Emotions in early confucianism 1

1

Chapter 1
Introduction
On the Subject Matter and Methodology Adopted

In recent years, the recovery of valuable manuscripts, largely from southwestern
China, has changed our perspective of ancient China considerably, making it necessary to
reassess what we know about the early Chinese intellectual landscape.
1
The Guōdiàn 郭
店 bamboo texts excavated in 1993 and published in 1998, and the Shànghǎi museum
corpus (commonly abbreviated as the Shàngbó 上 博 texts) retrieved in 1994 and
published from 2001 to 2008 afford fresh insight into the formation of early
Confucianism.
2
Specialists have carbon-dated these newly recovered texts, which are


1
The renowned sinologist, Lǐ Xué Qín , for example, raised the possibility of a new Chinese
intellectual history based on the vast amount of archaeological findings unearthed in recent decades. With
these finds, he envisages the end of “doubting antiquity (yígǔ ),” which was the dominant
hermeneutical position of 20
th
-century sinology, and the dawn of “interpreting antiquity (shìgǔ )” in
the study of ancient Chinese history. See Lǐ Xué Qín, Chóng Xiě Xué Shù Shǐ (Shíjiāzhuāng:
Héběi Jiàoyù chūbǎnshè, 2002), and his Zǒu Chū Yí Gǔ Shí Dài (Shěnyáng: Liáoníng
Jiàoyù chūbǎnshè, 1997).
2
Jīngmén Shì Bówùguǎn , Guōdiàn Chǔmù Zhújiǎn (Běijīng: Wénwù

chūbǎnshè, 1998); Mǎ Chéng Yuán ed., Shànghǎi Bówùguǎn Cáng Zhànguó Chǔ Zhú Shū
vol. 1-7 (Shànghǎi: Shànghǎi Gǔjí chūbǎnshè, 2001-2008).
2

largely Confucian in nature,
3
to the late Warring States period (476 - 221 B.C.E.), and not

3
The Guōdiàn and the Shànghǎi Museum corpus also contain texts that are not Confucian in nature. For
instance, the Guōdiàn corpus contains a shorter version of the Daoist text Lǎozǐ , an exceptionally
interesting cosmological account, the “Tàiyī Shēng Shuǐ (The Tàiyī Begets Water),” as well as a
body of sayings now given the title “Yǔcóng Sì (Collected Sayings IV),” which can be considered
eclectic (zá ) in nature. These formed the focus of my Masters thesis. As for the Shànghǎi Museum
corpus, some of it are better classified as eclectic and military texts (bīngshū ); there is also a
complicated Daoist cosmological account, the “Héngxiān (Eternal Origins),” as well an alleged
Mohist text, the “Guǐshén Zhī Míng (The Revelations of the Spirits).” For the purposes of this
dissertation, I focus on the Confucian texts. Besides the Guōdiàn and Shànghǎi Museum collections, other
important Confucian archaeological finds in the past few decades include the texts discovered at Bājiǎoláng
, Fùyáng and Yínquèshān , but they are dated to the early Hàn (206 B.C.E. – 220
C.E.) period, and are thus not included in the pool of primary sources examined here. For these Hàn
Confucian texts, see Dìngxiàn Hànmù Zhújiǎn Zhěnglǐ Xiǎozǔ

, “Dìngxiàn 40
Hào Hànmù Chūtǔ Zhújiǎn Jiǎnjiè 40 ,” in Wénwù 1981 vol. 8, 11-13;
Fùyáng Hànjiǎn Zhěnglǐzǔ , “Fùyáng Hànjiǎn Jiǎnjiè ,” in Wénwù 1983 vol.
2, 21-23; and Yínquèshān Hànmù Zhújiǎn Zhěnglǐzǔ , Yínquèshān Hànmù Zhújiǎn
vol. 1 (Běijīng: Wénwù chūbǎnshè, 1985), respectively. Our understanding of early
philosophical Daoism has been greatly enhanced by the discovery of Mǎwángduī silk manuscripts,
which contain not only a version of the Lǎozǐ, but also a number of texts collectively titled Huángdì Sìjīng

, the latter of which have given us a new understanding of “Huánglǎo Daoism.” See
Mǎwángduī Hànmù Bóshū Zhěnglǐ Xiǎozǔ , Mǎwángduī Hànmù Bóshū
vol. 1 (Běijīng: Wénwù chūbǎnshè, 1980).

3

later than 278 B.C.E.,
4
which suggests they probably predated the Mencius or Mèngzǐ 孟
子. Hence, these new findings provide great potential in bridging the gap between the
thought of Confucius or Kǒngzǐ 孔子 (551 - 479 B.C.E.) and that of Mencius or Mèngzǐ
孟子 (372 - 289 B.C.E.).
This dissertation seeks to understand the nature of early Confucianism through a
study of the early Confucian interpretations of emotions. It draws its sources from the
Analects or Lúnyǔ 論語, the Mencius or Mèngzǐ 孟子, and the recent archaeological
manuscripts, with cross-references to the pre-Confucian classics so as to situate these
early Confucian texts in their proper context. It is important to note that this dissertation
is focused on two aspects: what the early Confucian understanding of emotions reveals
about early Confucianism and the ethical implications involved.







4
There is no consensus on the precise date of these findings. Historically, the Guōdiàn corpus was likely
buried before 278 B.C.E. when the Qín general Bái Qǐ

(? - 257 B.C.E.) invaded the Chǔ capital of
Yǐngdū (where the tomb in which the Guōdiàn bamboo slips were found was situated) and changed
the nature of Chǔ burials. This dating does not necessarily apply to the Shànghǎi museum bamboo slips as
they were not found in Chǔ tombs but were retrieved from the antiques market in Hong Kong. However,
archaeologists have carbon-dated the latter as 2257 ± 65 years ago, which places them at 313 - 183 B.C.E
4

This chapter looks at early Confucianism in general; I also discuss the main
methodological principles. Chapters 2 to 5 form the core chapters of this dissertation,
each investigating a group of emotions. Because of the strong connection between
philology and philosophy in Chinese thought, each of the subsequent chapters addresses
these two distinct yet closely connected aspects separately, with greater emphasis on the
philosophical understanding of the emotions in early Confucianism. Chapter 6
summarizes and concludes the dissertation.
1.1. The subject matter
1.1.1. The early Confucians
My area of interest is “early Confucianism,” which should be understood as the
thought of Confucius, as well as the interpretation of his thought, mainly in the Warring
States period. As is well known, the most reliable source on the teaching of Confucius,
the Analects, was unfortunately, not written by the master himself. Instead, it comprises
Confucius’ sayings and doings, recorded by his disciples and, in the last five chapters, the
sayings and doings of Confucius’s disciples recorded by the next generation of disciples;
5


5
Most scholars agree that the last five books of the Analects were written much later, mainly because they
are accounts of what Confucius’ disciples said and did, rather than the Master himself. See, for example, D.
C. Lau trans., Confucius the Analects (Hong Kong: Chinese University Press, 2002), 265-270. This was
also the view of Brooks and Brooks, who attempted to reorganize the Analects in a chronological manner.

See E. Bruce Brooks and A. Taeko Brooks, The Original Analects (New York: Columbia University Press,
1998). All English translations of the Analects and the Mencius in this dissertation are taken from D.C.
Lau’s, namely Confucius the Analects and Mencius (Hong Kong: Chinese University Press, 2002)
respectively, unless otherwise stated. All names, except the more commonly used Latinised “Confucius”
and “Mencius,” are in pīnyīn.
5

likewise, the Mencius is a record by Mencius’ disciples of what he said and did.
6
In other
words, although Confucius and Mencius may be the source, the evidence we have for
their teachings is derived from a group of interpreters who were instrumental in the
passing down of a “Confucian tradition.”
7
Who, then, are these interpreters, namely, the
“early Confucians”? What is distinctive and central in the collective memory of this
group of people?

6
As D.C. Lau has pointed out, although there are scholars who maintain that the Mencius was actually
written by the Master himself, there are reasons to believe the otherwise. Lau concludes, based on the
posthumous titles of feudal lords mentioned, the use of the appellation “tzu” [z
ǐ
] and the nature of the
variant passages in the book, that Mencius was compiled by more than one of Mencius’ disciples. However,
he also added that this “does not detract Mencius from its authoritativeness” as there “is rarely any
divergence in the text where the same passage is found more than once. The words of the Master were, as is
to be expected, sacrosanct in the eyes of the disciples, and the greatest care must have been taken to
preserve them as they were spoken by him.” See D C Lau trans., Mencius, 347-349.
7

Confucius was one of the most important thinkers in Chinese Philosophy, especially if we consider
Chinese thought to be predominantly “Confucian” in nature. However, “Confucianism” is not, in a strict
sense, synonymous with “Confucius’ thought.” Rather, the term Confucianism encompasses different
interpretations of the Master’s thought by subsequent thinkers over a two thousand-year span. Departing
from Confucius’ original writings and thinking when he was alive, Confucianism has developed into
different and diverse traditions both in content and orientation. For example, Confucianism in the Hàn
(202 B.C.E. - 220 C.E.) took on elements of correlative yīn-yáng philosophy; neo-Confucianism in
the Sòng (960 - 1279) and Míng (1368 - 1644) took on elements from religious Daoism and
Buddhism; and contemporary “new-Confucianism” looks at how Confucian cultivation fits within the
present world context.
6

Strictly speaking, if we are to adhere closely to the Chinese language, this group
of intellectuals should be called the “rú-ists,” taken from the term “rú 儒” and the school
of thought it represents, namely, “Rú-ism” (Rújiā 儒家), for the term “Confucians” seems
to suggest that this group of intellectual only exist after Confucius. In actual fact,
Confucius himself belonged to this social group by the name of Rú, which existed before
him. However, there is an important reason why we can yet call the particular group of
intellectuals “Confucians,” for by the time of Confucius, there is no evidence suggesting
that they are still performing the traditional duties which formerly makes them distinctly
rú.
8
Instead, what characterizes them after Confucius is, rather, the fact that they were
followers of Confucius’ teachings, and thus, could be legitimately regarded as
“Confucians.” In other words, the development of “Rú-ism” seems to take on a major

8
The origin of rú is still unknown and heavily speculative. In 1954, Wing-Tsit Chan identified its origins
as one of the eight basic unsolved problems in Chinese Philosophy. The traditional interpretation was that
of Hú Shì (1891 - 1962), who explained rú as “weakness” and the rú-ists as “people in ancient China

who wore antiquated-style garments and whose expression indicated meekness.” In this regard, they were
seen as remnants of the overthrown Shāng (17
th
- 11
th
Century B.C.E.) empire and their role was to
perform the Shāng traditional religious rites. Fung Yu-lan, on the other hand, saw these people instead as
Zhōu (11
th
Century - 256 B.C.E.) officials who lost their official posts upon the disintegration of the
Zhōu feudal system. See Wing-Tsit Chan, “Basic Problems in the study of Chinese Philosophy,” in
Philosophy East and West vol. 4 no. 1, (April 1954), 157-166. In this article, a number of other
interpretations like those of Guō Mò Ruò (1892 - 1978) (discussed along Marxist line of class
struggle) and Herrlee G. Creel (1905 - 1994) (rú being a derived sense from the non-combatant nature of
these people) were also identified; but I agree with Chan’s assessment that they are too narrow, and Fung’s
interpretation can complement Hú’s in our understanding of what really constitutes rú.
7

change with the flourishing of Confucius’ thought, the impact of which is such that it
renders a nominal change from “Rú-ism” to “Confucianism” reasonable.
9

Regardless of Confucius’ hereditary roots, his ideal political system was largely
based on the Zhōu feudal system, which by his time, was already disintegrating and in
decline. In line with the common belief then, the Zhōu was the legitimate authority that
received the Heaven’s mandate (Tiānmìng 天命 )
10
to rule when they succeeded in
overthrowing the last king of Shāng. In a way, the Zhōu feudal system was similar to the
Western medieval systems, where the ruler was hierarchically above a bevy of landlords

who had hereditary rights over fiefs and vassals.
11
With this well-knit and interlocking

9
This transition is of paramount importance. Paraphrasing Hú Shì’s position, Wing-Tsit Chan has pointed
out that Confucius, being a Shāng descendant, “represented a new type of ju [rú], firm, active, and
progressive [instead of the stereotyped ‘weak,’]. Thus, Confucius… revived an old tradition, injected new
blood into it, and raised it to new heights.” See Wing-Tsit Chan, “Basic Problems in the study of Chinese
Philosophy,” 161.
10
Tiānmìng is the Zhōu house’s political banner when they sought legitimacy for their taking over of
political leadership from the Shāng, with the Zhōu ruler as “Son of Heaven (Tiānzǐ ).” For a detailed
account of the astrological background of Zhōu’s notion of Heaven’s Mandate, see David W. Pankenier,
“The Cosmo-political Background of Heaven’s Mandate,” in Early China vol. 20, (1995), 121-176.
11
The hierarchy, from top down, comprised gōng , hóu , bó , zǐ and nán , which A.C. Graham
thinks is translatable to Duke, Marquis, Earl, Viscount and Baron respectively. See A. C. Graham,
Disputers of the TAO: Philosophical Argument in Ancient China (Illinois: Open Court, 1989), 2.
8

system,
12
the Zhōu ruling house was able to hold on to power for around four centuries,
before a succession of weak rulers led to its decline and eventual sacking of its capital by
the Quǎnróng 犬戎 in 771 B.C.E
With this defeat and the shifting of the Zhōu capital to Chéngzhōu 成周 (in
present-day Luòyáng 洛 陽 ), the Zhōu ruling house never regained its authority;
eventually, its power was reduced to nominal; dukedoms started taking the name of
overlordship (bà 霸) resulting in conflicts and wars.

It was during these chaotic times that Confucius entered the historical scene.
In a way, Confucius represented the conservative voice of his time: his path to the
attainment of his ideal political system was essentially a return to the golden age of Zhōu,
where everyone used to abide by a proper code of conduct.
13
This set of ethical rules
which provides guidance on the propriety in social behavior was set out in the rites (lǐ 禮)

12
Functioning under a feudal system where the empire was ruled by the Tiānzǐ and governed by a balance
of kinship units comprising surnames that were the same as (tóngxìng ) and different from (yìxìng
) the ruling house, the Zhōu empire demonstrates a well-organised political system in which we see a
systematic fiefdom and titles succession system; an established tax and tribute system; as well as a check-
and-balance system between the ruler and the landlords. For a good description of the Zhōu political
system, see Cho-yun Hsu and Katheryn M. Linduff, Western Chou Civilisation (New Haven and London:
Yale University Press, 1988), chapters 5-7.
13
This is of course, a romanticized portrayal of the past. In fact, it was common for many early Chinese
philosophers to fall back on history and tradition, even if it meant fabricating the truth, to bring about
changes in the present, tuō gǔ gǎi zhì (making use of past traditions to change the present
system); many instances can be found in early Warring States texts such as Mencius.
9

of Zhōu, and passed down in written form by texts accumulatively known by the same
name, which, with cross reference to the other traditional classics,
14
helped Confucius
derive notions of yí 宜 (generally translated as propriety) and yì 義 (rightness or
righteousness).
15

However, given the changing circumstances, he also saw a need to
provide a universal moral principle for the people, centered on what was arguably the








14
They are namely, Shī , Shū , Yuè , Yì and Chūnqiū . Contrary to past speculation among
sinologists that the classification of these six classics as the “six canons (Liù Jīng )” came much later
in history (as the term Liù Jīng was only first mentioned in the “Tiānyùn ” chapter of the Zhuāngzǐ
), the “Liù Dé

(Six Virtues)” in the Guōdiàn corpus listed the name of these six classics in
sequence (though the term Liù Jīng was not mentioned). (See “Liù Dé” slips 24-25, Guōdiàn Chǔmù
Zhújiǎn, 188.) Given this, it is possible that the six classics had already been canonized at the time of the
early Confucians. But this does not mean the five classics we have now (Yuè is no longer extant) are the
same as what was available to the Confucians. There are several texts in the present five classics that could
have been later additions or even forgeries.
15
For a good examination of the etymological roots and historical interpretation of yì, see David L. Hall
and Roger T. Ames, “Getting it right: On saving Confucius from the Confucians,” in Philosophy East and
West vol. 34 no. 1, (Jan. 1984), 3-23.
10

predominant notion in Confucius’ thought: the notion of rén 仁.
16

16
The relation between rén and lǐ is a dynamic one. Wei-Ming Tu, for example, sees rén as “a principle of
inwardness” and not a quality acquired from outside, whereas lǐ is “an externalization of jen (rén) in a
specific social context,” or “a principle of particularism.” He added that: “A Confucian may very well
refute an established lǐ by exposing its incompatibility with jen [rén].” See Wei-Ming Tu, “The creative
tension between Jen and Li,” in Philosophy East and West vol. 18 no. 1-2, (Jan April 1968), 29-39. This
rén-centred interpretation is different from Herbert Fingarette’s which placed his emphasis on lǐ instead,
seeing lǐ as a “holy rite,” by which an adherence is central in importance. See Herbert Fingarette,
Confucius: The Secular as Sacred (New York: Harper and Row, 1972). Kwong-Loi Shun, in a separate
article on the Analects much later, identified two traditional positions, the “instrumentalist,” where rén has
an evaluative priority over lǐ; and the “definitionalist” where rén is defined by an adherence to lǐ, seeing the
latter as “conservative” as rén “no longer provides a standard against which one can assess the justifiability
of revising or departing from a rule of lǐ.” He proposes an interpretation that lies “between the two
extremes,” where “rén is shaped by the existing lǐ practices in that it is not intelligible and cannot be shown
to have validity independent of lǐ,” and yet there is “room for departing from or revising an existing rule of
lǐ,” although revision “has to proceed against the background of a general acceptance of the existing lǐ
practices and has to be based on good reasons,” adding that Confucius probably believes “that the existing
lǐ practice do function well and that there is little basis for departure or revision.” (Kwong-Loi Shun, “Jen
and Li in the Analects,” in Philosophy East and West vol. 43 no. 3, (July 1993), 457-479.) This position is
later described by Chenyang Li as the “constitution” position, who, instead of using Shun’s analogy of
marriage where lǐ is compared to the rites involved in a marriage, necessary and sufficient for rén
(analogically, the marriage), recommended his grammar-language metaphor to illustrate the lǐ-rén relation.
In his article, he listed six reasons why his analogy was aptly construed, and also explained how, by seeing
lǐ as “cultural grammar” and rén as “the mastery of a language,” one avoids two areas where Shun’s
analogy breaks down. See Chenyang Li, “Li as Cultural Grammar: On the Relation between Li and Ren in
Confucius' Analects,” in Philosophy East and West vol. 57 no. 3, (July 2007), 311-329.
11

It is difficult to translate rén into English, as Confucius had meant it in different
senses dependent on context, of which some were near to what we understand as
“benevolence” or “humaneness.”
17
In fact, one can even say that rén, when fleshed out in
term of shù 恕,
18
is just another way of phrasing the Golden Rule.
19

However, it should be pointed out that the early Confucians were not armchair
philosophers. With the political upheaval and social turmoil, which predictably became
worse in the centuries after Confucius’ death, what they sought was essentially a solution
to what was going on in their lives at the time. By following Confucius, they were giving

17
For a detailed analysis on the different senses of rén, see Wing-Tsit Chan, “The Evolution of the
Confucian Concept Jen,” in Philosophy East and West vol. 4 no. 4, (Jan. 1955), 295-319. As rightly
pointed out in the article, rén, besides carrying its different “particular” senses, can also take on a general
sense of an encompassing notion for all virtues.
18
See Analects 4:15. Philip Ivanhoe has written an interesting article which investigates the relationship
between zhōng and shù, as interpreted by various prominent scholars; he then offers his own view: that
zhōng is adherence to lǐ; with shù providing the flexibility when applying lǐ in practical situations. In a way,
this parallels Wei-Ming Tu’s and the “instrumentalists” interpretation of the relation between rén and lǐ as
mentioned earlier in the footnotes. See Philip J. Ivanhoe, “Reweaving the ‘one thread’ of the Analects,” in
Philosophy East and West vol. 40 no.1, (Jan. 1990), 17-33.
19
It is interesting how similar versions of the Golden Rule have emerged as a universal ethical principle
across different civilizations. For a good summary of various interpretations of the Confucian version of the

Golden Rule, as well as the similarities and differences between it and the Christian version, see Sin Yee
Chan, “Can shu be the one word that serves as the guiding principle of caring actions?” in Philosophy East
and West vol. 50 no 4, (Oct. 2000), 507-524.
12

recognition to his proposal, which necessitated a reversal to the “imagined” traditions
upheld by Zhōu and an appeal to a virtue known as rén.
20

We do not know for sure how many disciples Confucius had in all, but a certain
part of the Shǐjì 史記 recorded seventy prominent immediate ones,
21
who after the death
of their master, “broke up and scattered among the feudal lords, the more important ones
becoming tutors and high ministers to the rulers, the lesser ones acting as friends and
teachers to the lower officials, while some went into retirement and were never seen
again.”
22
As far as this dissertation is concerned, I do not focus on any particular disciple
or, for that matter, any early Confucian “schools” following the demise of Confucius,

20
In this light, Graham is right when he pointed out that Chinese philosophy in general is more concerned
with “the Way” than “the Truth.” (See A. C. Graham, Disputers of the TAO, 3.)
21
This record of seventy is found in the “Biographies of Confucian scholars (Rúlín Lièzhuàn ),”
which could be only an approximate round number. The exact number of these prominent ones actually
varies from seventy to seventy-two to seventy-seven even within the Shǐjì. The “Hereditary House Of
Kǒngzǐ (Kǒngzǐ

Shìjiā ),” for example, records for us three thousand disciples, of which seventy-
two are well-versed in the six arts (Liù Yì ); the “Biography of Zhōngní’s Disciples (Zhōngní Dìzǐ
Lièzhuàn ),” on the other hand, quoted Confucius as saying he had seventy-seven disciples
who studied under him and had extraordinary capabilities, and even breaking the latter according to their
strengths in four specializations, namely virtues, political matters, words and literature.
22
Burton Watson trans., Sīma Qiān, Records of the Grand Historian of China (New York and London:
Columbia University Press, 1961), 396.
13

because there was not sufficient evidence to substantiate such speculations.
23
Instead, my
interest is the intellectual landscape of largely, Warring States Confucianism, treating the
Analects, the Mencius and recently excavated early Confucian writings as an embodiment







23
With the retrieval of recent archaeological findings, many scholars have speculated the existence of a
Confucian school of thought known as the “Sī Mèng Wǔxíng school” (Sī Mèng Wǔxíng Xuépài
). The “Sī Mèng Wǔxíng school” started receiving much attention after the excavation of a text,
named “Wǔ Xíng (The Five Virtuous Conduct),” in the Mǎwángduī findings, which shifted
our attention from the traditional five-phases form of “wǔxíng” to another possible interpretation of seeing
“wǔxíng” as five Confucian virtues, which, in turn, explains why “wǔxíng” was mentioned when the writer
of the “Fēi Shíèr Zǐ ” chapter in Xúnzǐ criticizes the thought of Zǐsī (483 - 402 B.C.E.)

and Mencius collectively. It so happened that a similar “Wǔ Xíng” (without its commentary) is also found
in the Guōdiàn corpus, and thus renewing our interest in this mysterious “Sī Mèng Wǔxíng school.” This
connection is further substantiated by texts in the Guōdiàn corpus that were most likely written by Zǐsī and
perhaps more probably his disciples, the “Zīyī (The Black Robes)” and “Lǔmùgōng Wèn Zǐsī
(Duke Mù of Lǔ asks Zǐsī)” being good candidates. I am receptive to this possibility. However, this
dissertation does not entail a conclusive position with regard to this issue, neither is there a need for me to
see the thought propagated in, for example, the Guōdiàn findings as belonging to a particular school
exclusively.
14

of the collective memory of this group of people known as the early Confucians.
24

1.1.2. Emotions, not qíng
This dissertation looks at emotions as mentioned by this group of intellectuals in
extant written texts, and how these narratives, in turn, shed light on what we understand
as early Confucianism. Contrary to our common understanding, the word “emotion” is
not equivalent to the Chinese notion qíng 情, for qíng means much more than emotions.
Most discussions on the notion of qíng in ancient China takes flight from A. C.
Graham’s insightful generalization in 1986 that qíng never meant “passions,” even as

24
I am interested in early Confucian thought as a whole, and the Guōdiàn findings are a part of early
Confucian texts available to me. For a good book that sees the Guōdiàn findings in light of the “Sī Mèng
Wǔxíng school,” see Liáng Tāo , Guōdiàn Zhújiǎn yǔ Sī Mèng Wǔxíng Xuépài
(Běijīng: Zhōngguó Rénmín Dàxué chūbǎnshè, 2008.) On that note, I am neither interested in who was the
“successor” of Confucius’ thought. There are some articles written in the past on whether Mencius was the
one, especially given his reputation as the “second Confucian sage (yǎshèng )” in the Confucian
tradition. See for example, Philip Ho Hwang’s article and Martin Lu’s interesting comments in Philip Ho
Hwang, “A New Interpretation of Confucius,” in Philosophy East and West vol. 30 no. 1, (Jan. 1980), 45-

55; and Martin Lu, “Was Mencius a true successor of Confucius?” in “Comment and Discussion,”
Philosophy East and West vol. 33 no. 1, (Jan. 1983), 79-86. Any such endeavor eventually runs into
problems like the difficulty in defining “successor” and even the speculation and reconstruction of a
distinct “tradition,” of which we have not enough evidence to make conclusive remarks. For one, we do not
know for certain the “Dàxué ” and “Zhōngyōng ” chapters of the Lǐjì , classified by Zhū Xī
(1130 - 1200), together with the Analects and the Mencius as the “Four Books” (Sì Shū ), were
from the Warring States or, like most other chapters of Lǐjì, came into being only in the Hàn. In this
dissertation, I adopt a safe approach by not considering both chapters as early Confucian texts and, thus,
not as part of my primary sources.
15

late as in the Xúnzǐ. The “passions” sense of qíng was only found in texts after the
classical period, maturing in neo-Confucian thought, where it is contrasted with xìng 性,
commonly translated as “nature.”
25
He then went on to illustrate that qíng, taken as a
noun, means “the facts;” as an adjective, “genuine;” and as an adverb, “genuinely.”
This point was challenged by Chad Hansen nine years later, who saw Graham’s
explanation of a shift in the meaning of qíng from “reality” and “essence” to “passions”
as “unsettling.”
26
Instead, he proposed that there exists a single, common interpretation
across all the different senses, namely, “reality feedback” or “reality input.”
27
However, it
should be noted that Hansen was only questioning the semantic development model
Graham proposed, focusing instead on what was common (a basic sense) between the
two different senses and denying a shift from a sense that was “metaphysical and
objective” to one that was “subjective and psychological.”
28

He was not challenging
Graham’s viewpoint that qíng never meant “passions” in the Warring States.
In 2004, Michael Puett, making use of a Guōdiàn text, the “Xìng Zì Mìng Chū

性
自命出(Nature comes with allotment),” re-ignited this discussion by illustrating that qíng

25
A. C. Graham, Studies in Chinese Philosophy and Philosophical Literature (Singapore: Institute of East
Asian Philosophies, reprint, 1986), 59.
26
Chad Hansen, “Qing (Emotions) in Pre-Buddhist Chinese Thought,” in Joel Marks and Roger T. Ames,
Emotions in Asian Thought: A Dialogue in Comparative Philosophy (Albany: State University of New
York Press, 1995), 195.
27
Chad Hansen, “Qing (Emotions) in Pre-Buddhist Chinese Thought,” 183.
28
Chad Hansen, “Qing (Emotions) in Pre-Buddhist Chinese Thought,” 195.
16

in this text actually meant “idealized emotional dispositions,”
29
and pointed out that “any
attempt to translate qíng in such a text as either ‘reality response’ (with no affective
sense) or simply ‘emotions’ would be misleading,” adding that the “reality response”
sense was only distinguishable in the Huáinánzǐ and that of “emotions,” in the works of
Dǒng Zhòngshū 董仲舒 (179 - 104 B.C.E.). In other words, he challenged both Graham’s
semantic-shift model and Hansen’s proposal of a “single, unified meaning” and sees qíng
as having “a broad semantic range” and different “shades of meanings” selectively used
by different thinkers.

30

Along the same lines, Christoph Harbsmeier, in another article in the same book,
derived seven possible semantic senses of qíng in pre-Buddhist Chinese, substantiating
each with textual evidence, the seventh of which was closest to the sense of “emotions,”
though only found predominantly in lyrical poetry, especially the Chǔcí 楚辭
,
instead of




29
Michael Puett, “The Ethics of Responding Properly: The Notion of Qíng in Early Chinese Thought,” in
Halvor Eifring ed., Love and Emotions in Traditional Chinese Literature (Leiden: Brill, 2004), 50-51; 67.
This sense is mainly derived from how the sages were said to “[take] the worthy traditions from the past,
organized them, patterned (lǐ ) their qíng, and thereby made them available to educate the latter-born” in
the “Xìng Zì Mìng Chū,” (ibid., 50).
30
Michael Puett, “The Ethics of Responding Properly: The Notion of Qíng in Early Chinese Thought,” 67.
17

Confucian philosophical writings.
31

This short detour into the debate of qíng is to illustrate the complex notion of qíng
in ancient Chinese thought. In fact, any discussion of qíng will probably involve more
notions such as xìng and mìng 命, each with its wide semantic range and history of
academic debate.
Perhaps as Halvor Eifring suggested, the Wittgensteinian model of “family

resemblance” can help explain how seemingly unrelated senses of a word can be related
to one another:
Among a number of senses A, B, C etc., A may not be directly related to C, but
indirectly so, because A is related to B and B is related to C. In the case of qíng,
“basic facts” is not directly related to “love,” and there is no single basic meaning
underlying both interpretations. Still, there is a connection between “basic facts
(of a matter)” and “basic instincts (of man),” and between “basic instincts” and
“emotions,” and again between “emotions” and “love.” In this perspective, it
makes sense to treat qíng meaning “basic facts” as belonging to the same lexical
item as qíng meaning “love.”
32


31
They are (1) Factual: The basic facts of a matter; (2) Metaphysical: Underlying and basic dynamic
factors; (3) Political: Basic popular sentiments/ responses; (4) Anthropological: General basic instincts/
propensities; (5) Positive: Essential sensibilities and sentiments, viewed as commendable; (6) Personal:
Basic motivation/ attitude; and (7) Emotional: Personal deep convictions, responses and feelings. See
Christoph Harbsmeier, “The Semantics of Qíng in Pre-Buddhist Chinese,” in Halvor Eifring ed., Love and
Emotions in Traditional Chinese Literature (Leiden: Brill, 2004), 69-148.
32
Halvor Eifring, “Introduction: Emotions and the Conceptual History of Qíng,” in Halvor Eifring ed.,
Love and Emotions in Traditional Chinese Literature (Leiden: Brill, 2004), 9.
18

I will come back to the Wittgensteinian model of “family resemblance” in the
next section on methodology adopted in this dissertation; for now, it is sufficient to say
that for the purposes of this research, my focus is on how the early Confucians interpreted
specific emotions rather than on various meanings attributed to qíng.
1.1.3. Our interest

This leads, then, to what is understood by emotions in the context of early
Confucianism.
With regard to this dissertation, my emphasis was not in situating early Confucian
emotional theories in the context of contemporary debates, but more on what the various
narrative accounts of emotions in early Confucian texts reveal about the early Confucians,
especially their value system. This approach thus moves away from heavy theoretical
issues such as the validity of a cognitivist account of emotions, motivational theories, the
importance of social conditioning on emotions and the stages involved in an emotional
process. In other words, my focus is on “what a certain emotion means”—rather than
“what is an emotion”—in the early Confucian context. In other words, my focus is not so
much on what the emotions are in the ontological sense, but more of what the early
Confucian understanding of emotions can tell us about early Confucianism, as well as the
ethical implication involved.
However, there are certain underlying assumptions that can be made with regard
to the nature of emotions as understood by the early Confucians.
Firstly, the early Confucians regarded emotions as intentional states caused by an
external object or event. For example, a person’s joy could come from the sight of a
beautiful mountain in front of him; or his anger could be provoked when he witnessed a
19

tyrant’s cruel act. These emotional responses need not be rational or even morally
correct. For example, one may feel angry when slighted by another person
unintentionally; or one may feel joy knowing he obtained wealth not by righteous means.
Thus, to the early Confucian, emotions always involved a cognitive process, in contrast to
some moods and sensations that are seemingly unintentional or may have been caused by
non-cognitive factors.
Secondly, and following from the previous point, the early Confucians believed
what a person feels necessarily reflects who he is or wants to be. A virtuous person will
feel a certain set of emotions in a particular context, whereas a vicious person will feel
otherwise. Similarly, a person who aspires to become a virtuous person will also feel a

certain set of emotions whereas a person who does not do so will feel another. Having
said that, whether this necessarily leads to a set of corresponding behaviors was not
known, as in the case of hypocrisy, but it seems likely that the early Confucians agreed
that a person who underwent the right cultivation would necessarily feel the right
emotions. I will come back to the implications involved in this important assumption in
the concluding remarks of this dissertation.
The third point is about the difference between man and animal. It is difficult to
conclude whether the early Confucians believed that animals share some of our emotions
as they often projected human emotions onto animals when they saw parallel behaviors;
for example, sadness detected in the cries of a dying bird or fear in that of a goat about to
be slaughtered. It is also uncertain whether they considered these displays of emotions as
cognitively induced. Nonetheless, they drew a line between man and animal with regard
to moral emotions. To them, the mark of a human lies in one’s ability to differentiate
20

right from wrong, and thus one’s ability to feel joy when one does something morally
praiseworthy and shame and/or guilt when one does something morally blameworthy.
As far as this dissertation is concerned, these are assertions with regard to how the
early Confucians understood the nature of emotions. In subsequent chapters, I approach
the subject matter by structuring my investigation on some of the primary emotions in
Chinese philosophy, namely xǐ 喜, nù 怒, āi 哀 and lè 樂. The choice of these emotions is
by no means exhaustive. In fact, we cannot even say that these four emotions constitute
what is recognized as “a set of primary emotions” in the early Confucian context. The
“Xìng Zì Mìng Chū,” for instance, seems to consider xǐ, nù, āi and bēi 悲 together as a
“basic” group. In other words, the choice of xǐ, nù, āi and lè, is guided primarily by later
delineation of them in Chinese tradition as the primary emotions. In modern Chinese, of
course, xǐ nù āi lè serves as a convenient idiom to mean “emotions” in general.
33

I also looked at other related emotions in an attempt to pinpoint their meaning in

early Confucianism; for example, the difference between xǐ and yuè 悅 , and their
difference with lè; the difference between nù, yùn 慍 and yuàn 怨; or āi, qī 戚 and bēi. I
also explored fear-associated emotions such as jù 懼, wèi 畏 and huàn 患. Each of the
following chapters starts with a philological study of the concept represented by a
particular character, followed closely by the philosophical analysis on the implications

33
Ulrike Middendorf has consolidated a database of these “basic emotions.” According to her research, xǐ
nù āi lè

did not come to exist as a basic group until the second century B.C.E. in texts like the Guǎnzǐ and
the Zhuāngzǐ. See Ulrike Middendorf, “Basic Emotion Terms in Warring States Texts: Sequences and
Patterns,” in Paolo Santangelo and Donatella Guida eds., Love, Hatred, and Other Passions: Questions and
Themes on Emotions in Chinese Civilization (Leiden: Brill, 2006), 126–148.
21

when a certain character was used. By studying the choice of character used and the
implications that particular character carries, we step into the emotional world of the
early Confucians.
I should also explain why I chose not to discuss notions such as hào 好 (likes) (or
ài 爱, love of), wù 惡 (dislikes/ hatred of) and yù 欲 (desires).
34
Although I agree that an
investigation on these notions can aid in our understanding of the early Confucians, going
into detail also complicates matters. Notions such as preferences and desires differ from
our common understanding of emotions in the following two senses. One, they can be
understood as dispositions which are not triggered off by external circumstances. For
example, a man’s tendency to procrastinate can be explained by his penchant for idling,
dislike for work or desire for other activities, but we cannot flesh out these dispositional
attitudes as emotions. Alternatively, they are emotions only in a derived sense. For

example, my preference for coffee could be the result of enjoyable coffee breaks with
friends in the past, or my dislike for tea could be because of an embarrassing experience
where I spilled tea on myself at a social occasion.
Thus, though a study of dispositional attitudes is useful in examining the causes
and effects of emotions, it may not necessarily be a study specifically on emotions. In
addition, a preference for doing something or a desire for something is closely related to
the joy one hopes to get when doing or having something; conversely, a dislike for doing

34
The “Lǐyùn ” chapter of Lǐjì listed xǐ, nù, āi, jù, ài, wù, yù as “rén qíng .” As for the
“Zhèngmíng ” chapter of Xúnzǐ, two different combinations were mentioned: “hào, wù, xǐ, nù, āi, lè”
and “xǐ, nù, āi, lè, ài, wù, yù,” within the same chapter, commonly identified later as the six-qíng theory (liù
qíng shuō ) and seven-qíng theory (qī qíng shuō ) respectively.
22

another thing, or a distaste for it, is intimately associated with the irritation and frustration
one seeks to avoid when doing or having that particular thing. To write another chapter
on preferences and desires would result in unnecessary repetitions. In fact, we are
actually addressing dispositional attitudes such as preferences and desires when we
account for what causes joy or anger, sorrow or fear.
Indeed, there already exists a wide lexicon to choose from in identifying
emotional terms available to the early Confucians. By the time of the Analects, the
vocabulary of emotions had already expanded to an interesting spectrum containing
words with different and often subtle tones and shades. In fact, many of the terms were
also found in the pre-Confucius Chinese classics, to which I will cross-reference in
subsequent chapters.
35

More importantly and as mentioned earlier, given the close connection between
external turns of events and the emotions of the early Confucians, the coming to terms

with their own emotions has become part and parcel of the Confucian ethical enterprise.

35
With the new discoveries made in the archaeological field, we can no longer hold that emotional terms
were not found in the pre-Classical period. A search on the CHANT system (CHinese Ancient Text
Database) provided by the Chinese University of Hong Kong gives us, for example, 120 hits of xǐ and 12
hits of lè in the oracle bones written records, with 94 hits and 70 hits in the bronze inscriptions respectively.
In addition, in the bronze inscriptions, there are more than 16 instances where xǐ and lè were used within
the same text, with even two hits having xǐ and lè side by side with each other. It does seem that xǐ and lè
are the most frequently used emotional terms in the pre-classical period. Other emotional terms found
include 17 qī in the oracle bones; as well as 9 āi, 1 jù, 2 qī, 14 wèi and 11 yōu in the bronze inscriptions. In
addition, we do not dismiss the possibility of the existence of other emotional terms or the variants of these
terms in both records.
23

An investigation of their interpretation thus serves as a good platform in understanding
early Confucianism.
1.1.4. Conclusion
“Early Confucianism” is not a solitary theoretical concept with a single thread of
thought. Rather, it is a heterogeneous mix held by several generations of individuals,
spread over a considerably wide geographical area in space, who sought to interpret and
apply Confucius’ teachings in life. This section serves to clarify the subject matter in this
dissertation. I have explained why I chose not to address the encompassing notion of
qíng, together with dispositional attitudes, as well as the nature of emotions ontologically.
Instead, my focus is on how a group of intellectuals who readily identified themselves as
Confucius’ disciples and followers understood emotions, both descriptively and
prescriptively. Hence, I chose to restrict my analysis to a series of emotions, albeit not
all-encompassing, so as to shed light on the psychological and moral world of these early
Confucians who were living in chaotic times: on the basis that perhaps the best way to
understand their world was through their joy and anger, hopes and disappointments, to

better understand how they felt or how they thought they should have felt.
1.2. The Methodology
1.2.1. Philology as our point of departure
As far as the Chinese language is concerned, there is often an interesting relation
between the etymological roots of a character and its meaning. An understanding of the
etymological roots of a Chinese character is often helpful when we attempt to investigate
its semantic usage, especially with regard to the study of early classical texts where the
usage of individual characters had yet deviated much from its etymological origins.
24

This dissertation treats philology as a point of departure.
36
In subsequent chapters,
I start by identifying a few characters within a semantic range, seeking to understand the
original meanings of individual words as suggested by their etymological components,
before exploring their usages and setting them apart according to varying textual
contexts.
But this is not as simple as it seems.
Take Chapter 2, for example. I identified three main characters, xǐ, yuè and lè.
However, each of these characters comes with two attributes: a semantic scope and a
syntactic scope. I chose the words based on semantic scopes: xǐ, yuè and lè carry
semantic senses which overlap with one another; and in relation to their English
translation, they relate broadly to notions such as “happiness,” “pleasure,” and “joy,”
though not neat enough for a one-to-one transliteration.
In a way, the semantic scopes of xǐ, yuè and lè overlap each other like a Venn
diagram with all three domains cutting partially into one another, such that one of my
tasks was to see how this is done. In other words, there are contexts in which both xǐ and
yuè can be used; and there are cases where only xǐ can be used, but not yuè or vice versa.
This is further complicated by the syntactic scope of the individual word, that is, its
grammatical property. For example, in the case of xǐ and yuè, yuè can be negated (by

adding a bù 不 in front of yuè to get búyuè 不悅) but xǐ cannot. Thus in situations where

36
As far as this thesis is concerned, the word “philology” is used in a narrower sense as pertaining to
etymology or “word study”. I understand that it can also be used more broadly to include the close study of
texts.
25

one need to convey an emotion near to that of “unhappiness,” the usage of xǐ by negating
it is out of consideration.
However, philology is only a point of departure, as my main focus is not on the
etymologies of the individual character depicting an emotion. Instead, I am interested in
the implications of their different usages. Syntactic differences aside, in what contexts did
the early Confucian use xǐ, and what about yuè? How many shades and hues of
“happiness-related” emotions did they have? Were they necessarily aware of these
differences? Which particular emotion did they most desire? What were the implications
if one felt a certain emotion? And how did their attitudes towards such emotions tell us
about their value system? These were some of the questions I wanted to explore in this
dissertation; to do so, I chose to start with a philological “word-study” of each emotional
notion as a foundation for its philosophical study.
1.2.2. A thematic dichotomous pattern and the notion of qì
Besides using the philology-philosophy structure to facilitate discussion in the
core chapters, readers will also observe an underlying thematic dichotomous pattern,
notably with xǐ and nù as a semantically antonymous pair; and lè and āi as another.
Together, these emotions correspond to another pair of a similar nature, one that is
associated with external turns of events assessed from the agent’s point of view, namely,
dé 得 (gains) and shī 失 (losses).
To elaborate, I will start by categorizing emotions.