The meaning of Analects 2.21

Someone said to Confucius, “Master, why don’t you engage in government?” The Master said, “The Book of Documents says, ‘Filial! But be filial, and a friend to your brothers, thus contributing to government.’ Why then do that other kind of ‘engaging in government’?”


I’ll suppose for the sake of argument that the reported exchange is authentic, and argue that it is not significant evidence of Confucius’ views.  Confucius is not aiming to communicate his views here.

He seems to say here that filial piety (and fraternity) amounts to a kind of participation in governance.  But there is no apparent suggestion about how that would be so.  Confucius does not elsewhere display a view on that topic that could argue against the importance of his holding office.

Would it have been out of character for Confucius to deflect a challenge about his lack of political achievement, or a feeler about his interest in an appointment, by way of a hopelessly vague statement serving mainly as a smokescreen? The only three points the remark at 2.21 suggests with any clarity are points Confucius probably regarded as false:

(1)  Confucius had no strong reason to pursue public office.

(In fact he had long been deeply concerned to find high public office.)

(2)  Confucius expected to continue doing plenty in the way of filiality and fraternity.

(In fact there would be relatively little occasion: for his parents were long dead, and probably his only brother had also passed.)

(3)  A person’s ongoing practices of filiality and fraternity make it less important for that person to serve in actual office.

(In fact, according to the rest of the Analects, Confucius thought that one’s personal virtues greatly increase the importance of one’s holding high office, and vice versa. He thought it crucial that the worthy be promoted and the rulers be virtuous.)

We might be tempted to think that whatever Confucius thought about his own case, he accepted the view 2.21 displays about the general relation between family virtue and government.  But it seems to me that 2.21 does not display any view on that topic.

One could speculate that Confucius did not mean to suggest or appeal to point (2) about his own family relations, but rather was arguing in general terms from the premise that if people in general were filial and fraternal, then there would be no need for them to serve in government, or no need for any particular talented and virtuous person to serve in government (as there would be plenty of adequate candidates).  But (a) there is no indication elsewhere in the Analects that Confucius held such a view.  And (b) that premise could hardly support unconcern with office in a real world that needed improvement.  So if this is the suggestion he meant to make, then, I submit, his remark was not designed to communicate his actual views on the question.

Even granting all these negative points, one might object that in light of Confucius’ general view that the family virtues are the root of all social and political virtue or norms, we may fairly take him at 2.21 as alluding to that general view, putting an exclamation point to it without articulating it or applying it in an intelligible way.  The remark could have succeeded in making such an allusion if the questioner knew that Confucius held that core view (but then why make the allusion?).  For my part, I do not myself know that Confucius held that view; I think he did not.  This post is part of my argument for that.

More in the Comments below.

Hannah Pang detail

7 replies on “The meaning of Analects 2.21”

  1. A. Background speculations

    I have read somewhere that the use of the name ‘Kongzi’ by the narrator in passages in later Books of the Analects is a sign of the relative unreliability of those Books as reports of Confucius’ words. But since my thesis is that 2.21 is not illuminating about Confucius’ views, I’ll assume for the sake of argument that it is authentic.

    Legge’s note says that the use of the name ‘Kongzi’ in 2.21 indicates that the questioner was not one of Confucius’ students. The Brookses say that it “indicates a high-ranking questioner.” These inferences suit my reading of the passage; but I do not understand their basis, as it seems to me the name is used here by the narrator, not the questioner.

    An indication that the questioner was not one of Confucius students is that he is not named. Another indication is the question itself. A new student would perhaps not ask such a question of the master, but would ask it of other students, after which he would not ask it of the master—not so bluntly.

    The question suggests that Confucius has an established reputation, as does the sententiousness of his answer. And the generality of the questioner’s phrase wei zheng 為政 suggests that the questioner is asking not about some particular job offer, but rather about Confucius’ continuing absence from government. Thus it suggests that Confucius is settled in one place and not pursuing office. Confucius’ reply suggests that the conversation occurs in a place that one could suppose to be not very far from the home base of Confucius’ family. Hence the conversation, if real, would seem to have taken place in the final two or three decades of Confucius’ life, after his return from Qi (as Lau suggests) or after his return from Wei.

    Perhaps he should look young enough that one would not assume his parents are long dead. But not necessarily. If the remark was an empty deflection of an insult, or a disavowal of any intention to join or oppose the current leadership, then its rationale need not have been completely plausible. (17.20 reports that Confucius sent word to a caller that he was not at home, but then played and sang so the man could hear.)

    B. What general doctrine?

    Confucius at 2.21 suggests that filiality and fraternity amount to some kind of wei zheng 為政. This might mean that they amount to “doing government business” in some expanded sense. At any rate it seems to suggest that they have a high degree of import that is in some way related to governance. Degree aside, what kind of connection might he have in mind here between governance and a non-officeholder’s being filial and fraternal? That these virtues are contagious and thus promote public order? That they make one an obedient instrument of the government? That one proper end of government is the people’s virtue, so that one’s own virtues are constituent parts of one of the ends of government?

    His remark at 2.21 seems not to articulate any particular kind of connection (at least for an audience unfamiliar with the document he is quoting, now generally thought to be lost). So far as we can see, it presents no such idea. One might speculate that it is trying to suggest some such idea. What idea?

    Let us set that hard question aside for a moment, and consider first the specific points that Confucius does seem to aim to impart at 2.21, the clear suggestions in what he says.

  2. C. One clear thing: Confucius and office

    One point, I think, is quite clear in Confucius’ remark at 2.21. Confucius intends to suggest to the questioner that in Confucius’ view, Confucius’ holding a position in government is of little or no importance for Confucius’ purposes.

    Granted, he only suggests this by way of a rhetorical question. Still, he says it more clearly and definitely than he says anything else in 2.21. So if we are to take him at his word on anything in 2.21, we should start with this point.

    Yet Confucius seems to have held the opposite view. Confucius’ teaching of others was primarily training for their government service. And he seems to have thought it was very important that he himself attain high office. What held him back was the lack of opportunity to serve under a ruler who would accept what Confucius had to offer.

    He said that instead of worrying about not having an official position, one should worry about being suitable for a position (4.14). He told Zigong that he was eager to sell his gem, and was waiting only for the right buyer (9.13). He said that if anyone were to use him, he would do great things in a short period (13.10). He lamented that nobody knew him (14.35). Strangers knew of him as someone who kept trying despite the hopelessness of the quest (14.38, 14.39, 18.5). He had a conversation based on the premise that he was anxious to get a government job (17.1). He even toyed with the idea of serving under some plainly illegitimate rulers (17.5, 17.7).

    Further, for what it’s worth, 18.7 reports Zilu saying to a hermit that it is wrong not to serve in government (不仕無義). The passage reports Zilu saying this while traveling with Confucius, though not in his presence. Zilu would have mentioned it to Confucius, and presumably he would not have preserved it for posterity if Confucius had disagreed at the time. So if the passage is authentic, the view is likely to be Confucius’ own.


    Against all this, in connection with 2.21, Hall and Ames cite Confucius’ statement at 8.13 that when the Way does not prevail one should conceal oneself (Thinking Through Confucius, p. 185), hence not work in a bad government:

    The Master said, “Uphold the faith, love learning, defend the good Way with your life. Enter not into a country that is unstable; dwell not in a country that is in turmoil. Shine in a world that follows the Way; hide when the world loses the Way. In a country where the Way prevails, it is shameful to remain poor and obscure; in a country which has lost the Way, it is shameful to become rich and honored.” (Leys)

    I think we should not take this remark as a renunciation of the project of seeking to lead a country toward the way by working toward finding a high position in which one’s advice will be respected by the leadership, and training others to do the same. The point seems rather that just as one should not risk one’s life foolishly by involvement in violent situations, one should not risk one’s life foolishly by vocal public criticism of the leadership; nor should one be a stable part of a corrupt establishment, invested in its continuation and serving as its ornament.

    Confucius is likely to have thought he should not speak very openly about his reasons for not holding office in Lu, late in life. His real reasons are likely to have been one or both of two related kinds of reason. First, he might have thought himself above any formal office that would actually have been open to him (cf. 18.3). He probably would not want to give this explanation to outsiders. Second, and perhaps more importantly, he may have had reasons that amounted to serious criticisms of the leadership. And he is likely to have thought he should refrain from expressing such criticisms, especially to people outside his close circle, whether in or out of office (consider 1.10, 7.11, 7.15, 7.31, 8.13, 8.14, 14.3, 15.1, 15.7, 16.10). Once, when challenged about flitting from place to place, he said the reason was a dislike of stubbornness (14.32). The opposite answer, that he was stubborn, would have been less misleading; but it would have been a criticism of his various hosts.

    We should take seriously the idea that 2.21 is an act of self-concealment, concealing Confucius’ general views and concerns. Indeed, at 11.22 he admits using rhetorical questions to give even his own students disparate impressions about his general view of the relation between family deference and broader aspiration.

    For these reasons I think the presumption should be that at 2.21 Confucius is mainly just deflecting, speaking to someone he is willing to mislead on at least one core point of his own moral outlook: the importance of government service properly so-called, among the aspirations of a talented and virtuous person in his milieu.

    D. The second-clearest thing: Confucius and his family

    If anything else is clear in Confucius’ statement in 2.21, it is that he there suggests that his own filial piety (and perhaps his friendliness with his brothers) is in some sense an adequate alternative to his holding government office. He is at least suggesting that he is doing, now and in future, plenty in the way of filial piety.

    Amy Olberding comments that his remark at 2.21 “serves to remark the limitations of what he can offer. Despite his considerable learning, Confucius’ contribution to governing is indistinguishable from what any unlearned yet filial citizen can offer” (“Confucius’ Complaints and the Analects’ Account of the Good Life,” Dao Dec. 2013, p. 423f).

    But Confucius probably did differ from most people, by having far less occasion than most to exercise filiality or fraternity.

    There is a report that Confucius’ father died when Confucius was two, and that his mother died before she was 40, when Confucius was no older than 23. That has been the traditional understanding, and even the Brookses report it simply as a fact (The Original Analects, Columbia 1997, pp. 10, 270; and B. Brooks, “The Life and Mentorship of Confucius,” , Sino-Platonic Papers No. 72 (May 1996) p. 27—see n.42 on the early death of Confucius’ mother). Robert Eno answers some doubts about the tradition, in “The Background of the Kong Family of Lu and the Origins of Ruism,”, Early China 28 (2003), and reports it as a fact in his teaching translation of the Analects.

    The fact that Confucius is said to have chosen a husband for his elder brother’s daughter (5.2, 11.6) strongly suggests that Confucius’ father and elder brother were both deceased by the time Confucius reached middle age (but the Brookses propose that the brother may only have been handicapped: OA, p. 269). Confucius never mentions either of his parents, and neither Confucius nor, I think, any source, mentions any interaction or event involving Confucius and a parent (unless we count 9.16, discussed here ). I think there is no report of any other brother.

    Thus it would seem that when the conversation at 2.21 took place, Confucius had long had no prospect of acting in a filial way toward living or recently deceased parents, and probably no occasion for friendliness with any brothers. He could only sacrifice to his long-dead parents, including a father he had never known.

    Now, when Confucius describes filial piety 孝 in the Analects, he presents it mainly as a way of relating to living parents (2.5 to Meng, 2.6, 2.7, 2.8, 11.5; cf. 1.6). Several of his other surviving remarks speak of how to treat living parents (4.18, 4.19, 4.21, 13.18). There is a secondary special emphasis on recently deceased parents (1.11, 2.5 to Fan Chi, 17.21, 19.17, 19.18), which would seem not to apply at 2.21. There is one laudatory mention of someone who did not change his father’s policies or officials (19.18), a kind of filiality available only to holders of high office. There are one or two distinct mentions of sacrifice to dead parents (2.5, 2.24; cf. 3.12, 6.22, 11.12; but see the end of 11.26); Confucius could still sacrifice. In office he might have been in a better position to do that.

    Perhaps in office one is in a better position to find prudent matches for one’s family members (cf. 5.1).

    The tradition about Confucius’ parents argues for either rejecting the authenticity of 2.21 on the grounds that it claims a living parent, or else reading Confucius’ remark at 2.21 as mainly a smokescreen.

    (Slingerland translates 2.21’s “友于兄弟” as “friendly to one’s elders and juniors.” I think he is alone in that. He may have based this choice on his understanding of the noun you 友 “friend,” about which he says elsewhere that

    in the Confucian context, you refers not to random acquaintances or childhood buddies but, rather, to a subset of the professional colleagues of a given “gentleman”—the cultivated scholar-politician who is the target of Confucian education—who are more or less of the same rank/seniority, and with whom the gentleman, to borrow a concept from Aristotle, shares a vision of the “Good.” You marks out those colleagues whom one finds personally amenable and who share one’s moral aspirations.
    (“The Situationist Critique and Early Confucian Virtue Ethics,” Ethics 121:2 (2011), p. 402n.31)

    But offhand it seems to me that the standard case of a professional colleague at that time would have been a colleague in government. It is hard to think that Confucius would have expected the general phrase “友于兄弟” to be understood in that way in the conversation at 2.21.)

  3. E. Unclear general doctrine

    How could filiality and fraternity be an adequate alternative to government service? What could Confucius have been thinking in 2.21, if his parents and brother were alive and he thought his good treatment of them made it unimportant that he hold office?

    Let’s look at the passage again:

    Someone said to Confucius, “Master, why don’t you engage in government?” The Master said, “The Book of Documents says, ‘Filial! But be filial, and a friend to your brothers, thus contributing to government. Why then do that other kind of ‘engaging in government’?”

    The question “Why do that other kind of ‘engaging in government’?” (奚其為為政) amounts to the suggestion that filiality and fraternity amount to “engaging in government” or “doing government business” in some extended sense (at least if we assume that the duplication of 為 was in the original remark, as I shall assume here). It thus suggests that we are to read the quoted string “施於有政” in a way that suggests the same. What exactly does this quoted string mean?

    Authorities understand the string in diverse ways:

    Soothill: “are shewn forth in the public service”
    Legge: “These qualities are displayed in government.”

    Ames & Rosemont: “is carrying out the work of government.”

    Waley: “You will be contributing to government.”
    Watson: “this contributes to government.”
    Sarkissian: “You will be an asset to those governing.”

    Slingerland: “exerting an influence upon those who govern.”
    Lau: “can exert an influence upon government.”
    Chin: “this has an influence on the way of government.”

    Another candidate might be, “treat the governors that way.” For at least Slingerland and Sarkissian read “有政” here as meaning “those who hold government positions.” (In general, I gather, “有政” means governmental policymaking. Still, perhaps the absence of “者” is not dispositive against the Slingerland and Sarkissian readings: cf. e.g. 1.14.) If they are right about that, then perhaps the default hypothesis about the meaning of “施於有政” in the original Document should be that it was speaking simply of filiality and fraternity toward family members serving in government, and was not a description of any general effect of those virtues. Thus at 2.21 Confucius might have meant, stretching the original a little, “be filial and fraternal (as it were) toward the government. Obey the government with a smile, and in that way help in governance.” On the other hand, the distinction within the quotation between “于” and “於” seems to argue against this sort of reading for the original Document.

    The original Document is lost to us now. One passage in the “Old Text,” generally thought to be a counterfeit from many centuries after Confucius’ time, has something close to the quotation in 2.21. Appointing someone to be a local ruler, the king says:

    It is you who are possessed of excellent virtue, filial and respectful. Being filial, and friendly with your brethren, you can display these qualities in the exercise of government. (Legge)

    Here the string “施有政” may mean “exercise [these qualities] in office,” in line with Legge’s rendering of 2.21. Or might it simply mean “do policymaking” or “carry out [our] policy”? “Filial and fraternal, you are qualified to carry out governance,” qualified to hold office.

    (If, as is generally thought, this part of our Book of Documents is a fake, why would it not reproduce exactly the received wording of Confucius’ quotation, and why would it seem to oppose his argument?)

    But even an accurate reading of the document Confucius had in mind would be of limited value for understanding his remark. Although Confucius chose the quotation, that point gives us no basis for confidence that he meant the same as what the passage originally meant, or even what he thought it originally meant. (The word ‘filial’, for example, is thought to have meant something different in the Documents from what it meant in the Confucian group.)

    In sum, so far as I can see, the phrase “施于有政” in Confucius’ mouth here gives us no clear indication about what kind of connection he might have meant to indicate on this occasion, between the family qualities and government. What more it might have meant to his original questioner, we do not know. I suspect it was not very intelligible.

    F. Speculations on the family/state doctrine behind 2.21

    In evaluating proposals, we should distinguish two questions:

    How might someone’s filiality and fraternity constitute “serving in government” or “working for the government” or “doing government business,” in a broad sense?

    For any proposed answer to the first question — How might that feature of the pair of virtues support the point it is used most directly to support in 2.21? That is, how might it imply that someone who is filial and fraternal is thereby relieved of important or compelling reasons to serve in public office?


    Let’s review some hypotheses.

    1. . . It doesn’t mean much.

    This is the reading I favor. Confucius was deflecting an insult or a feeler from the governors, by a remark that on the one hand sort of suggests that one’s family virtues make a broad contribution to the success of governance, but on the other hand commits him to nothing at all, or to something very minimal at most.

    One such minimal thought might be that in being filial and fraternal one is in that respect doing what the government presumably asks of one, doing one’s part to serve the government’s program—just as one does in obeying any law. The basic idea would not be specific to the family virtues (though we could take the mode of expression of that basic idea here as indicating that Confucius assumed that good family relations were generally seen—were presumably seen by the questioner—as the main duties of a private person as such).

    Another such minimal thought might be that in being filial and fraternal one is setting a good example for the neighbors. Perhaps Soothill was right that “施於有政” means that those good ways of relating to one’s kin are “shewn forth” in public service: i.e. that high office makes one a moral model. Perhaps Confucius expected the questioner to understand that lords and high officials are traditionally supposed to be visible models of virtue. Perhaps then the thought at 2.21 is that even without holding office, Confucius would be a visible paragon, displaying to the neighbors those same ways of relating to family. The neighbors might be inspired to be more filial and fraternal themselves. Now, it may be hard to adduce this point to argue for the unimportance of taking office, simply by quoting an authoritative statement of the idea that office makes virtue visible. Still, one might conceivably suggest by such a quotation that private virtue could have an effect similar in kind, if not comparable in degree.

    Perhaps for any grand theory we might want to read into 2.21, there is a minimal version of it, a modest common-sense view, that Confucius might have thought his remark would suggest, a view that does not really support unconcern with office. Any such reading would be consistent with my main point that Confucius’ remark at 2.21 was mainly a smokescreen and tells us nothing of interest about his general views.

    Again: I favor the minimal reading of 2.21 because I think Confucius’ parents were long gone, he thought it very important that he find high public office, he thought a person’s virtue (e.g. filiality) would increase rather than decrease the importance of her serving in government, he believed in dissembling about his reasons for not holding office, and 2.21 articulates no general view.

    But what are the alternative readings?

    2. . . The contagion of a civilian’s virtue.

    Perhaps Confucius thinks his filiality (and perhaps fraternity) will be emulated by others, and the ripple effect will be so great as to compensate for his not working in government.

    But as I argued recently, the evidence outside of 2.21 is that Confucius probably did not think that the particular virtues of someone out of office would tend to inspire emulation on a large scale. He certainly held that the personal virtues of people in high office make a major contribution to society. (Part of how the virtue of governors leads to virtue in the people is that the people respond well to being treated well; they reciprocate. But one’s filiality and fraternity cannot be reciprocated by the neighbors.) So a concern for the spread of virtue would argue for the urgency of getting the virtuous into office, promoting the worthy to high office, and leading the lord to virtue (for which one needs his ear); it would not argue against that urgency, as the contagion reading of the remark at 2.21 would have it.

    Even if Confucius did have a grandiose view of the contagion of private virtue, his own in particular, I suppose he would not have expected an outside questioner to share that view, or to read it into a remark that at most alludes to the view very indirectly.

    3. . . One’s own broader virtue.

    On this reading, Confucius’ thought is that one who is filial and fraternal will have the other virtues too, and then these will have some good effect without his holding office. De Bary writes (in “Why We Read the Analects of Confucius,” in De Bary ed. Finding Wisdom in East Asian Classics (2011), p. 53, perhaps following Daniel Stevens in “Confucianism, Pragmatism, and Socially Beneficial Philosophy,” JCP (2009:1), p. 59):

    Confucius’s answer [at 2.21] is somewhat terse and a little oblique, but it takes us back to where we started in the Analects [i.e. 1.2]: filiality as the value underlying all social and civic virtue. Public service is not performed only by those in office; anyone who practices and promotes such civic virtues is rendering a public service. And indeed the practice of such values is the precondition for anyone who might qualify for office.

    (1.2 does not single out filiality.)

    I find it hard to think that Confucius would have thought his questioner would catch the Youzian line of thought from Confucius’ remark, or add the leap to “and promotes.”

    De Bary speaks of values. At 1.2, Youzi says that one’s filial piety helps root one’s own civic virtue psychologically, apparently on grounds of some similarity to civic virtue. But Youzi does not say that the values specifically embodied in the two root virtues ground the value of, or are the main values served by, civic virtue. And I think Confucius never suggests that.

    De Bary may be proposing further that having the virtues reduces the importance of holding office because the civic virtues are a necessary qualification for office. But he does not say how the latter point would support the former. I cannot see how it would. (Others who seem to find the point about qualifications at the heart of 2.21 are Li Chenyang (”Equality and Inequality in Confucianism,” Dao 11.3 (2012), p. 307) and Ni Peimin (Confucius: The Man and the Way of Gongfu (2016), p. 90f.; cf. Analects 13.20; but they too offer no explanation of how this point would relate to Confucius’ surface conclusion.)

    Slingerland proposes in his note under 2.21,

    There are probably two layers of meaning here. The more general point is that one should “do government” through “not doing” (wu-wei): that is, by perfecting oneself—as Master You puts it in 1.2, establishing the “root” of virtue—and letting the rest follow naturally through the power of one’s personal example and Virtue. Some commentators also see here an indirect criticism of the Ji family, whose usurpation of power in Lu involved shocking mistreatment of parents and brothers. Cf. 12.11 and 13.3

    Slingerland’s second layer, that the remark involves a very veiled criticism of the Ji family, strikes me as a plausible speculation, assuming that the circumstances fit.

    Slingerland’s first layer may be interestingly different from De Bary’s reading. The idea here may not be that the questioner is supposed to know or hear Youzi’s developmental theory; rather Slingerland may be taking filiality and fraternity simply to stand for personal virtue, especially the virtue of someone in private life. This they could do simply by being virtues, even if Confucius’ had never heard of Youzi’s theory. For, first, there is a certain latitude because Confucius makes his point here mainly by quoting a text, no doubt constrained by the need for an echo of wei zheng and perhaps by the concerns of the second layer, about the Ji family. (Indeed the concerns of the second layer could have led him to use a text.) Second, for people in general, private life would have meant mainly mainly family life. (Granted, Confucius usually emphasized virtues other than filiality and fraternity in talking with his followers about greatness; but theirs was a special project, of training for public service, while Confucius’ questioner here was probably from outside the group.)

    Indeed, if the questioner was a person of political importance (as the Brookses think is signaled by the narrator’s use of the name ‘Kongzi’), it is possible that the question was a probe to discover where Confucius stood, whether he was a threat or potential ally. In that case, Confucius may have intended here mainly to suggest that he was focused only on humble private life. He wanted to “spend more time with his family,” as they say. I think he might have expected his remark at 2.21 to be taken as a credible disavowal of an intent to make trouble for the establishment, even if he expected the questioner to know that his parents were both long dead.

    Still, while Confucius may indeed have had half a mind to think that one should pursue office without pursuing office, i.e. by pursuing virtue and hoping for the best, that is a far cry from thinking that doing so would be likely to generate the benefits of good government even without office. We do not have an account here of a general view that could make that idea plausible.

    4. . . The people’s virtue is the wind.

    Perhaps Confucius meant that if the people in general, including himself, were to focus on being good homebodies, the governors would be inspired or liberated to rule well. But such a thought would seem to go against his usual line.

    Hall and Ames found this thought in 2.21:

    “Sociopolitical order is ultimately derived from and hence must be restored at the most immediate level, moving from the more distant political order toward its ground in familial and personal order.” (Thinking Through Confucius (1987), p. 185.)

    But elsewhere in the Analects, Confucius’ view seems to be that the current disorder could be remedied if a ruler were to become virtuous or heed a good minister.

    Also 2.21 does not suggest this theory of the primacy of the people. An argument relying on an unstated premise can be taken to suggest the missing premise, if it is easy to think that the premise would complete the argument. But the above theory does not on its face support the conclusion of Confucius’ argument at 2.21: that it is not incumbent on him to seek or hold office.

    In any case, the interpretive argument here seems to rely on one or the other of two premises, neither of which is explicit: either (1) the premise that Confucius thought, and expected his questioner to undersand, that the contagion effect of private family virtue is comparable to the effect of an official’s family virtue; or else (2) the premise that Confucius meant, and expected his questioner to understand, only that if all or most were filial and fraternal, there would be good political order. I argued against (1) not far above, and argued against (2) in the Post proper.

    Erin Cline writes, elaborating on 1.2 (in Confucius, Rawls, and the Sense of Justice (2012), p. 126):

    In particular, we should notice that 1.2 mentions that one who is filial does not stir up a rebellion. The idea that filial piety constitutes the roots of political order prevails in the Analects. One of the clearest formulations of this idea is found in 2.21.
    [Slingerland’s 2.21 is here quoted.]
    It is difficult to imagine a clearer indication of the connection Kongzi sees between the cultivation of filial piety and political order. He thinks that the government alone cannot bring about a stable and harmonious society; rather, such stability must be initially cultivated in the context of the family. Members of society must develop certain dispositions, learning to think and feel for others in certain ways, if there is to be political order. The family serves as the model for the ideal state in this regard.

    Youzi does say that one who is filial and respectful of elders rarely stirs up rebellion. But none of the points Cline presents here is formulated in 2.21, and none of them appears in Confucius’ other remarks in the Analects. (I plan to argue in a later post against the idea that for Confucius, the family was the model for state.)

    [Correction: One of the ideas Cline presents may appear in 15.25. In Confucius’ words: “…The common people today are the same people who allowed the Three Dynasties to put the upright Way into practice” (Slingerland) 斯民也,三代之所以直道而行也. But those people are already at hand in 2.21; they are not to be created by Confucius’ filiality and fraternity. –12/19/16]

    One could read Confucius’ “Why do that kind of governing?” as proposing that formal governance need not be done at all: where there is family virtue, government is unnecessary tout court. But I do not see any real connection between this point and the idea Cline finds in 2.21: that when the populace does not have the Way, the mere addition of good governors would be insufficient to remedy the problem. Confucius seems often to deny this latter point, presumably with reference to the same society, and the same sort of society, that the conversation at 2.21 contemplates.

    But suppose we take Confucius’ “why do that kind of governing?” as proposing only that it is not important that he himself hold office. How then would his remark at 2.21 suggest the view Cline sketches—that good government “alone” cannot bring about social harmony? Maybe the thought is that somebody has to be working the populace directly, as one of the people, and that’s where Confucius sees the most need for him. But the thought would have to be that private virtue is somehow distinctly more contagious than is virtue in office. I think Confucius never suggests anything along those lines.

    5. . . Sustaining the institution of the family.

    The sociologist Edward Shils proposes what may be a slightly different reading.

    Perhaps Confucius means that maintaining the family is a contribution to maintaining public order or social harmony and hence is a contribution to the work of the government. The family has its place in the Way, and whatever sustains the family is an act of adherence to the Way. This in turn fortifies adherence to the Way in the realm. Service to the state does not exhaust the performance of obligations of the gentleman-scholar in relation to the state. The gentleman-scholar ‘cultivates himself and thereby bring peace and security to his fellow men … He cultivates himself and thereby brings peace and security to the people.’”
    (“Reflections on Civil Society and Civility in the Chinese Intellectual Tradiiton,” in W. Tu ed. Confucian Traditions in East Asian Modernity (1996), p. 49)

    But the point that “service to [sic] the state does not exhaust the … obligations of the gentleman-scholar in relation to the state” does not imply or suggest that service in the state is not one of the gentleman’s obligations. A gentleman can have more than one obligation. Imagine arguing that you need not pay your debts because you should be brave – without pointing to any apparent conflict between the two.

    What does almost all the work in Shils’ argument is the tacit premise that strengthening one’s own family strengthens “the family” in society non-negligibly, and indeed significantly more than one could strengthen “the family” from high office, despite the policy tools available to a government and the contagion of virtue in office. Here the assumption would have to be that virtue out of office is distinctly more contagious than the same virtue in office—or else that for some other reason serving in government would be a bad idea: specifically, because the current rulers are just too bad. The more conclusive is this latter consideration, the less need there is to appeal to any other consideration to support the conclusion that Confucius may or should stay out of government.

    We never see Confucius in the Analects discussing “the family” as a kind of institution, mentioning its health or weakness, either as a general sociological fact or in regard to a particular family. In just one place he comes close to speaking of family virtues as spreading in society: at 8.2, where what may be family virtue by someone in high position is expected to generate something like kindness among the people. The account of the effect does not focus on the family.

    In the passage Shils quotes, where Zilu has asked about the gentleman, Confucius does not say that the gentleman cultivates his family (14.42). (Zilu may need to be redirected away from a focus on others, on outside enemies.) Confucius may well have thought that while out of office one should focus on one’s own virtue rather than focusing on seeking office, not because he was thinking of strengthening the one’s family, but rather because seeking the good opinion of others is a distraction from virtue, while the only person who can really bring peace and security is a virtuous person in office.

    Roger Ames writes in Confucian Role Ethics (2011), p. 167f.:

    Since family feeling is the ground of Confucian role ethics, and since polity in this tradition is a direct extension of the family as quite literally “country-family” (guojia 國家), Confucius can further claim that being a responsible and productive member of one’s family is tantamount to governing the country. This is the claim that Confucius intends when some mean-spirited person, aware of Confucius’s life-long frustration at never having been appointed to any important political office, feigns surprise at Confucius’ failure to secure a station appropriate to his gifts:

    Someone asked Confucius, “Why are you not employed in governing?”
    The Master replied, “The Book of Documents says: ‘It is all in family reverence! Just being reverent to your parents and a friend to your brothers is carrying out the work of government.’ In doing this I am engaged in governing. Why must I be employed in governing?”

    This passage can be easily misunderstood as minimalist—that is, each one of us in our families makes our own small contribution to the greater political order. I believe, however, that Confucius’s point is quite the opposite. Most of the real significance of our political lives (and our religious lives too) transpires close to home. If we ask after the relative importance of the state and the family in effecting cosmic harmony, we must allow that family is the ultimate source and ground of political order, and in the absence of the flourishing family and the thriving community it enables, political order is a sham or worse. It is for this reason that any formal pretense to be a strong state independent of the thriving community is an empty abstraction: what Whitehead would call “misplaced concreteness.”

    (The argument of the first sentence above would be a non sequitur even if being filial and fraternal amounted to governing one’s immediate family. Also the term 國家 does not appear in the Analects; nor does any reference to cosmic harmony.)

    Ames seems to acknowledge that the minimal reading is a natural reading of 2.21. If I understand correctly, there are two defining differences between what he calls the “minimalist” position and the “opposite” position here.

    The first difference is that the “opposite” position may involve a larger estimate of how big a fraction of good political order simply is good family life. Good political order is good relations among the people, and the main relations among the people are internal to their families. (Though relations among families were salient loci of political disorder in Confucius’ time.)

    But to understand 2.21, what we want is an explanation of how a person’s family virtues would promote political order outside that person’s family. And this first part of the “opposite” view does not address that point.

    The second difference between the “minimalist” position and the “opposite” position is that only the latter thinks the health of the family in society is a key support of the other aspects of political order (though Ames’ account of the minimalist position does not suggest to me that the minimalist would disagree with this).

    But to understand 2.21, what we want is an explanation of how one person’s family virtues would promote wider political order in significant measure. And this second part of the “opposite” view does not speak to that point.

    7. . . Teaching these virtues.

    Perhaps Confucius took it as understood by his questioner that his moral teaching centers on filiality (and maybe fraternity). His mention of those virtues might then almost have sufficed to indicate that Confucius is speaking here of his activity as a teacher. His point could then be that by spending his time teaching virtue rather than working in government, he has an important political or societal influence. On this reading, his remark at 2.21 was not meant to suggest anything about his own family relations.

    But I wonder whether teaching filial piety was such a prominent part of Confucius’ teaching project that he would have expected an outside questioner to understand his reply in this way. Teaching friendliness to brothers was not a prominent part of his project. I hope to show soon in another post or two that Confucius’ teaching did not not center on family virtues.

    Still, conceivably at 2.21 he meant only that teaching virtue, or teaching the family virtues, is worthwhile. Someone who teaches virtue can at least feel that he is making a contribution to society, even if he does not serve in office.

    This reading proposes no general view about the relation between family virtues and government. It does not imply that Confucius himself had much occasion to exercise the family virtues. On this reading Confucius may be dissembling about his political ambitions, or his remark may reflect that at an advanced age he thinks there is no longer any realistic prospect of realizing his lifelong aim of guiding a government.

    That’s the end of my prepared Comments.

  4. You have certainly pointed this out in your comments, but I think the key thing to recognize is that Confucius does not, at this point, want to ‘engage in government’ in the literal sense. It may in fact be the case that Confucius thinks for contingent reasons that literally engaging in government, at this point in time, does not express ‘filiality’ or contribute to good government, both of which the Book of Documents indicate the gentleman must do.

    In essence, Confucius is contributing to government, whether or not he “engages” with the governments which rule the various warring states. On my reading, being filial in accordance with the Book of Documents has little to do with having a living mother or father to venerate, but with taking one’s proper role in society in general (as constituted partially by family, and partially by the other communities, including the relationship to the ruler.)

  5. Hi William, thanks for the attention and the good point! And I’m pleased to meet you!

    You are right that I underemphasized the point that Confucius was probably not interested in a government position in Lu at that time.

    To be really careful and precise about it, I’d say:
    I agree that since Confucius was not working in government, he did not want a position of the sort that was available to him where he was settled. I don’t know whether any position was available to him at all in Lu around the time of the conversation. If the question at 2.21 was a feeler, then something was available to him; but maybe the question was just a “mean-spirited” taunt, or the opening gambit of a thinker hoping for a serious conversation, or an ignorant expression of adulation (as I might say to Wallace Shawn: “Why don’t you take more serious roles?”), or simply a naive inquiry. Confucius might still have wished for a high position, there or elsewhere, if he would have been allowed to perform its duties. At this point he presumably preferred to stay in Lu rather than to travel in search of unknown possibilities.

    My impression is that on the surface, Confucius’ rhetorical question at 2.21 is supposed to sound like it’s expressing a very general point that has application to him in particular; not just expressing a point about him in particular. Hence my point “(3)” in the post proper (which I did not formulate well). The scope of the actual or putative general point is left unclear: to what sort of circumstances would it apply? Nothing in the phrasing suggests any narrow limitation, e.g. to circumstances where those in power are invested in opposing traditional principles of legimitacy, etc. Hence Confucius’ interpreters tend to read fairly grand theories into the remark.

    I think we don’t know how xiao 孝 was conceived by the Document from which Confucius was quoting. But if we read “友于兄弟” as referring to actual brothers, living ones, then I think it would be strange to read xiao 孝 in that passage in the way you describe. Do you think Slingerland’s reading of “友于兄弟” is likely to be correct?

    We don’t know whether Confucius had reason to think that the questioner would have been familiar with the Documents and what the term xiao 孝 might have meant there, different from what it generally meant in Confucius’ time. But the fact that Confucius cites the “Documents” rather than a particular Document may be some slight evidence that he did not expect the questioner to know the Document in question??

    Be all that as it may, —
    I think maybe you agree that the passage is not making any big claim (or evidencing any big view) to the effect that relations among nearish living family members are the heart of governance or social order, or an adequate alternative to formal government? Whatever Confucius may think about those things, on your reading he is not displaying here that he holds a view about that that would be controversial today?

  6. My view is grounded in the fundamental assumption that whatever Confucius says (and whatever the recorders of the Analects cared to write down) is meant to serve the purpose of illustrating the concept of 君子 in order to educate. That is, by quoting the Document, he is not attempting to elucidate the Document itself, or appealing to it in order to support his view.

    Regardless of which motive we are to impute to the questioner, Confucius’ response is the proper response of a 君子. I believe that this incontrovertible fact is why people tend to construct “fairly grand theories” based on his remark. But in reality, I do not think it is important to be spending time trying to figure out the circumstances in which this Document would apply, because doing so would imply that I believe there are circumstances in which a 君子 ought to deny that the Document is wrong, and not merely inapplicable. The fact of the matter is that Confucius is applying the Document to his situation, right now, and if we are looking at the Analects to understand his viewpoint, we should first of all recognize that Confucius himself had better to be correct if we plan on getting anywhere.

    I do not think 友于兄弟 refers to Confucius’ living brothers. But re: your response to Slingerland’s interpretation, I want to ask, why should one’s elders and juniors be limited to those who are part of one’s social circle? Very few people interact only with those in their social circle, and I personally would not hold up any of these individuals as paradigmatic 君子. People ought to be friendly to all, not only to their brothers, and certainly not only to those who share our socioeconomic background, for instance.

    Nor do I know enough to offer a definitive comment on whether 孝 was differently understood from the time of the compilation of the Book of Documents to the time of Confucius. But the fact that Confucius saw himself as a transmitter of the wisdom of the Zhou, should be enough reason to believe that Confucius, at least, understood 孝 as well as anybody could claim to understand King Wu’s embodiment of 孝 towards his father, King Wen.

    孝 is a virtue which is certainly necessary for being a 君子. But being a 君子, while one might participate in government and carry out his responsibilities very well, does not necessarily include participating in government. Being one does, however, necessarily include participating in society in the proper way, in particular, 友于兄弟.

  7. Hi William,

    I think my reading of 2.21 does not require that Confucius is lying or mistaken, or that the lost Document is mistaken. My reading does require that Confucius is speaking misleadingly by way of a rhetorical question, but my reading does not necessarily imply that he is doing something he thinks a junzi would not do. He seems to think it is OK to mislead in certain cases, at least by rhetorical questions or silence. I think Confucius misleads in the initial conversations at 11.22. He does seem to think that a junzi should conceal himself when the way does not prevail, and that is a kind of misleading. He also thinks fathers and sons should cover up for each other.


    I’m not sure I understand your reply on this.

    My own knowledge of ancient Chinese is limited, so for fine points I rely heavily on reputable translators and occasional text searches at the CTP. So far as I know, almost all translators render 兄弟 at 2.21 as brothers. The only exception I have noticed is Slingerland, who translates it as “elders and juniors.” I wondered why. I looked at his translations of the same phrase in its other appearances in the Analects:

    12.5 “elder and younger brother”
    13.7 “brothers”
    13.28 “brothers”

    So I thought something special about the context at 2.21, something not found in the other cases, must have led him to think that the term was being used there in an unusual way. And I remembered his footnote about 友 from an earlier discussion of something else he had said in the article:

    Consider the closest thing we could find to a term for ‘honesty’ in Warring States (sixth century to third century B.C.E.) China: the virtue term xin 信, usually translated as “trustworthiness” or “reliability.” There is always a bit of debate concerning the precise connotation of traditional virtue terms such as this, and their usage also has varied somewhat from thinker to thinker and over time, but what is beyond dispute is that the scope of xin is confined to a gentleman’s professional behavior toward his colleagues, superiors, and/or inferiors, …

    I had found that claim troubling, both because (a) this article in a top mainstream journal would be one of the mainstream philosophers’ main encounters with Chinese philosophy; the piece received much on-line discussion at the time by mainstreamers; and (b) the claim amounted to a grave moral charge against early Chinese moral thought. Also I had never before encountered this idea about xin 信, which Slingerland reported as the exceptionally solid consensus interpretation, and I suspected that Slingerland was radically misrepresenting the consensus: that in fact his view of xin 信 was beyond dispute only in being beneath it. But not much is beneath dispute by me, so I asked the Profession and did my own research. I seemed to find that the claim was wrong, but I have never got response to my evidence&arguments, nor any response directly to my question about what other people think on this point. (My interaction with the profession is basically only on this blog.) It’s all nested under David Elstein’s comment 11 in this Jan. 4, 2012 thread. What do you think?

    This idea about xin 信 might be relevant for our discussion about 2.21. If the virtue name xin 信 in Confucius’ mouth referred only to professional interactions, then perhaps we do not have a record of him saying that honesty is an important virtue for non-professional interactions, such as (probably) 2.21. So even if the purpose of the passage is to illustrate the junzi, the remark could still be meant to be seen as misleading.

    So anyway—I thought this idea that 友 is only about professional relationships was probably what led Slingerland to avoid the “brothers” translation of 兄弟 at 2.21. Which would mean that so far as I knew, nobody was thinking that兄弟 here refers to elders and juniors in general. So I never considered that as a candidate interpretation of the term here. But I’m interested, especially because of this.

    So my question to you or anyone is: in early texts, e.g. the New Text Documents, was兄弟 sometimes clearly used in that very wide sense, as referring to elders and juniors in general?

    And do you agree with Slingerland about 友?

    “孝” in the Analects

    William, do you think that in Confucius’ remarks in the Analects (when he is not quoting others), he usually or ever uses “孝” to mean specifically “ taking one’s proper role in society in general (as constituted partially by family, and partially by the other communities, including the relationship to the ruler)”?

Leave a Reply

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

This site uses Akismet to reduce spam. Learn how your comment data is processed.