Warp, Weft, and Way

Chinese and Comparative Philosophy 中國哲學與比較哲學

Confucius on the family as model

Many hold that for Confucius the family is the model for organized political society in some sense; that Confucius regarded the norms for relations beyond the family as largely based on the norms for relations with kin.  Here I follow Joseph Chan in challenging that view.

Chan argued against the idea that for classical Confucianism generally, the state is modeled on the family, is “the family writ large.”  The differences are too deep to allow that reading. For example, in classical Confucianism, the reverence we owe to our lord is conditional upon his character and conduct, while the reverence we owe to our parents is not.  In the family the emphasis is on the duties of a son to his parents (a lower position relating upward), while in the state the emphasis is the duties of governors in relation to the people (an upper position relating downward, or at least displaying downward). Chan makes this case, and says much more of great interest, in “Exploring the Nonfamilial in Confucian Political Philosophy,” in H. Chaihark and D. Bell eds. The Politics of Affective Relations: East Asia and Beyond (2004), pp. 61-74.

Here I look at Confucius only.  The argument is in my prepared Comments below this post, in several Parts.  Part A tries to catalogue the similarities Confucius saw between family and state—looking mainly for similarities he actually notes or alludes to.  Of course, one should expect him to see major similarities even if he never thought of the family as a model for the state, because family and state must each reflect the general human condition.  Indeed, major similarities would be consistent with the idea that the family is modeled on the state.  Part B describes some major differences between Confucius’ view of the family and his view of the state. Part C explores what philosophical use Confucius may have made of the similarities, by looking critically at a detailed statement by P. J. Ivanhoe on that topic. Part D sums up.

I argue based on Confucius’ own statements in the Analects. I adopt the working hypothesis that his statements there, especially in Books 1-15, are mostly authentic and largely representative of his views.  A better study would consider other reliable ancient records of his words; I’m not sure what those would be.

I think the idea that Confucius’ statements in the Analects are representative of his views is more plausible insofar as the views expressed there are consistent—especially if the view we find in those statements differs from the recorded views of his successors, who are the main potential source of inauthentic sayings in the Analects, and differs from what we find in the general run of probably inauthentic statements elsewhere.

Even a general presumption of authenticity cannot justify our putting great interpretive weight on any one remark by Confucius, nor especially on a bold reading of one remark.  Even if we could be certain of the authenticity of every passage not bearing positive evidence of inauthenticity, nevertheless in most cases we do not know the context or addressee to which Confucius’ suggestive remarks may have been tailored.  We know that he sometimes tailored his remarks on filial piety to his interlocutor in ways that could potentially be very misleading about his general view of the relation between filial piety and broader aspiration (2.25, 11.22, 2.21).

One point commonly cited to show that Confucius saw family and state as similar or analogous is that his tradition saw them as similar or analogous. Indeed, power was held by families as much as by individuals: a point on which Confucius did not comment, and on which one can well imagine him having mixed feelings. Unfortunately, I am not competent to discuss such features of the tradition before Confucius; I must focus on the Analects.  Granted, Confucius broadly endorsed his tradition.  But surely we should not conclude that a certain idea has a certain leading role in Confucius’ thought if, say, we do not find Confucius himself alluding to that role nor even with certainty find him alluding to that idea, but do find him accepting a number of points that seem to oppose the idea.

As usual, I take all my Chinese texts, and Legge when I use him, from the Chinese Text Project (ctext.org).Hannah Pang detail

June 17th, 2016 Posted by | Analects, Chinese philosophy - 中國哲學 - 中国哲学, Confucianism, Confucius, Filial piety, Roger Ames, Role Ethics | 22 comments

22 Responses to Confucius on the family as model

  1. Bill Haines says:


    Confucius never directly says that the family is analogous to the state, nor that some aspect or feature of the family is similar to some aspect or feature of the state (government or society). But there are other ways to allude to the idea that two things are similar in some respect, or to display that one thinks of them as being similar in some respect.

    A.1 . . . . Both important

    In 11.24 Confucius speaks of “assassinating father or lord” (弒父與君) as two especially heinous acts. That is one similarity between family and state, but it is too universally acknowledged to indicate any distinctive philosophical view (cf. also 14.15-17).

    Confucius sometimes mentions the clan (家 or 宗族) alongside the village (鄉黨) or country (邦) as arenas in which one can be famous or distinguished, or act well or badly (12.2, 12.20, 13.20). He sees such very extended family groups and political units as being similar in that each is a significant kind of group in which a person might act and be known.

    But it might be misleading to use the term “family” in discussing such a view. When we think of “family” today, for most of us the conventional core image may be of two notionally equal parents and the children they are raising, sharing a home as its only residents, and joined by a selection of next-closest kin on holidays.

    Maybe the conventional core image is mainly of a mother and father, children, aunts and uncles, and grandparents; rather than of, say, a husband and wife along with their parents, siblings, in-laws, and children; or of a set of adults along with their children and children’s spouses, and their grandchildren, nieces, nephews, great-nieces and great-nephews. That is to say, maybe our core image leans toward being from the point of view of children. The child’s family is more matrix than responsibility or pride.

    I wonder whether Confucius had a word for “family” that conjures up an image remotely resembling ours. His core image of clan 家 may have been a number of men united by lines of patrilineal descent, occupying a number of households (along with any dependents) — yes? I gather his term 室 meant house or household (cf. the Greek oikos); I suppose even one person living alone (with servants) would constitute a household 室 — yes? The term 室, like the English “house”, could also mean a powerful or noble clan. Unlike Aristotle, Confucius seems not to have a word or phrase for “community” or “social group” that could apply to kin-groups as such and could also apply to some non-kin units such as villages or states.

    I gather that for Confucius, filial piety 孝 was mainly about how a man relates to his father or parents, living or recently deceased. We might call it a virtue toward parents and, at the semantic periphery, toward ancestors. But it does not follow that he thought of filial piety 孝 as a family virtue. Is there any reason to think that Confucius thought in terms of a category “clan virtues” or “household virtues” or “kin virtues”? Beyond the word xiao 孝, I do not know that he has or uses any other word for a virtue that would fall into such a category, (unless we count the use of 父 and 子 as honorary verbs at 12.11).

    Any important department of life, I suppose, would be subject to some rules of ritual, perhaps even ritual specific to that department. Confucius thought there were ritual standards relevant to relations among kin. He occasionally speaks of sacrifices and mourning rituals, and these at least would have been specific to kin. In just one four-character string he implies that there is ritual applicable to some relations among living family members in non-mourning contexts—specifically, there is ritual applicable to a man’s serving his parents (2.5)—though it is unclear how much of the ritual Confucius has in mind here was specific to family.

    A.2 . . . . Family imagery used for non-family relations

    One way to point to an analogy or similarity between A and B is to use the language of A in discussing B, as when we speak of founding fathers, the mother country, sister cities, or brotherhood; or, conversely, when we speak of governing our children, or say that a person’s home is her castle. If someone speaks of A in the language of B, presumably she is thinking of A on the model of B, at least for the nonce.

    Confucius uses various kinds of figurative language in discussing governance, public life, and general moral education. He speaks of the North Star (2.1), wind and grass (12.19), plants (9.22, 9.28), a sacrificial calf (6.6), vessels (2.12, 5.4), pulling a cart (2.2), sharpening tools (15.10), carving, troweling and trimming (5.10, 5.2) and stuff v. pattern and adornment (6.18, 9.2).

    He uses family imagery too, though there may be room for argument about what terms we should count as family imagery.

    One way Confucius arguably uses the language of family for something in government is to use the term “son of Heaven” (天子) as a term for the emperor (once in a quotation at 3.2, and once at 16.2).

    Aside from 天子, there is just one place in the Analects where Confucius uses a family term to speak of something pertaining specifically to government, states, or office.

    The Master said, “Lu’s governance and Wei’s governance are brothers.”

    It would seem that Confucius’ main point here is that Lu and Wei have similar traditions of governance, or similar current policies. The governance in Lu is kin to the governance in Wei. (Confucius might have said this toward suggesting that understanding Lu can help qualify someone to serve in Wei. He makes a similar point about the relation between Lu and Qi at 6.4, without the family metaphor.) “Brothers” here seems to stand mainly for similarity—and perhaps also for common origin as suggesting similarity.

    Still, there may be a background allusion to the idea that since the states are both notionally fiefs of Zhou, they must not be enemies. (Confucius may have wanted to suggest that someone who has served one of these states can be trusted in the other.) This possible background point would indeed involve using a family relation as a model for conceiving a political relation.

    We have completed our survey of how Confucius applies the language of family to governmental matters.

    (Conversely, Analects 2.21 arguably shows Confucius using governmental language for family relations, saying that filial piety and friendliness with brothers amount to “participating in government” in some figurative or extended sense. The suggestion might be that there is some important (but unspecified) similarity between filiality and fraternity on the one hand, and participating (well) in government on the other. But I argue here that this passage is not good evidence of Confucius’ views.)

    There are several passages in the Analects where Confucius uses a family term to refer to ways of relating that are outside the family, if not governmental. He speaks of his own followers as 二三子 and 小子, and (as I think) 弟子. And he complains that although Yan Hui thought of him as a father, he was not able to bury Yan Hui as a son (11.11).

    I am inclined to think we should not regard 君子 as a figurative use by Confucius of family language for a non-family thing. I am not sure of an original literal meaning involving “son” — does the term originally refer to the son of a lord, hence perhaps to members of powerful clans, powerful people in general?

    Aside from 君子 and those cases involving the master-disciple relation, there are just three passages where Confucius uses what is arguably a family term to speak of non-governmental relations outside the family. The terms are 弟 and 孫 (in senses later standardly marked by 悌 and 遜, which do not appear in our Analects; at least not at the Chinese Text Project):

    The Master said, “At home, a young man must respect his parents; abroad, he must respect his elders. ….” (Leys)
    子曰:「弟子入則孝,出則, …。」

    … “And next to that, if I may ask?”
    “His relatives praise his filial piety, and the people of his village praise the way he respects his elders. ” … (Leys)
    … 曰:「敢問其次。」曰:「宗族稱孝焉,鄉黨稱焉。」…

    Yuan Rang was squatting on his heels, and so waited the approach of the Master, who said to him, “In youth not humble as befits a junior; in manhood, doing nothing worthy of being handed down; and living on to old age – this is to be a pest.” With this he hit him on the shank with his staff. (Legge)

    In these three passages, Confucius uses two terms (弟 and 孫) that may embody conventional family metaphors for ways of relating outside the family. Of course, the mere fact that he uses these terms in a common application is not evidence of his placing any weight on any conventional metaphor they may embody, which in turn may be thin.

    There is no evidence that the terms 弟 and 孫 were important for Confucius as names of ways of relating specific to family roles, such as the roles these graphs can name. The three passages just quoted are the only Analects passages in which Confucius uses the term 弟 to name a way of relating to others. In at least two of the three, he uses the term specifically for relations outside the family. (In the Analects I suspect there is no case of anyone using 弟 for a virtue special to brothers; indeed I wonder whether the term was ever used in that sense as early as Youzi’s time.) As for 孫, there are other Analects passages where Confucius uses it to name a way of relating to others: 7.36, 14.3, 15.18, 17.25; and Zigong uses it at 17.24. Throughout, the reference is pretty clearly not to family relations as such.

    We have completed our survey of how Confucius uses family language for non-family relating. So far as we can see in the Analects, he uses live family imagery for things outside the family only when (a) in a quotation and a late book he refers to the emperor as the Son of Heaven, (b) once he calls two approaches to governance “brothers,” and (c) several times he uses the father-son metaphor to describe his relation to his students.

    Thus I would disagree with Roger Ames’ suggestion that for Confucius, family is “the governing metaphor” (Confucian Role Ethics (2011), p. 96).

    A.3 . . . . Parallel phrases juxtaposed

    Two passages quoted in the previous section, 1.6 and 13.20, may suggest an analogy or similarity otherwise than by mere vocabulary. These passages each juxtapose parallel phrases involving filial piety 孝 and elder-respect 弟. Thus Confucius’ remarks at 1.6 and 13.20 may well aim to remind his auditors of the existence of some kind of similarity between filial piety and elder-respect—perhaps the obvious one, that each involves respect for some elders.

    (An interesting question for another day would be: When Confucius juxtaposes grammatically parallel phrases, as at e.g. 2.24, to what extent and in what ways are the juxtapositions significant? How does he tend to use that trope?)

    One interesting possible difference between filial piety and elder-respect is that while filial piety pertains to a relationship, elder-respect might apply even as a way of relating to a person one has not seen before and does not expect to see again. But perhaps that situation would be very unusual; or perhaps that difference was not a major concern for Confucius.

    It is not clear in these passages that Confucius is thinking of elder-respect on the model of filial piety (or vice versa). What seems clear is at least that in the first two of the three passages he is supposing that there is at least some thin similarity, and using that at least to give his statements some elegance, which should aid memory and suggestion.

    A.4 . . . . Service as a point of similarity between some upward roles

    In two remarks, Confucius uses parallel phrases in juxtaposing serving 事 superior officials with serving 事 one’s father or one’s 父兄 (father and elder brothers, or certain males of the father’s generation). In this way Confucius notes that relating upward to officials has that point of similarity with relating upward to certain elder kin.

    The Master said, “Abroad, to serve the high ministers and nobles; at home, to serve one’s father and elder brothers; in all duties to the dead, not to dare not to exert one’s self; and not to be overcome of wine – which one of these things do I attain to?” (Legge)

    The Master said, “My children, why do you not study the Book of Poetry? The Odes serve to stimulate the mind. They may be used for purposes of self-contemplation. They teach the art of sociability. They show how to regulate feelings of resentment. From them you learn the more immediate duty of serving one’s father, and the remoter one of serving one’s prince. From them we become largely acquainted with the names of birds, beasts, and plants.” (Legge)

    One might think we should give these passages little weight; for 17.9 is in a late book and arguably shows internal signs of inauthenticity (though translation to English seems inevitably to exaggerate that appearance). And there is a special difficulty about the authenticity of 9.16.

    But Confucius does use the term “serve” 事 elsewhere in connection with each relation separately:

    Inside the family:
    In two other passages Confucius speaks of serving 事 parents (2.5, 4.18). Also, after Zilu asks about serving the spirits of the dead, Confucius mentions serving people 人 (11.12), and may be thinking of family (cf. 11.22).

    Outside the family:
    In many other passages Confucius speaks of serving a superior in government (3.18, 5.16, 8.20, 11.24, 15.10, 15.38, 17.15, and arguably 1.5; and at 13.25 as a correlative to 使, and at 3.19 following an interlocutor duke who uses 事 and 使 as correlative terms). At 18.2 Confucius speaks of serving people 人 and seems to envision working in government.

    (It seems possible that at 1.5 and 18.2, Confucius is thinking of an official serving the people 民. Confucius also uses the term lao 勞. He speaks of laboring for parents at 4.18; cf. 2.8. At 13.1 he probably speaks of laboring for the people (but cf. 19.10). But there is no passage that juxtaposes laboring for parents with laboring for the people.)

    I have now named every Analects passage in which Confucius arguably uses 事 as a verb.

    Now, it seems offhand that while “serving” 事 the lord is the specific role of a minister or aide, it may be the role of a subject only occasionally. Perhaps a subject need only serve when called upon 使 (see 1.5, 5.16, 12.2, 14.41; cf. 13.25). Thus insofar as Confucius is thinking of a family-state analogy in connection with service, it seems likely that he is thinking mainly of an analogy between a son’s relation to his father and a minister’s relation to his lord, or an aide’s relation to his superior.

    Thus Confucius’ conception of a man’s relation to his father (or some set of elder kin) and his conception of a man’s relation to his superior in government certainly have at least this one point of similarity, in his view: a man serves his father (or some set of elder kin), and an official (or perhaps any person) serves his superior in government. This similarity or analogy seems to be about the fact that both family and state involve vertical relationships.

    (That idea is starkly different from the proposal in the Mozi that the ren 仁 [ruler] relates to the people as a good son relates to his parents. And Youzi may have in mind a mixed comparison at Analects 1.2. If and insofar as Youzi’s point is that the root of humanity for a ruler is a man’s proper attitude to his parents and other elders, for Youzi in that point the ruler would seem to be the analog of the younger parties, while the people or the whole country would seem to be the analog of the parents and elders.)

    A.5 . . . . A second point of similarity: special respect 敬

    We can add a detail to the similarity in respect of service. Though Confucius never suggests a parallel between family and governance while mentioning special respect 敬, we have seen that he probably recognizes a parallel regarding service 事. And he elsewhere several times associates service 事 with respect 敬: once in connection with family (4.18) and four or five times in connection with government (1.5, 5.16, 13.19, 15.38, and probably 16.10; though in at least two of these five passages, Confucius seems likely to be speaking of respect for the task or service). Thus we should take respect as a point of similarity between the son-parent relation and the minister-lord or subject-lord relation as Confucius thinks of them. In each relation, the former party should serve the latter with respect.

    We have now have found Confucius assuming at least a superficial similarity between certain family relationships and certain state relationships. In what we have found so far, the similarity is always only in how the lower party in each relationship should relate upward.

    One might be inclined to say that these similarities amount to no more than the point that family and government each involve internal ranking, or authority relations, a point that everyone everywhere may be presumed to notice. For interpretive purposes it is a triviality.

    But that dismissal is not quite right. For (1) the family relations Confucius has in mind here are presumably between adults, and I suppose it is not universally recognized that there is a solid authority-ranking between fathers and adult sons. And (2) rank or authority is about who should obey whom; it is not necessarily about serving someone. Thus the similarity we have found, in Confucius’ view, between the practical norms for the lower parties in certain family and state relations, is not entirely trivial. And (3) rank or authority is about who should obey whom; outside the family it is not about special respect or palpable reverence.

    Still, we should not overstate the significance of those first two de-trivializing points. It is unclear how significant they are. For (A) when we think of the authority of a parent, we think of childhood; and we think the main purpose of a parent’s authority over a child is the well-being of the child (and the integrity of the china). So we do not think of family rank and authority as a matter of service to parents. Though if we were asked to think specifically about the relation between parents and adult offspring, we might think more in terms of service—depending on the ages involved. (B) When we think of the authority of officials over citizens, we do not associate this with the project of serving the officials. But when we think of the authority of officials over their aides, we do think more in terms of service. Not personal service, of course, but service to the officials qua officials; service to their official projects. We may assume that that is what Confucius too has in mind for ministers (though it is unclear what he might regard as a family analog of this limit). (C) It is a little unclear, at least to me, how Confucius conceives service 事 here. Toward the end of the Euthyphro, Socrates distinguishes two concepts of service (therapeia) to the gods: benefitting them and obeying them. We might add a third candidate conception for service 事: serving X is performing the labors standardly associated with supporting someone in X’s role.

    In any case, the similarity we have found is merely in certain practical norms for the lower parties in certain relationships. And on the state side the relations Confucius is referring to may only be relations among officials in the government.

    It is one thing to see Confucius alluding to the fact of this similarity, or at least the service part of it, as we have done; it would be quite another thing to see him suggesting that the fact of this similarity itself has some functional significance (such as enabling us to grasp the state by applying the family as a metaphor, or enabling us to explain the forms or norms of the state as natural developments from family roots, either historically or continually through developmental psychology).

    A.6 . . . . 12.11: 君君,臣臣,父父,子子。

    We have so far found Confucius noting three special analogies or similarities between aspects of the family and aspects of society outside the family, or government:

    1. The master-disciple relation is similar to the father-son relation, in largely unspecified ways.

    2. Elder-respect in society is similar to filial piety in the family, in unspecified ways.

    3. An official should serve his superior with special respect (and, as we shall see, occasional criticism in case of disagreement); and similarly in the family a man should serve his father and some other elder kin with special respect (and, as we shall see, occasional criticism in case of disagreement).

    Of these three similarities, the first does not involve the state. The second and third are, on their face, peripheral to Confucius’ primary concern regarding the state—because the second and third are similarities between lower roles relating upward. Confucius’ primary concern regarding the state is rather about what constitutes good governance by officials in general, what kind of governance to aim at as a participant in governing: mainly the conduct of an upper position in relation to the people. Even when he is giving advice to his students rather than to a ruler, he far more often mentions norms for governing than norms for relating to a superior official.

    Does Confucius ever suggest a similarity between some family norms and norms for governing ? There are four passages in which he might be thought to do so.

    First, 2.21 may suggest an unspecified similarity between governing and upward relating in the family. I argue against the significance of that passage here.

    Second, 1.6 is often read as suggesting, without saying, that Humanity 仁 or general social virtue is an extension of the attitudes of filial piety to the people at large. I shall argue against this reading in another thread.

    Third, 16.1 may be taken to give a detailed description of what is at once the task of a ruler in relation to his people and the task of a clan head (not: a father) in relation to his clan’s members. Here Confucius is reported to say to two disciples who were serving the head of the Ji clan:

    …I have heard that those who possess a state or noble house are not concerned about whether their people are scarce, but rather [about whether their people are content; they are not concerned about poverty, but rather concerned] that what wealth they have is fairly distributed. If wealth is fairly distributed, there should be no poverty; if your state or house is in harmony, there should be no scarcity; and if your people are content, there should be no instability. This being the case, if those who are distant will not submit, simply refine your culture and Virtue in order to attract them. Once you have attracted them, you should make them content. … (Slingerland; I have marked with brackets the part that Slingerland plausibly adds)
    … 丘也聞有國有家者,不患寡而患不均,不患貧而患不安。蓋均無貧,和無寡,安無傾。夫如是,故遠人不服,則修文德以來之。既來之,則安之。 …

    But surely here Confucius is not talking about how the head of a powerful house should govern the various households of his clan; rather he is talking about how he should govern the whole population over whom he and his clan rule, as the Ji clan ruled over Lu. For, first, Slingerland’s translation “your state or house is in harmony” does not reflect a distinct mention of the house (clan) in the text; the whole phrase simply translates the word “harmony” 和. Second, it seems to me a little odd to say that a head of a family does not worry about the paucity of family members, but worries about their being content or feeling secure. (The oddity is much greater if we do not amend the text as Slingerland’s translation does.) Third, the explanation that follows, about how good government will increase population by attracting people from afar, surely does not refer to attracting people to become members of one’s clan.

    I think the reading of 16.1 that has it describe how a clan head should govern his clan members (the various households) is a minority reading, and I shall refer to it later as such.

    Fourth, 12.11 may suggest a similarity between governing and a father’s downward relating in the family.

    The duke Jing, of Qi, asked Confucius about government. Confucius replied, “There is government, when the prince is prince, and the minister is minister; when the father is father, and the son is son.” “Good!” said the duke; “if, indeed, the prince be not prince, the minister not minister, the father not father, and the son not son, although I have my revenue, can I enjoy it?” (Legge)

    The remark’s very abstract form makes it hard to know what sorts of activities or qualities Confucius has in mind, and what we are to make of the juxtaposition.

    What is a 臣 in this context? Most translators render 臣 here as “minister”, though Lau and Leys have “subject”. (Sam Crane discusses this interpretive question here.) Mencius at 2A1 uses臣 to say of the tyrant Zhou that he was ruler of all China: “一民莫非其臣也.” Mencius sometimes calls himself “臣” in addressing a ruler; if in some of those cases Mencius did not hold office under that ruler, then Legge’s translation “your servant” may get the idea right. To refer to oneself as 臣 in addressing a lord might be a figurative way to stress one’s devotion or subservience, without implying that one holds the position of minister; for the idea may be that it is the specific role of a minister to assist the lord.

    (The word臣 can at least sometimes be used to mark the vassal relation, but I do not know how closely the lord-vassal relation was seen as kin to the ruler-minister relation, nor what sort of relation was the original home of the term 臣.)

    The form of Confucius’ remark may suggest an emphasis on role-names, and hence suggest that he would not be using the word 臣 in what would have been recognized as an extended sense. Compare 9.12, where he refused to let his disciples pretend to be his 臣. (But I argue here against Hansen’s parsing, which makes the passage directly about the correct use of names.)

    A consideration possibly favoring the wide reading of 臣 as “subject” is that the duke has asked about governing, and governing is, on its face, a relation between ruler and subjects as such. One might think governance is not covered unless subjects in general are mentioned.

    Alternately, one might think that governing is, on its face, what the governors do, the lord and his officials. This team is mentioned only if we read 臣 as “minister” or “subordinate official.” (Perhaps subjects in general are mentioned in the mention of fathers and sons; or perhaps the fathers and sons Confucius has in mind are in the governing classes.)

    (Is it possible that zheng 政 was here understood to mean something like policy or good order, so that Confucius’ reply should be taken more as describing the order at which the government should aim, than as describing the activity of maintaining order?)

    The broad framework idea in the remark may be that of everyone fulfilling all their roles, with the four roles mentioned standing for all roles, or taken to be the main important roles in society. The vision then would be of society as a construct of vertical roles. Family and state have in common that the main relation in each is vertical.

    Moss Roberts reads the remark as assigning the responsibility for both parties to the upper party: “Let ruler rule as he should and then the ministers will serve as they should; let the father guide as he should and then the sons will serve as they should” (“Foreword” to K. A. Besio and C. Tung eds. Three Kingdoms and Chinese Culture (2007), p. ix). If we understand the statement this way and if we translate 臣 as “subjects” rather than “ministers,” then the first half of the remark fits the view Confucius expresses elsewhere in the Analects. But if we understand the second half of the remark in Roberts’ way, it does not seem to fit Confucius’ view elsewhere in the Analects. (See §B.4 below.)

    Duke Jing’s reply may allude to the idea that his aides are supposed to fill the treasury and the larder, while it is a stereotypical responsibility of a filial son to set food before the parents (cf. 2.7, 2.8). Perhaps it is fair to sum up the duke’s broad point this way: “Aha, I see! The father-son relation is part of governance too!” Or, perhaps better: “Aha, I see! Those who govern not only need their aides to be aides, they need their sons to be sons too!” For what the duke hears in the comment is mainly that for governance to work, sons as well as ministers have to do their part. Where Moss Roberts reads the remark as mainly about the responsibility of the upper parties, Duke Jing seems to hear the remark as mainly about the responsibilities of the two lower positions. And surely that is a natural enough way to understand the remark. The statements, “I’m supposed to give the orders and you’re supposed to follow them!” and “It’s my job to give the orders, and your job to follow them!” are all about the responsibility of the lower party.

    The usual view of the context of the remark seems to fit that reading. According to Soothill, Legge, Slingerland, and Riegel , the context of the remark was Confucius’ early visit to Qi in the (large) company of a temporarily exiled Duke of Lu. Slingerland’s note elaborates:

    [Duke Jing’s] nominal minister, Chen Qi, had usurped control of the state, and the Duke’s plans to pass over his eldest son for the succession had set off contention among his sons.

    The other authorities I listed give roughly the same picture of the situation in Qi (but place more stress on the role of a concubine as pushing for one son). Annping Chin thinks the conversation at 12.11 must have taken place during a later visit to Qi; I think Sima Qian says Duke Jing also made a visit to Lu. I can’t sort it out. But the context reported by Slingerland does seem to fit the idea that the main focus of Confucius’ remark was on the lower parties’ knowing their place—or at least, that was the main focus of his remark as he should have expected the Duke to understand it. It’s not the lord’s job to serve the ministers; it’s the ministers’ job to serve the lord. It’s not the Duke’s job to serve his sons; it’s the sons’ job to serve the Duke. Nothing prevented Confucius from giving an answer specially tailored to the Duke’s circumstances, so long as he could phrase it in an abstract way that guarantees that it is not wrong, and that it is generally suitable to be meditated on.

    This point is important, because Confucius’ remark at 12.11 is the one passage in the Analects where there is a significant likelihood that he is suggesting that the role responsibilities of a father in relation to a son are similar to those of a lord in relation to someone below (probably ministers, possibly subjects). If it’s not happening here, it’s not happening anywhere. Even if we take the remark that way, the suggestion is only of similarity in the abstract. There is no suggestion about how the two roles might be similar. (And as we shall see, his vision of the role of a father toward a son seems to be profoundly different from his vision of the role of a lord toward his subjects.)

    The context as reported by Slingerland also suggests that by 臣 Confucius meant “ministers”; and it suggests that by fathers and sons Confucius may have meant the lord and his sons, or the families of those engaged in governance, rather than families throughout the population. Thus the whole remark would have been about relations internal to the governing circles.

    Another possibility is that Confucius juxtaposition of family and political relations may have been a suggestion that the two kinds of relation should be kept separate. Duke Jing might have felt his messy situation as an unwelcome mixing of family relations and political relations. The point of juxtaposing 君君 to 臣臣, we may say, is to advise against muddying the distinction. As we might say: “The lord should be the lord and the minister should be the minister,” or “A lord is one thing; a minister is another.” (Cf. “Boy, you knew who you were then / Girls were girls and men were men.”) Similarly, the point of juxtaposing 父父 to 子子 may be to advise against muddying the distinction. The point of juxtaposing 君君臣臣 to 父父子子 may have been the same: don’t muddy the distinction between these two relations. As we might say: “The lord-minister relation is one thing; the father-son relation is another”; or, “Lord and minister should be lord and minister; father and son should be father and son.” If that is how Confucius meant the remark, he had his mind here on the distinction between family and political relations, more than on any similarity or analogy. (Offhand it seems to me that the “Don’t muddy” reading is more likely to be true of the two small juxtapositions, or of the one large one, than of all three.)

    Supposing that in this remark Confucius has in mind an analogy between the two relations, I think there is no significant evidence here that he is thinking of one of the two relations on the model of the other. But if he is doing so: is there any clue here about which of the two would be the model? There are perhaps two small clues, pointing in the same direction. One is the order of presentation: what is first is more likely to be the model. And there is the question the statement is answering, which suggests that the point is that the personal is political, as it were; even the relation between father and son should be seen as a political relation. These two clues suggest that in this remark, the state is serving as a model for the understanding of family relations. But these clues are slight.

    In sum:

    12.11 is the only passage in which, I think, there is a significant likelihood that Confucius is comparing the role of a father (or any family role) to the role of a lord. 12.11 is also the only remark in which Confucius may suggest that family and state are similar in respect of both positions in the constitutive relations of each, not just that family and state are similar in respect of the responsibilities of lower parties to higher in each. But (a) there is significant reason to think that Confucius has in mind here only a comparison between the father-son relation and the lord-minister relation; hence nothing about responsibilities between neighbors or between governor and subjects in general, i.e. no comparison between parenting and governing. And (b) there is significant reason to think that the main focus here (as in all the other comparisons-by-juxtaposition) is on the responsibilities of the lower parties in each relationship. Finally, (c) there is no strong reason to think that we can infer from the parallelism in Confucius’ remark that he is suggesting any similarity between state and family beyond the universally recognized fact that each involves authority relations. His remark here seems to display at most a simple conservatism, an acceptance of whatever consensus can be taken for granted among people without special education, about the main standard responsibilities of the main roles in society.

    A.7 . . . . Positive Attitudes

    I have now completed my survey of similarities between family and state that Confucius alluded to as similarities, or showed some identifiable sign of noticing. But of course there are countless similarities between any two things. There is one more broad similarity between family and state that may be worth discussing whether or not Confucius ever alluded to it.

    It is widely thought nowadays that we should have “care and respect” for people in general. To abbreviate here, I’ll say one should have a positive attitude toward others.

    Whatever we may think about Confucius’ particular remarks and concepts, we may fairly suppose that Confucius thought one should have a positive attitude toward one’s parents. We need no special evidence that he thought so. Pretty much everybody thinks so.

    I suppose there is adequate textual evidence that Confucius thought one should have a positive attitude toward people in general, to one degree or another. In that too he was not taking a very controversial position. Indeed the very wide consensus on this view is part of why we should be willing to take a few quick statements as evidence that he held this thin view. (Incidentally, Confucius’ acceptance of this general point is an additional reason, were one necessary, to think that he thinks one should have a positive attitude toward one’s parents.)

    Therefore Confucius’ view of a man’s relation to his parents and Confucius’ view of a man’s relation to people in general are similar in that he thinks each relation should be characterized by a positive attitude. (That is not to say that the degrees should be the same. In general when someone proposes that Confucius thought similar norms apply in and out of the family, she is not proposing a similarity in degree.)

    We can broaden the point. Confucius presumably also thought one should have a positive attitude toward family members in general (clan or household), not just toward parents. For everybody thinks this; and also it is a direct inference from the more general view. We may even add that he thought one should have a stronger positive attitude toward kin than toward non-kin, other things equal. No direct textual evidence is needed to show that he held this view; virtually everyone holds it.

    (In fact there is little or no direct textual evidence to support the claim that Confucius thought one should have a positive attitude toward any family members other than parents, aside from his universal statements that we should have a positive attitude toward everyone. There are two passages that have him saying that one should serve (9.16) and maybe heed (11.22) elder brothers, or perhaps he means paternal uncles. As we shall see in §C.3 below, there is some dubious direct evidence that Confucius thought one should be friendly with one’s brothers, and some evidence that he thought most parents do in fact care about their infant children. And 8.2b may may suggest very indirectly that the right way to treat family is not entirely dissimilar from kindness, small ren 仁. (Or maybe the term 親 there doesn’t mean family; maybe it means parents, or close associates.) But again: no direct evidence is necessary.)

    Hence, presumably, Confucius’ view of family and his view of the state (society) were similar in that in each case he thought each member should have a positive attitude toward the other members (and more than toward nonmembers, other things equal).

    This feature of Confucius’ thought is not at all distinctive to him. That is our main assurance that it is a feature of his thought. Also, this similarity between (his conceptions of) family and of society is not special to those two groups; it would presumably hold between any two significant groups.

    What might be distinctive is what Confucius does with this similarity, if anything.

    We shall turn to that sort of question in Section C below.

  2. Bill Haines says:


    But let first survey some of the main differences between family and state as Confucius saw them, without limiting our survey to differences Confucius alluded to as differences, or even to differences he showed some identifiable sign of noticing as differences.

    B.1 . . . . Difference: conditions of service

    The one passage that mentions both service and respect in connection with a man’s relation to his parents is specifically about remonstrance:

    The Master said, “In serving your parents you may gently remonstrate with them. However, once it becomes apparent that they have not taken your criticism to heart, you should be respectful and not oppose them, and follow their lead diligently without resentment.” (Slingerland)

    Confucius similarly thinks an aide should express his disagreement with a ruler (14.22, 16.1, and arguably 14.21).

    The point that a man may remonstrate with his parents at all may not be Confucius’ main point here. Confucius’ usual concern regarding parents is to correct a deficiency of respect for them, not an excess. And it may not have been at all unusual for a man in the ruling or educated classes to voice to his parents a disagreement with them. In fact we do that for anyone we truly care about, whatever the relationship (Confucius makes a similar point at 14.7). It’s not a special feature of the relation to one’s parents.

    Confucius’ main point at 4.18 may be that any remonstrance should be gentle, and (more importantly) that we should continue in respectful service to parents with whom we disagree, not repeating our objections. A good son serves his father’s way even long after the father’s death: 1.11, 4.20.

    Confucius emphasizes the opposite view for governmental relations (as Joseph Chan points out). A main structural feature of the ministerial role, in Confucius’ view, is that when a ruler persists in acting wrongly, a minister should not continue to serve him (5.19, 11.24, 15.7, 16.1; cf. 8.13, 14.37, 15.1, 15.8, 18.4; but see 14.16, 14.17, 14.19). And a ruler should aim to rule well enough to induce even the subjects of other rulers to forsake them: 13.4, 13.16.

    There may also be a certain shallowness to the respect that a minister should have for his lord on the job. According to 16.1, in upbraiding two of his disciples for their inadequate control over the lord they are advising, Confucius is reported as saying,

    …“…When a tiger or rhinoceros escapes from his cage; when a tortoise or piece of jade is injured in its repository – whose is the fault?”… (Legge)

    B.2 . . . . Difference: grounds of position

    Close kin to this dissimilarity is another, pertaining to the proper grounds of rank. We may assume that Confucius thinks that throughout the family (among males), rank or relative authority is determined by age and descent. Now, regarding government, Confucius is approximately silent on how the highest members of the feudal hierarchy should be chosen. At least for the rest of the government he favors the promotion of the worthy (2.19, 2.20, 12.22, 13.2), largely because what makes government good and effective is the virtue of the governors (e.g. 2.1, 2.3, 8.2, 12.7, 12.19, 13.6).

    Still, perhaps we see Confucius appointing the worthy to his family, or at least appointing the innocent, when he chooses husbands for his daughter and his niece (5.1, 5.2, 11.6).

    B.3 . . . . Difference: grounds of relationship

    The disanalogies sketched above are reflections of a far broader and deeper difference between Confucius’ views of family association and outside association.

    Confucius may not explicitly emphasize close personal relationships within the family. If he does not emphasize them in fact, that is a difference between family and society as he views them.

    Insofar as Confucius recommends close personal relationships within the family, based on kin relationships, ascriptive relationships, he approves continued close association with a number of people on grounds quite independent of their moral qualities. But when Confucius is not referring specifically to family, he puts great emphasis on the point that one should choose one’s associates based on their qualities (1.6, 1.14, 4.1, 9.30, 15.10). One should have no friend inferior to oneself (1.8, 9.25). If a friend will not listen to one’s remonstrance, one should leave that friend (12.23). Perhaps this is what Confucius means when he says approvingly that someone who hates un-renness will not let the un-ren approach his person (不使不仁者加乎其身: 4.6). (Presumably 不仁 is worse than simply not 仁.)

    Indeed, these views about relationships outside the family seem to argue against the idea that Confucius emphasized close personal relationships within the family, based mainly on kinship relations such as being a brother or son. The views about outside relationship might be easier to reconcile with the idea that one should serve parents with obedience and shows of respect, than the idea that it is essential to be close friends with one’s brothers. On the other hand, Confucius does recommend that a man accept the moral leadership of his father just because he is his father, for time (1.11, cf. 11.22). In any case, it would appear that for Confucius, the proper modes and aims of relationships with kin are profoundly different from the proper modes and aims of non-kin relationships in private life and public life, for people with high aspirations.

    B.4 . . . . Difference: the locus of responsibility and the social psychology of virtue

    The Analects gives the impression that Confucius thinks that among living adults, the main responsible agents are at the top of the state (society) and the bottom of the family.

    Regarding the state, organized political society, Confucius’ main topic is virtue for governance, and for governors. The character of the governors, and their treatment of the people, determine the character of the people and their treatment of the governors. For a lord, the people’s dereliction is no excuse.

    Regarding relations with kin, almost Confucius’ only topic is how a son should treat a parent, living or recently deceased. For a son, a parent’s dereliction is no excuse.

    In another thread recently I speculated about the main reasons why, as Confucius might have thought, a lord’s virtue might have more effect on others than a neighbor’s virtue would. The reasons I listed would seem to apply much more strongly as reasons to think a parent would have a powerful impact on a son.

    Did Confucius think a parent’s character and his treatment of his son determines the character of that son and his treatment of his parent? No remark preserved in the Analects comes near this question (except that Confucius says that if a man is filial, he will follow his father’s way for a while). And yet Confucius was a father, and his students were or expected to be fathers. In speaking to the powerful, Confucius made heavy use of the idea of the moral power of a leader’s virtue as a way of encouraging his interlocutors to virtue. One might expect him to have used the same approach in speaking with prospective fathers, if he thought the premise would be credible.

    On the other hand, Confucius may have noticed in history that sons are often markedly better or worse than their fathers. I imagine this was a familiar thought among all those concerned with government’s history or prospects, and especially those who gave some thought to rules for succession to high office. Confucius may have looked at his own acquaintances and noticed the same mismatch between fathers and sons. He may have been less than well satisfied with his own son, or with his students’ fathers. Confucius’ students too may have noticed all these things. Perhaps for such reasons Confucius did not have a high opinion of the moral power of fathers in relation to their sons; at least not a view parallel to his view about the moral power of exemplary lords over the people, which is arguably the leading principle in his theory of the state. Perhaps he felt that he had not quite worked out how his general philosophy would apply to the family. Or perhaps he thought that by the time a son is old enough to begin to be serious about virtue or vice, he is already subject to powerful influences outside the home.

    (Whether it is true that a virtuous father will have virtuous sons depends, of course, on what we are counting as virtue, on how many sons the father has, and on whether he is focused on other projects, distant from his sons.)

    Again: the Analects gives the impression that Confucius thinks the main responsible agents in vertical relationships are in the upper positions in the state (society); and the lower positions in the family. Confucius’ focus is much more sharply on a single position in the family, than on a single position in the state. But while his justification for the focus is salient regarding the state, it is unrecorded regarding the family.

    One might suspect that the general impression that Confucius assigns responsibility to opposite poles in these two arenas does not reflect Confucius’ actual views. One might argue for this point as follows.


    Confucius may have been focusing on those positions mainly because of whom he was addressing. On the one hand, he was addressing governors and those in training to advise governors and participate in governing. To all these he would stress mainly the responsibilities of upper state parties. On the other hand, most of his students were young men with responsibilities to living parents, and not many family responsibilities to younger kin. To these he might stress mainly the responsibilities of lower family parties.

    Further, Confucius likely occupied a potentially awkward position, as a quasi-father figure to young men with living fathers, and thus in some sense a rival to their fathers, whose Ways the young men were duty-bound to follow (4.20). He took some of his trainees far from home, though he thought a filial son should as a rule not travel far from his parents (4.19). He exposed some to potential death from starvation, though it was generally expected that a good son would be able to support his parents (2.7). Confucius’ position in relation to the fathers of his students may have been awkward or even precarious. We may see a sign of such tension in 11.22, where Confucius advises Zilu to go slow on the moral training out of consideration for his father and elder brothers or uncles.

    Thus whatever views about filial piety that Confucius may have had, such as that a good son should show respect, these views are likely to have been of lively and immediate practical interest on all sides, and thus may be represented in the Analects out of proportion to their philosophical importance in his overall view.

    Further, the selection in the Analects must reflect the interests current in the group not only during Confucius’ lifetime, but also after his death, when the materials were compiled. Two people in particular may be worthy of note here. First, Youzi, who stressed filial piety, probably joined the group shortly after Confucius’ death, and led part of the group for a time. He is a likely figure to have begun a collection of Confucius’ sayings, and we have a record of him soliciting Confucius’ sayings from a disciple. (For evidence and arguments, see my paper “The Purloined Philosopher,” PEW 58.4.) Second, Zengzi, whose recorded sayings suggest that he stressed filial piety more than did Confucius or Confucius’ other followers, was well-placed to have had his ears filled with a double or triple dose of filial piety from an early age, even if Confucius did not regard that virtue as being of great political importance. For (a) Zengzi was a son of Confucius’ student Zeng Xi, (b) Zeng Xi’s interests seem not to have bent toward politics (11.26), and (c) Zengzi may not have been a promising politician (11.18).


    The above considerations may be reason to think that the proportion of passages on filial piety in the Analects (a proportion that may appear larger than it is, because they are mainly in early Books) may exaggerate the importance of filial piety in Confucius’ philosophy, and thus exaggerate the importance of filial piety in his philosophy of the family (if he has general ideas about the family).

    One these grounds we might hold that the impression that the Analects locates the main responsibility at opposite ends of the state (society) and the family, top and bottom respectively, is at least a distorted picture.

    But this line of thought comes with a price. For the sheer number of passages on filial piety is our best reason for thinking family was in some sense the basis of his overall philosophy. If we lose that reason, what reasons are left, for thinking that family is at the core of Confucius’ philosophy?

    The other main contributory reasons, I suppose, are these:

    (1) Youzi’s statement at 1.2, a reason I have addressed by arguing at PEW 58.4 that Youzi was an outsider who had his own ideas, and arguing here that 1.2 is not about the primacy of family—though the latter point, if true, weakens one of my arguments for the former point. [See also this. –BH 8/8/2016]

    (2) Confucius’ highly obscure statement at 2.21, whose significance I challenge here; and

    (3) a popular but very bold reading of Confucius’ remark at 1.6, which I shall attack soon. Here.

  3. Bill Haines says:


    Most of the rest of my discussion is in the form of a set of worries about a particular elaboration of the family-as-model reading of Confucius by the first-rate philosopher and scholar P. J. Ivanhoe. On pages 1-3 of Ethics in the Confucian Tradition: The Thought of Mengzi and Wang Yangming (2002), Chapter 1, “Kongzi’s View of the Way,” he writes,

    At the heart of Kongzi’s conception of the proper life for human beings—the “Way” (dao 道) – is a model of a harmonious and happy family, one whose different members each contribute to the welfare and flourishing of the whole, according to their role-specific obligations. These obligations—serving as a mother, a father, an elder brother, etc.—and the practices and norms associated with them were the primary guides to the moral life. In this sense, the family served as the basic paradigm for the well-lived life. However, the moral life did not end with the family. One had roles to fill in society as well. There were parallel obligations to king and state which, though never superseding, were fashioned on the model of those to parents and family. …
    … The obligations to one’s family were most important because for Kongzi they were the earliest and strongest bonds human beings form. They also were regarded as the source of our social obligations; our obligations to others were developed out of and modeled on the family. … (p.1)

    The strongest feelings are originally and forever within the family. The virtues of “filial piety” (xiao 孝) and “respect for an elder brother” (ti 悌) are the source from which one draws in extending and developing such feelings for others and are the most profound examples of the type of concern that characterizes those who are ren. …

    In the well-ordered state, society becomes the family, writ large. The king, as father—and according to some accounts both father and mother—to his people, must care for them and as a proper parent provide for them in the broadest sense of the term. Rulers must enrich their people—materially, ethically, and spiritually. The ideal state is like an extended family, providing for the needs of all its members and working for the common benefit of all while preserving a strict hierarchy in its structure and clear divisions among its constituent roles and their respective obligations and norms. Most important of all, the enlightened state, like the ideal family, is permeated at every level with a deep, lively, and particular concern for the well-being of each of its members. (p.3)

    (Cf. also Ivanhoe, Confucian Reflections (2013), p. 66.)

    C.1 . . . . Happy harmonious families

    While it would have been reasonable of Confucius to have been inspired by the idea of a father (or higher family or clan member) ensuring the material and spiritual happiness of his family or clan, such a thought would have been, in certain ways, uncharacteristic of him.

    Confucius does sometimes mention or generalize about families as entities. At 12.2 and 12.20, he speaks of countries (邦) and clans (家) in parallel terms, as arenas in which one might be famous, distinguished, or resented; and he mentions clans (宗族) in a similar way at 13.20. At 5.8 he mentions the position of steward (宰) for a clan of a hundred chariots. At 5.8 and 5.28 and perhaps 14.9 he characterizes the size of a town by the number of houses or households (室) it includes. At 16.3 Confucius speaks of the state revenue’s not being in the hands of the ducal house (祿之去公室). At 14.11 he mentions the office of chief steward for this or that ruling clan (趙魏老).

    Confucius has some interest in in the personal virtue of filial piety, but I see no sign that he has much interest in the quality of family life as such, or recognizes that as a topic. I think Confucius never speaks of a family as being harmonious or happy, or as flourishing, or as having the Way or any moral quality, or anything like that. (I’ll discuss the two arguable exceptions in a moment.) Beyond his mention of “a clan of a hundred chariots” at 5.8, I think he never speaks of a family as having any quality or condition.

    Beyond the passages listed just above, Confucius directly mentions or generalizes about families as entities in two more places at most. These are the only other places where he might be thought to speak of a family as having some quality or condition. One is at 16.1, where Confucius arguably speaks of a clan’s membership as being harmonious, prosperous, etc. But he does so only on the minority reading I argued against in §A.6 above.

    The other place where Confucius may be thought to allude to the quality or condition of a family is at 13.8, where he praises a prince of Wei for being good at householding (善居室) on the grounds that the prince finds even a little wealth to be splendid. (The prince must not have lived with his father.) One might think there is an implicit reference here to a good condition of a family. Translators render 善居室 variously:

    showed a laudable attitude towards a house as a place to live in (Lau)
    made the most of his living accommodations (Ames & Rosemont)
    dwelt well content in his house (Soothill)
    dwelt as a man should dwell in his house (Waley)
    knew how to live in his house (Brookses)
    good at managing his household wealth (Watson)
    good at running a household (Slingerland)
    good at household management (Huang)
    knew the economy of a family well (Legge)
    showed a sensible attitude toward running a household (Chin)

    If the reference here is to the wealth available for the prince’s household, understood to include his immediate family if any, we might think it odd that Confucius adduces only the prince’s own expressions of satisfaction. We might think that it is an essential part of the duty of a householder to see to the economic security of his dependents, including dependent kin. Granted, the prince’s position may have been secure, and his accommodations perfectly adequate. But elsewhere, when we see Confucius discussing the proper attitude toward wealth, we never see it cross his mind that one of the reasons a man might seek to avoid poverty or starvation is that he has obligations to his own family (see 1.14, 4.5, 4.9, 6.5, 7.16, 8.21, 14.2, 15.2, 15.32; contrast 11.23). Thus even if here at 13.8 Confucius does mean home life, and assumes that the residents include at least a spouse and children, it is still likely that he is not thinking of family life, the quality of family interactions.

    By contrast, Confucius is occasionally explicit about a governor’s duty to see to the economic welfare of the people.

    Confucius does sometimes mention the conditions or qualities of non-family social groups. He is quoted as doing so at 16.1 (see §A.6 above). In earlier Books he speaks of a country (邦) as having the Way or not (5.21, 8.13, and 15.7), and says something similar about Qi and Lu (6.24). At 16.2 he speaks of the world (天下) as something that can have or lack the Way. He seems to use ren (仁) to name a quality of neighborhoods at 4.1, of states at 13.12, and possibly of the world at 12.1. He speaks in many places of qualities of the people (民).

    It is unclear whether we should say that Confucius ever speaks of an organized group as being happy (but see 14.17, 16.1). Among the many unique features of 16.1, it is the only passage in which Confucius uses ‘harmony’ to name a condition of a social aggregate.

    Confucius often describes a state in good condition by breaking it down into two main roles, saying what each does—in happy harmony, if you like. The north star stays in its place and the others bow around it (2.1); the ruler rules by virtue and ritual, so the people keep themselves in line (2.3); the ruler approaches the people with dignity, is filial and kind, and promotes the worthy, so the people will have respect, trust, and enthusiasm (2.20); a ruler should employ ministers according to the ritual and ministers should serve a ruler with fidelity (3.19); the ruler is good to his kin and longtime associates, so the people are virtuous and honest (8.2); the ruler is straight, so others will be straight (12.17); the ruler loves justice and good faith, so the people follow enthusiastically (13.4); the rulers love ritual, so the people are easy to employ (14.41); etc.

    There are only three places where Confucius mentions reciprocal family parties each doing something. At 12.11 he speaks of fathers’ fathering and sons’ sonning (saying it is part of governance). But the point may just be the need for obedience. At 13.18 he says fathers cover for sons and sons cover for fathers. But he is probably not thinking of their doing this together in conspiracy. At 17.21 he adduces parents’ care for infants as a reason for extended mourning after the parents die.

    C.2 . . . . Contributions to family flourishing

    Again, Ivanhoe speaks of

    a harmonious and happy family, one whose different members each contribute to the welfare and flourishing of the whole, according to their role-specific obligations.”

    As Confucius barely mentions the qualities or conditions of clans or households, he does not mention inputs to those qualities or conditions, as such. But he does sometimes mention role-related benefitting of particular kin, beyond 13.18 and 17.21. As we have seen, he speaks on perhaps four occasions of “serving” parents, and once of serving other elder males, perhaps brothers. On specifics, he mentions supporting and feeding parents (2.7, 2.8), albeit as foils for main the points he wants to get across, which are not as obviously about benefit. In one place he probably says that the only worry parents should have regarding their offspring is that their offspring might become ill (2.6). Here the idea seems to be that good and filial behavior comprehensively supports, protects, and comforts parents. But we may still wonder whether that contribution is offered here as the purpose of filial piety, or only as a sign or measure of it.

    Confucius also sometimes speaks of benefitting kin younger than oneself, as surveyed exhaustively in the following Section.

    C.3 . . . . Mother, father, elder brother

    Ivanhoe elaborates:

    These obligations—serving as a mother, a father, an elder brother, etc.—and the practices and norms associated with them were the primary guides to the moral life.”

    Confucius makes very little mention of role-specific obligations or practices of a mother, a father, or an elder brother. Those are upper positions, relating downward. They are also the only upper family positions Confucius mentions (unless by 父兄 he means something like uncles, in which case Confucius does mention uncles (but says nothing about their obligations), and never mentions elder brothers at all except in the general phrases 兄弟 and 昆弟). [Oops: Confucius is reported as discussing the role of a clan head at 16.1, though (as I have argued above) not the intra-clan role. -1/1/17] Let us review all that Confucius says about the obligations or typical practices of the three positions Ivanhoe names.

    … “The Documents say [approvingly], ‘… a friend to his brothers …’ …”
    [Slingerland has “friendly to one’s elders and juniors”]
    … 書云:…友于兄弟 …

    Zi Lu asked, saying, “What qualities must a man possess to entitle him to be called a scholar?” The Master said, “He must be thus: earnest, urgent, and bland – among his friends, earnest and urgent; among his brothers, bland.” (Legge)

    … “… a father should act as a father; a son should act as a son.” …
    … 父父子子 …

    … “… [When a ruler is virtuous], the people from all quarters will come to him, bearing their children on their backs – what need has he of a knowledge of husbandry?” (Legge)
    …「… 夫如是,則四方之民襁負其子而至矣,焉用稼?」

    … “Where I come from, uprightness is different. A father covers up for his son, a son covers up for his father; there is uprightness in this.”
    “… 吾黨之直者異於是。父為子隱,子為父隱,直在其中矣。”

    … “… only after three years does a child leave the embrace of its father and mother …”
    …”… 子生三年,然後免於父母之懷 … “

    When Yan Yuan died, Yan Lu begged the carriage of the Master to sell and get an outer shell for his son’s coffin. The Master said, “Whether he has talents or has not talents, everyone calls his son his son. There was Li; when he died, he had a coffin but no outer shell. I would not walk on foot to get a shell for him, because, having followed in the rear of the great officers, it was not proper that I should walk on foot. (Legge)

    When Yan Yuan died, the disciples wished to give him a great funeral, and the Master said, “You may not do so.” The disciples did bury him in great style. The Master said, “Hui behaved towards me as his father. I have not been able to treat him as my son. The fault is not mine; it belongs to you, O disciples.” (Legge)

    (11.11 suggests a certain casualness about whom one may regard as one’s father or son, contrasting with Confucius’ refusal at 9.12 to let his disciples act toward him as though they were his aides or ministers 臣.)

    It all adds up to very little. Regarding norms and practices for parents between their offspring’s infancy and their offspring’s death, Confucius mentions only one practice: that a father should conceal his son’s crimes. Confucius is likely to mean that any family member should conceal the crimes of any other, so that this point is not about fatherhood or parenthood as distinct from any other family roles.

    Confucius discusses the norms for an elder brother only in discussing brothers in general. And the two minimal comments about brothers in general (quoted above) may be no evidence of Confucius’ views even about brothers in general. For the approval of friendliness toward brothers expressed at 2.21 is a secondary feature of a quoted text, within a remark that is distinctly unreliable as evidence of Confucius’ views. (And Slingerland translates 2.21 without any mention of brothers.) The other passage, 13.28, sets out a difference between relations with friends (or colleagues) and relations with brothers. Translators render this difference variously. For example:

    earnest in spirit, persuasive in speech / of gracious bearing (Soothill)
    earnest and urgent / bland (Legge)
    critical and exacting / indulgent (Waley)
    earnest and keen / genial (Lau)
    particular, punctilious / agreeable (Brookses)

    I wonder whether the line of 13.28 that mentions brothers and distinguishes them from other associates, is authentic. It reads like a scholium attempting to resolve what a reader mistook for a contradiction. If the line is not authentic, then at 13.28 Confucius did not mention brothers at all. But if the line is authentic, does it oppose the point about brothers at 2.21, by telling us not to treat our brothers as we treat our friends? And then what in the family would be a model for friendship?

    Whatever the status of the records of what Confucius may have said about brothers in general, the idea that we should be friendly or at least genial with our brothers is in no way distinctive.

    The Analects does show Confucius acting as an upper or quasi-upper party in family relationships. He chooses a husband for his daughter (5.1) and for his older brother’s daughter (5.2, 11.6), and recommends books to his son (16.13, 17.10). His treatment of his son struck a contemporary as very distant, and his choice of a husband for his daughter seems risky.

    C.4 . . . . Son, younger brother

    Before continuing with the material on p. 1, we should note that on p. 3 Ivanhoe stresses mainly lower family positions relating upward.

    The strongest feelings are originally and forever within the family. The virtues of “filial piety” (xiao 孝) and “respect for an elder brother” (ti 悌) are the source from which one draws in extending and developing such feelings for others and are the most profound examples of the type of concern that characterizes those who are ren. …

    (Ivanhoe then quotes Youzi’s statement at 1.2.) But Confucius is at least not explicit about any of this. He never compares the strength of feelings within and without the family (except perhaps at 11.10 and 19.17, two passages that seem at cross purposes on the point). Confucius never mentions the feelings of children; there is no indication that he regarded the matter as important. His remark at 16.7, on the feelings to guard against at different ages, begins with boys around puberty guarding against lust, perhaps because one does not expect anyone younger to be guarding against errant feelings.

    As we saw in §A.2 above, Confucius says nothing at all about the virtue of 弟/悌 in the sense of “respect for an elder brother.” Nor, I have argued, does anyone else in the Analects. There are just two passages where Confucius might or might not mention the specific role of a younger brother in relation to an older: see here; unless we also count his tolerant comments surrounding Duke Huan, who secured his dukedom by having his elder brother killed (14.15-17).

    When Confucius is teaching his followers about filial piety, he sometimes stresses feelings; but he does not suggest that parents are natural objects of early strong feelings. The main thing he says about feelings toward living parents is “色難”—the hard part is to show respect, to seem to care, to serve with visible love (2.8; cf. 2.7). The 21st century might wonder why that would be hard, or would need mentioning at all, especially to the kind of people who had signed up for study with Confucius. Perhaps the answer is that the concrete burdens a Spring and Autumn father put on a son were burdens indeed, straining positive feelings; and personal closeness may have been lacking. In any case, the picture we seem to see in this teaching is not that visible care and respect are naturally early and strong.

    The other thing Confucius says about feelings toward living parents is that one must know their ages. “On the one hand, it is a source of joy; on the other, of trepidation”一則以喜,一則以懼 (4.21, A&R). His point is perhaps that a man should care enough about his parents to know their ages. This remark might be a rebuke of someone who did not know, or it might be in effect a prescription to think on their ages, in order to cultivate one’s feelings for one’s parents. In any case the assumption seems to be that one’s parents are old enough that their death is in view: Confucius is talking about the filiality of an adult.

    I shall say a little more about filial piety and childhood below.

    Now back to Ivanhoe’s p. 1.

    C.5 . . . . The primary guides to the moral life

    Again, Ivanhoe writes,

    These obligations—serving as a mother, a father, an elder brother, etc.—and the practices and norms associated with them were the primary guides to the moral life. In this sense, the family served as the basic paradigm for the well-lived life.

    Confucius never says that any family obligations or practices are guides to the moral life outside the family; I think he never suggests anything like that (unless at 1.6).

    In addition to that apparently telling absence, I think there are other absences that tell powerfully against the claim that for Confucius, norms of family relating were the main guide or paradigm for the moral life. Confucius’ account of his own moral progress makes no mention of family. He never mentions family in connection with the golden rule or his occasional claim that his doctrine has a unifying feature. Further:

    Confucius is thrice quoted as saying “Put your main emphasis on trustworthiness and fidelity” 主忠信 (1.8, 9.25, 12.10; cf. 5.28, 7.25, 15.6). He uses the term zhong 忠 to sum up how a minster should serve a ruler (3.19), to sum up how any government official should act (12.14), and to sum up the way to relate to people in general (13.19). He uses the term xin 信 to name a main virtue, to sum up how he wishes to relate to friends 朋友 (5.26) and to name the ineliminable core concern of the state (12.7). But he never uses either of these terms, zhong 忠 or xin 信, in connection with family. (The closest he comes is to say at 2.10 that if a ruler is filial and kind 孝慈, the people will be loyal 忠 to him.)

    Confucius often uses the term yi 義 to name the prime concern of the junzi (4.10, 4.16, 15.18, 17.23; cf. 5.16). But he never uses this term in connection with family.

    Ritual 禮 might be regarded as Confucius’ main guide or paradigm for the moral life. In Book 3, the only mentions of family are about mourning or sacrifice. Book 10 is full of detail about relations with rulers, friends and strangers; food, underwear, and the pajamas of home; but mentions family only in that it twice mentions a government-related ancestral temple and once mentions that Confucius paid for the funeral of a friend whose family could not.

    Outside Books 3 and 10, Confucius mentions ritual 禮 by name in 30 passages, and just three of these connect ritual with family in some way. They connect it with living family. But in none of the three does the connection seem clearly significant.

    In the first (2.5), the main emphasis regarding ritual is on relations with deceased parents, and the surface point is that ritual is a guide to filial piety (not vice versa). Further, one has to suspect that the only reason Confucius mentions ritual at all here is as a device to avoid overemphasizing the wishes of Fan Chi’s living parents (cf. 11.22).

    In the second (7.31), the connection is that Confucius either overlooked a ritual rule pertaining to family, or (more likely) suppressed the issue for political reasons.

    The third (8.2) seems to comprise two independent passages placed together, the first of which is about ritual and the second of which may well mention family. No relation between the two points is suggested or apparent.

    In sum, Confucius seems never to suggest that family relations are models or guides for ritual. (See also 3.8 and 16.2.)

    Of the 52 passages in which Confucius mentions ren 仁, 47 make no mention of family, and five mention family in some way. I have promised a separate post on 1.6. Let’s look now at each of the other four passages where Confucius mentions ren and something about family.

    “… When those who are in high stations perform well all their duties to their relations, the people are aroused to virtue. When old friends are not neglected by them, the people are preserved from meanness.” (Legge)
    … 君子篤於親,則民興於仁;故舊不遺,則民不偷。

    Here the connection to family is very indirect. There is no suggestion here that the attentiveness to one’s relations (or parents or intimates) inspires one toward ren仁 oneself. Also, ren 仁 here would seem not to be the great virtue of Humanity.

    Zhong Gong asked about perfect virtue. The Master said, “It is, when you go abroad, to behave to every one as if you were receiving a great guest; to employ the people as if you were assisting at a great sacrifice; not to do to others as you would not wish done to yourself; to have no murmuring against you in the country, and none in the family.” Zhong Gong said, “Though I am deficient in intelligence and vigor, I will make it my business to practice this lesson.” (Legge)

    This passage does not suggest that the norms of Humanity 仁 arise out of family life. It says that Humanity is largely modeled on state ritual, and that being thoroughly inoffensive to one’s clan is an attractive consequence or minor part of Humanity.

    Zi Zhang asked, “What must the officer be, who may be said to be distinguished?” The Master said, “What is it you call being distinguished?” Zi Zhang replied, “It is to be heard of through the state, to be heard of throughout his clan.” The Master said, “That is notoriety, not distinction. Now the man of distinction is solid and straightforward, and loves righteousness. He examines people’s words, and looks at their countenances. He is anxious to humble himself to others. Such a man will be distinguished in the country; he will be distinguished in his clan. As to the man of notoriety, he assumes the appearance of virtue, but his actions are opposed to it, and he rests in this character without any doubts about himself. Such a man will be heard of in the country; he will be heard of in the clan.” (Legge)

    Here Confucius mentions the clan 家 because Zizhang brought it up; he mentions ren only in a description of some fakery, and not in connection with family as distinct from state; and this ren is probably a particular rather than a general virtue.

    In 17.21, Zai Wo argues against mourning to a third year, on the grounds that for the junzi (nobles?) to mourn for three years would be harmful. Confucius replies that the junzi could not abide mourning for only one year.

    “… But now you feel at ease and may do it.” Zai Wo then went out, and the Master said, “This shows Yu’s want of virtue. It is not till a child is three years old that it is allowed to leave the arms of its parents. And the three years’ mourning [by nobles?] is universally observed throughout the empire. Did Yu enjoy the three years’ love of his parents?” (Legge)

    Here Confucius is not saying that if Zai Wo were more filial, he would have been ren. Rather he seems to be saying that if Zai Wo were ren he would, from gratitude or for some other reason, be more filial. That is not the point that filial piety would guide us to ren; it suggests rather that ren would guide us to filial piety.

    In sum, though Confucius’ students seem almost desperate in the way they press him for illumination about the virtue of Humanity 仁, and he gives them many clues, still there is no passage where he suggests that something about family would be a clue for understanding Humanity.

    In late books Confucius catalogues the junzi’s concerns at the three ages of life (16.7), three authorities the junzi holds in awe (16.8), ten things of which the junzi is mindful, (16.10), and four things the junzi hates (17.24)—all without any mention of family.

    Confucius lists five or six features of the junzi at 1.8, five at 1.14, four at 5.16, three at 14.28, and four at 15.18—all without any mention of family.

    Throughout the Analects, of the 66 passages where Confucius speaks of the junzi (君子), 63 make no mention of family at all. Another (16.1) mentions family and mentions the junzi, but quite separately. The two passages where Confucius mentions family in connection with the junzi are 8.2b and 17.21. The strongest connections we might see in these passages between being a moral junzi and fulfilling one’s family roles are that a junzi is good to his family and grieves long for his deceased parents. Confucius thinks a good person will do these things. But there is no suggestion that such things have a special or unique role in the virtue of a junzi.

    Several of Confucius’ remarks on the junzi suggest that family virtue might be a poor basic paradigm for the virtue of a junzi. For example (Legge throughout):

    The Master said, “If the junzi be not grave,…. Have no friends not equal to yourself. …”
    子曰:「君子不重 … 無友不如己者,…。」

    The Master said, “The superior man is catholic and not partisan. The mean man is partisan and not catholic.”

    The Master said, “The superior man, in the world, does not set his mind either for anything, or against anything; what is right he will follow.”

    The Master said, “The superior man … is sociable, but not a partisan.”
    子曰:「君子… 群而不黨。」

    Six passages in the Analects give short lists of the main matters that Confucius spoke of, or took very seriously, or topics that someone found it striking that he did not much talk about (5.13, 7.14, 7.18, 7.21, 7.25, 9.1). Nowhere on any of these lists do we find anything about family, except perhaps the point that Confucius did not talk about spirits (神).


    The sum of all these absences is the strong positive appearance that for Confucius, the practices and norms of the main family relations were not the primary guides, nor even among the primary guides, to the moral life; that for him the family did not in this sense serve as the basic paradigm for the moral life.

    C.6 . . . . No trump

    Ivanhoe continues,

    However, the moral life did not end with the family. One had roles to fill in society as well. There were parallel obligations to king and state which, though never superseding, were fashioned on the model of those to parents and family. …

    If Confucius thought obligations to king and state never supersede family obligations, that would be a sign that he thought family obligations are fundamental; which in turn would suggest that he might have thought other obligations are patterned after family obligations.

    I shall argue that we have no prima facie reason to ascribe to Confucius the radical view that the moral pull of king or state never supersedes the moral pull of family, and (separately) we have concrete reason to ascribe to him the denial of this view.

    (It is of course hard to pin down the conceptual distinction between saying that a state duty supersedes a family duty, and saying that because of the state duty, the putative family duty was never a real family duty. By speaking in terms of “moral pull,” I mean to sidestep that real issue. Note that if there were no real but superseded duties, then the claim that family duties are never superseded would suggest nothing about the primacy of family, because no duties of any kind would ever be superseded.)

    There are two kinds of question about evidence to consider. (a) Is there positive evidence that Confucius thought public concerns could limit or supersede the claims of family? And (b) is there positive evidence that he thought they could not? The idea that the claims of family cannot be superseded is a radical view. To ascribe a radical view, one needs evidence.


    Let us first survey the best suggestive evidence that Confucius thought the claims of family might indeed sometimes be superseded by public concerns. There is not a great deal of material on this; Confucius may not have been one to focus on potential internal tensions in his view. Still, there is enough material leaning in this direction to require us to seek strong evidence before asserting the contrary view.

    (i) At 11.8, quoted earlier, Confucius says that when his son Li died,

    “… he had a coffin but no outer shell. I would not walk on foot to get a shell for him, because, having followed in the rear of the great officers, it was not proper that I should walk on foot.”

    (ii) Confucius suggests that the duty to follow one’s father’s way can give way to other considerations after three years (1.11 and 4.20).

    (iii) He may think the rules of ritual limit the claims of family (2.5, 3.1, 3.2).

    (iv) He seems to think that the extent to which one should heed one’s family elders is a line that should be drawn in different places for different people (2.5, 11.22). At 11.22 the concern that locates the line seems to be internal to training for office.

    (v) As we saw, he says the junzi is not partisan (2.14, 4.10, 15.22).

    (vi) Of the three passages recording Confucius’ choice of husbands for his daughter and niece, two seem to present the matter as though the woman were awarded to the man as a kind of emphatic comment on a specific minor feature of the man’s public life (5.1, 11.6; cf. 5.2). If that was indeed the attitude, might it have come from Confucius?

    (vii) Confucius says it was wrong of a feudal sub-lord to use a veiled threat in asking that a successor to him be appointed from his own family (14.14).

    (viii) Confucius says of the emperor Yu that he “lived in a mean hovel, expending all of his energies on the construction of drainage ditches and canals. I can find no fault with Yu” 卑宮室,而盡力乎溝洫。禹,吾無間然矣 (Slingerland; 8.21).

    (ix) For the sake of a political relationship, Confucius seems willing to suppress mention of someone else’s violation of a rule about family names (7.31).

    (x) Duke Huan was thought to have had his older brother killed to secure the dukedom. Confucius approves Guan Zhong’s willingness to serve the fratricidal Duke, for the sake of the greater good (14.16, 14.17). Confucius calls the Duke himself “straightforward” 正 (14.15).

    (xi) At 12.4, Confucius says that the junzi has neither anxiety nor fear, because “when internal examination discovers nothing wrong, what is there to be anxious about, what is there to fear?” (Legge) 內省不疚,夫何憂何懼? What indeed?


    Is there prima facie evidence that Confucius held the radical view that the claims of family are never superseded by the claims of king or state? The main candidate for that role is 13.18; let us consider it. In this passage, echoing the Duke’s word, Confucius says that there is uprightness in sons who cover for their fathers and vice versa, on such a matter as the stealing of sheep.

    Now, suppose the only way you can raise the money to pay your taxes is to sell your mother’s ears (against her will)—and suppose, if you can, that selling her ears doesn’t happen to be against the law. If Smith said you shouldn’t do it, we would not take that as prima facie evidence that Smith thinks the family always trumps the state, or even that she would disagree with us about where to draw the line.

    Similarly, I think, Confucius’ view at 13.18 agrees with what most people in the post-industrial world would say today, at least outside of Kant class. If it’s our view too, then it is not prima facie evidence of radical familism. 13.18 is not even evidence that Confucius would disagree with us about where to draw the line.

    Confucius need not have been thinking here mainly about Family versus State. He seems to disapprove in general of airing other people’s dirty laundry in public. He thought the state ideally should press its claims otherwise than by way of penal law or judicial tort proceedings (2.3, 12.13). He may have thought that turning in one’s father is a sign of base motives. Confucius may have been thinking of the severity of the punishment for stealing (the Duke’s attitude suggests it may have been severe), and the relative unimportance of sheep.

    In any case I see no reason to doubt that he was simply making a judgment of common sense about the case at hand, the same judgment you or I might make against the Duke’s radicalism, rather than applying some universal radical principle himself.

    There are passages in the Analects that, unlike 13.18, should suggest that Confucius’ estimate of the moral strength of the claims of family on adults was greater than ours. But I am not aware of any Analects passage that could be prima facie evidence that Confucius thought the claims of family always trump the claims of king or state.

    C.7 . . . . Earliest and strongest

    Ivanhoe writes,

    … The obligations to one’s family were most important because for Kongzi they were the earliest and strongest bonds human beings form. They also were regarded as the source of our social obligations; our obligations to others were developed out of and modeled on the family.

    It is possible that for Confucius, our upward role obligations to king and government were modeled on those to elder kin; or vice versa. At least he thought both kinds of upward obligation are similar in that they involve service and respect, and at least some expression of any serious disagreement. Strictly speaking, that similarity is also compatible with the idea that he thought our family bonds are modeled on state bonds, or each modeled on something else, or each formed in similar ways by the same human nature and human condition.

    Here I want to comment on the apparent suggestion that Confucius thought the reason why the two kinds of relation are similar is that family obligations are the “earliest and strongest” bonds we form.

    It is not clear that Confucius placed such weight on the time order. Confucius never says or suggests that family bonds are earlier or stronger than all other obligations—except that there may or may not be a faint suggestion in 1.6 that we have filial duties before becoming seriously involved in public affairs, which of course we do.

    Did Confucius think the obligations of filial piety take hold of potential public servants much earlier than non-family bonds? In one late passage, as we have seen, Confucius identifies something in infancy as a ground of a later filial obligation, presumably by way of our much later awareness of it (17.21). But it seems clear that when Confucius discusses filial piety in practice, he is not thinking of the obligations of children. He is not addressing children (as Ames and Rosemont point out in “Family Reverence (孝) in the Analects,” in Olberding ed. Dao Companion to the Analects (2013), p. 125). Confucius is teaching filial piety to people who are already involved in life outside the home (1.6), people who have strong bonds with a master, people who might support parents (2.7), people who might remonstrate with their parents (4.18), people who might travel (4.19), people who might do their parents significant service (e.g. 9.16) and learn how to do so by reading the Odes (17.9), people whose general life intentions can be observed (1.11); he is thinking also of what people do after their parents have died.

    If Confucius thought the importance of filial piety lay very much in developing that virtue in childhood, he should be talking about how one should train children. But that is a topic he never touches on. So he was not thinking of children.

    Aristotle thought children cannot have real virtues or vices because their capacities are not sufficiently well developed for them to have settled habits of considered intention (1100a1, 1119b5, 1144b5, 1172a21). Now, Confucius never says anything about any virtues or vices of children (unless in public life, at 14.44). But if Confucius did think children can have filial piety, then he may have thought filial piety is simply one of many virtues that children can have. In fact, if Confucius considered children capable of virtues, he may have thought children are somewhat less capable of handling the obligations of filial piety than of, say, elder-respect in the street, trustworthiness, fairness, or friendship with other children; for many or most of the concrete duties of filial piety that Confucius mentions are beyond the capacity of children.

    I have argued that when Confucius actually alludes to analogies between family and state relations, the lord is the analog of the father (or other elder kin), at least as a patient of duties. If Confucius were to develop the analogy, the virtue of a lord or of an official qua participant in governing would presumably be the analog of the specific virtue of a father toward his offspring; not the analog of the specific virtue of a son toward his parents. But it would seem that the downward obligations of the upper family roles are mostly not among the earliest bonds we form. (True, elder brothers become so in childhood, if childhood bonds count. But many males never become elder brothers; we have no report that Confucius was ever an elder brother. Was the typical man of his time and class an elder brother? That is, did the typical family of his time and class have more than four children, more than two males surviving into adulthood? I don’t know.)

    Is there a sense in which parenthood is among our earliest and strongest bonds? A picture can be painted; for arguably no bonds are strong or even genuine but those we have as fully-formed adults. Perhaps (1) in Confucius’ milieu most people who became involved in state affairs were men who had children in early adulthood, before developing many significant bonds outside the family. And perhaps (2) during those children’s first ten years these fathers were closer to their young children than the Analects might suggest, so that not only would the sons be prized possessions, but the early parental obligations would have been among life’s strongest bonds. That is, parenting would have been an absorbing practice. And perhaps (3) Confucius thought a father’s task of governing his young children was interesting or important in some way. But if these three premises were true, then probably (4) many of Confucius’ disciples were fathers who had conversations with Confucius about parenting and were deeply interested in what he said. Yet it would seem that (4) is false. Hence it would seem that at least one of the three premises must be false. But the falsity of any one of them would be a serious problem for the idea that Confuicus thought family roles are key models for public roles because family roles are earlier and stronger.

    The fact that we have no significant record of Confucius’s views on the role obligations of a father, mother, or elder brother may suggest that he did not think these are among our strongest bonds or strongest-felt obligations; or at least that he did not give their strength the foundational importance that Ivanhoe says.

    We might also question how certain it is that Confucius would have accepted the general theory that whatever our earliest and strongest bonds are, these will or should be the model on which state relations are based. It is not something Confucius ever suggested.

    Granted, the Analects represents him as describing filial piety, or norms for relating to parents, on a number of occasions. He would appear to have thought it is one of the important virtues. But that need not mean that he thought filial piety was a natural root shaping the forms of all good relating in society and state.

    He might have thought instead, for example, that (a) our earliest bonds, perhaps weak ones, are formed when we are most malleable by circumstance: when we are children, and (b) that our earliest serious and stable bonds are formed when we are most susceptible to training: young enough to be malleable, but old enough to listen and learn. He might have thought that a good mature culture will train its young men into the sort of bonds that will serve its own broader purposes—for example, by teaching filial piety to young men. Confucius might have thought that filial piety is important (for adults) because it is a kind of dramatization of general respect for tradition and authority. Filial piety could be a spiritual exercise in support of a staunch general conservatism, without being the original blueprint of the forms and duties of state and society. Indeed, it could be true that any individual’s good family relations would prepare her well for public life, even if the basic forms of family life were designed to be introductory versions of the forms of the state, so that the family was modeled on the state.

    Or Confucius might have thought that the moral value of being an upstanding family member, and perhaps of other things too, is that in this way we train in ourselves a certain basic general strength of character, available to be directed to important purposes and practices (3.8, 5.10).

    C.8 . . . . Ideal family

    Here is the remainder of our selection from Ivanhoe.

    In the well-ordered state, society becomes the family, writ large. The king, as father—and according to some accounts both father and mother—to his people, must care for them and as a proper parent provide for them in the broadest sense of the term. Rulers must enrich their people—materially, ethically, and spiritually. The ideal state is like an extended family, providing for the needs of all its members and working for the common benefit of all while preserving a strict hierarchy in its structure and clear divisions among its constituent roles and their respective obligations and norms. Most important of all, the enlightened state, like the ideal family, is permeated at every level with a deep, lively, and particular concern for the well-being of each of its members

    (In fact Confucius never calls the ruler the father, or father and mother, of the people. He never uses family imagery for relations of political authority among people. See Part A above. –BH 11/10)

    There are two main areas of similarity proposed here between the ideal family and the ideal state: mutual concern and clear distinct roles.

    Mutual concern

    I think there is no place where Confucius actually mentions the ideal family. Still, as we have seen, there are a very few places where Confucius more or less suggests that a good son or brother is concerned for his parents or brothers. And 12.11 may reflect that Confucius has a vision of an ideal son and an ideal father.

    It seems to me offhand that Ivanhoe’s broad claims about the concerns in the ideal family must lean heavily on the “minority” reading of 16.1 that I argued against in §A.6 above (or else lean mainly on our shared assumptions about what would be ideal). That minority reading would make 16.1 the one place where Confucius comes close to actually mentioning the condition of the ideal family. At least it would make the passage speak of how a clan head (not: a father) ideally should order his clan members. And it would make 16.1 the one place where Confucius mentions something like a concern for the well-being of all or many of a family’s members.

    Even on the minority reading I think there is no suggestion in 16.1 that a deep, lively and particular concern for the well-being of family members permeates the well-governed clan at every level. The concern is depicted in 16.1 as the clan-head’s concern. It is blanket concern rather than particular concern, and it is partly indirect. Further, since 16.1 speaks of distribution of wealth (if that is how we are to understand 均), I suppose the concern would have to be understood as a concern primarily for the households of the clan, or heads of households, not for individual family members as such. Of course, Ivanhoe’s whole picture need not be painted within 16.1, for the picture to be justified as an interpretation of Confucius’ views. If the minority reading of 16.1 can be defended, one should then bring in other passages to elaborate the picture.

    I had hoped to discuss here the claim about clear distinct roles. But I’ll postpone that to another day, another thread.

  4. Bill Haines says:


    It seems to me that all the main similarities Confucius notes between family and wider society are specifically between the son-father relationship and other relations. He makes no significant use of family as a metaphor except in using the father-son relation as a metaphor for the master-disciple relation. Aside from the master-disciple relation, all the similarities he notes between the son-father relation and other relations pertain only to the responsibilities of the lower party: (a) filial piety is somehow similar to elder-respect, and that (b) as a son should serve his father with special respect and some remonstrance in case of disagreement, so an official (and perhaps any subject) should serve his ruler with special respect and some remonstrance in case of disagreement.

    But in various ways Confucius does not think son is to father as minister (or subject) is to lord. The loyalty of a son should not depend on the father’s character or conduct; but the loyalty of a minister should (and the loyalty of a subject may) depend on the character and conduct of the lord. One official should outrank another because of greater worth; but a father’s authority over his adult son is not founded on greater worth. While in the family Confucius seems to locate the main responsibility for the relationship with the son, in society at large he locates the main responsibility with the governors (though he is looser as to where to locate it between superior and inferior officials). Confucius shows no sign of thinking that a father’s character determines his son’s character (and only perhaps at 17.21 does he show some sign of thinking that a father’s treatment of his son tends to influence the son’s treatment of the father–though that is not how the passage is usually read); but Confucius emphasizes that a governor’s character and his treatment of the people determines the people’s character and their treatment of the governor.

    More generally, if and insofar as he values close personal relationships with family members (and it is not at all clear that he does), Confucius is valuing close relationships with certain people on grounds unrelated to their character; while outside the family or in general, Confucius stresses that one should pursue close relationships with the virtuous, and avoid close relationships with one’s moral inferiors.

    It is unclear that Confucius has an idea similar to our picture of “the family.” He does not discuss the family (or the clan or household) in general, nor discuss the quality of family life (moral or otherwise), as he discusses the quality of society. He does not suggest that good family relations are among the main models or guides for other relations, and there is ample indication that he did not hold that view. There is no indication that he regarded the family as the model for the state.

    [Prompted by a comment below, I would like to clarify: my proposal is that insofar as the statements attributed to Confucius in the Analects are authentic, the text gives us sufficient evidence that Confucius did not regard the institution or norms of the family as the model for the institution or norms of the state. –12/21/16]

  5. Loubna says:

    This analysis is very insightful. Thank you for posting it. I’ve been thinking about the family-state analogy in relation to Mencius, so this will be particularly helpful (Can I cite it as is?)

    • Bill Haines says:

      Hi Loubna,

      Thank you! Yes you are welcome to cite it as is (and attack it with abandon).

      I understand that it may be in harmony with your recent book, but I didn’t have a chance to look at that.

  6. Kaaazuo says:

    Why should there be a distinction between the ethical principles of a father/son relationship and those a ruler should live by?

    • Bill Haines says:

      Hi Kaaazuo,

      I didn’t mean to suggest that Confucius was right; I didn’t mean to get into that question at all above. (Though the rightness of a view is at least some slight evidence that people hold the view.) Also I’m inclined to disagree with Confucius about what kinds of rulership there should be—what kinds of political office, what forms of government—and about how fathers and sons should relate to each other.

      I think how a father should relate to his son, and how a son should relate to his father, changes with their ages. I think the role of an official toward a subordinate (an aide, a citizen, or a subject) differs from the role of a father toward a son, and each differs from the role of a son toward a father, partly because of differences in what the former parties know about the latter ones, and differences in the powers the parties hold over one another. An official and a subordinate can change places; a father and a son cannot. Etc. etc.

      I’m inclined to assume that ethics or morals has some fundamentally unified explanation, so that at some very abstract level just one principle or theory applies in all cases. But I think such a principle is not going to be such that any of us could easily see what course of action it recommends in any given case. It’s not going to be handy for use in deliberation. (This is a point people sometimes make about utilitarianism and should make about Kant’s universalization test.)

  7. Kaaazuo says:

    Bill Haines said: “I’m inclined to assume that ethics or morals has some fundamentally unified explanation, so that at some very abstract level just one principle or theory applies in all cases. But I think such a principle is not going to be such that any of us could easily see what course of action it recommends in any given case.”

    This is a marvelous statement. Just when I felt that you had trivialized Chinese thought with your analysis, you totally redeemed yourself.

  8. Kaaazuo says:

    Bill Haines said: “I didn’t mean to suggest that Confucius was right; I didn’t mean to get into that question at all above.”

    Chances are, “Confucius” wasn’t right. What we believe to have been transmitted from this fictitious person is a body of hallowed texts and interpretations produced by bureaucratic folks held in high regard by those who came after. It led to the ruin of imperial China.

  9. Kaaazuo says:

    Bill Haines said: “An official and a subordinate can change places; a father and a son cannot. Etc. etc.”

    This is a remarkable observation. In the former relationship, the parties can trade places; while in the latter, they cannot (exchange identity). And why not? Such a question would never arise in a traditional Confucian world. We, however, are studying Chinese philosophy in a modern society where no relationship boundary are exempt from the test of reason.

  10. Kaaazuo says:

    Bill Haines said: “There is no indication that he regarded the family as the model for the state.”

    What is implied here is that there is no proof, no textual evidence to support the default view. But who holds that view and how did it take root in people’s minds? And why is there an investment in shifting such a conclusion away to the one you (and Professor Chan) hold?

    There are some (and one of them is Professor Tu Weiming) who expound the view that it all begins, not with the family but, with the person. He (or she) stands at the center of society where it all begins. In (western)Confucian theory, there is no “me” without “you”. Therefore, the starting point is essentially the family as one’s transformational influence emanates like concentric rings from oneself to family members, to clan, the entire country and beyond (if he ever becomes a 天子).

    The default “Tu Weimeng” view is aspirational and served as a teaching tool for parents and teachers to guide the young in Confucian societies since time immemorial. Your view is unsentimentally “cut and dry”. It could be what we now need in a modern society.

    • Bill Haines says:

      I don’t know much about the history of efforts to characterize the views of the man or the fictional character. Joseph Chan seems to say in his article that the family-model understanding of early Confucianism comes from Liang Qichao. I’m not trying to push any philosophical agenda.

  11. Kaaazuo says:

    Bill, as a “scholar” in the Confucian tradition, any view you hold is an “agenda”.

    It’s just unfortunate that not everyone can “think”. And those who can’t (and that’s about 99% of the population), rely on those who can. Time and time again throughout history and across cultures, folks had been betrayed by the “thinkers”. How would your conclusion on the family not being the model for the Chinese state be helpful in improving the governing of China?

    Confucianism is important only because it held together the Chinese society for more than 2000 years. How it did that (despite the rule of brutal Sons of Heaven and a corrupt scholar-official bureaucracy) is not understood (at least, not by me).

    Do you think that it is ethical to study a Chinese philosophy that bears no fruit?

    • Bill Haines says:

      I am not a scholar in the Confucian tradition; I’ve hardly been exposed to it. My roots are in a different philosophical and cultural tradition, one in which the mere word of any one philosopher (as distinct from the reasons she offers) counts for approximately nothing, and one in which Confucius doesn’t have any significant symbolic weight (nor do other old Chinese figures or all but a small handful of Western figures). Most Anglophone scholars of the Analects are rooted in the same tradition.

      My idea of scholarship is that it should be a collective enterprise. One tries to contribute one’s bit of evidence and argument in the hope of furthering general understanding. See here , and my several comments starting here. Maybe part of the problem and stability of Confucianism is that it tended to be unfriendly to that process, partly by favoring permanent vertical relationships, a structure that inherently tends to give cultural dominance to leaders who, because of their power, are ignorant, insensitive, and unwise.

      I don’t know how much weight Confucius’ word carries in and around the PRC today, or how much attention mainland thinkers pay to Anglophone scholarship. But I think a prime social and political problem today and in the Imperial tradition has been the inadequate scope for open discussion, discussion concerned with truth more than with rhetorical effect. And I think people who do things just because they think Confucius said so are not thinking as much as the average person has a duty to do.

      As for the project of constructing a fictional Confucius and endowing him with symbolic authority in order to push a certain philosophical agenda, today—I suppose there are two main ways in which a community of scholars might do that. One way is to keep the fictional construction—perhaps a document adjusted once a year, called “Our Confucius”—separate from the non-fictional academic discussion of what fictions should be told and why. The other is to mix it all together, in such a way that even the participants in the project can’t quite tell what the other participants really believe and what they’re just saying (or who is and isn’t a knowing participant in the special game), and the general reading public is likely to mistake the fiction for nonfiction. I think both approaches are hopeless. Regarding the latter – people’s generally not doing that sort of thing is what allows us to have reasonable opinions about the world around us—hence about what sort of philosophical shifts might be needed. Knowledge on such topics can exist only if there is a vast horde of journalists and thinkers aiming resolutely at the truth.

      Still, the fact that Confucius is regarded as an important figure has something to do with why I think the interpretive enterprise is worthwhile. But I haven’t quite articulated for myself how that line of thought works exactly. (For example: why shouldn’t we look for philosophical ideas by working up a philosophical interpretation of any one of several hundred novelists from around the world, as a higher priority than Confucius? I haven’t worked out my answer.) As for me personally, I find myself an ex-academic with a few things left to say about Confucius, as by-products of another old project, so I figure I might as well say them. Besides, it’s fun.

  12. Kaaazuo says:

    Bill Haines said: “I am not a scholar in the Confucian tradition; I’ve hardly been exposed to it. My roots are in a different philosophical and cultural tradition, one in which the mere word of any one philosopher (as distinct from the reasons she offers) counts for approximately nothing, and one in which Confucius doesn’t have any significant symbolic weight (nor do other old Chinese figures or all but a small handful of Western figures). Most Anglophone scholars of the Analects are rooted in the same tradition.”

    So, how would teachers of Chinese philosophy, such as you, pose yourselves to prospective students (supposedly) enthralled with the prospect of getting educated on the subject? Or should I rephrase my question and ask, ‘what do you surmise is the expectation of the student herself in this regard?’.

    There are Anglophone professors of Chinese philosophy who would take exception to what you say. They have gone “native” (or appears to have) and expound on Confucianism with conviction and authoritatively stamp it with phrases in Chinese – both spoken (even if heavily accented) and written, on the chalk board (for illustration and effect).

    It is unfortunate that my roots are in the Confucian tradition and I am compelled to take every word a teacher says seriously. And you are a teacher by virtue on your encyclopaedic knowledge on the subject regardless of your disinterest and conviction.

    While I share your fun in the study of Chinese philosophy, I cannot say that it amuses me (the way it does you?) considering my belief that it holds the key to a social harmony that could end global misery.

  13. Bill Haines says:

    When I was a teacher I didn’t teach the Confucian tradition—by which here I have meant a thing of thousands of years mostly after the Analects. What I find fun in trying to figure out Confucius’ views is not that the activity is “the study of Chinese philosophy” but largely that it is like a puzzle. I think the enjoyment of puzzle-solving is a large part of the pleasure of interpreting old philosophical texts from anywhere, for the people who do it, at least in the Western tradition.

    “There are Anglophone professors of Chinese philosophy who would take exception to what you say.” Would take exception to what? I don’t imagine that e.g. Ames would disagree with anything I said in my paragraph “I am not … the same tradition.”

    It’s not clear to me whether you think the long Confucian tradition mainly takes the man Confucius seriously.

    What’s the key?

    • Kaaazuo says:

      The key is relationship of the familial kind. Everybody is somebody’s child until they start cloning us. When that happens, all bets are off. In a modern society, where traditions don’t matter, all bets are off.

      The family is the model for the state in a Confucian society no matter what Professor Chan says. This is true in the “spiritual” sense. It is not possible to learn from the Classics if one is obtuse. The “master” said as much. And yet, more than a billion Chinese in China live the tradition without understanding a single phrase of classical Chinese. If you try to understand it (Analects), you can’t live it. And your philosophical tradition doesn’t accept magic. Thus, a whole nation is held together by a teaching regardless of (inapt) ruler and (bad) government.

      Seriously, is this funny, Bill?

  14. Kaaazuo says:

    Bill Haines said: “My idea of scholarship is that it should be a collective enterprise. One tries to contribute one’s bit of evidence and argument in the hope of furthering general understanding. See here , and my several comments starting here. Maybe part of the problem and stability of Confucianism is that it tended to be unfriendly to that process, partly by favoring permanent vertical relationships, a structure that inherently tends to give cultural dominance to leaders who, because of their power, are ignorant, insensitive, and unwise. “

    For the Confucian society, unwise leaders – like diseases – are inconsequential as they come and go. It is an imperative to hang tough (忍) for the greater good. This comes from the teaching of filial piety ( 孝道 ) when a son must be a son (regardless of the father) so as to keep the family together (and society intact). It takes moral fibre to face adversity and forbearance builds character. This is a Confucian ethic to be internalized through self-cultivation.
    Apparently, the Chinese have internalized that as a people.

    Confucianism as practised by the scholar-elite did favor vertical relationships. Theirs was also a collective enterprise to stay dominant, as an upper-class, at the expense of the populace. Surely, you can tell that they were not into role-ethics of the kind the Analects speak to. The mandarins were practicing Confucianism the way Senators and Congressmen are practising John Stuart Mill to uphold the US Constitution in the service of the American people.

    “Thinking together” sounds good only if everyone can think and participate in the process. East or West, this is not a practicable reality.

  15. Bill Haines says:


    My claims about Confucius are in no sense claims about “Confucianism” (or vice versa), no matter whether we take “Confucianism” to mean the main lines of the long philosophical tradition after him or the pattern of East Asian society. I’m inclined to think Confucius’ views (insofar as these are discoverable in the Analects) were misread from pretty early on, and I suspect that this is partly because Youzi organized the beginning of the Analects for that purpose. Chan is talking about “early Confucianism,” the cluster of pre-Qin (?) philosophies, rather than the long social or philosophical tradition.


    I shouldn’t have said so simply that permanent vertical relationships are “a structure that inherently tends to give cultural dominance to leaders who, because of their power, are ignorant, insensitive, and unwise.” I should have said: a system of permanent vertical relationships has certain strong tendencies to (a) corrupt leaders’ sources of information, strip them of empathy, and rob them of adequate critical feedback, and (b) heighten the cultural and moral influence of whatever leaders there are. There could at the same time be countervailing institutions and tendencies. For example, it has sometimes been argued in favor of permanent vertical relations that they give the dominant parties (patriarchs, brahmins, plantation owners, slaveholding citizens of Athens) leisure to study and think.

    As for whether there is any workable way to organize post-industrial society, that question seems harder every day.


    When I said above that scholars’ mixing fabrications inextricably with their actual views about Confucius for social effect, in writing for each other, is a “hopeless” project I added this comment: “people’s generally not doing that sort of thing is what allows us to have reasonable opinions about the world around us—hence about what sort of philosophical shifts might be needed. Knowledge on such topics can exist only if there is a vast horde of journalists and thinkers aiming resolutely at the truth.” To this point one might object that it is possible to do the one thing in the one area and the other thing in all other areas.

    My main reply is that what makes a somewhat reliable wide discussion of general issues possible is certain general norms and practices, and intellectual institutions that encourage intellectual creativity and encourage vigorous testing against evidence (e.g. by rewarding people for thinking up new theories and refuting popular ones). Because of these norms, practices, and institutions, tale-telling about Confucius or the Analects would be caught out. Therefore, to survive, the mixing team would have to suppress the general institutions, practices, and norms. That’s not possible (in some places today), but if it were possible it would have the effect of stunting knowledge generally.


    Kaaazuo: “ ‘Thinking together’ [is] good only if everyone can think and participate in the process.”
    “Only” — That seems to me obviously false. I say more at the first link in my July 5 Reply above.


    I have some scattered thoughts on the relation of family and society in general, but I don’t have a developed view on the topic, and I don’t have time to enter on a vast exploration of it. I hardly have time over the next few months to do the other things I’m obligated to do. If you have a developed view about family & society you might lay it out and defend it in the Discussion section, and I and others might have some comments.

  16. Kaaazuo says:

    Bill Haines said:

    Kaaazuo: “ ‘Thinking together’ [is] good only if everyone can think and participate in the process.”
    “Only” — That seems to me obviously false. I say more at the first link in my July 5 Reply above.”

    I did review your first link. When I said “everyone”, I did mean everyone impacted by the collective thinking process and not just scholars or “experts” in the “think-tank”. I am talking about Confucian scholarship in the business of statecraft that has consequences throughout society – this means every man, woman and child affected by governmental decrees. In any case, “thinking together” among the few for the many is an abomination.

    East or West, scholars are not wise men. And yet, they would insinuate themselves as intellectuals in public affairs either as government officials or voices of popular dissent. In doing, so they, invariably, become a nuisance to society.

    I enjoyed our chat but will not take up your invitation to pursue a discussion on the family and society. Although I have the time, I don’t have your rigor that is necessary for an informed debate.


Leave a Reply

Your email address will not be published. Required fields are marked *