Comparing Ourselves with Others: Which Measure Have I Quietly Accepted?

Epigraph

“they do not all specify the same sort of merit”

Aristotle, Nicomachean Ethics V.3. A labelled clause from W. D. Ross’s English translation in the Internet Classics Archive.

What this article is about

Comparison often begins with a brief observation: someone else has more, works faster, or receives more attention. Then the subject quietly changes. Instead of completed actions, we discuss ability; instead of ability, our standing among people; instead of standing, our personal worth. Between these conclusions lie decisions about what to count, whom to compare, and why a ranking is needed. Those decisions can remain unnamed even when the numbers are correct. This article examines an accepted measure before it becomes a general verdict. Through an invented case, we distinguish volume, accuracy, speed, working conditions, and the purpose of assessment. Primary texts help pose different questions about comparable people, merit, the cost of an advantage, and available means. The aim is to understand what a comparison tells us, what information it lacks, and which next decision it can reasonably support.

I. Forty Records against Seventy-Two

Imagine a voluntary group preparing a catalogue for a small exhibition. Irina processes forty records; Pavel processes seventy-two. At a group discussion, Pavel’s result is mentioned first. Irina hears more than a report of quantities: she thinks she is less useful to the group. She disputes neither her forty records nor Pavel’s honesty. Yet a number becomes a description of a person before anyone names the question it answers. For the organiser, quantity may indicate progress towards a completed catalogue. For a participant, it may imply a rank. For a future visitor, the important question is whether an exhibit’s description can be trusted. These interests overlap, but calling everything “better” loses their differences.

The case is invented for reasoning, rather than offered as a portrait of a typical person. Irina works for two hours; Pavel works for four. An independent check finds thirty-eight of Irina’s records and sixty of Pavel’s usable. For now, assume equal task difficulty and one known standard of correctness. These are strong assumptions, which we will examine later. They reveal several orders within the same observation. Pavel contributes more records and more usable records. Irina contributes more usable records per hour and has a lower error fraction. None of these facts needs to disappear out of consideration for another participant. Their coexistence is precisely what a single ranking conceals.

The purpose is not to find a measure that makes Irina win. That would continue the same contest with a more convenient rule. We need to understand each measure’s use. If sixty descriptions must be added tomorrow, total output matters. When allocating a short shift, speed may matter. When a visitor might receive a misleading label, accuracy matters. Different decisions can draw on different evidence. The error begins when a local answer expands into an answer about the whole person: “fewer records” becomes “less ability,” then “less reason to belong here.” Those additional conclusions require grounds that the count itself does not provide.

II. Why This Person Became the Reference

In his 1954 article, Leon Festinger distinguishes evaluating opinions from evaluating abilities and discusses turning to other people when an objective, nonsocial basis for evaluation is unavailable. The opening also considers the choice of a suitable comparison person. Someone whose ability is too far removed may offer little information about one’s own. This is a theoretical account discussing research of its period. It establishes neither the cause of Irina’s experience nor a rule for every contemporary group. What helps here is the role of selection: choosing Pavel is already part of the assessment. The number of records does not make him the only possible reference.

Pavel may be relevant if both participants perform similar work, learn the same action, and have comparable experience. He may be less relevant if one enters checked, short labels while the other investigates difficult archival information. The difference in results remains real, but its significance changes. Irina might discover a need to organise her search better rather than evidence of lower ability. Moving from observation to explanation requires more information. A count does not disclose the tasks, who prepared the material, how many corrections were needed, or what help was available. Missing information gives us no warrant to replace it with a convenient story about either participant.

Checking comparability need not diminish another person’s achievement. We can recognise Pavel’s substantial contribution and ask what Irina wants to learn from him. Perhaps he identifies repeated fields quickly and follows an agreed sequence of entry. That is a specific skill he could demonstrate. “He is more successful than I am” leaves much more unspecified. Successful at what, over which period, by whose criterion, and for which purpose? A clearer question asks less of the other person’s position. Sometimes comparison remains useful. Sometimes it becomes unnecessary because the task already has a testable standard of completion. We can then check the work without placing the participants in a general order.

III. The Denominator Makes a Decision Too

Calculate the measures in our case. Forty entered records are fewer than seventy-two; thirty-eight usable records are fewer than sixty. Per hour, however, Irina produces thirty-eight divided by two, or nineteen usable records. Pavel produces sixty divided by four, or fifteen. For unusable records, Irina’s fraction is two out of forty, or five per cent. Pavel’s is twelve out of seventy-two, approximately sixteen point seven per cent. These calculations are compatible. A contradiction appears only if we assume they must all produce one overall rank. They answer different questions about the work, even though they use the same underlying counts. Changing the denominator changes what relationship becomes visible.

Hours reveal the rate of usable output. All entered records provide a basis for the error fraction. Yet usability itself needs examination: who checked correctness, which errors mattered, and was the rule applied equally? Giving a typo and a fabricated attribution the same “unusable” mark hides different consequences. More detailed categories could help, but a sophisticated table still cannot choose the organiser’s aim. A number becomes informative through its definition, rather than through extra decimal places. We need separate checks for arithmetic accuracy and the appropriateness of the relationship we chose. An exact calculation may answer a question that nobody actually needs answered.

Nineteen against fifteen also cannot tell us what the participants will produce tomorrow. We described one period under particular conditions. Task difficulty, materials, and available time may change. A bounded conclusion is that this measure was higher for Irina during this period. Assessing a stable rate would require further observations under comparable conditions. Completing today’s exhibition may not allow time for a long series, so the group might decide while acknowledging uncertainty. Honesty does not demand postponing every decision. It demands matching the decision’s scope to what is known, including what a short observation cannot establish about future performance or general ability.

IV. Agreeing about Merit Does Not Define It

In book V of the Nicomachean Ethics, Aristotle notes that people can agree that shares should follow merit while differing about what merit means. Ross’s translation names contrasting ancient political criteria. We do not adopt freedom status, wealth, or birth as modern measures of human worth. The useful point is the structure of disagreement. A shared word can conceal different rules. Two people say “fair,” but one means contribution, another need, and another an agreed equality of shares. Until those rules are named, an argument about the result may actually be an argument about what should count. Repeating the result more precisely will not settle that prior question.

Our organiser proposes thanks based on total usable output. Another participant suggests accuracy; a third suggests the time each could contribute. These are three proposals, rather than three readings of one compulsory number. The group can discuss the catalogue’s needs and the terms of voluntary participation. If equal thanks were promised beforehand, announcing a competition afterwards changes those terms. If the task is choosing a checker, accuracy may deserve particular attention. Aristotle supplies no ready modern procedure for the group. His distinction helps locate where the rule needs justification, rather than another calculation within a rule that has already been accepted without discussion.

A rule becomes clearer when its connection to a decision can be checked. “For the next checking task, we need someone who detects errors accurately” says more than “we choose the best.” Even a past error fraction may be insufficient: entering correctly and checking another person’s entries are different actions. A relevant trial or other evidence is needed. This clarification protects the person assigned on poor grounds as well as the person ranked lower. Pavel need not want to check records simply because he processed many. A broad compliment cannot establish the relevant skill, consent, or available time. The assignment requires its own conversation and evidence.

V. Equal Means and Different Opportunities

In Equality of What?, Amartya Sen distinguishes resources from what people can do with them. He retains difficulties of selecting and weighting capabilities and presents a partial moral guide. For our topic, a short question follows: giving the same means does not fully describe the same opportunity to perform a task.

Add a detail to the invented catalogue. Everyone receives access to one database, but it is available only in a reading room at specified times. Irina has two suitable hours; Pavel has four. Access is equal in a narrow sense: both are registered and can see the same fields. Possible use time differs. We need invent no illness, family circumstances, or hidden motivation. The known schedule is enough. Total usable records fit the question of contribution over a whole shift. Work per available hour requires another denominator. Choosing a question does not remove unequal opportunities, but it can prevent us from treating them as an already established difference in ability.

An assessment need not become a complete account of everyone’s life. A decision calls for conditions that could change its grounds. If correctness depends on an archival card, check the card. If a shift is being assigned, establish available time. If help with a particular technique is needed, discuss that technique. Additional questions gain relevance from their connection to the task. Gathering endless personal details can cross a participant’s boundaries without providing a complete causal picture. Someone may give only the necessary information: “I am unavailable during those hours.” A timetable need not become a biography offered in defence of the person’s right to participate.

VI. An External Advantage Can Have a Price

In section 25 of the Enchiridion, Epictetus considers preference at an entertainment and an invitation to an influential person’s supper. The actual version labelled with Elizabeth Carter’s name describes its price as praise and attendance. Someone who did not pay has retained the choice not to praise a person they dislike. This does not mean all recognition is bought with flattery. It is a particular ancient example that makes an exchange visible. Comparing the external outcome alone can conceal the means of acquiring it. Missing an invitation then appears as pure loss, although declining its terms may preserve another good. We need to consider what is acquired and what is given up.

In our catalogue, the price can be ordinary and honest. Pavel spends two extra hours in the reading room. They help the group and leave less time for another activity. Irina chooses a smaller contribution alongside another task. We do not know whose choice is better for an entire life, and need not decide. She can acknowledge Pavel’s advantage in total output without expecting the same output without the same expenditure. He can acknowledge her accuracy without treating his extra time as a mistake. Considering costs enriches comparison when it describes actual means, rather than hunting for a moral defect in the more visible participant.

This does not assume that every price is freely chosen. Duties, limited access, and institutional rules can impose real constraints. Stoic reasoning does not replace their examination or make exclusion harmless. In a voluntary case, however, we can distinguish wanting a result from wanting to preserve an acceptable way of participating. Sometimes these aims fit together; sometimes one must choose or seek a different arrangement. Wanting another person’s outcome may be reasonable, but it does not tell us whether we would accept its conditions. That question often disappears when we ask only why someone else has something we do not. Naming it does not settle the answer for us.

VII. Repairing an Error Differs from Distributing Honour

Aristotle distinguishes the distribution of shared goods from rectifying injury between people. In the selected V.4 passage, rectification follows a different principle and considers the injury’s character. We do not turn this ancient argument into present legal advice. The catalogue needs only the conceptual distinction: thanking a contributor and correcting a false label serve different purposes. Even a participant with high total output must correct a particular inaccurate entry if they undertook its preparation. Their rank does not erase the error. Nor must someone with lower output treat the other’s mistake as proof of their own usefulness. The record needs repair regardless of their relative positions.

Suppose one of Pavel’s labels names the wrong artist. Repair requires establishing the correct name, replacing the label, and, where necessary, notifying anyone who used the misinformation. Calling Pavel a generally careless person clarifies none of these operations. Such a verdict might appeal to Irina if she felt ranked lower, but it does not fix the catalogue. A broad compliment cannot fix it either. A useful description remains local: this attribution is wrong, here is the checked basis for replacement, and here is the person who will make the change. A real problem receives a specific address instead of becoming another occasion for allocating overall personal judgements.

This distinction does not remove responsibility; it makes responsibility more precise. Repeated substantial failures can justify revising the checking process or someone’s role. That requires relevant evidence and a clear rule, rather than protecting everyone from every consequence. Respect can coexist with restricting an assignment. Thanks for completed work can coexist with requiring unfinished work to be corrected. A single overall rank can make us choose between unconditional approval and complete rejection. Separating the questions leaves more specific actions, each justified through its own object. The group can address what happened without having to settle a total assessment of the participant first.

VIII. Another Person’s Rank and My Task

In section 23, Epictetus warns that seeking to please an external observer can redirect one’s life. This is a Stoic requirement for philosophical conduct, rather than an experiment about praise. Our editorial application is narrower: comparison can quietly change a participant’s task. Irina came to help produce a reliable catalogue but begins aiming primarily to outrank Pavel. Working on descriptions becomes a means to another result. Perhaps competition suits her and was agreed beforehand; then it is a separately chosen aim. The difficulty arises when an unnamed new aim directs her effort and the original purpose of participation disappears from discussion. A scoreboard alone does not reveal that change.

Look at the next action. Irina selects the easiest cards because they quickly increase her count. If the group needs difficult, unverified descriptions completed, high personal speed can contribute less to the current bottleneck. The arithmetic remains correct; what is being optimised has changed. Entered rows are convenient to count, but their relationship to a finished exhibition depends on the remaining work. Someone taking a difficult case may look worse on the counter while helping more with the actual obstacle. Recognising this requires knowledge of the process and its intended result. Positions in a table cannot provide that knowledge on their own.

Returning to the task does not mean Irina’s feelings must immediately change. She may still want recognition and notice Pavel’s result. Yet different actions become available: asking for feedback, proposing that difficulty be considered, learning a useful technique, or clarifying her available time. Each addresses a particular question. “I am worse” provides no equally clear address. If the group actually humiliates participants or changes promised rules, examining an internal measure cannot exhaust the problem. Relationships and rules need discussion. Checking a criterion can help name that problem, but cannot substitute for other people’s responses or guarantee agreement. A clear question still has to enter a real conversation.

IX. Discussing the Measure without Reading Minds

Suppose Irina asks why quantity was mentioned first. She need not begin by claiming she was deliberately devalued; the intention is unknown. She can describe what she heard and ask a specific question: “We listed volume. When choosing a checker, will we consider accuracy and checking experience?” This does not forbid the organiser from valuing volume. It asks how a report connects to a future decision. If no roles were being allocated, part of Irina’s inferred conclusion loses its basis. If roles were being allocated, there is a concrete subject to discuss. Either factual answer provides more useful information than a guess about the organiser’s private attitude.

Her request could be narrower still. If she wants to understand her work, she can ask to see the two errors and the correctness criterion. Feedback on a record reveals an action she can change. If she wants to enter records faster, she can ask Pavel to demonstrate his sequence. He may agree or decline; declining does not by itself prove a negative opinion of Irina. Different requests have different recipients and possible answers. A clearer measure helps keep them from merging into a demand for a skill, reassurance about personal worth, an equal role, and guaranteed goodwill from every participant. No single answer is responsible for supplying all of these.

Sometimes no adequate answer is available. The rule is vague, data are incomplete, or participants will not agree. The measure can then remain provisional and the decision bounded. The group may use total output for today’s planning without inferring lasting ability, then clarify checking later. Irina may decide how much more time to offer from the known remaining work. Lacking an ideal scale does not force her to accept any offered scale as truth. It also does not make all results equally good. Between those extremes remains work with specific information, possible action, and the limit of the conclusion that the information supports.

X. Comparison after Its Purpose Is Checked

Return to forty and seventy-two. These counts remain correct in the invented case. We have neither diminished Pavel’s contribution nor declared Irina the winner under a nicer rule. We distinguished volume, speed, accuracy, and available time; connected a measure to a decision; examined conditions needed for explanation; and separated repair from thanks. Comparison can now be a useful instrument. It may show where a skill, help, time, or clarified rule is needed. It becomes less useful when asked for an answer absent from the data, such as an overall measure of a whole person. Better arithmetic cannot supply that missing connection on its own.

The primary texts perform different work. Festinger offers a psychological theory discussing research of its era. Aristotle examines normative relationships and distributive grounds. Epictetus considers a chosen good and its price. They are not three confirmations of one modern causal law. Bringing them together helps separate questions that an everyday phrase merges. We read selected passages rather than claiming entire works. The catalogue’s numbers and conversations are invented for this article. They establish nothing about the frequency of such experiences, the behaviour of all groups, or the effects of the optional practice. Their value lies in making the reasoning inspectable, with its assumptions and steps stated.

“Which measure have I accepted?” preserves both standards and the freedom to reconsider. Where accuracy matters, acknowledge an error. Where time differs, avoid treating time as ability. Where a role is at issue, test that role. Where a rule lacks grounds, ask its purpose. If competition is chosen, name its object and the cost of joining. Another person’s result can then remain their result: something to learn from, admire, feel conflicted about, or recognise as evidence of a real problem. Making it the measure of an entire life requires a further decision. A count does not make that decision for us, however forcefully it is announced.

What this means in practice

If you wish, choose a recent comparison with modest consequences. You need not select a painful situation or disclose personal details. Describe the observation without an overall rank: what the other person did, what you did, and which data are known. Name the unit and the question it can answer. “Number of completed tasks” and “readiness for a particular role” may require different evidence. If division is involved, write out the numerator and denominator. Check whether choosing a convenient measure has changed the original question.

Add one known condition and one unknown. Do not fill the unknown with a story about another person’s ability, effort, or attitude towards you. Ask which clarification could alter the next decision. It might concern correctness, available time, or a demonstration of the relevant skill. Choose a small action: check an error, ask to see a technique, clarify a rule, or keep the conclusion bounded. You can also recognise another person’s achievement without drawing a general verdict about yourself.

This is no psychological questionnaire and promises no change in feeling. You can stop if the question is unsuitable now. A sufficient result is a more precise account of the measure and its purpose, even without a final answer. Keep two sentences nearby: “This measure establishes the following” and “The remaining questions still need grounds.” They help identify both the information a comparison supplies and the point at which it is asked to decide more than has been checked.

Key quotation

“It is sold for praise; it is sold for attendance.”

Epictetus, Enchiridion 25, actual MIT rendition labelled as Elizabeth Carter’s translation. “It” refers to the influential person’s supper: a particular example of an invitation’s price.

Connections between traditions

1. Observed difference and selected merit. Festinger distinguishes objects and grounds of evaluation in the opening; Aristotle V.3 notes differing meanings of merit. Together they let us name the assessed property before discussing an order. A psychological account of turning to others does not establish the moral correctness of a criterion.

2. A suitable person and the result’s price. Festinger’s selection of a comparable person and the invitation’s price in Enchiridion 25 reveal different conditions of a result. An outcome is more informative alongside known means and costs. This proves neither that success requires flattery nor that differing costs make every comparison unfair.

3. Exact proportion and justified measure. Aristotle V.3 examines four terms in a proportion; Festinger discusses an objective basis for assessing ability. Defined quantities allow an exact local answer, but selecting the quantities needs separate grounds. A ratio becomes neither a whole-person score nor a calculation of justice through psychological theory.

4. Repair and a chosen role. Aristotle V.4 separates rectification from distribution; Epictetus 24 considers useful participation compatible with fidelity. A specific error can be repaired while a future role is discussed separately. Stoic choice does not erase real loss, and an ancient account of rectification does not replace present law.

5. Means and opportunity. Sen’s selected argument distinguishes resources and capabilities; Festinger examines comparable people. Their connection suggests checking known task conditions. It computes no capability index and establishes no individual cause.

6. A conventional measure and a governing aim. Aristotle V.5 connects exchange comparability with demand and agreement; Epictetus 23 and 25 consider external approval and its price. A common measure can work for a named task without becoming a life’s governing principle. This is an editorial comparison of different questions, asserting neither historical influence nor a single causal mechanism.

Primary sources

Leon Festinger, A Theory of Social Comparison Processes, Human Relations 7(2), 117–140, 1954, DOI 10.1177/001872675400700202. The opening of the retyped primary text, selected qualifications, and complete Summary were actually read; this article reuses genuine I05 reading. Underlying experiments were not separately read. The theory establishes no personal cause or universal rule for contemporary groups.

Aristotle, Nicomachean Ethics, V, W. D. Ross’s translation in the Internet Classics Archive. V.2–4 and the central opening of V.5 were actually read, including merit, proportion, rectification, and exchange. The epigraph is a labelled V.3 clause. This electronic version’s print edition is unverified; Greek was not collated. Ancient political criteria are described within their context.

Epictetus, The Enchiridion, Internet Classics Archive rendition labelled as Elizabeth Carter’s translation. Actually read sections 23–25 are used; the invitation-price passage was freshly read for this article. Modern wording prevents treating the page as a diplomatic reproduction of Carter’s eighteenth-century edition. The Russian key quotation is the author’s rendering of this English text.

Amartya Sen, Equality of What?, Tanner Lecture, Stanford, 22 May 1979, original English. The opening and selected continuous final argument, printed pages 215–220, were read. A brief distinction retains its partial status. The middle and cited works were not separately read. No direct Sen quotation is used.

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.