cannot relieve that person of the obligation to decide.

The leadership industry has blurred this boundary by concentrating heavily on how leaders should present themselves and how employees should experience them. Leaders are encouraged to show vulnerability, communicate empathy, demonstrate curiosity, invite participation, create belonging, acknowledge emotions, and make people feel valued. Each practice can improve the human experience of work.

Together, however, they have moved the centre of leadership away from institutional obligation and toward interpersonal reception.

The leader increasingly becomes responsible not only for behaving fairly and respectfully, but also for securing a favourable emotional interpretation of difficult decisions. Discomfort becomes suspicious. Disappointment becomes evidence that the process was deficient. Direct correction risks being interpreted as psychological threat. A rejected recommendation can be treated as proof that the employee was not heard. A clear boundary can be interpreted as exclusion. An unpopular judgment can become evidence that the leader failed to create safety.

Under this standard, the legitimacy of authority becomes contingent upon the response it produces.

This creates an impossible obligation. Responsible leadership frequently disappoints people.

A decision selects one direction and rejects another. A performance standard separates acceptable work from unacceptable work. A resource allocation funds some priorities and delays others. A strategy excludes plausible alternatives. An investigation reaches a conclusion that one party will dispute. A restructuring removes responsibilities, positions, or people. Accountability establishes that an obligation was not fulfilled and that a consequence must follow.

These decisions should be evidence-based, proportionate, respectful, and open to legitimate challenge. They cannot be made without producing frustration, disagreement, uncertainty, or loss.

An organization that treats those reactions as presumptive evidence of unsafe leadership makes responsible authority increasingly difficult to exercise.

Leaders respond predictably. They extend consultation after the necessary information has been obtained. They soften expectations until meaning becomes negotiable. They replace correction with continued coaching, decision with dialogue, and consequence with another opportunity for reflection. They describe performance deficiencies without concluding what must change. They repeatedly invite perspectives without establishing where participation ends and decision authority begins.

The organization appears more considerate while becoming less governable.

The right to speak gradually becomes confused with the right to prevail. Being heard becomes confused with having one’s position adopted. Inclusion in the discussion becomes confused with shared authority over the decision. The absence of agreement becomes evidence that the process remains incomplete.

Yet psychological safety was never meant to establish these entitlements.

Employees should be able to challenge a decision without being punished for the challenge. They are not entitled to have the challenge accepted.

They should be able to admit an honest mistake without humiliation. They are not entitled to exemption from every consequence of the mistake.

They should be able to explain the circumstances surrounding poor performance. They are not entitled to redefine the required standard through that explanation.

They should be able to disagree with authority. They are not entitled to make authority disappear.

A workplace cannot become psychologically safe by making every conclusion provisional, every requirement negotiable, and every consequence dependent upon the recipient’s approval.

That does not produce safety.

It produces institutional uncertainty.

How Weak Leadership Acquired the Language of Care

The most consequential effect of this substitution is not that leaders have become excessively kind. Kindness does not weaken leadership. Respect does not prevent accountability. Empathy does not require indecision.

The problem is that leadership avoidance has acquired morally respectable language.

A delayed decision can be described as further inclusion. An ambiguous expectation can be described as preserving flexibility. Reluctance to correct poor performance can be described as empathy. The refusal to establish a consequence can be described as maintaining psychological safety. Continued discussion can be presented as evidence of openness even when the discussion no longer informs anything.

The leader avoids the burden of authority while appearing to practise an advanced and humane leadership philosophy.

This is why the misuse of psychological safety is more consequential than a simple misunderstanding of terminology. It can conceal institutional failure.

Consider the manager who inherits a team member whose work repeatedly falls below the required standard. The deficiencies are known. Colleagues compensate for them. Deadlines slip. Customers experience the consequences. Previous managers have held numerous supportive conversations, offered coaching, clarified expectations, and invited the employee to explain what additional assistance might help.

Each conversation demonstrates care. None resolves the condition.

The employee facing accountability is treated as the principal subject of psychological safety. The people carrying the additional workload are not. The manager whose credibility is deteriorating is not. The employees who met the standard without receiving special accommodation are not. The customers exposed to preventable failure are not.

When accountability is withheld, its burden does not disappear.

It is transferred.

An institution cannot evaluate safety solely from the perspective of the person facing correction. It must also examine the conditions imposed upon everyone else when correction does not occur.

The same failure appears at higher levels of authority. Organizations encourage employees to speak candidly but retreat when the truth implicates a powerful executive, commercially important leader, or politically protected function. They conduct surveys, facilitate listening sessions, and train managers to receive feedback more constructively while leaving the institution’s actual protections unchanged.

Employees notice.

They see who is permitted to violate the declared standard. They see which concerns produce action and which produce scrutiny of the person who raised them. They see managers reverse legitimate decisions after escalation by influential employees. They see corporate values applied rigorously downward and interpreted flexibly upward.

These employees do not remain silent because their manager failed to ask an inviting question.

They remain silent because they have correctly understood the institution.

This is not principally a psychological safety problem. It is an authority and governance problem.

No amount of interpersonal warmth can overcome institutional evidence. Employees will not believe that candour is protected merely because leaders adopt more receptive language. They judge safety by what happens after someone tells the truth.

Does the organization examine the concern seriously? Does it protect the person from retaliation? Does it distinguish an inconvenient claim from an irresponsible one? Does it act when the evidence warrants action? Does the standard constrain powerful people? Does the institution accept a cost to uphold what it publicly claims to value?

When the answer is no, another workshop on psychological safety does not address the failure.

It disguises it as a communication problem.

Psychological Safety Cannot Govern an Organization

Psychological safety answers an important question: Can people contribute relevant information without expecting illegitimate interpersonal harm?

Leadership must answer a different question: What will authority do with the information once it has been contributed?

The first question governs access to candour.

The second governs the institution.

Psychological safety cannot decide which competing priority should receive resources. It cannot determine whether a performance explanation is sufficient. It cannot allocate decision rights, reconcile incompatible obligations, define acceptable risk, distinguish intelligent failure from negligence, or determine when a repeated deficiency requires consequence.

It can help those matters enter the discussion.

It cannot resolve them.

This boundary is essential because contemporary organizations face decisions for which no psychologically neutral answer exists. Closing a facility, correcting an executive, changing strategy, denying a promotion, restructuring a team, ending employment, rejecting a proposal, or enforcing a disputed standard will affect people unequally. Some will experience the decision as fair. Others will experience the same decision as unnecessary, threatening, or wrong.

Leadership cannot make its legitimacy dependent upon achieving emotional consensus.

The stronger standard is whether authority was exercised responsibly.

Did leaders obtain relevant information before acting? Could employees challenge assumptions without retaliation? Were competing interests considered rather than merely acknowledged? Was decision authority clear? Was the conclusion supported by evidence and legitimate organizational purpose? Was the standard applied consistently? Were affected people treated with dignity? Did leaders accept responsibility for the foreseeable consequences?

These questions protect people without surrendering the organization’s capacity to govern.

They also prevent a false choice between psychological safety and accountability. The issue is not whether organizations should choose candour or standards, voice or authority, respect or consequence.

A serious institution requires both.

Protected Candour and Responsible Authority

The leadership standard that replaces the current substitution is protected candour with responsible authority.

Protected candour means that employees can raise concerns, admit errors, ask questions, seek help, challenge assumptions, disagree with authority, and present inconvenient evidence without humiliation, retaliation, or damage imposed merely because the contribution was unwelcome.

Responsible authority means that leaders remain obligated to evaluate what they hear, exercise judgment, make decisions, define expectations, maintain legitimate standards, protect responsible conduct, impose proportionate consequences, and remain accountable for the institutional conditions their decisions create.

Protected candour without responsible authority produces disclosure without resolution.

People speak, leaders listen, concerns accumulate, and the organization congratulates itself for openness while the conditions generating those concerns remain unchanged.

Responsible authority without protected candour produces decisions without sufficient truth.

Leaders act, but the information reaching them has been filtered by fear, hierarchy, self-protection, and the perceived danger of disagreement.

Neither condition is sufficient.

Together, they create dependable leadership.

This stronger standard also clarifies what employees can legitimately expect. They can expect protection for raising a concern responsibly. They cannot expect every concern to be validated. They can expect their evidence to be considered. They cannot expect their recommendation to govern. They can expect fair treatment during accountability. They cannot expect accountability to disappear. They can expect leaders to explain consequential decisions where appropriate. They cannot expect every decision to secure their agreement.

These boundaries are not hostile to psychological safety.

They preserve its value.

Psychological safety becomes meaningful when it allows the organization to gain access to truth that hierarchy, fear, and self-protection would otherwise suppress. It becomes distorted when it is used to prevent authority from reaching a conclusion about that truth.

The proper sequence is straightforward.

People must be able to speak.

Leaders must be willing to hear.

Authority must still act.

Restoring Leadership to Its Proper Work

The leadership industry has spent years teaching leaders to pay closer attention to their effect on other people. That correction was necessary. Authority exercised without regard for dignity, voice, power, or interpersonal consequence can become destructive.

But a correction becomes a new failure when it displaces the responsibility it was meant to improve.

Leadership is not merely the ability to create a positive emotional climate. It is not validated by whether people felt inspired, reassured, included, comfortable, or safe in every interaction. Those experiences matter, but they remain incomplete evidence.

Leadership must ultimately be judged by what people can depend on.

Can they depend on expectations being clear? Can they depend on decisions being made before delay becomes its own decision? Can they depend on legitimate standards applying across differences in rank and influence? Can they depend on responsible action receiving institutional support? Can they depend on concerns being examined rather than merely welcomed? Can they depend on leaders following through when the answer becomes uncomfortable or expensive?

These are harder obligations than creating psychological safety because they require more than interpersonal skill.

They require judgment, authority, courage, consistency, and consequence.

The objective is not to return organizations to fear, humiliation, command-and-control management, or the suppression of dissent. It is to reject the equally serious proposition that leadership becomes responsible merely because it is experienced as receptive.

A leader can listen respectfully and still fail to lead.

A leader can invite candour and still avoid the truth it produces.

A leader can make people feel safe while leaving them inside an institution they cannot depend on.

Psychological safety remains an important condition for learning, challenge, and responsible participation. Research continues to support its relevance in complex and interdependent work. But no enabling condition should be permitted to assume authority over the entire institution.

Psychological safety should protect the entry of truth.

Leadership must govern what happens next.

Psychological safety asks whether truth can enter the room.

Leadership begins with what authority does once it arrives.

Build Leaders Who Can Hear the Truth—and Still Lead

Psychological safety should help leaders gain access to the information, disagreement, and concerns they need. It should not leave them uncertain about their responsibility to decide, clarify expectations, uphold standards, address performance, and follow through.

The 5 Essential Skills for Leading People and Teams helps managers exercise authority with greater clarity, judgment, consistency, and respect—especially when the necessary decision will create discomfort or disagreement.

Organizations do not need leaders who merely invite candour. They need leaders who know what responsible authority requires once candour has been expressed.

Explore The 5 Essential Skills for Leading People and Teams.

Jim Woods is the founder of Seattle Consulting Group, originator of The Woods HR Power Model™, and co-creator of The Five Decisions of Leadership™. His work examines leadership judgment, HR governance, institutional accountability, organizational trust, and the management failures prevailing doctrine often misdiagnoses.