Warp, Weft, and Way

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

Is Analects 1.2 about family?

Here are some reasons to think that Youzi did not regard family as the root of humanity or of the Way.  (I used to think he did.)

Most of my argument focuses on defending a view held by Soothill, Leys, Chin, and maybe Lau and Slingerland: that by 弟 in Analects 1.2, Youzi meant elder-respect, a virtue commonly associated specifically with life outside the family.  It would follow that according to 1.2, only one of the two parts of the root of humanity is specifically a family virtue.  If 孝 and 弟 have something relevantly in common for Youzi, family isn’t it.


Youzi says, “One who is filial and respectful of elders, but likes to go against his superior, is rare. One who does not like to go against his superior, but likes to stir up trouble, has never been. The junzi works on the root. When the root is established, the Way grows. Filiality and elder-respect are the root of his practice of ren, yes?”


My main argument for this reading of 弟 is simple: it seems unlikely that Youzi would offer as the root of the junzi’s humanity a combination of virtues that is obviously unavailable to eldest sons.  We should not read Youzi such a way as to imply that eldest sons cannot have the root of humanity, if there is any other way to read him.

I’m not sure why this argument hasn’t impressed itself on me earlier.  Is it new?

I’ll supplement it in prepared comments below.   Hannah Pang detail

May 16th, 2016 Posted by | Analects, Chinese philosophy - 中國哲學 - 中国哲学, Chinese Texts, Confucianism, Confucius, Education Models, Filial piety, Moral Psychology, Roger Ames, Role Ethics, Ruism, Self-Cultivation | 22 comments

22 Responses to Is Analects 1.2 about family?

  1. Bill Haines says:

    1a. Linguistic evidence: the Analects

    Quite aside from the above consideration, I think the linguistic evidence from the most nearly contemporary texts leans in favor of elder-respect as the right reading at 1.2, the main virtue Youzi had in mind as 弟.

    I believe we have no record of Youzi using the term (or discussing these topics) on any other occasion. So the most directly relevant linguistic evidence about what 弟 might mean in 1.2 would seem to be usage by other speakers in the Analects, even speakers Youzi may not have met, when they are using 弟 as a name for a trait, and especially when they are speaking of it as a companion virtue to 孝.

    The character 悌 never appears in the Analects, at least not at the Chinese Text Project.

    Only one other person in the Analects uses 弟 to name a character trait or a way of relating to others. Confucius does so in three places. In two places, he presents 弟 as a companion virtue to 孝, one whose main arena is outside the family. In Legge’s translation (which I’m not endorsing):

    The Master said, “A youth, when at home, should be filial, and, abroad, respectful to his elders. … “

    … Zi Gong pursued, “I venture to ask who may be placed in the next lower rank?” And he was told, “He whom the circle of his relatives pronounce to be filial, whom his fellow villagers and neighbors pronounce to be fraternal. ” …
    … 曰:「敢問其次。」曰:「宗族稱孝焉,鄉黨稱弟焉。」…

    Perhaps 弟 as elder-respect made a familiar pair with 孝, as the non-family companion virtue of 孝.

    Here is the third and last place where Confucius uses the term, this time paired with 孫:

    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)

    The other Analects passages where 孫 is used to name a trait are: Zigong at 17.24, and Confucius at 7.36, 14.3, 15.18, 17.25. Throughout, the reference is pretty clearly to adults, and not to family relations. (The character 遜 never appears in the Analects, at least not at the Chinese Text Project.)

    Thus, even apart from the question whether half the root of humanity might be unavailable to firstborns, usage in the Analects outside of 1.2 strongly suggests that 弟 in 1.2 refers to elder-respect, regarded as a virtue whose main arena is outside of the family.

    • Bill Haines says:

      I want to elaborate on my argument here a little.

      Youzi seems to give little overt indication as to whether by 弟 he means (a) subfraternity or (b) elder-respect or 遜. That is one reason to think the term did not need disambiguation in his time, was not simply ambiguous in his time—which would mean that usage in contemporary or earlier texts would match what he meant. So he meant (b).

      But let us suppose for the sake of argument that the term was ambiguous as between (a) and (b); that is, let’s suppose that despite the meagre evidence about contemporary usage that I’ve collected in this string, in fact the term was commonly used both ways.

      Then we should look for signals within 1.2 as to what Youzi meant there.

      (How much reliance we should place on those signals depends on whether we can suppose that Youzi thought he was giving adequate signals as to which he meant, signals internal to the remark. Now, I think as a general rule one should suppose that an author thinks she’s giving adequate signals. And the Analects seems to display the genre of Sayings-to-Be-Cherished-Over-a-Long-Period, hence presumably through different contexts; such a saying should be such as to travel well (at least for one person). Further (or alternately), as I suggest in “Purloined,” Youzi’s remarks seem to be writings rather than mere remarks someone took down; this too suggests that they may have been composed with the intent that they travel well from context to context.)

      What signals do we find within 1.2, as to whether 弟 there means (a) or (b)?

      One might think: we find none. Insofar as that is true, that would be evidence that the term was not ambiguous, so that Youzi meant (b). But we’re hypothesizing now that the term was ambiguous.

      And indeed there are maybe some signals. The most salient is that the term is paired with 孝.

      I can think of two ways in which that might be a signal.

      First, it might have been standard to pair 孝 with 弟 in one sense of 弟, but not in the other sense. So far as I can see in contemporary texts (as reported above and below in this thread), this was the case; suggesting that Youzi meant (b). (Indeed the premise that Youzi thought his signals sufficient, plus the premise that his main signal was the pairing with 孝, implies that if he meant (a), we would be unlikely to find 孝 paired with 弟 in sense (b). But we find plenty of the latter. So the premises would imply that he meant (b).) Indeed, a leading idea of the pairing of the two virtues in other texts is to make it explicit that one is covering both family life and non-family life.

      Second, one might feel that the pairing of 孝 with 弟 makes better sense, makes up a more coherent whole, in one sense of 弟 than in the other.

      Now, “respect for elders inside and outside the family” looks like a nice coherent whole, yes? And we do occasionally find Confucius and others occasionally mentioning this pair.

      How about “respect for parents and for elder brothers?” This looks like an odd choice from the point of view of the kind of reading that sees 1.2 as suggesting that family role virtue in general is the root of ren—the kind of reading that avoids a narrow focus on upward relations.

      Objection: From the point of view of that kind of family reading, one might defend the intrinsic importance of precisely that pair of virtues on the grounds that they are the first family virtues a child might need—or indeed the first virtues period.

      Reply: That point has no application unless Youzi was in fact thinking of childhood. But his approximate contemporaries such as Confucius don’t seem to have been thinking of childhood (nor especially early childhood) when they were thinking of filial piety. (We don’t see them thinking about subfraternity.) And when Youzi says the junzi attends to the root, I take him to be speaking of attending to one’s own 孝弟.

      But suppose it is wrong to see Youzi as trying to emphasize family role virtue in general. Suppose it is wrong to see him as offering 孝弟 as a sort of sample to suggest family virtues in general. Maybe the pair “filiality and subfraternity” can make sense as a coherent or complete whole of some other kind?

      How might we characterize that whole?

      The pair doesn’t amount to “respect all the elder males in one’s family,” because (a) it includes one’s mother, (b) it probably excludes uncles and elder cousins, and (c) it doesn’t certainly include grandparents. And (d) someone who respects all the elder males in his family may still be ineligible to do subfraternity.

      The pair doesn’t amount to “respect all the elder males in one’s family who are in one’s household.” For, again, Youzi is probably not focusing on the virtues of children in nuclear family households. One might not be living with one’s father or elder brother, and one might be living with uncles or cousins. And there’;s still the problem about mother, and the problem about eldest sons.

      The pair might almost amount to “respect your closest male relatives,” depending on how closeness is calculated; except that of course there’s still the problem about mother, and the problem about eldest sons.

  2. Bill Haines says:

    1b. More data from old texts

    Here are a few further amateur forays into the pre-Qin linguistic data, based on searches at the Chinese Text Project. I would be grateful for counsel from the better-qualified.


    None has 悌.

    In the Documents, 弟 never names a trait.

    In the Odes, 弟 names a trait only as part of a compound term “豈弟” (愷悌)that Legge translates as “delighted and complacent” or “happy and courteous” or “easy and self-possessed,” applied mostly to lords.

    In the Book of Changes at the CTP, 弟 simply means “younger brother” except in one passage in the Tuanzhuan (Warring States period?): “父父,子子,兄兄,弟弟,夫夫,婦婦,而家道正; 正家而天下定矣。” But insofar as this passage is evidence of the use of 弟 as a familiar adjectival verb for the way a younger brother should relate to an older brother, the passage is equally strong evidence that all the the other verbalized nouns in its list were familiar as adjectival verbs for the special virtues of their respective roles. And I gather the other terms were not at all familiar as adjectival verbs.


    Here are the passages in the Mozi where 弟 (or 悌) refers to a way of relating to others, given with the CTP’s headings and numbering, and Mei’s translation from the CTP.

    Both meanings are present. “Elder respect” is the meaning in at least 2 out of 3 places in the “I” essays, 1 out of 2 in the “II” essays, and 0 out of 3 in the “III” essays—as though the meaning were shifting toward what etymology of the written form suggests.

    Anyway the following rule seems to account for all cases in the Mozi: Wherever just two basic roles are listed, one characterized at least in part by “孝” and the other characterized at least in part by “弟”, this latter character refers to elder-respect outside the family.

    – – – – Elder-respect – – – –

    Exaltation of the Virtuous II, Par. 4
    … At home the vicious are not filial to their parents, and, having left their home town, they would not recognize their elders. …
    … 入則不慈孝父母,出則不長弟鄉里 …

    – – – – Both meanings in one paragraph – – – –

    Anti-Fatalism I, Par. 5
    … the people were filial to their parents at home and respectful to the elders in the village or the district.…
    …入則孝慈於親戚,出則弟長於鄉里 …

    … the people would not be filial to their parents at home, and respectful to the elders in the village or the district. …
    …入則不慈孝於親戚,出則不弟長於鄉里 …

    … the father would not be affectionate, the son would not be filial, the elder brother would not be brotherly, and the younger brother would not be respectful. …
    … 為父則不慈,為子則不孝,為兄則不良,為弟則不弟 …

    In the last instance above, for the proper trait of an elder brother, the text uses a word (良) that is not special to the case. Might the word used there for the proper trait of a younger brother (弟) also not be special to the case, but simply an application of a general term for respect for elders to the case of a younger brother’s respect for his older brother—especially as the term was already used for elder-respect twice in this paragaph? And may something similar be going on in other passages below?

    – – – – Being a good younger brother – – – –

    Universal Love III, Par. 12
    Therefore, universal love is really the way of the sage-kings. It is what gives peace to the rulers and sustenance to the people. The gentleman would do well to understand and practise universal love; then he would be gracious as a ruler, loyal as a minister, affectionate as a father, filial as a son, courteous as an elder brother, and respectful as a younger brother. So, if the gentleman desires to be a gracious ruler, a loyal minister, an affectionate father, a filial son, a courteous elder brother, and a respectful younger brother, universal love must be practised. It is the way of the sage-kings and the great blessing of the people.

    Simplicity in Funerals III, Par. 6
    … When there is insufficiency, the undutiful younger brother will ask his older brother for help, and when he does not receive it he will hate the elder brother. …
    … 若苟不足,為人弟者,求其兄而不得不弟弟必將怨其兄矣 …

    On Ghosts III, Par. 1
    … father and son, elder and younger brother are no longer affectionate and filial, brotherly and respectful, virtuous and kind. …
    … 父子弟兄之不慈孝弟長貞良也 …

    Against Confucianism II, Par. 7
    … Such a man will not be loyal as a minister, filial as a son, respectful in serving an elder brother or gentle in treating the people. …
    … 以是為人臣不忠,為子不孝,事兄不弟,交,遇人不貞良。…


    Outside the string 孝弟 or 孝悌, there is only one place in the Mencius where 弟 or 悌 refers to a character trait or a way of relating to others:

    …Here now is a man, who, at home, is filial, and abroad, respectful to his elders … (Legge)
    … 於此有人焉,入則孝,出則悌 …

    In all the other places where 弟 or 悌 refers to a trait or a way of relating, it is part of the string 孝弟 or 孝悌. Such a string appears six times in the Mencius. 孝悌 in 1A, and 孝弟 in 6B and 7A.

    In 1A3 and the similar passage in 1A7, the context suggests elder-respect.

    In 1A5 the context leans toward brothers.

    In 6B2 the 弟 in 孝弟 clearly refers to elder-respect.

    In 7A32 the context may leave the matter open. On the one hand the compound is offered as applicable to young people generally (hence presumably including firstborns); on the other hand the term used for young people is 子弟.

    In 7A39 the context suggests but does not require that 弟 in 孝弟 pertains specifically to younger brothers.

  3. Bill Haines says:

    1c. Authorities

    Authorities are divided on how to understand 弟 in 1.2.

    Some who have translated 1.2 understand its 弟 as elder-respect: Chin, Leys, Soothill, Lau pretty much, and Slingerland perhaps. Others prefer the virtue of a younger brother: Ames & Rosemont, the Brookses, Chan, Dawson, Huang, Ivanhoe, Knapp, Legge, Shun, Waley, and Watson. (In Confucian Role Ethics (2011), Ames gives both readings on p. 88 without noting that they differ; cf. pp. 176, 187f.)

    Lau translates 弟 at 1.2 as “obedient as a young man.” Is elder-respect 弟 a virtue only or mainly for young men? I think we have no reason to ascribe such a view to Youzi. Confucius at 13.20 gives elder-respect 弟 as one of the two virtues of the second-best kind of government official. The Mozi regards elder-respect 弟 as a virtue (and its absence a vice) in the people in general. In the Mencius at 1A5, elder-respect 悌 is a trait for adults (壯者) to cultivate. At 1A3 it is a trait people in general should have; at 1A7 it is a virtue that should be taught in the schools. At 3B4 it is a trait for those who should be supported by the farming class because they preserve and transmit the way of the former kings. At 6B2, elder-respect 弟 is a trait people in general should have.

    Keith Knapp, in the article “Ti 悌 (Fraternal)” in X. Yao ed. The Encyclopedia of Confucianism (2003), p. 604, takes the view that the basic meaning is the virtue of a younger brother (e.g. at 1.2, which Knapp attributes to Confucius), while elder-respect is a kind of extended or figurative application. If he gives an argument here, it is the etymology. He reports instances from the Liji where the meaning appears to be elder-respect.

    • Bill Haines says:

      Additions: Robert Eno and David Hinton read 弟 at 1.2 as elder-respect.

      I have just now come across Douglas Berger’s proposed translation of 孝弟 at 1.2: “filial deference.” On this reading, 1.2’s root is all about the family even though 弟 is not read as subfraternity (in “Relational and Intrinsic Moral Roots: A Brief Contrast of Confucian and Hindu Concepts of Duty,” Dao 7 (2008), p. 158).

    • Bill Haines says:

      Other authorities who render Youzi’s 弟 as elder-respect: Kim-chong Chong, Erin Cline before 2015, Sam Crane, Hongkyung Kim, Ryan Nichols, Charlene Tan, and Bryan Van Norden.

      But it is unclear to me whether those who choose to render Youzi’s 弟 as elder-respect see the choice as having any significant implications, or even see elder-respect as a virtue that could apply outside the family at all (much less as applying primarily outside the family, as we see in at least three early texts). For example, when Van Norden translates Youzi’s 孝弟 as “filiality and respect for elders”, he says in the same breath that 1.2 is a suggestion “about the importance of the family” (Virtue Ethics and Consequentialism p. 128, and similarly in the intro to his Mengzi tranlsation, p. xxi). And while Kim-chong Chong translates Youzi’s 孝弟 as “being filial toward parents and respectful of elders” in more than one work, he also uses “filial love” as a convenient shorthand for the whole (Early Confucian Ethics 151 n.54—I did something similar myself in “Purloined”). And while Hongkyung Kim consistently translates Youzi’s 弟 as compliance with one’s elders, he also takes one’s elder brother as the relevant core case of an elder (Analects of Dasan vol. 1 p. 38). [Oops, it’s Dasan who says that’s the core case.] Indeed, sometimes authorities seem to think no choice is involved at all.

      [After writing the above I saw Bryan’s new essay on the Mencius, and discussed the question a little further, under this.]

      [Heiner Roetz translates Youzi’s 弟 as “fraternal deference”, but explains that he means a general respect for elders: Confucian Ethics of the Axial Age (1993), p. 55 . But he comments at p. 309 n. 90 that 1.2 suggests that “other virtues can be conceived of as expansions of family morality.” –12/22/16]

  4. Bill Haines says:

    1d. Semantic shift?

    From what little I can see (above), the appearance is that 弟 (as a trait) was “elder-respect” first, and perhaps came to mean the virtue of a younger brother after Confucius’ time or Youzi’s time. I gather Confucian texts after Mencius often use 弟 to mean the virtue of a younger brother, when they are talking about the value of 孝 and 弟; though a proper survey is beyond my ability.

    Might there have been a shift of meaning over time, toward the virtue of a younger brother?

    Is it possible that later interpretations of Youzi were influenced by such a shift, despite the extreme philosophical awkwardness? I think it is much easier to overlook the problem about eldest sons when you are referring to a hallowed saying than when you are composing it in the first place.

    Is there a record of the phrase 孝弟 antedating Youzi?

    I suppose the prima facie etymology of the written form of the trait-name 弟 (or 悌) might have fueled a shift over time in how the word was understood. The etymology of the written form is always highly visible, whereas actual usage in Confucius’ day was not easy to observe a few centuries later. Indeed even today we may see scholars trying to push the meanings of ancient words in directions suggested (even slightly) by etymology.

    (I wonder whether another possible cause of a shift from elder-respect toward youngerbrotherliness might have been the political frustrations of Ruism (and other intellectual movements). People in training to be Ru are presumably self-selected to be people whose fathers and elder brothers (if any) are not against that sort of thing; but one cannot select the elders one encounters outside the home. As Ruism became a larger, more self-conscious and more peculiar project, might it have come to look silly and pretentious (cf. 14.9)? Might it have been increasingly unpopular among village elders, village worthies? Or even if there was no such change, might the Ru have been increasingly sensitive to a tension between their pedantic political pretensions and local elder-respect?)

  5. Bill Haines says:

    2a. Objection: 父兄

    In two passages in the Analects, someone pairs “父兄” as though the position of older brother were closely comparable to that of father. (It’s Confucius each time.) Such comparability might make sense: one’s older brother is the only other older male in the nuclear family. Thus we seem to have at least some contemporary evidence that one’s duties to father and elder brothers already made up a recognized key pair of duties, however unavailable the whole pair may have been to eldest sons. This fact weighs in favor of reading Youzi’s 弟 as the virtue of a younger brother.

    Here are the two passages:

    … “Your father and your elder brother are still alive; how could you practice at once what you have just learned?” … (Legge)
    … 「有父兄在,如之何其聞斯行之?」…

    The Master said, Outside, he serves Prince and nobles; inside, he serves father and elders. In funeral services he does not dare to be remiss; he does not get into trouble over wine. What difficulty has this for me? (Brookses)

    This not a great deal of evidence, especially when we note that aside from reports that Confucius found a husband for his elder brother’s daughter (5.2, 11.6), there is no other mention in the Analects of elder brothers as distinct from younger brothers, or vice versa. We find only “兄弟” and “昆弟”—and there are problems about both times Confucius uses one of these strings, as I’ll argue on another occasion (viz. section C.3 under my 6/17/16 post). (At 14.16 and 14.17, Confucius endorses a certain tolerance, for sufficiently important political ends, of the murder of an older brother; but he does not mention that it is an older brother.) This paucity of mentions would seem to be evidence that Confucius did not regard the specific fraternity of a younger brother as a key virtue.

    Nor is it clear that 9.16 and 11.22 actually show Confucius holding that view.

    [ CORRECTION: Confucius uses “兄弟” or “昆弟” in four remarks. Only two speak of role responsibilities, and these are problematic. ]

  6. Bill Haines says:

    2b. Problems about 9.16

    There are difficulties about the authenticity of 9.16.

    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 accept it without hesitation (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.

    For this and other reasons, the Brookses argue that Confucius would have been well known not to have done most of the good things listed in 9.16, so that “the ‘Confucius’ of this passage must thus be fictive, and seen at the time as fictive” (Original Analects, p. 54). That is, whoever wrote the remark expected all Confucian Ru around that time to regard it as fiction.

    If we decide the remark is inauthentic, I imagine we can have no strong opinion about its date, so that it would have little significance as evidence about what Youzi meant.

    But it is hard to accept the idea that the passage is so obviously inauthentic that it must always have been meant to be seen as such. Is there a good defense of the remark’s authenticity or prima facie authenticity? I am not sure.

    One way to defend the remark’s authenticity is to find an interpretation of the closing string “何有於我哉” that does not seems to say that Confucius did the things listed (or had occasion to). The string “何有於我” appears also in 7.2, and in no other early text at the CTP.

    The Master said, “The silent treasuring up of knowledge; learning without satiety; and instructing others without being wearied – which one of these things belongs to me?” (Legge)
    (“… these present me with no difficulties.” —Lau)

    (何有於X appears in a few other old texts, none of which strikes me as helpful here.)

    Where 9.16 seems an almost dismissive list of conventional good points, perhaps of an official, 7.2 lists activities distinctive of Confucius in his view—though the activities mentioned are among the humbler points he might have mentioned as distinguishing him. But translators aim to interpret the string in the same way in both passages; some read it Lau’s way and some Legge’s. Annping Chin adds a note to 9.16 with an idea I have not seen elsewhere:

    Most scholars agree that the weight of this remark lies in the last part of the sentence, which could mean that these four things are so fundamental to human conduct that either “I have no trouble with them” [heyou yu wo] or most people could do these things and so “what need would there be for me” [heyou yu wo]?”

    I can offer some linguistically ignorant speculations about the closing string that might at least save 9.16 from being prima facie fictional.

    Conceivably in 7.2 and 9.16, Confucius is saying, “What objection can I have to any of these things?”

    Conceivably the “我” in 7.2 and/or 9.16 is purely notional and simply means “a person,” like the placeholder “you” in conversational English or the placeholder “I” in much philosophical writing. Is there any evidence of such a usage elsewhere in pre-Qin Chinese? It might not be easy to recognize.

    Conceivably the difficult string simply meant “What’s the big deal about these? Why should I be concerned?”—that is, denying some possible ground of concern supplied by the context, now lost.

    I am not competent to evaluate these proposals; I do not know if any of them can suffice to avoid the worries about the authenticity of 9.16.

  7. Bill Haines says:

    2c. Who are my 父兄?

    In searching the CTP for early uses of the string “父兄” I found that this string is absent from the Documents and Changes, and appears in the Odes only as part of the longer string “父兄弟”. Later, “父兄” was often used in contexts where (I gather) it cannot mean father and elder brothers, e.g. in one passage in the Mencius and usually in the Liji. Legge’s translations: aged relatives, uncles and elder cousins , uncle or elder cousin, uncles and cousins (父兄 or父兄弟), uncles, cousins, and uncles, brothers, and cousins.

    This material suggests to me that perhaps at 11.22 and even 9.16, 兄 need not mean older brothers. Service to 父兄 need not have been an activity unavailable to eldest sons.

    At 11.22, is the clause “your 父兄 are still alive” (有父兄在) a clue that the 父兄 contemplated are all of the previous generation?

    At 9.16, could a wide reading of “父兄” as “elder male kin” defend the statement’s authenticity, or at least its prima facie authenticity?

    I hope someone better-qualified can help out here.

    At present I find myself without any solid reason to think that Confucius ever mentioned or alluded to a specific duty toward older brothers.

  8. Bill Haines says:

    3. On translating xiao 孝 as “family reverence”

    It is very common for scholarly writers mentioning Analects 1.2 to report falsely that it says that filial piety is the root of humanity—even immediately before or after quoting Youzi’s claim that the root is 孝弟. I have done it myself.

    Because we so often tend to overlook the 弟 half of Youzi’s root, the impression that 1.2 is about the importance of family-as-such can be massively strengthened by translating 孝 as “family reverence” (i.e. revering the family ) or even “family feeling,” as Ames and Rosemont sometimes do.

    That kind of translation invites us to read into 1.2 a claim about the importance of family relations in general, all family relations, rather than just the one or perhaps two kinds of upward family relating that Youzi actually mentions. We might be glad of the invitation, for we might feel that the upward relating of good offspring (and younger brothers) does not make a good model for the virtue of a leader; we may feel that a much better model would be the downward relating of good parents. Or we may feel that what grounds our humanity is the whole panoply of family relations. We might want to read that idea into Youzi and thereby into Confucius. Henry Rosemont writes,

    For Confucius, the roots of ethical conduct are grounded in the family, and he focuses on the responsibilities of family members as determined by their role(s) in family life: Father, mother, son, daughter; grandmothers and grandfathers; brothers and sisters; aunts and uncles.
    A Reader’s Companion to the Confucian Analects (2014), p. 40

    In fact, in the Analects, Confucius says a good deal about the responsibilities of a son; says almost nothing about the responsibilities of father, mother, and brother; and makes no mention of the other six positions.

    [ CORRECTION: He mentions uncles if that’s what he means by 父兄, but says nothing of their conduct. ]

    On p. 1 of The Chinese Classic of Family Reverence (2009), Rosemont and Ames give what I think is their main defense of their new translation of 孝 as “family reverence” or “family feeling”—translations that at least strongly suggest that 孝 is equally apt in all directions within the family, and is not mainly an attitude toward one’s parents.

    The replacement of “piety” by “reverence” or “feeling” is defended on the grounds that “piety” suggests theism and sanctimony. Indeed I have observed that to non-native speakers, “filial piety” can suggest theism or something analogous; but I think “filial piety” is unlikely to carry that suggestion for native speakers. To my ear, “reverence” carries a distinctly stronger taint of sanctimony than does “piety.” And surely xiao 孝 is more than a feeling.

    The replacement of “filial” by “family,” which is the main novelty in the Rosemont/Ames translation and the important point for us here, seems to be defended as follows (on the same page):

    Ideally, each generation instructs and inculcates in the succeeding generation a reverence for the family by modeling the appropriate conduct toward the generation that preceded them, thus suffusing the family with unconditional love and a sense of belonging.

    It is unclear to me exactly what the argument is—whether, for example, it assumes that xiao 孝 should be identified with its own ideal consequences. And I do not know whether the authors are supposing that we have sufficient reason to think that, say, Youzi or Confucius had these consequences in mind. I don’t know what the evidence would be for that.

    On p. 59 the authors seem to suggest that xiao 孝 is reverence for family in the sense of the linear patrilineal chain of one’s progenitors and one’s mother—a “family” that does not include oneself, nor one’s siblings, daughters, or uncles.

    The new translation is likely to mislead, because the word “family” does not normally suggest to Anglophone readers a chain or tree of patrilineal descent extending far into the past and maybe the future, in which each male unconditionally loves and belongs to his parents. Instead it suggests something very different: an interacting community of living sisters, aunts, etc., in which each party has a sense of everyone’s secure mutual commitment and love. That sense is a natural reading of the term “family feeling.” It is in significant part a sense of the attitudes of others, which I suppose xiao 孝 is not.

    The terms “family reverence” (parsed as revering the family) and “family feeling” do not suggest an attitude that is more toward elders than toward juniors or peers. Rather, these English phrases are explicitly neutral in that respect, because family is what family members have in common.

    We have little direct evidence about precisely what Youzi meant by xiao 孝, but it seems clear that the in the Analects, Confucius meant by it mainly a way of relating to one’s parents, not mainly a way of relating to family members in general, nor to any long chain. He seems to think mostly of living parents. And he seems to say at 2.7 that in his day most people understood xiao 孝 as a matter of material support, presumably of one’s parents. (Perhaps Confucius is here simply forgetting to take account of what people mean by 孝 toward deceased parents; or perhaps sacrifice was recognized as support?) Confucius’ claim about general usage or general opinion might have some evidential value as to who are the objects of xiao 孝 as Youzi meant it in 1.2.

  9. Bill Haines says:

    4. The root of what?

    Another possible reason for doubting that Youzi thought of family as the root of the Way is that Youzi does not quite say that 孝弟 is the root of the Way. That might be what he means, and it might not.

    The junzi works on the root. When the root is established, the Way grows. Filiality and elder-respect make up the root of his humanity, yes?”

    Youzi proposes at 1.2 that 孝弟 grounds ren 仁. Your practice of 孝弟 will give you a strong push in the direction of developing ren 仁, and if you’re serious about ren 仁 you’ll need to be punctilious about 孝弟. Part of why this is so is that 孝弟 has a certain similarity to ren 仁. On my reading, Youzi says at 1.13 that trustworthiness 信 similarly grounds rightness 義, and that respectfulness 恭 similarly grounds ritual propriety 禮. (And in turn, the collective practice of ritual propriety, he says at 1.12, similarly grounds social harmony 和.) Thus I read Youzi as giving roots for three of the four virtues for which Mencius gave duan 端 at 2A6. Mencius’ doctrine of the duan 端 is sometimes associated with his celebrated barley argument at 6A7 for human equality, which is foreshadowed by Youzi at 2A2.

    So it seems possible that Youzi did not mean to commit himself to the claim that 孝弟 is the root of a humanity that is the whole Way of the junzi. He may have thought instead that the Way of the junzi includes humanity, rightness, and ritual propriety, each with its own root. One of the roots is partly constituted by a certain family virtue; but family isn’t what that root is about.

    Against this idea is the fact that Youzi begins 1.2 with “其為人” and echoes this at the end with “其為仁”, as though to suggest that 仁 is the virtue for human beings.

    That’s the end of my prepared comments.

  10. Ben Hammer says:

    Dear Bill,

    Thanks for sharing your insight. I think it is interesting, but I cannot say I fully agree with it. If you will indulge me I will list a few simple reasons below.

    1. After reading your own supplements, I gathered that your main point can be summed up when you stated:

    “From what little I can see (above), the appearance is that 弟 (as a trait) was ‘elder-respect’ first, and perhaps came to mean the virtue of a younger brother after Confucius’ time or Youzi’s time. …….
    Might there have been a shift of meaning over time, toward the virtue of a younger brother?”

    From what I see you are not proposing an entirely new definition of ti. You recognize that it could mean respect from younger to elder brother, or respect for elders in general, and there has always been the consensus that it contained these two meanings. You argue that, especially in the Analects, it was first and foremost a general respect towards elders.

    I personally hold to the traditional understanding that it was first and foremost a younger brother’s respect to the elder brother. The first reason is based on the composition of the character itself, so you could call it an etymological or paleographic reason.
    The standing heart radical indicates that this word pertains to feelings or some form of activity in our heart-mind, and the left hand “di” is its phonetic component. But as is the case with many other characters, the phonetic component also holds a concurrent signification for the character’s meaning. In classical terminology this is referred to as 会意兼形声字. Some other examples are 仁,贱、浅,姓,婚,娶,功,均,返,etc. In the case of 悌, it would be glossed as 悌,从心从弟,弟亦声. The “di” phonetic here does not just determine its sound, it also modifies its meaning, circumscribing a general “feeling” of the heart into one that specifically pertains to younger brothers. Of course from here it is easy to extend this meaning to a feeling of respect to elders in general, but etymologically speaking, it is only possible that the “younger brother->elder brother” meaning came first, and the more general meaning came later.That is why it is written the way it was.

    2. Also regarding this idea of a “shift”, I do not believe that going from one definition of ti to another is such a significant paradigm shift as you propose. In Confucianism, going from respecting your elder brother to respecting your elders is not greatly conceptually different. The point of the original passage you quoted (Analects 1.2), along with so much else in the Analects and early Confucianism, is that you use the same respect for your elder brother, that you learned at home, to respectfullly treat elders outside of your home. This gradual outward radiation of ethical feelings is, as you already know, at the center of the work of peopel like Hall, Ames, and Rosemont Jr. So I guess what I am saying is that, the juxtoposition that you are proposing between the old reading and the new one, I don’t think exists. Ti can have a narrow object (elder brother) or a wider one (elders), but the core idea of respect for those older than you remains the same.

    3. Regarding your opening argument:

    “It seems unlikely that Youzi would offer as the root of the junzi’s humanity a combination of virtues that is obviously unavailable to eldest sons. We should not read Youzi such a way as to imply that eldest sons cannot have the root of humanity, if there is any other way to read him.”

    In ancient China, which was rural and agricultural, having many sons/many brothers was common. Stressing the concept of xiao was Confucius’ way of urging the entire younger generations (children) to respect and care for the entire previous generations (parents); but that was not enough to account for the respect and care that should be shown among people of close to the same generation, but where the idea of older/younger still existed (siblings). We shouldn’t look at it as Youzi trying to exclude the eldest child, but rather as trying to include all of the others.
    The traditional reading does not deny the oldest son the chance to be ethical or virtuous. He still has every opportunity to cultivate his xiao, at home and outside, and to practice being a good older brother to all his siblings. And all these ideas of brotherly love and respect which he experiences at home, he can apply those to his social relations out of the home.

    4. Finally, on a textual note, you asked if there were any appearances of xiao ti that antedated the Analects. The Book of Rites was created piecemeal, like the Analects, with no single or certain date of creation, like the Analects. But it was around the same time. In it we see: 《仪礼•士相见礼》“言孝弟于父兄。”唐代的贾公彦正义曰:“孝弟,事父兄之名,是人行之本。”In this early example of xiaodi/xiaoti, it is used in exclusive reference to sons treating fathers and younger brothers treating elder brothers, and the Tang Dynasty exegete concurrs.

    Those are the main reasons why I beleive that the traditional reading of ti, first as younger brother respecting elder brother, is not problematic. All the other meanings and aspects of it you bring up are real and valid. As I see it, they are already existing extensions of the old reading.

    I did not have time to address all the good points you brought up, but this should suffice to start a conversation.

    Ben Hammer
    Shandong University, Advanced Institute of Confucian Studies
    Jinan, China

  11. Bill Haines says:

    Thank you, Ben! I’m grateful for your marvelous response. It is really helping me think. I feel a little silly discussing these things in detail from a position of ignorance, but it’s all I can do. I think you hae understood me exactly, except that maybe I’m not thinking of the possible semantic shift of this word as something as radical as a paradigm shift.

    [Correction: I may have read Ben carelessly. My thought is not that 弟 always had both meanings and the elder-respect meaning is at the fore in Confucius and Youzi. Rather my proposal is that the subfraternity meaning was absent or far less common in the time of Confucius and Youzi; that Confucius and Youzi meant the term unambiguously as “elder-respect,” understood either as applying primarily outside the family, or as not distinguishing. -12/22/16]

    To your 1, 2, and 4 (on the word):

    In the Yili passage you cite, 弟 is certainly used in contemplation of family, so this is an item in the scales against my proposal—though because of the prepositional phrase [于父兄], the meaning of the statement would be the same if 弟 simply meant respect, or simply meant elder-respect (compare the Mozi’s “為人兄必友”, quoted above).

    From the very little I can find on line, I gather the dating of the Yili is very much up in the air. Yes?


    I am under the impression that the 悌 form may have come in after Youzi’s time, perhaps not too long after; so I think what’s relevant is the conditions under which the very same graph could be used. The etymological argument against my proposal is that it’s the same graph, not just a similar graph. But did it never happen that one graph was used for two distinct words just because of the phonetic similarity, and not because of any conceptual relatedness?

    My proposal about 弟 doesn’t require that to have happened. So long as the sound was the same and the ideas had some kind of arguable kinship – as, for example, general humility and younger brother do – might the same graph have been used for two different (spoken) words on that ground alone?

    Also I understand that there is every degree of possible gradation between a word’s having just one meaning and two quite distinct meanings. One has to look and see. And any meaning is a matter of core and fringe. My proposal that Youzi mainly had in mind elder-respect doesn’t require that the word had two quite distinct usages. But if we find that in uses at and before Youzis’ time the main concern on each occasion seems to be broad elder-respect, that amounts to support for my proposal about what Youzi had in mind; especially if that broad elder-respect is commonly taken as a companion virtue of xiao 孝.

    I do think any term meaning youngerbrotherliness (subfraternity?) could become extended to encompass elder-respect in general. And I think any term meaning elder-respect in general could be extended, or indeed directly applied, to the case of younger brothers vis-a-vis older. Just a priori, the latter move seems the simpler easier move.

    Especially toward evaluating the etymological or paleographic argument, it seems relevant to ask whether it was generally the case that graphs for family positions such as 弟 and 兄 were also used for adjectival verbs. It seems plausible a priori that a term for a relative position would be used as a term for the virtue of the proper conduct of that position (as a core meaning that could later be extended). To put the same point another way, it seems plausible a priori that as a general pattern, terms for relative positions would also be used as a terms for the virtue of the proper conduct of those positions. But my impression is that in fact there was no such general pattern, except within the occasional special “父父…” list. So the idea that using 弟 for the virtue of youngerbrotherliness is a linguistically natural thing, to be expected on general principle, may be mistaken.

    It seems possible that (before or after the introduction of writing) a word for younger brother could, by direct metaphorical extension, come to generate an adjectival verb for the trait of elder-respect, without there being much use of that word in the narrow sense or application “goodyoungerbrotherliness.” Perhaps it is more plausible to think that the word 孫
    (遜) for humility originated in this way, than to think that grew as an extension of a common word whose main meaning was the proper virtue of a grandson?

    (I find myself feeling that in “父父…” lists—Analects 12.11 and the Tuanzhuan passage I quoted above—the relation between each noun and the verbalized noun that accompanies it is somehow different from the relation between the noun 弟 “younger brother” and the adjectival verb 弟 in e.g. the Yili passage as you read it. I can’t quite put my finger on the somehow. Maybe it’s that in the lists, the verbs feel to me like transitive verbs of action, while 弟 (“is youngerbrotherly”) doesn’t. But why do I feel this way? Maybe it’s just because of English-language patterns, and therefore irrelevant. A hunter hunts, a father fathers—these English verbs are action verbs, indeed transitive ones, even when we read “fathers” not as mere siring. But I think 孝 as a verb is stative (is an adjective, as Pulleyblank would say—p. 23) and grammatically intransitive. In the Mozi’s phrase “不弟弟” (quoted above) and Youzi’s “其為人也孝弟”, 弟 looks adjectival. I don’t know. Anyway I’m not suggesting here that elder-respect is more adjectival or intransitive than youngerbrotherliness. Rather I’m maybe feeling that the verbs in the “父父…” lists are not instances of the kind of relation between noun and verb that would directly fit the (eventually familiar) use of 弟 to mean youngerbrotherliness.)

    To your 3:

    On the question which or what meaning is first and foremost, your point is well taken that there are more kinds of evidence than just records of early uses of 弟. For example, there is the importance of family and the number of brothers in agricultural society, and there are the general emphases of early Confucian or Ruist thought.

    On the number of brothers: I’ll take your word for numbers of sons among farmers in Youzi’s time. I agree that having large numbers of sons is common among farmers in general, at least where land is plentiful or survival uncertain. I don’t know anything about the ownership or inheritance of farmland in Youzi’s time. I would like to hear about that.

    I am under the impression that the 人 from whom the Ru were drawn, the kind of people the Ru would have seen as potential junzi, were in general not farmers who could benefit from the extra hands. I think it is a general fact that people from better-off socioeconomic classes tend to have fewer children. Perhaps declining elites have even fewer?

    So far as we know, Confucius had one brother and one son. Do we know of the existence of any brothers of any of his disciples, or any other early Ru?

    I imagine lords in power would have had many sons, partly in order to make sure there would be at least one viable successor. But in a lordly family, the specific position of eldest (living) son would have been highly distinctive. It would have been on everybody’s mind. Maybe the position of heir to the lordship is the root-image of the term junzi; I don’t know.

    It seems to me that the importance of respect for elder brothers, or of any punctilious concern about age-order, would make the position of eldest son distinctive and important. The leader tends to be a model, and the model should be capable of the root.


    The Mencius and other texts do stress the idea of the outward extension of ethical feelings and patterns from family to others. I don’t quite see the idea in Confucius’ remarks in the Analects, as I hope to explain soon in other posts.

    I think I see something similar to that extension idea in Youzi. But I think that for Youzi the operative general idea is not the extension of feelings from family to more people. It seems to me that “fewer-to-more” or “family-to-more” is not the operative idea in 1.12 and 1.13. Instead the operative idea seems to be something like the way concrete patterns of face-to-face relations serve to shape the abstract form of our relations with society at large (which may be the same people, as when ritual models social harmony at 1.12).

    For this Youzian dynamic, I think it doesn’t matter too much whether the 弟 part of the root is respect for elder brothers, or respect for any and all elders one might meet. But it seems to me that his choice of a root is much more plausible, and fits usage better, if he was thinking mainly of respectful relating to elders one encounters, in general.

    Best regards,
    Bill Haines
    no academic affiliation

  12. Hi Bill,

    From Axel Schuessler’s ABC Etymological Dictionary of Old Chinese (2007, p. 210)

    d1 弟 – ‘Younger brother’ Shijing]; ‘younger secondary wife’ 娣 [Shijing].
    — endoactive of di 第 ‘order, sequel’, literally ‘the person who is following in sequence’.
    — Etymology: Sino-Tibetan *dwi: Proto Tibeto-Burman *doy ‘younger brother’ > Written Burman thweB ‘be youngest’, Japanese sedoi ‘last born child’; also found in Mon de? ‘younger sibling’ (<- Tibeto-Burman?).

    di 弟
    'Respectful toward elder brothers, fraternal' 弟悌 (Mand. ti) [Mengzi].
    [<] either exoactive or late Old Chinese general tone of di 弟.

    di 第
    'Order, sequel' [Zuozhuan]. Karlgren (1956: 14) connects this word with ti 梯 'ladder'.


    Regarding “It seems unlikely that Youzi would offer as the root of the junzi’s humanity a combination of virtues that is obviously unavailable to eldest sons."
    — I would suggest that ti 悌 could be interpreted as referring to acting as a younger brother should act toward his elder brother. An elder brother will likely know what is this is like, and could express it toward his elders in general.

    • Bill Haines says:

      Just to clarify the state of the discussion as I understand it:

      Ben writes , “it could mean respect from younger to elder brother, or respect for elders in general, and there has always been the consensus that it contained these two meanings.”

      (I agree that it had both meanings before the Han; perhaps that is the consensus Ben means. I don’t know if anyone has ever thought that it always had both meanings–that it acquired them simultaneously. As for which came first, a priori either order seems easily possible.)

      Scott quotes Schluesser’s ABC dictionary, which dissents even from that modest consensus by not offering “respect for elders” as a possible meaning. Though Schluesser takes a controversial position, he does not offer evidence.

      I seem to find a number of cases of the use of the term as “elder-respect,” probably antedating—at least possibly antedating–any uses as “subfraternity” that I can find. I do not think it would be reasonable to take Schluesser as an authority outweighing that apparent evidence.

      For my part, alas, I’m unlikely to have any further evidence to offer on this matter. I can’t read Chinese well enough to make much use of tools that don’t use English to help me. I would be out of my depth with oracle bones, bronzes or bamboos even if I knew where to find them on line. Also I am a complete ignoramus about the evidence and issues regarding the dating of old texts, and about what material has been dug up other than what I can find at the Chinese Text Project.

  13. Bill Haines says:

    Hi Ben, I owe you more reply; I’m working on it.

    Hi Scott,

    Thank you!

    When Schuessler says 弟 (younger brother) is “ endoactive of 第,” I think he means it derived from a usage that would then be written 弟 and would now be written 第. Yes?

    Schuessler doesn’t mention a meaning “elder-respect” – are you offering him as an authority for the idea that the word never really took that sense (so that when it seems to, it’s always a bit of ad hoc figurative language)? Schuessler doesn’t suggest a text for the virtue-of-subfraternity earlier than the Mencius, but we have texts where the word seems to mean “elder-respect” earlier than that.

    I think probably your final proposal is specifically about 1.2, so I think what you mean by “acting as a younger brother should act toward his elder brother” is “acting [toward one’s elder brother] as a younger brother should act toward his elder brother.” So I should add your name to the long list of those who favor that reading at 1.2. Are you saying Schuessler too is such an authority because he doesn’t offer elder-respect as even a possible meaning for that word?

    At first I thought you were proposing that the word 弟/悌 should interpreted pretty much across the board (not just in 1.2) as referring to “acting as a younger brother should act toward his elder brother.” (So I didn’t assume this formula meant “acting [toward one’s elder brother] as a younger brother should act toward his elder brother.”) I thought a lot about that as a general proposal. Here are those thoughts—from here to the end of this Comment:

    I think there are different ways to understand “acting as a younger brother should act toward his elder brother,” based on different answers to the question “acting that way toward whom?”

    A. acting toward one’s elder brother as a younger brother should act toward his elder brother.
    B. acting toward anyone & everyone as a younger brother should act toward his elder brother.
    C. It kind of has to take an object: being 弟/悌 to X is
    acting toward X as a younger brother should act toward his elder brother.

    If Youzi meant A, there’s the problem about eldest sons.
    If Youzi meant C, he did not write well; besides, we commonly see the term without an object.

    B involves acting toward everyone, older and younger, in that way: as though one had begun with A as one’s root, and then extended it to everyone, or to everyone one meets. (Some people might say that if this is what Youzi meant by 弟, he was confusing root and branch: to whom would one extend one’s B?) We could call it a kind of general humility, sort of like 恭 (and 孫/遜?). That reading would be friendly to the idea that Youzi’s root was not specifically about family (though there may be a family metaphor behind the word he used). At least some of the texts I cited for the “elder-respect” meaning also fit B. Are there any early texts in which B is the plain meaning?

    If we narrow B down a little, to D

    D. acting toward one’s elders as a younger brother should act toward his elder brother.

    then we have elder-respect, described by the definer using a simile of subfraternity, perhaps to say that this comparison originally motivated the usage. Going around being like a “little brother” to his elders. Which is not the same as saying that the word was commonly (or ever) used for acting toward one’s big brother the way a little brother should.

    I think B is an interesting idea. I hadn’t thought of it. I was mainly just trusting the standard view that 弟/第 [CORRECTION: 弟/悌] can mean either A or D (a view that does not specify how distinct are the two usages).

    If we think B or D arose as a kind of metaphor from something specific to junior brothers, we could think it arose as a kind of extension of a familiar word for the virtue of subfraternity: treating one’s elder brothers as one should. Alternatively, we could think it arose directly from the noun “junior brother”. B or D is going around acting like a little brother, acting like people’s little brother. There need not already have been a familiar word for the virtue of treating one’s elder brother as one should.

    It also seems offhand that either usage, B or D, could have arisen directly from the word 弟/第 (as Schuessler says the meaning “junior brother” did), without going by way of the meaning “junior brother” at all. Partisans of this hypothesis can be happy that we don’t see a whole bunch of names of virtues based on family positions: being 父, being 兄, etc. This sort of move looks like it could lead to usage D only if there’s a general recognition of seniority as a standard determinant of rank, at least in unofficial settings. But there would be no such precondition for this sort of move leading to usage B.

    (My efforts to draw distinctions don’t reflect a view that the meanings of ordinary terms draw sharp lines.)

    Here’s another distinction. I think 恭 is not so much something you do toward someone, not a way of treating someone; it’s more about general demeanor. More or less. Maybe your original proposal about 弟/悌 was along those lines, and in talking about choosing an object I’ve been missing the point. Maybe the thought was that there’s a certain way we expect junior brothers to act (at least when they go in 入), something like 恭; and to be 弟/悌 is to act that way in general (in or out)?

    • Hi Bill, I thought my last sentence indicated I meant D. That’s what I meant.

    • Bill Haines says:

      Oh, sorry! So you’re saying you read 弟 at 1.2 as elder-respect?

  14. Bill Haines says:

    Hi folks,

    How can a glaring problem be overlooked?

    So, I’m saying that if the language and context were such that IF 弟 in 1.2 would originally have been understood as being specifically about subfraternity (a younger brother’s proper relating to an elder brother), then there would have been a glaring problem. I think that in composing 1.2 and deciding whether to mention 弟, Youzi would certainly have noticed that problem, if that’s what the word in that context would tend to mean. And I think the early audience would have noticed the problem. Some would think, “I have nobody to be subfraternal to—what about me? And hey, that reminds me: what about legitimate lords in general?” 1.2 would have amounted to a moral criticism of current principles of legitimacy. The composer would have anticipated this problem.

    Yet WHEN we have understood 弟 in 1.2 as being about brothers, or when it has been translated for us that way, the problem hasn’t glared at us, Anglophone scholars. Why not? I think there are many factors that would reduce the glare for us, that would not have operated for Youzi in composing the statement, or for the audience he would have envisioned or encountered.


    For one thing, age-distinctions among siblings are not on our moral or political radar; we are not primed to notice implications in that area. The position of “eldest son” is not among the concepts necessary for understanding the tragedies of our recent history, the ideals toward which we work, and the virtues relevant to governance. In those respects we differ from Youzi and Confucius’ circle.


    Translations commonly render 弟 as “fraternity” or something age-neutral like that. Such a neutral translation removes the appearance that there might be a special problem about eldest sons; if anything there would be a problem about only sons (or, if we erase the sexism, which is often a reasonable approach, only children). And that’s a much smaller problem, for three reasons: (1) Only sons are a much narrower class than eldest sons. Anybody who has any kind of brother, or sibling, can be fraternal. (2) “Only son” or “only child” is not a key special position socially and politically. (3) If Youzi were implying that to be a good person, it is important to have siblings, that wouldn’t be a howler.


    When people are accustomed to a text and regard it with a certain reverence, they tend to overlook problems. How many generations have read Genesis over and over without noticing the contradiction between the first two chapters? Or in the Sermon on the Mount: compare Matthew 5:14-16 with Matthew 6:1-4, or compare Matthew 7:1-6 with Matthew 7:7. I’m not saying these are irreconcilable; I’m saying people simply don’t see the prima facie problems, which should be glaring.


    In connection with 1.2, we tend to overlook 弟 entirely, thinking of the passage as proposing that the root is filial piety. Scholars even do this in print, in reporting the content of 1.2. That makes us unlikely to notice problems about 弟.

    Why do we let ourselves be distracted by filial piety?

    A main reason may be that we work gradually toward an understanding of any given passage in the Analects, by reading the whole collection over and over. Confucius says a fair amount about filial piety in the Analects, indeed not too far from 1.2 in the book, while Confucius doesn’t mention subfraternity. (Of the four statements where Confucius can be taken to be speaking of norms for brothers, two draw no age distinctions (2.21, 13.28), and the other two (11.22 and 9.16) can easily be read as de-emphasizing the whole matter.)

    We don’t read 1.2 to understand Youzi. We read it to understand Confucius (and early Confucianism). Indeed we tend to rely quite heavily on 1.2 to understand Confucius, partly because Confucius never says why filial piety might be distinctively important. Also Mencius specially stresses filial piety in connection with ren.

    Another reason may be that readers find it a little hard to think of relations with brothers (the usual way 弟is represented in translation) as remotely comparable in moral or psychological importance to filial piety. We do not think of siblings as having significant authority over other siblings. If we think of 1.2 as referring to roots in childhood, then we will think the difference in significance between a parent’s authority and a sibling’s authority is quite radical. If we think of 1.2 as referring to roots in adulthood, we today may find it especially hard to think of siblings as very important.

    The English words “fraternal” and “fraternity” are far more commonly used not in the family sense but in a wider sense, where they are about egalitarian community and fellow-feeling. There they carry a positive aura—and even this is something we mostly just allow to make us feel good without really thinking about it. Fraternity is the part of “Equality, Liberty, and Fraternity” that for most Anglophones just adds metric balance and historical sheen. People used to give lip service to “brotherhood.” We may tend to avert our mind’s eye from words like “fraternal” and “brotherhood” because we try to look past sexism.

    Translating 孝 as “family reverence” or “family feeling” can make it easier to lose sight of 弟 in the root. For example, Rosemont and Ames write of

    xiao, conventionally rendered “filial piety” in English (but we render it as “family reverence,” or “family feeling” to avoid the close Christian associations with the word “piety” today). …

    … In the Analects 1998 1.2, Youzi claims that becoming ren begins at home. That is, the ground of our consummate humanity is family feeling:

    Exemplary persons (junzi 君子) concentrate their efforts on the root, for …

    (“Family Reverence (xiao 孝) as the Source of Consummatory Conduct (ren 仁)” (Dao 7.1, 2008), pp. 9, 16.)

    Youzi could have given the root simply as 孝。That’s simpler, more elegant, more memorable, and was easier to think of in the first place. But he decided to include 弟. He must have thought it was important.

    Does subfraternity really matter?

    Ben, I agree that on the subfraternity reading, 1.2 does not imply that eldest sons can’t have any virtue. (And I agree that as a matter of fact, eldest sons presumably can have plenty of virtue; that’s my premise.) But on the simple subfraternity reading, 1.2 does distinctly imply that eldest sons cannot have the root of the junzi’s humanity.

    Youzi is arguing that 孝弟 is important. He’s saying, “If you want to be a humane junzi you have to do this.” The more important he’s saying that root is, the bigger the problem about people who can’t quite have it.

    One might reply: But isn’t it enough that we be filial and subfraternal insofar as there is occasion? Wouldn’t that be enough to serve as the root of ren? So if there is never occasion, I automatically have the root?

    That depends on how the rooting is supposed to work. The rest of Youzi’s surviving work suggests to me that one would actually have to do things for the root to work. Maybe there’s no disagreement on this.

    One might reply, clarifying: But if I’m an eldest son, and so have no occasion to be subfraternal, isn’t it enough that I be filial? I mean, how important can subfraternity be anyway?

    In truth maybe filial piety is more important than subfraternity. The greater that difference is, the odder it would be that Youzi thought writing subfraternity into the root was worth the loss of elegance. I think elder-respect has a far better claim to parity of importance with filial piety. It involves way more people. It is part of public life, shading more directly into official life. There is reason to think Confucius would agree that elder-respect has a better claim than subfraternity to be comparably important with filial piety. Confucius certainly mentions elder-respect, and does not certainly mention or allude to subfraternity. Confucius says elder-respect is important. That is, he lists filial piety and elder-respect side-by-side at 1.6 and 13.20, as if to say that they are comparably important; and he seems to elaborate on the developmental importance of elder-respect at 14.44.

    Ben, you write, “We shouldn’t look at it as Youzi trying to exclude the eldest child, but rather as trying to include all of the others.”

    I think including 弟 in the root does not move Youzi in the direction of including all of the others. With 孝 alone he would already be there, as you said earlier. Adding 弟 might be thought to make the inclusiveness more salient: Half the root is special to younger brothers. But that’s just salient exclusion.

    Scott, is it your proposal that 1.2’s account of the root crucially involves playing on the ambiguity of one of its two terms? Are you proposing that according to 1.2, for eldest sons (only sons and eldest brothers) the 弟 part of the root is elder-respect rather than subfraternity; while for younger brothers it is subfraternity and not elder-respect? I think this reading could be improved a bit by revising it to say that for younger brothers the 弟 part is subfraternity and elder-respect. And I think this new reading could be improved by making it simpler: for everybody, the弟 part is elder-respect.

  15. Bill Haines says:

    When I said above that the concept of “eldest son” was important for historical tragedies, I had in mind struggles over the succession, e.g. Duke Huan’s killing of his elder brother for the succession (mentioned at 14.16 and 14.17). I’m just assuming that that sort of struggle was a continuing problem; that whenever a lord had several sons, the issue was on people’s minds.

    One might object that this point implies that the virtue of subfraternity is very important, whereas I held above that it is not so important as compared to elder-respect.

    My main reply is that there’s a difference between saying (as I do) that subfraternity is an important virtue for key people because of its direct political consequences, and saying (as I do not so much) that subfraternity is an important virtue for all 人 because it grounds other virtue (psychologically or however).

    Incidentally, regarding this duke who killed his elder brother, Confucius says at 14.15 that he was “straight” 正. Presumably he isn’t endorsing the fratricide; and granted, he is drawing a particular contrast. Still, the remark does not sit well with the idea that Confucius thought subfraternity is comparably important with filial piety as a ground of general virtue.

  16. Bill Haines says:

    In a footnote to a discussion of the Analects in Ethics in the Confucian Tradition (2002), p. 168, P. J. Ivanhoe writes,

    Dora Dien suggests that the virtues of filial piety and respect for one’s elder brother may have enjoyed a special place in Confucianism because of the structure of the family and the broadly observed rules governing inheritance. Because the Chinese typically divided all family property equally among the sons, in order to avoid a tendency toward diminishing wealth and power, extended families needed to maintain very high levels of cohesiveness and coordinated activity. …

    I can’t tell from what I can access, but it appears to me that Dien is thinking of a later period.

    Based on some floundering about in cyberspace, I gather that in the Western Zhou and Spring and Autumn periods, lordship passed by primogeniture in the central states, and large landholdings were held by lineages corporately, whose heads were normally defined by primogeniture. There is some indication in cyberspace that the principle of equal inheritance of land by all brothers was introduced by Qin in the 300s BC; but really I just have to hope that an expert will speak up here.


Leave a Reply

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