Confucius, Mencius, and Xunzi: A Disagreement about Human Beings

Epigraph

“If there are faults, one should not fear correcting them.”

Confucius in the Analects, 1.8, final sentence. Editorial English translation of our Russian rendering of Columbia’s English selection, Sources of Chinese Tradition, 1999; not a verified published English quotation. An individual translator for this excerpt was not established on the page.

About this article

A child sees another child hurting and reaches out. A minute later, they refuse to share a favorite toy. This hypothetical observation readily confirms any belief about humanity. One adult sees natural goodness; another sees selfishness requiring correction. But what do they mean by nature—a first feeling, a lasting capacity, or education’s outcome?

Confucius, Mencius, and Xunzi belong to one broad tradition of ethical education but explain its foundations differently. Confucius emphasizes learning, relationships, and ritual form. Mencius defends good human nature through the beginnings of ethical capacities. Xunzi maintains that untrained desires produce disorder and goodness requires deliberate work. Behind “humans are good” and “humans are evil” lies disagreement about what is innate, where norms originate, and how people change. We will examine this dispute in historical context, then consider its contemporary significance separately: for education, rules, hope for change, and responsibility toward others.

I. A shared question: becoming able to live with others

Early Confucian thought benefits from being read against political fragmentation and struggles for order. Its authors discussed people within relationships, education, and governance. Nature therefore was not a matter of rating moods: how many good and bad feelings do we have? It concerned arranging life so people could fulfill obligations and distinguish worthy action.

We know Confucius through complex transmission of sayings and narratives. The Lunyu, or Analects, is not a diary written by the teacher himself. The academic account of Confucius warns about varied early portrayals and difficulties reconstructing a single doctrine.

This does not make the text useless. It changes reading: we can examine attributed instruction, its argument, and later influence while remaining cautious about biography. Understanding an assertion matters more than adorning it with a confidently imagined scene from a sage’s life.

Mencius and Xunzi continue and contest this inherited work. They cannot conveniently become an optimist who loves people and a pessimist who wants to punish them. Both regard ethical change as possible and important; they differ in explaining initial capacities and their development.

Our contemporary example makes the question tangible. An organization wants employees to report errors honestly. It may appeal to responsibility, create procedures, or lead by example. These approaches are compatible, but their selection depends on how people and conditions of action are understood.

The organization is not an ancient state, nor a modern employee a participant in traditional hierarchy. Analogy explores a question without proving identical social orders. This distinction will matter throughout.

II. Confucius: ethics appears in the form of relationships

Two important terms are ren, humaneness, and li, ritual order. Li here means more than a grand state ceremony: it includes behaviors making respect, role, and relationship visible. In Columbia’s Analects selection, passages 1.1, 2.3, 2.7, and 3.4 connect learning with practice, governance with ethical example, care with respect, and mourning with actual feeling.

Is performing an outward act sufficient? Someone can provide a relative everything necessary while humiliating them in conversation, or perform an apology without acknowledging consequences and immediately repeat the action. Form is needed but does not exhaust relationship.

Conversely, someone may be certain of their love and deem expressions of attention unnecessary. Others cannot access that inner conviction; they encounter lateness, unanswered messages, and broken promises. Form is not an empty shell here. It helps feeling become a discernible human action.

This reading does not make Confucius a contemporary advocate of fully equal relationships. The historical system included hierarchies, family duties, and political roles. We may examine the value of form without declaring the old order binding today.

Ethical example matters too. In a hypothetical family, an adult demands respectful speech from a child while continually humiliating others. The spoken rule and operative example convey different messages. Our contemporary conclusion requires no myth of flawless adults: example’s influence makes acknowledging one’s own errors particularly important.

The Confucian question can therefore be heard as one of consistency. Which forms sustain a worthy relationship? What do our actions actually transmit? Where has teaching remained speech alone? Attention turns toward shared life rather than private beliefs about personal goodness.

Book two adds a distinction: governance through punishments is compared with ethical example and ritual form. In the first case people may merely avoid punishment; in the second, shame and inner change become questions. The Analects passage, 2.3, shows this concerns more than good manners.

Philosophically, external sanction addresses what follows a violation. Ethical education should change how the agent perceives the act itself. Stop deceiving only under observation and the conditions of advantage changed. Regard deception as an unworthy relationship and the foundation of choice changed.

This transition does not make law and sanctions unnecessary. Someone may reject the ethical ideal or exploit absent protection. “A good example replaces rules” would be premature. The passage’s interesting question concerns the quality of order: what guides people when immediate control weakens?

Ritual form is not covert surveillance. It makes relationships recognizable and structures participation. Yet the order itself needs assessment: teaching shame for opposing injustice produces obedience, not yet virtue. We must distinguish internalizing a worthy relationship from internalizing fear of rejection, rather than use Confucius to endorse any education.

III. Mencius: virtue’s beginnings still need cultivation

In the famous passage 2A6, Mencius offers a child threatened with falling into a well. Initial alarm and compassion are explained through immediate response rather than calculations of advantage or reputation. The institutional Mencius excerpt preserves this argument about responding to another’s suffering.

“Beginning” matters. Four beginnings correspond to humaneness, righteousness, ritual propriety, and wisdom. A sprout is not a stable mature virtue. The academic overview of Mencius emphasizes extending and developing these tendencies.

A contemporary reader may take an unwarranted step: if compassion is natural, simply leave it alone and everything will work out. But beginning and outcome differ. Spontaneously pitying a familiar child is easier than accepting an inconvenient obligation toward a stranger. One response does not reveal whether care survives fatigue, distance, and personal advantage.

Our hypothetical example is someone moved by another’s misfortune. They may feel strongly and do nothing, help quickly and lose interest, or discover what is needed and accept a manageable commitment. These differences show why ethical capacity is not reducible to emotional intensity.

The idea of sprouts preserves hope without a guarantee. We can seek grounds for development without declaring every action good already. Good nature here neither excuses inflicted harm nor obliges someone harmed to await unconditional change.

Moreover, the well argument is a philosophical example, not a modern experiment proving identical reactions in everyone. We can discuss its intuition’s persuasiveness and limits while distinguishing argument from empirical research.

IV. Extending care beyond the nearest case

Mencius gives ethical response important work of extension. In 1A7, a ruler can pity an animal while overlooking his subjects’ suffering. Connecting an already operative response with another significant situation is more than ordering greater feeling. It requires seeing a similarity previously unnoticed.

Why does nearby, intelligible pain move us while distant pain becomes a number? Distance, habit, insufficient information, or reluctance to change a comfortable order may matter. But we cannot determine a particular person’s cause beforehand. We must see how they describe both situations and where the difference really matters.

A hypothetical manager supports an exhausted friend yet treats employees’ overload solely as deficient motivation. They can understand fatigue within a particular relationship. Our contemporary question is whether an acknowledged criterion can extend to another role without erasing differences.

Extension does not mean identical help for everyone. Closeness, obligations, and resources remain significant. We can care for many while fulfilling different commitments. Interpret extension as immediately saving everyone and the impossible demand quickly becomes a reason to do nothing.

There is political significance too. Mencius considers living conditions: deprivation makes sustained ethical behavior difficult. The IEP account of Mencius places his teaching beside governance, not merely inner exercises in kindness.

A contemporary boundary matters here. Circumstances can explain difficulty without abolishing responsibility. Demanding personal responsibility must not conceal conditions systematically preventing worthy action. Both sides keep the question of humanity from conveniently blaming only the vulnerable.

For Mencius, care’s political meaning exceeds a ruler’s kind mood. The IEP discussion of governance considers the people’s priority and a ruler losing the ethical grounds of status, citing 7B14, 5B9, and 1B8. This is not modern democracy, but neither unconditional justification of power merely because it exists.

Our comparison highlights movement from a person’s quality to a role’s quality. Someone may occupy a ruler’s place without doing what ethically justifies it. “Ruler” therefore describes power and demands a use of it. Authority becomes assessable, not the final authority over every assessment.

This also changes good nature’s meaning. If an ethical beginning is only a private ability to pity, it can be separated from society’s arrangement. If it must develop through governance, decisions’ consequences for people beyond immediate feeling must be noticed. Moral growth demands a wider scale of attention.

Connecting good governance with personal virtue invites an objection. A benevolent ruler can err, receive distorted information, or depend on those interested in distortion. Contemporary application must therefore add checks and correction to questions about intention. We add this ourselves, not attribute a ready-made theory of modern institutions to Mencius.

The distinction preserves his demand’s force: position does not exempt power from ethical judgment. It also avoids making an ancient philosopher the author of every political solution we desire. A historical argument can be a substantive interlocutor even where its framework differs from ours.

V. Xunzi: goodness arises through deliberate work

Xunzi uses xing, nature, differently: what is given at birth without learning. Desires and initial tendencies alone do not create orderly shared life. Following them without measure leads to conflict; goodness requires deliberate activity, education, and form. Chapter 23 presents this in the Columbia excerpt.

“Human nature is evil” easily transfers an alien religious framework into the dispute. The SEP discussion of Xunzi specifically distinguishes his concept from Augustinian evil. It is neither inherited original sin nor innate enjoyment of inflicting pain.

Accurate comparison first asks whether the authors dispute one definition of nature. If one includes development’s direction under suitable conditions and another isolates untrained initial desires, “good” and “evil” concern different descriptions. This does not eliminate disagreement but changes its structure.

Xunzi can be read as defending the possibility of learning what is not ready-made. That differs from insisting “you are corrupted.” A contemporary person cannot argue calmly; this does not make them forever unworthy of relationships. Perhaps knowledge, forms, and practice are needed that “be yourself” cannot supply.

Education does not guarantee change through any pressure. A hypothetical adult suppresses a child and calls it correcting nature. One philosophical word does not turn power into wisdom. Means, consequences, and relationships need separate assessment. A teaching’s historical value grants no contemporary permission for arbitrary conduct.

Xunzi’s thesis is therefore more interesting than everyday pessimism. It investigates how human activity creates the possibility of worthy order. Even readers rejecting his assessment of initial nature can take seriously what cannot be expected without learning.

VI. Ritual and music: form can change desire

For Xunzi, ritual and music participate in ethical education. They do not merely constrain an already completed person externally; they shape discernment, feeling, and action. The IEP overview of Xunzi describes assimilating and performing form: a long path should make worthy action one’s own way of living.

In a contemporary hypothetical example, someone learns to listen through a rule: restate what was heard before objecting. Initially external, the rule gradually helps them notice another’s meaning before a ready-made reply appears. More than demonstrable behavior changed; the structure of conversation available to them changed.

This is our analogy, not proof of Xunzi’s doctrine. It only shows why form may exceed restraint. An organized action can teach us to see what habit previously missed.

Form also carries the opposite risk. Someone performs every required gesture while retaining contempt: thanks others formally while exploiting them, apologizes to close an issue, or observes meeting procedure without allowing objection. Outward correctness alone does not prove an ethical result.

Xunzi places strong trust in a particular cultural system created by sages. Contemporary readers can ask why this system deserves transmission, what happens when norms conflict, and how to assess a form sustaining injustice. These questions deserve arguments rather than disappearance beneath “respect tradition.”

A useful distinction remains: form can support feeling, educate it, or conceal its absence. Determining its function requires examining participants’ experiences and possibilities of correction. A historical term cannot perform that check for us.

In “Encouraging Learning,” Xunzi contrasts knowledge remaining between hearing and speech with learning changing the whole way of acting. Repetition is connected with understanding, surroundings, and steadiness; proximity to an educated person matters alongside reading. The Columbia primary selection shows a teacher helping interpret forms, while learning for outward display contrasts with work on oneself.

This clarifies deliberate activity. It is not merely a command of will to suppress desire. For a stable result, people need ways to discern, understand, and practice. Knowledge travels from a heard formula to something one can do without constant prompting.

Repetition and reflection cannot occupy opposite camps. Repetition without understanding may fix an empty habit; understanding without acquisition may remain beautiful speech. Together they explain why a developed capacity differs from initial inclination and momentary inspiration.

A teacher’s role matters equally. Text preserves a model without always explaining a new case. Someone demonstrating application adds what short prescriptions rarely supply. For contemporary readers, however, this does not make every teacher infallible. What they teach and whether they can explain correction must be established separately.

Xunzi links education with a cultural ideal; his position therefore differs from neutral skill training. He does not discuss equally successful acquisition of any life form. Borrowing repetition’s technique does not mean accepting his whole philosophy. Which culture deserves assimilation remains central.

Read more closely by distinguishing results: can someone display signs of learning, or has their action become more integrated? This protects education from competition in outward signs. But the demand must be checked in life, not by the self-designation “educated.”

VII. Where the real disagreement lies

Comparison becomes clearer through separate questions. What is initially given? Which capacities already have an ethical direction? Where does worthy behavior’s criterion originate? What conditions support growth? What work do habit, teachers, and social form perform?

For Confucius, relationships, learning, and practice form an important field. Mencius emphasizes developable beginnings; Xunzi consciously created ethical culture. These emphases reveal defects in contemporary discussion: feeling demanded without form, form without understanding, and development hoped for without conditions.

“All three really agree” would nevertheless be too strong. Substantive differences remain. Remove them and historical names become respectful symbols attached to our convenient system. Comparison matters more when it reveals each author’s principal problem.

Imagine a school. One teacher appeals to compassion, another creates rules letting everyone be heard, and a third shows how to acknowledge error. All may participate in one process. But if the first rejects rules, the second ignores understanding, or the third violates their own example, different difficulties appear.

Their compatibility within one school does not resolve philosophical disagreement. In our hypothetical case, a child learns to leave others time to answer. We observe a new action, but cannot infer whether an initial ethical beginning was cultivated or a previously absent cultural form acquired. Explaining change needs further arguments about what was already present and what learning contributed. A successful lesson cannot decisively prove any favored theory.

Different definitions of nature matter without making the Mencius–Xunzi dispute empty wordplay. Behind definitions remains the origin of ethical direction: why should development count as goodness growing rather than any inclination strengthening? Our example keeps this question separate from a rule’s usefulness. Practical cooperation is possible; declaring the explanations of humanity identical is premature.

Contemporary decisions do not require choosing a final theory of human nature beforehand. Sometimes identifying existing capacities, missing skills, obstructing conditions, and a form for testing change is more useful. This applies questions rather than declares an ancient school victorious.

Philosophy cannot be replaced by a practical recipe either. Historical disagreement still matters for ethics and understanding people. Our limited application gives it a place in life without claiming to settle the entire dispute.

VIII. Good nature, habit, duty, and grace answer different questions

“Good” and “evil” become clearer when we distinguish norms’ source from the path of change. Comparison is useful as a test of what each theory asserts, not a vote about humanity. Similar attention to education can conceal incompatible foundations.

Aristotle, Nicomachean Ethics II.1, connects moral virtue with habit. This offers a substantive meeting with Xunzi: developed qualities do not follow from simply approving initial desires. Habit itself, however, does not explain why this life deserves acquisition. Xunzi’s answer involves sages’ cultural work and ritual; Greek terminology cannot replace these foundations while preserving apparent agreement.

Another question connects Mencius with Kant: is immediate sympathy sufficient for moral worth? The academic account of Kant distinguishes duty’s motive from sympathy alone. Kant does not thereby become an opponent of care. The comparison reveals different grounds of moral assessment. For Mencius, response belongs to a developable ethical beginning; for Kant, obligation does not depend on that feeling’s presence. Both challenge kindness disappearing with favorable disposition, but explain the criticism differently.

Finally, “evil” does not make Xunzi a Chinese Augustine. The discussion of later Augustinian doctrine, §7.6 links ethical difficulty with the Fall and liberating grace. Xunzi’s dispute does not presume Adam’s inherited guilt. Combining these frameworks loses both deliberate cultural work’s meaning and Christianity’s own problem of salvation. One disturbing adjective does not establish a common anthropology.

This changes our approach to the opening child. We can notice compassion, study acquired habits, ask what is owed and how change is possible. We cannot derive a whole theory from one gesture and let it dictate every subsequent gesture’s meaning. Comparison opens questions rather than grants adults final authority to judge a person completely. The three Confucian thinkers’ disagreement should remain, even when their emphases help reveal one situation.

IX. Conclusion: nature’s question should not close learning

The three thinkers leave different paths into one difficult field: how people become capable of worthy action among others. Their dispute shows that ethical life cannot be fully explained by initial feeling, external rules, or a favorable self-opinion.

Mencius does not promise sprouts grow unaided. Xunzi does not declare people hopelessly corrupted. Confucian form is not substantive merely because correctly performed. Each conclusion entails work contemporary simplification readily omits.

We can preserve hope and checking, demand action without declaring someone an unchanging embodiment of deficiency, and learn form without justifying humiliation. These distinctions do not settle the ancient dispute but prevent premature verdicts through it.

The opening child still can pity and still can withhold a toy. Both observations matter. Adults must do more than choose a fact confirming their worldview: notice capacity, help it acquire form, check conditions, and show through action what they teach. The question of humanity becomes open to work again.

What this means in practice

Choose a small recurring episode: difficulty listening, a postponed promise, or a rule existing only on paper. Avoid beginning “Am I a good person?” It too quickly turns investigation into personal defense or accusation.

First ask what ethical distinction you already notice. Perhaps you understand why waiting without a reply is unpleasant. That is an existing ground of care. Then ask what action lacks: brief-reply skills, a clear deadline, permission to decline, or distributed duties. Feeling and performing are different capacities.

Choose a simple form expressing the needed relationship. In a hypothetical example, someone agrees to answer requests within a day, even if only to clarify a deadline. The form must not disguise a knowingly impossible promise. It has a purpose, conditions, and revisability.

Assess both sides. Did action change? Did the relationship become clearer to the other person? A completed procedure accompanied by more irritation and deception is not automatically useful. Stronger feeling with continued broken commitments has not become stable form.

The exercise does not judge another’s nature from one episode. Broken agreements may reflect confusion, inadequate resources, or unwillingness. These explanations lead to different decisions. Do not fill unknowns with an ancient name and call it understanding.

The right to stop remains. Hope for someone’s development does not oblige endless waiting. We can acknowledge possible change while setting boundaries on present participation. This distinguishes respect for human capacity from promising realization here, on our preferred timetable.

Key quotation

“Whoever holds firmly to inner strength can order themselves; able to order themselves, they can respond to others.”

Xunzi, “Encouraging Learning,” chapter 1, paragraph on inner strength; editorial English translation of our Russian rendering of Columbia’s selection, not a verified published English quotation. Ethical results face toward relationships here. This does not prove every strict education produces such capacity.

Connections across traditions

These are our analytical comparisons. Shared questions and limits are shown; matching formulations alone establish no historical influence.

Positions	Similarity	Boundary
Xunzi, nature ↔ Aristotle, II.1	Ethical qualities require acquisition.	Sages’ cultural work is not the Greek theory of virtue.
Mencius, 2A6 ↔ Kant	Morality cannot be judged only by convenient outcomes.	Sympathy’s beginning and duty’s motive ground it differently.
Xunzi in SEP ↔ Augustine, §7.6	Initial human conditions require ethical change.	Xing/e are not doctrines of inherited sin and grace.
Confucius, 2.7/3.4 ↔ Xunzi, learning	Form needs substance in relationships and action.	Ritual appropriateness and deliberate cultural activity have different emphases.
Mencius: developing beginnings ↔ Xunzi: education ↔ Confucius: transmission	Change is possible and needs conditions.	Good beginnings, untrained nature, and inherited practice are not one definition of nature.

Sources

Xunzi. Encouraging Learning, Asia for Educators, Columbia University. Chapter 1 selection from Sources of Chinese Tradition, 1999, pp. 161–164.
Confucius. Stanford Encyclopedia of Philosophy. §§1–2: varied sources and historical reconstruction’s complexity.
Selections from the Confucian Analects. Columbia University. 1.1, 2.3, 2.7, 3.4: learning, example, respect, and ritual.
Selections from the Mencius: On Human Nature. Columbia University. 2A6: the child by a well and ethical beginnings.
Mencius. Stanford Encyclopedia of Philosophy. Sprouts, extension, 1A7, material and educational conditions.
Mencius (Mengzi). Internet Encyclopedia of Philosophy. Ethical and political context.
Selection from the Xunzi: Human Nature Is Evil. Columbia University. Chapter 23: nature and deliberate activity.
Xunzi. Stanford Encyclopedia of Philosophy. §§2–3: xing/e, ritual, and music.
Xunzi (Hsün Tzu). Internet Encyclopedia of Philosophy. §§3–5: education, assimilating form, and cultural activity.
Aristotle. Nicomachean Ethics, II.1 and 6, translated by W. D. Ross. Primary text: habit and the mean relative to the person..
Kant’s Moral Philosophy. Stanford Encyclopedia of Philosophy. Academic discussion of duty and sympathy in the Groundwork..
Augustine of Hippo. Stanford Encyclopedia of Philosophy, §7.6. Academic discussion of sin and grace..

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 coauthor 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.