Perfectionism: when quality becomes a condition for the right to act

Epigraph

“to be good”.

Aristotle, Nicomachean Ethics, II.9, translated by W. D. Ross. The Russian rendering is an original editorial translation. Text.

What this article is about

Imagine a professional preparing a proposal for a client. The content has been checked, the calculation agrees, and the email can be sent. Yet a feeling remains: it is not good enough. He rearranges paragraphs, seeks a more convincing headline, and checks what has already been checked. By the end of the day, the email remains unsent. The work really has become slightly neater, but the client still has nothing. Where does care for quality end and a condition that forbids action begin?

This article considers perfectionism through the structure of a standard: what counts as fulfillment, when stopping is allowed, and what a mistake comes to mean. High demands are not declared a problem in themselves. The question arises when the result's quality must also prove the person's flawlessness, protect them from every evaluation, and remove all uncertainty. Contemporary research helps distinguish aspects of this question; philosophical texts provide a language of measure and action. We preserve the boundaries between them and examine what a sufficient standard looks like in ordinary work, learning, and relationships.

I. A high standard does not yet explain the delay

A person can work carefully and on time. They may also be delayed by a complicated task, an unavailable answer, or a serious error they have discovered. “Perfectionism” should therefore not replace examination of the circumstances. If the proposal depends on confirming a price, waiting is reasonable. If the calculation is wrong, another check is necessary. An unpleasant feeling before sending does not by itself prove that the standard is excessive, and speed does not prove professionalism.

In our illustrative scene, it helps first to reconstruct the sequence. What was ready by noon? Which changes followed? What did they improve for the recipient? Had a deadline been agreed? The missing action becomes visible without explaining the whole personality. We can then ask why another check appeared after the previous criteria were fulfilled. A possible transition emerges: the demand placed on the document expanded into a demand to eliminate every unpleasant impression of its author.

The same high standard has different costs at different stages. Before sending a commercial proposal, important terms need to be accurate. An exploratory conversation may need only a clear hypothesis and a list of unknowns. If an early conversation must already contain the final solution, the person loses the opportunity to obtain the information needed for that solution. The standard blocks its own means of fulfillment. Rather than labeling it high or low in general, we can ask whether it fits this particular stage.

A similar mistake arises in relationships when discussion is permitted only after someone fully understands their feelings. They want to speak responsibly but wait for language that excludes every misunderstanding. The other person encounters silence and must interpret it. A limited, honest sentence can begin the exchange: I do not understand everything yet, but I want to discuss yesterday's conversation. It is not suitable for every case and guarantees no good response. It illustrates a principle: a beginning may need to be clear enough for continuation rather than finally perfect.

II. When evaluating work becomes evaluating the whole person

In Shafran, Cooper, and Fairburn's 2002 model, clinical perfectionism is considered through excessive dependence of self-evaluation on demanding standards despite adverse consequences. The authors also describe raising the standard after achievement. The abstract was read here; the model is not used to diagnose the reader. Our subject extends beyond clinical work and is limited to the structure of a demand in a particular situation.

The difference can be heard in two questions. “Has this calculation been checked sufficiently?” concerns the document. “If the calculation is disliked, do I have the right to call myself a professional?” concerns belonging to a role. The second question can introduce demands that technical checking cannot resolve: pleasing everyone, never encountering rejection, and never looking like a beginner. The work consequently carries a burden for which its criteria were not designed.

Imagine a young translator. An editor proposes replacing two inaccurate phrases. In one reading, this is part of professional exchange. In another, it proves that the translator was only pretending to be competent. Now the task is to defend an entire biography rather than understand the correction. The translator might spend a long time proving that the phrases are permissible, although correcting them is straightforward. Or decline the next commission to avoid another such exposure. Neither option follows automatically from two comments.

This working distinction does not declare a professional role an empty convention. Skills, duties, and sometimes official requirements stand behind it. A substantial mistake may indeed reveal a gap in competence. But the gap needs to be named, assessed, and addressed rather than used to infer universal unfitness. Specificity is especially important when decisions affect other people: insufficient skill cannot be corrected solely by a kind attitude toward oneself. It requires narrowing the task, training, assistance, or declining work that cannot yet be performed responsibly.

III. Research disagreement preserves an important distinction

Models of perfectionism are not all structured alike. Dunkley and colleagues, 2006 challenge combining high personal standards and self-critical evaluation into one construct. The available abstract presents analyses of two samples: 527 students and 236 patients. The authors emphasize self-critical evaluative tendencies. The full paper was not read; causality and individual predictions are not inferred from these results.

For the reader's question, this disagreement is more useful than a ready-made label. Someone can value quality, pursue large goals, and avoid turning every failure into a moral catastrophe. Someone can perform a modest task while demanding unattainable protection from evaluation. The goal's absolute height therefore does not reveal the whole construction. Its connection to self-treatment, stopping conditions, and consequences of the chosen working approach needs consideration. Here the article offers its own editorial distinction rather than reproducing a research scale.

Consider two musicians. One prepares a difficult program for months, accepts comments, and schedules a performance. Another does not permit themselves to play a short piece for friends until every possible mistake has disappeared. An observer might call both perfectionists. But understanding their behavior requires asking which standard helps action, which makes action impossible, and which risks a person reasonably accepts in a particular performance.

Of course, not every difficult program should be performed on its scheduled date. Preparation is sometimes genuinely insufficient. This is examined through relevant criteria: consistency in particular passages, shared rehearsals, program length, and venue conditions. This article does not propose replacing professional evaluation with a wish to feel freer. It proposes keeping professional evaluation distinct from a demand to become someone who can never receive a negative judgment. The latter has no achievable technical completion because another person retains their taste, expectations, and possibility of refusal.

IV. A standard needs a recipient

Quality relates to the result's purpose. A client proposal should make the terms understandable and allow a decision. A learning draft should give the teacher material for feedback. A letter to a friend should convey an important thought with sufficient honesty. If all these results are evaluated against one ideal of flawlessness, their different purposes disappear. The person perfects the form without asking who needs it and why.

The recipient is not always the best judge of quality. A client may request the impossible, a teacher make a mistake, or a friend misunderstand the chosen language. But the recipient's need helps establish the work's subject. We can discuss what is required, which properties are mandatory, and which improvements may follow later. The standard then becomes an agreement or a deliberate professional choice rather than an endless internal feeling. Concrete examples of the expected result and a clear distinction between necessary and desirable help here.

In an illustrative scene, a designer prepares an interface prototype. The user needs to test whether the next step is understandable. The designer spends the evening on shades and subtle animation because showing unfinished work feels frightening. Those improvements may be valuable in a finished product. But they do not answer the prototype's central question now. Sometimes an unfinished form makes a meaningful defect visible before it becomes expensive. This is an editorial working example, with no claim to a measured universal effect.

A useful standard can face a counterquestion: if this improvement disappears, what does the recipient lose at this stage? A concrete answer warrants consideration. If the only answer is that it will reveal the author's imperfection, the burden has shifted onto the author. This shift does not make the desire for excellence ridiculous. It simply shows that the document is doing two jobs. They can be named separately, and the reasonable cost of the second can be decided. Often that is enough to stop presenting personal protection as a mandatory property of every working version.

V. Measure differs from a convenient compromise

In Nicomachean Ethics, II.2 and II.6, Aristotle connects an appropriate measure with the circumstances of action. The mean relative to us is not simply an arithmetic midpoint. This philosophical argument supplies no formula for work quality and does not permit seeking a midpoint in every moral violation. We use a limited thought: more effort does not automatically mean a more appropriate action.

Checking half of a dangerous calculation does not make arithmetic compromise virtuous. Producing a tenth version of a safe introductory email does not guarantee value. Measure depends on the subject and consequences. “How much harder should I try?” can therefore be replaced with “Which uncertainty must still be removed?” Sometimes the answer requires lengthy work. Sometimes the remaining uncertainty concerns another person's future response and cannot be removed technically.

In an illustrative example, the organizer of a small seminar wants to eliminate every possible mishap. The room, time, access, and materials can be checked. Every participant's mood and an ideal discussion cannot be guaranteed. Attempts to manage the latter consume resources needed for the former. Adequate preparation also includes a reasonable response to a possible problem: a backup contact, flexibility in the order, and clear notification. It does not require proving that no problem will ever occur.

Yet measure cannot justify a decision afterward merely because someone was tired. Tiredness matters as a circumstance, but consequences and duties to others remain. If the work requires an unavailable check, its boundary can be honestly stated or delivery postponed by agreement. A substantial standard cannot quietly be reduced and this labeled maturity. The idea of measure thus becomes more demanding: it asks what each criterion protects, which risks remain, and who has authority to decide about them. A personal feeling of sufficiency does not replace authority in shared work.

VI. A standard that moves after achievement

Sometimes someone formulates a clear criterion beforehand, fulfills it, and immediately declares it too weak. This change may be reasonable when new information has emerged. Without new information, its function deserves examination. In an illustrative scene, a professional decides to send a proposal after checking the calculation and agreeing the deadline. Both conditions are met. Sending is now permitted only after proving that competitors have not prepared a better proposal.

Changing a standard need not be forbidden altogether. Living work often requires refinement. It helps to preserve a history: what changed, when, and why. The initial and new criteria can be placed side by side. If the new criterion protects a different interest, that interest should be named. Calculation safety and winning a competition concern different tasks. They may connect, but victory is not fully controlled by the document's quality. Even an excellent proposal sometimes loses for reasons the author does not know.

An impossible new criterion conveniently hides behind the word “professional.” But professionalism also includes finishing within known constraints, communicating unknowns, and making verifiable decisions. Someone who indefinitely holds onto a document is not necessarily caring more for the client. Perhaps they encounter less external feedback. This is a hypothesis, rather than an explanation of every delay. Asking what new information each additional delay actually brings helps test it.

Sometimes achievement leaves an unfamiliar emptiness: the task no longer demands effort, and the permitted pause feels unclear. There is no need to explain this automatically through a deep fear or childhood history. In the selected episode, noticing that the completion rule has again been canceled may be enough. It can be allowed to operate once, and its consequences examined. If the result is insufficient, that is new information. If things go acceptably, the earlier anxiety remains an experience that did not accurately predict the outcome. Both possibilities should stay open until the actual result.

VII. A draft is a distinct kind of good work

A draft has its own quality: it should reveal the thought, unknown, and location of a question clearly enough. Its shortcomings can be productive if they are precisely what needs discussion. But a draft does not release the author from honesty. A document on which another person will make an irreversible decision cannot simply be labeled preliminary while substantial uncertainty is concealed. An early version's usefulness depends on an agreed way of using it.

In Tao Teh King, §63, translated by James Legge, attention turns to dealing with difficulty before it becomes large. We use this normative image to compare early action. The Daoist text has its own language of Dao and non-action; it is not a project-management study and does not prove prototypes effective. Resemblance does not require erasing different foundations.

Imagine an author who does not know how to organize an article's central section. They can write several paragraphs and show the editor that particular fork. Or spend days perfecting the introduction, where uncertainty is already small. In the second case, busyness is not progress on the central question. An appropriate draft reduces the uncertainty for which it was created. A polished introduction can conceal that uncertainty, making the article look more complete than it is in substance.

In relationships, an early version of a thought also needs clear identification. “I am trying to understand what hurt me” differs from “I already know your true reason.” The other person may disagree or decline a conversation now. Draft status does not render all words harmless. It simply permits not demanding a final interpretation before encountering a response. Sufficient quality here means accurately naming boundaries: what I know, what I suppose, what I want to ask, and which conclusion I am not yet making. This language is sometimes harder than categorical accusation because it preserves actual openness.

VIII. A counterexample: flawlessness can be a required criterion

Some actions require very strict criteria, and this article does not propose experimenting with them for internal freedom. The rule for sending an ordinary draft cannot be transferred to operations where an error creates significant harm. The relevant work and authority determine the exact requirements. A general distinction suffices here: experiencing excessive demand does not cancel an actual mandatory standard. Sometimes the right response is to refrain from acting until a substantial check is complete.

The difference between justified and endless checking appears in its structure. A justified check has a subject, method, responsible person, and completion condition. If a new defect appears, the criterion reopens for an explainable reason. Endless checking can blur its subject: first a date, then style, then assurance against any evaluation. No result counts as sufficient. The number of checks alone cannot establish which construction is present.

A substantial error can also be distinguished from aesthetic dissatisfaction. In an illustrative example, a teacher prepares a learning resource. An incorrect rule must be corrected before class. An imperfectly composed illustration can probably wait if it does not obstruct understanding. This decision depends on the particular resource; the article supplies no universal list of acceptable defects. Order matters: properties on which use depends are protected first, then improvements considered if time and resources allow.

This counterexample does not weaken the central thought. It makes it more precise: the question is not whether any imperfect work may be permitted. The question is whether a sufficient standard for this action can be named and acknowledged as fulfilled. If it is unmet, further work is needed. If met, a wish to remove the remaining unpleasant feeling should not quietly become a new mandatory condition. Personal anxiety deserves attention but does not automatically have authority to rewrite the whole team's agreement, deadlines, and criteria.

IX. The cost of improvement belongs to a shared life

Every additional revision uses time and attention. Sometimes the cost is justified: correcting an important term protects the recipient. Sometimes the value is small, while the cost extends far beyond the document. Someone misses sleep, cancels a meeting, or delays a colleague to remove an almost invisible imperfection. This article infers no medical consequences from such an episode. It proposes including the resource spent and impact on others when assessing the quality of the decision itself.

In an illustrative scene, a professional makes an email more elegant while an assistant cannot start the next stage without the answer. The additional hour looks to the author like care for the result; for the shared work it may be an obstacle. If quality is evaluated only within one document, the cost disappears. It helps to ask whose action is delayed, which deadline is missed, and what improvement is obtained in return. Sometimes the answer shows that the best result for the whole requires completing one part earlier.

A similar cost appears in friendship. Someone wants to visit only in a good state, speak interestingly, and be entirely attentive. The wish is understandable, but if meetings are constantly postponed, the friend receives absence. Perhaps a brief, honest visit would now be more useful. Perhaps a pause really is needed. The decision depends on the agreement and both people's resources. An ideal of perfect presence should not conceal that absence also has consequences and may not be the other's preference at all.

The cost is not always measured in hours. Sometimes it is losing the opportunity to learn from a response unavailable before action. An author cannot know whether readers understand a thought until someone reads it. A teacher cannot see a learner's difficulty before setting a task. Waiting for complete certainty blocks a source of future clarification. But feedback is useful only under suitable conditions: the recipient knows the result's status, consequences are limited, and the author is prepared to consider the response. An unprepared action does not become reasonable merely because it is called learning.

X. Completion can be designed beforehand

A stopping condition is best discussed before the moment of greatest anxiety. Mandatory properties, desirable improvements, and time limits can then be distinguished. In an illustrative scene, a professional writes beforehand: I will send the proposal after checking price, scope, and deadline; I will check the email's clarity once; I will identify unknowns in a separate question. These rules do not guarantee a perfect document. They make the decision verifiable and prevent each new feeling from quietly becoming an order.

It helps to include a reopening rule. Which new information really requires returning to the document? An error in an important figure, a changed agreement, or a clear comment from the recipient. What is insufficient? Merely renewed doubt with no new subject to check. The distinction can be difficult, and someone may request a colleague's working assessment. But the colleague should not become an endless guarantor of infallibility. An agreed examination of the particular result is enough.

After sending, it helps to preserve what actually happened. The email was accepted, clarification requested, the client declined, or an error discovered—these are different results. Rejection does not automatically prove insufficient quality; acceptance does not make every line true. We can examine which comments concern the standard and which reflect an external choice. Such an examination refines requirements using information rather than raising the standard after every uncomfortable feeling.

Sometimes completion leaves anxiety. This need not mean that the decision was wrong. Someone can send a sufficiently checked document and continue doubting. This article does not demand ignoring doubt; it proposes establishing whether doubt has a new subject. If so, that subject is considered. If not, the selected action can remain complete for now. The stopping condition operates because relevant work has been done, rather than because absolute confidence has finally been felt. This is the distinction an endless standard often erases: it confuses the completeness of an action with a person's final protection from life.

What this means in practice

Choose a safe, bounded task that has been difficult to finish, and clarify its recipient and purpose. Write down several mandatory criteria and explain what each protects. Separately name desirable improvements. If the required standard is unknown, the first step may be clarifying requirements rather than trying to work even harder. Do not apply an experiment where you lack authority to change a substantial check.

Then consider the last revision you completed. Which uncertainty did it reduce? What changed for the use of the result? If the answer concerns only the impression of you, identify that separately. Wanting to look competent has an understandable place, but should not secretly govern every sending condition. One sufficient action can be chosen, with the new information that would reopen it for consideration specified beforehand.

After acting, compare prediction and fact. You may have expected general exposure and received two clarifying questions. You may have expected smooth acceptance and discovered a substantial error. In either case, preserve the actual event without forcing it into a story of completely overcoming perfectionism. Correct what really needs correction and refine the future standard. The experience of sending does not release you from further responsibility.

The professional at the beginning eventually sends the checked proposal. Several sentences remain that could be more elegant. The client receives the terms on time, and the author receives a response without which further progress was impossible. This is a bounded result, but it changes how work is organized: quality again belongs to the document's purpose. It no longer has to fulfill an impossible promise—to make a person invulnerable to every evaluation before granting them the right to act.

For recurring work, one example of a sufficient result can be retained as a reference next time. It is not a permanent ideal: it is refined when the task changes. Its usefulness lies in making the completion condition visible before another internal dispute begins. Sometimes the example also reveals which improvements belong to another stage and may therefore honestly be left until later.

Key quotation

“appropriate to the occasion”.

Aristotle, Nicomachean Ethics, II.2, translated by W. D. Ross, Internet Classics Archive. The Russian rendering is an original editorial translation. Together with the epigraph, fewer than 25 words are reproduced from this page.

Connections between traditions

Comparison	Resemblance and boundary
Shafran and colleagues ↔ a moving standard	Dependence of self-evaluation helps frame a question; the clinical model does not diagnose an everyday delay.
Dunkley and colleagues ↔ high standards	Goal height should be distinguished from self-critical evaluation; an abstract does not establish causality.
Aristotle ↔ sufficient measure	What fits depends on circumstances; virtue is not half of a check.
Tao Teh King ↔ early work	Difficulty is considered before expansion; the Daoist image is not an experiment with prototypes.
Professional checking ↔ permission to finish	Strictness is compatible with a stopping condition; safety is not lowered to obtain a feeling of freedom.

Sources

Shafran, R., Cooper, Z., Fairburn, C. G. (2002). Clinical perfectionism: a cognitive-behavioural analysis. Behaviour Research and Therapy, 40(7), 773–791. DOI 10.1016/S0005-7967(01)00059-6. Abstract read in the primary bibliographic resource's result; full text unavailable.
Dunkley, D. M. et al. (2006). Personal standards and evaluative concerns dimensions of clinical perfectionism. Behaviour Research and Therapy, 44(1), 63–84. DOI 10.1016/j.brat.2004.12.004. Metadata and abstract read, including the disagreement and samples; full paper not read.
Aristotle. Nicomachean Ethics, II.2, II.6, II.9; W. D. Ross translation. These sections read; used for the philosophical distinction between measure and circumstances.
Tao Teh King, §63; James Legge translation, Project Gutenberg No. 216. §63 and neighboring §64 read; comparison limited to early action, without transferring the whole doctrine of non-action.

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.