Authority and delegation: who owns the decision

Epigraph

“Genuine power is not coercive control”.

Mary Parker Follett, Creative Experience, Introduction; the first short part of a contrast on physical PDF page 2, web extraction lines 91–95. The context continues with coactive control. This is a historical social position, rather than a validated law of every organization. Primary text.

What this article is about

In an invented scene, Maya coordinates a small museum exhibition. Engineer Pavel proposes a system for maintaining conditions in the gallery. His assessment shows advantages of the chosen option, but installation would close part of the exhibition and change the budget. The curator worries about visitor access; the director wants a quick decision. Maya hears that specialists should be trusted and prepares to tell the team everything is settled. Yet Pavel supplied a technical assessment rather than an agreed decision about money, programme, and authority.

The scene is invented: this article establishes no museum engineering requirements and selects no equipment. We consider the transition from recognizing competence to transferring a decision. What exactly do we trust someone to do? Where do they have knowledge? Who determines the goal, accepts the cost, and answers for a subsequent promise? Expert conclusions, organizational roles, and personal trust may overlap without becoming identical grounds. Polite agreement does not establish that participants understood the same extent of transfer.

Delegation can be sensible. We need not repeat every specialist's work or reject help to prove independence. Yet transferring a task does not require our question to disappear. The article considers boundaries of authority and responsibility: what was assigned, what remains open, which answer is needed, and what should be checked afterwards. It does not promise to remove team conflict or make every decision safe through one clear formulation.

I. Competence concerns a particular subject

Pavel may assess a system's technical properties well without knowing which exhibition section matters most to the team. Maya may know the programme without being able to check the calculation professionally. This demeans neither role. It lets us ask what the recommendation actually supports. A calculation under particular conditions does not automatically identify the best option for every museum goal. Equally, not knowing the programme does not make a technical conclusion mistaken. It needs consideration within its domain while questions outside that domain remain visible.

Competence cannot be identified solely through confidence or a familiar title. Relevant qualifications, experience, access to information, and work limits matter for the task. This article checks no actual specialist and offers no universal examination. It distinguishes grounds of trust: familiarity, reputation, management appointment, or ability to assess this subject. These can support each other while providing different information. Popularity replaces no qualification, a position confirms no calculation automatically, and friendship does not remove the need to understand a professional conclusion's conditions.

A competent answer may contain uncertainty. Pavel says information about part of the gallery is insufficient. That is not necessarily poor work. If the director removes the qualification for a confident announcement, transmission changes the conclusion's scope. Maya can retain it without assessing every technical detail. She needs to know which question still requires information. A clear limit permits help to be accepted at its actual scale, rather than searching for a specialist who promises to settle everything without grounds.

The opposite error demands complete knowledge of every domain before recognizing expertise. No one must independently reconstruct Pavel's professional education. Reasonable trust can include establishing a relevant role, understanding the goal, and having a route to clarify important matters. This is an editorial distinction rather than a guarantee. High risks need appropriate professional checks determined by actual circumstances. But repeating all the work personally is not the only way to retain participation, and help is not a threat to freedom by default.

II. A recommendation is not always decision authority

Pavel may recommend an option without authority over expenditure. The director may authorize expenditure without knowledge to make an independent technical assessment. The curator may determine content without controlling all premises. Roles should be identified before one answer becomes the shared decision. Otherwise, the specialist receives a hidden assignment to choose for everyone, while management can describe its choice as an inevitable calculation result. The technical assessment is then used for a question it did not resolve or an authority it never claimed.

Our scene needs to establish what Pavel actually received: a request to assess options, an assignment to organize installation, or authority to select a system within a budget. Unstated terms cannot be assumed identically understood. One offer of help can mean consultation to a sender and acceptance of the entire task to its recipient. Clarification concerns object and authority, rather than personality. It can prevent an accusation about a broken promise one participant never accepted, while leaving real accepted commitments available for consideration.

Actual authority also has limits. Choosing a system within budget need not include cancelling an exhibition section or publishing a new opening date. If execution requires changed conditions, that question should return to whoever can accept the additional cost. This does not require asking about every detail. Small decisions within the task can belong to its executor. The boundary follows the agreed object and consequences, rather than a universal checklist of questions that must always go back to management before work can continue.

The arrangement could be broader: the team delegates the whole choice to Pavel and accepts known consequences. Then the transfer should be named honestly rather than presented as the absence of choice. Participants chose to entrust a domain. That act has grounds and can later be assessed. An assignment cannot be narrowed retrospectively when the executor acted within it, or expanded conveniently when an important decision remained with management. Accuracy works in both directions rather than protecting whichever side now regrets the outcome.

III. The goal is not derived entirely from the calculation

An option is assessed relative to a task. Object preservation, gallery access, cost, and timing may matter differently. The article chooses no museum priority. It shows that professional comparison requires identifying conditions. If Pavel seeks the best technical outcome with unlimited funds, that is not yet the answer about an available solution for this exhibition. If management limited funds beforehand, the condition should not be hidden behind his name as though the nature of the technical task itself imposed that budget.

Clarifying a goal permits no disregard for actual risk. Maya cannot simply prefer convenient opening if professional checking establishes unacceptable conditions. Mandatory requirements and who sets them need appropriate actual information, which is not established here. Moral distinctions do not replace technical assessment. They prevent another substitution: an obligatory condition and a manager's personal preference cannot become equally unquestionable merely because both appear in one confident answer. Their grounds and implications need to remain distinguishable to those making the organizational decision.

The curator may accept a necessary restriction while questioning duration or arrangements. Disagreement need not reject expert knowledge; perhaps another option satisfies the same requirements. Perhaps it does not. It needs subject-based consideration rather than being labelled resistance to expertise. Equally, supporting an important value does not make an option technically acceptable. The value explains why a question matters, while suitability requires appropriate checking. Neither relevant knowledge nor a significant organizational concern can do the other's entire job by itself.

The team can thus retain separate responsibilities. Pavel supplies technical grounds and limits; authorized people determine resources and programme while taking substantive requirements into account. Even if one person holds both roles, the questions differ. A director with engineering qualifications can calculate and authorize a budget without the first automatically proving the second. Naming roles identifies when they report information, assess cost, or exercise authority. It does not require multiplying participants artificially or abandoning a practical decision whenever responsibilities overlap.

IV. Follett preserves the individual within joint action

In the selected introduction to Creative Experience, Mary Parker Follett distinguishes coercive power from coactive control and connects human relations with individual integrity. This is historical social thought, rather than a proven universal management mechanism. In that same introduction she warns against replacing observation with easy analogies from another domain. That limit matters: her text validates neither our museum scene nor an editorial delegation exercise as a scientifically established technique. Source.

A limited question helps here: does a participant remain an acting person when a task is transferred? Maya can entrust calculation to Pavel while answering for her own team announcement. Pavel can accept professional work without having to guess every unnamed wish. This is an editorial comparison, not a transfer of Follett's entire system. It helps identify participation where trust sometimes demands disappearance from the question, and where together sometimes conceals one person's work under a convenient collective name.

Preserving the individual does not mean every decision is unanimous or every restriction coercive. Organizations can have defined authority and mandatory conditions. Those should be named rather than replaced by language of unrestricted freedom. If the director can decide the budget, that can be said directly. If someone cannot decline an assignment without substantial risk, the word yes does not establish willingness. Conditions of response matter even when management is polite and everyone visibly participates in the same meeting.

Follett's text does not permit a promise of no conflict. Participants may understand boundaries while retaining incompatible priorities, find that an assignment was unaccepted or impossible, or need another decision process. A historical idea remains significant as a question without replacing information about the current team. The article examines limited observable results: a clearer task, named condition, or clarified promise. Inner trust, harmony, and organizational growth are not declared established through one helpful formulation or the apparent smoothness of a meeting.

V. Executors need to know what is available

Maya transfers the task with information about the gallery, deadline, and programme constraints. If important information remains with her, Pavel cannot be assessed as though he possessed it. This does not require indiscriminate disclosure. Some material has access conditions, and only particular information may be needed. Sufficiency for the accepted work must be established without promising impossible completeness. Insufficient data becomes part of the task rather than a hidden defect in its executor's character or proof that expertise has somehow failed.

Resources matter too. A request for a solution may conflict with unavailable time, access, or funds. An executor can identify this before accepting or after discovery. Management can change conditions, narrow the result, or recognize current impossibility. The article sets no working-time standard and settles no employment dispute. It distinguishes a demand from an opportunity: insistence creates no missing resource, and a specialist's reputation does not remove consequences of inadequate access to important information. The arrangement needs assessment through what it actually makes possible.

After transfer, Maya can allow Pavel to choose a professional method within the agreed domain. Requiring approval for every step may make delegation formal: he answers for results without sufficient method freedom. Detailed restrictions can be necessary and need named grounds. Alternatively, management may retain the method and do the task through another person's hands. That is a possible arrangement, not a diagnosis of controlling behaviour. Distribution should then be reconsidered rather than blaming the executor for independence they were not allowed.

Pavel answers for the accepted part too. If the relevant conditions were available to Pavel, he cannot invoke general uncertainty about instructions to conceal a specific failure to perform. Discovering a problem requires communicating it within actual agreements rather than waiting for a convenient project ending. Delegation makes an executor neither an irresponsible instrument nor guilty for everything. Responsibility's object follows accepted action, accessible information, and authority. Their work remains a specific question alongside the quality of the assignment, with neither question cancelling the need to examine the other.

VI. Disagreement need not mean distrust

Maya can ask about recommendation conditions without accusing Pavel of incompetence. Pavel can clarify the task without rejecting the project. Questions help establish scope without guaranteeing a friendly reply. One person may see clarification as a status threat; another may use questions to postpone decisions endlessly. Neither follows from a first reply. The exchange's object matters: is a substantive ground unanswered, or does the checking demand change after each sufficiently clear answer? The observed exchange gives more information than an assumed motive.

Clarification should correspond to error cost and our competence. If Maya does not understand a technical answer, she can request significant conditions in accessible terms or seek an appropriate further assessment when actually needed. Her lack of understanding does not prove that the answer is false, but need not be hidden in deference to a title. A specialist can explain limits without making a reader a professional. Particular high risks need relevant checking; the article appoints no universal number of experts or mandatory second consultation for every situation.

Disagreement may concern acceptable cost rather than calculation. The curator recognizes necessity but requests different timing. The director recognizes accuracy but considers funds unavailable. These decisions need naming rather than treating rejection of a project as refutation of an engineer. Participants should understand what remains unresolved when an option is declined. No funds does not mean an obligatory condition is fulfilled. It may mean this exhibition is unavailable in its present form, rather than that another equally safe option has automatically been chosen.

Ordinary discussion has limits. Humiliation, threats, or substantial dependence may make direct dispute unsafe. Public proof of independence is not required for protection. Available support and another communication route may be chosen according to actual conditions. The article replaces neither professional nor legal help. Important questions cannot always be resolved by a clear meeting remark. Absence of a bold objection establishes no inner agreement and does not release a more powerful participant from examination of their own actions or the conditions they create.

VII. An assigned task does not cancel our own promise

If Maya promised an opening date, assigning calculation to Pavel does not make her promise his. He may accept an assessment deadline without accepting the exhibition deadline. If his recommendation makes the original plan impossible, she needs to revisit her announcement. A specialist cannot be made responsible for everything simply because their work affects the outcome. Tasks genuinely depend on each other, but different promises do not become one collective debt that management can reassign without the executor's and others' agreement.

Correction may be uncomfortable. Maya hoped to announce that everything was on schedule; she must now identify an open question and possible change. This is present responsibility for accurate words, not a verdict on her entire coordination ability. It does not require a new unchecked deadline. She can report what is known, what depends on a decision, and when another definite answer is available. Honest limits support planning more than a confident new date chosen solely to end an awkward conversation or preserve an appearance of control.

Pavel can promise within his own role: an assessment by a date, identification of important limitations, or notice of missing data. He answers for that rather than someone else's opening announcement. If he actually accepted broader organizational work, its scope must be retained instead of narrowed retrospectively. Accuracy is not a permanent escape from responsibility. It reveals both actual obligation and unsupported expansion through accessible words and conditions, without preferring whichever participant now finds a narrower account more convenient.

New information may lead the team to redistribute tasks. The new transfer and effects on previous promises should be named. Change does not erase existing expectations but can provide more honest grounds for subsequent work. Revision need not be a collapse of trust; it can clarify roles after discovery. Sometimes actual unreliability needs a separate response. Delegation language should not swallow these differences: clarification, breach, impossibility, and a new agreement have different objects and require different grounds rather than one general assessment of team spirit.

VIII. Checking results is not repeating the whole work

Afterwards Maya needs to know whether the promised result was obtained and under which conditions it can support a decision. She need not reproduce the professional method independently. Task correspondence, material limitations, and an understandable next step can be checked. Technical suitability needs relevant competence rather than a coordinator's everyday confidence. These roles can be separate. Checking should not imply management personally verified everything when they actually read a conclusion and made an organizational decision on its basis. The real verification boundary should remain visible.

Pavel supplied an assessment, but some initial information later changed. The result need not have been mistaken when prepared; its application domain may change. The relevant condition and need for new work must be established. An earlier conclusion cannot automatically be transferred to another situation for convenience. Nor does every new fact require repeating everything. Material-task correspondence needs competent consideration. Maya can identify the question and organize appropriate clarification while retaining her knowledge limit rather than claiming a technical determination she cannot make.

Checking may reveal nonperformance. An agreed comparison of several options might become description of one without explanation. That specific gap can be discussed. Reputation does not make it invisible, and an error does not prove incompetence in every domain. Accepted scope, accessible data, and possible correction matter. Changed assignment conditions should be named too. Result checking becomes more useful when it does not require advance confirmation of either the expert's perfection or the manager's correctness in having doubted them from the beginning.

A result can be obtained while relationships remain strained. The assessment is understood and budget chosen, but the curator still disagrees about programme restrictions. Unhappiness does not mean that no result was obtained, just as a signed decision does not prove restored trust. Technical assessment, organizational action, and relationships have different indicators. Our invented article measures none through research. Obtained and remaining results can be described without inventing total success, after which a separate question may be selected for further attention if its conditions are actually available.

IX. Status supplies no permanent answer to everything

Pavel's authority can save team time: a relevant answer need not meet universal suspicion. Yet successful gallery work does not make him sole judge of every future museum decision. A later task may concern another domain. Maya can return to him and clarify it without pretending no previous relationship exists. Trust can retain relationship history while preserving grounds' scope. One confirmed result is no permanent certificate for any subsequent conclusion voiced confidently by the same person, especially when the subject or underlying conditions have changed.

Holding a management position does not give someone knowledge of every subject either. The director can determine process without becoming a professional source of technical fact. Demanding acceptance without grounds merely through hierarchy closes competence questions through power. Exercising actual authority under limited information is a different act that should be named with risk and limits. The article decides no universal acceptability for such choices. It distinguishes a knowledge statement from an instruction so neither masks the other in the account given to participants.

Personal admiration can make distinction difficult. Someone values an expert or manager and avoids an awkward question. That relationship is understandable but establishes no recommendation scope. Respect can coexist with not transferring decisions outside the accepted domain. Choosing another executor for a task requires no general moral rejection of the first. Relationships are not guaranteed unchanged; clarification can have a cost. Independence involves no mandatory argument, but the ability to see what is transferred and on which grounds participation is accepted.

Authority also has value: relying on another's work without anxious repeated checking. The article does not seek to destroy that support; it makes it definite. Maya can trust Pavel's technical assessment, accept its limits, and organize other decisions. Each project need not become an endless investigation of personality. Without substantive new grounds, repetition may add no knowledge. But new risk or mismatch information deserves consideration; previous respect does not require closing an accessible question merely to preserve someone's convenient reputation or a smooth appearance of agreement.

X. Delegation can preserve joint responsibility

At the scene's end Pavel supplies an assessment, the team clarifies restriction costs, and Maya corrects her original announcement of complete readiness. This is an invented outcome, not evidence of effectiveness. Each act has a participant and defined result. The article selects no system, and actual engineering conditions remain outside its scope. Competence did not swallow every role, and an assigned task did not vanish into general trust language. Participants can identify their part and remaining uncertainty without presenting their work as proof of universal organizational success.

Joint responsibility does not mean identical shares. Managers, engineers, and curators have different powers and commitments. A problematic act need not be distributed across everyone for team language. Linked decisions should be examined without selecting a convenient sole culprit. A project can have a common result and separate action grounds. This editorial distinction calculates no actual liability. It preserves questions about who knew, accepted, communicated, and what was actually assigned, rather than assuming either complete collective guilt or complete individual separation in advance.

A task can finish without full value agreement. The curator regrets a closed section, the director questions cost, and the engineer retains a conditional technical assessment. If a decision was properly authorized and suitability appropriately checked, unpleasant feeling alone does not cancel it. Yet the article does not certify these actual conditions for participants. Completing a task does not prove trust or require anyone to welcome the restriction. An honest result can remain mixed while leaving the possibility of separately returning to unresolved disagreement under suitable conditions.

Authority can be a definite support and delegation a considered act. A question nevertheless does not disappear with assignment. We can identify knowledge limits, transfer methods, retain our promises, and decide within our roles. We can acknowledge that conditions remain unavailable. Freedom does not demand doing everything alone; responsibility does not demand possessing all knowledge. Their practical connection appears where participants understand the object transferred and can identify which next answer truly belongs to them rather than to an unnamed collective or assumed all-purpose expert.

What this means in practice

For a small, safe task, voluntarily write the object of transfer. This is an editorial exercise, not professional verification, a trust test, or guaranteed management method. Name the result, executor's role, information available, authority, and freedom of method. What do they recommend, decide, and not promise? The note need not be public or used for risky objection; under threats actual support and protection options come first rather than proving independence through confrontation.

Separately name the goal and your own promise. Which requirements need professional information, and which priorities remain organizational choices? Who accepts cost, and whose participation cannot be promised for them? Where conditions are insufficient, clarify uncertainty before confidently announcing readiness. Assignment should create no hidden demand to decide everything or responsibility for unnamed expectations. Yet accepted commitments cannot be narrowed retrospectively for convenience either; changing them requires a definite new communication that preserves what others already relied on.

After completion, check correspondence with the agreed task and important limitations. If suitability needs professional checking, organize it for the actual task without presenting your reading as comprehensive expertise. Name the result obtained, nonperformance, and remaining question. Checking can end without new grounds for repetition. The exercise does not guarantee calculation accuracy, agreement, safety, or team trust; it helps identify more precisely what was transferred, accepted, and which current action someone actually answers for.

Key quotation

“Integrity of the individual”.

Follett, the same introduction, physical PDF page 2, lines 96–104; a phrase from her own discussion of human relations. It requires neither rejecting expertise nor declaring every organizational role coercion. Primary text.

Connections between traditions

Follett's historical social thought considers individuality and joint action. It supplies a limited moral perspective rather than evidence of our work distribution's effectiveness.

A professional assessment concerns properties and conditions; organizational authority concerns decisions within an accepted role; a promise concerns future support for another person. This is an editorial map, rather than evidence of agreement among all traditions.

Joint work does not erase different powers, and expertise does not establish every value. Respect permits clarification of subject; repeating checks requires actual grounds rather than serving as a compulsory ritual of independence.

Sources

Mary Parker Follett. Creative Experience, 1924. Accessible PDF identifies author and year; the exact print basis of this digital version is unestablished. Personally read selected continuous introductory physical PDF pages 0–4, extraction lines 0–181; quotation locations on page 2, lines 91–104. No whole-book reading or page-image inspection is claimed. Primary text.
Mary Parker Follett Foundation. Follett Writings. The personally opened bibliographical page confirms Creative Experience's identity and 1924 date; it is not independent validation of the philosophical position. Bibliographical guide.

Short Russian translations were prepared editorially from the English PDF. The museum scene is invented; the source establishes no engineering requirements, verifies no participant's qualifications, and validates no exercise.

About the authors

Andrey Zarinsh is the author and founder of United Wisdom. He explores inner transformation, the unity of traditions, intercultural dialogue, and contemporary spiritual discipline. Instagram: @andrey.zarinsh.

Maria Evilkut is a co-author of United Wisdom. Alongside her research, she seeks to test these ideas in practice and observe how principles of unity and mutual understanding work in real life between people and cultures. Instagram: @ilkutki.