Encyclopedia of Chinese Language and Linguistics
Volume 3
Men–Ser
For use by the Author only | © 2017 Koninklijke Brill NV
General Editor
Rint Sybesma
(Leiden University)
Associate Editors
Wolfgang Behr
(University of Zurich)
Yueguo Gu
(Chinese Academy of Social Sciences)
Zev Handel
(University of Washington)
C.-T. James Huang
(Harvard University)
James Myers
(National Chung Cheng University)
For use by the Author only | © 2017 Koninklijke Brill NV
ENCYCLOPEDIA OF
CHINESE LANGUAGE
AND LINGUISTICS
Volume 3
Men–Ser
General Editor
Rint Sybesma
Associate Editors
Wolfgang Behr
Yueguo Gu
Zev Handel
C.-T. James Huang
James Myers
LEIDEN • BOSTON
2017
For use by the Author only | © 2017 Koninklijke Brill NV
Typeface for the Latin, Greek, and Cyrillic scripts: “Brill”. See and download: brill.com/brill-typeface.
ISBN 978-90-04-18643-9 (hardback, set)
ISBN 978-90-04-26227-0 (hardback, vol. 1)
ISBN 978-90-04-26223-2 (hardback, vol. 2)
ISBN 978-90-04-26224-9 (hardback, vol. 3)
ISBN 978-90-04-26225-6 (hardback, vol. 4)
ISBN 978-90-04-26226-3 (hardback, vol. 5)
Copyright 2017 by Koninklijke Brill NV, Leiden, The Netherlands.
Koninklijke Brill NV incorporates the imprints Brill, Brill Nijhoff, Global Oriental and Hotei Publishing.
All rights reserved. No part of this publication may be reproduced, translated, stored in a retrieval
system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording or
otherwise, without prior written permission from the publisher. Authorization to photocopy items for
internal or personal use is granted by Koninklijke Brill NV provided that the appropriate fees are paid
directly to The Copyright Clearance Center, 222 Rosewood Drive, Suite 910, Danvers, MA 01923, USA.
Fees are subject to change.
This book is printed on acid-free paper and produced in a sustainable manner.
For use by the Author only | © 2017 Koninklijke Brill NV
583
Rgyalrong Language
Rgyalrong Language
1. Int roduct io n
Rgyalrong is a group of four languages spoken
in Sìchuān province, People’s Republic of China.
These languages, along with Lavrung and Horpa/
Rtau, belong to the Rgyalrongic branch of the
Sino-Tibetan family (Sun 2000); they are likely
to be closely related to other languages such as
Pumi, Muya, Tangut, Queyu and → Qiang within
a larger Qiangic branch. The four Rgyalrong languages are known by various names, as some
scholars use the Tibetan place names of the area
where these languages are spoken, while others
use the Chinese transcription of these names.
Eastern Rgyalrong or Situ (Chinese Sìtǔ 四土,
referring to the four main Tǔsī 土司—a Chinese
administrative unit—governing the Rgyalrong
area) is the most widely spoken language with
about 100,000 speakers, and is the only one to
have been put to writing before 1949 (see Btsan.
lha 2010). It is spread over a large area from Lǐxiàn
理縣 county (Rnga.ba/Ābà 阿壩 prefecture) in
the east to Rong.brag/Dānbā 丹巴 county (Dkar.
mdzes/Gānzǐ 甘孜 prefecture) in the west. The
main references on this language are Lín’s (1993)
grammar, Huáng and Sūn’s (2002) dictionary,
Lin (2003) on the verbal system and Lin (2009)
on tone and intonation. The other three languages have fewer than 10,000 speakers each.
Japhug (Chábǎo 茶堡) is spoken in the
north-east of Mbarkhams/Mǎ’ěrkāng 馬爾康
county (Rnga.ba district). The main references
are Jacques’s (2008) grammar and Jacques and
Chen’s (2010) glossed text collection.
Tshobdun (Cǎodēng 草登) is located next to
Japhug in Mbarkhams county. No book-length
monograph on this language has appeared yet,
but numerous articles by Jackson Sun describe
various aspects of its phonology and morphosyntax (for instance, Sun 2000, 2003, 2006, 2007).
Zbu (Rìbù 日部, Showu) is located in the
northeast of Mbarkhams county in pockets in
neighboring Ndzamthang county. Few articles
have been published on this language (the
only reference speciijically devoted to Zbu is
Sun 2004).
The four Rgyalrong languages have many distinct dialects, each of which deserves detailed
documentation, including text collections, dictionaries and reference grammars.
In spite of being genetically related to Chinese,
the Rgyalrong languages are typologically quite
divergent from the isolating type. As we will see,
they are the only fully polysynthetic languages in
China, or even in all of Asia if we except isolated
languages of the sub-polar and polar regions
such as Ket, Ainu and Chukchi. These languages
are of exceptional interest for both linguistic
typology and historical linguistics.
2. Ph o n o lo g y
Unlike modern Chinese languages which have a
very simple syllable structure with few or no initial clusters, Rgyalrong languages have the most
complex syllabic structure of all Sino-Tibetan
languages.
Rgyalrong languages typically have large
consonant inventories; for instance, Japhug
has the following consonantal system with 49
phonemes:
p
pʰ
b
mb
m
w
t
tʰ
d
nd
ts
tsʰ
dz
ndz
n
s
z
l
ɬ
tʂ
tʂʰ
dʐ
ndʐ
tɕ
tɕʰ
dʑ
ndʑ
ʂ
ɕ
ʑ
r
c
cʰ
ɟ
ɲɟ
k
kʰ
g
ŋg
ɲ
ŋ
x
ɣ
q
qʰ
ɴɢ
χ
ʁ
j
More than 340 consonant clusters are attested in
Japhug, including clusters of up to three or even
four consonants. Some of the complex clusters of
Old Tibetan which are preserved in none of the
attested modern → Tibetan languages are still
present in Rgyalrong. For instance, Old Tibetan
bsgyur(d) ‘to change (past stem)’ is borrowed as
βzɟɯr in Japhug.
For use by the Author only | © 2017 Koninklijke Brill NV
Rgyalrong Language
Rgyalrong languages (except Japhug) are tonal
languages, and present morphological tone
assignment rules (see Sun 2005; Lin 2009).
Some of the consonantal contrasts found in
Rgyalrong languages are typologically quite
rare. We will only mention three of the most
unusual ones.
First, Tshobdun and Japhug have a series of /
fricative+prenasalized voiced stop+medial/ clusters, as in Japhug ʁmbɣi ‘sun’, ʑŋgri ‘star’ etc.
Second, Japhug has an intriguing ijive-way
contrast between /velar stops+j/ as in kjo ‘to
cause to glide’, /palatal stops/ co ‘valley’, /dental
affricates+j/ ɯ-mtsjoʁ ‘beak’, /alveolo-palatal affricates/ as in tɕoʁtsi ‘table’ and /uvular stops+j/ as
in qjoʁ ‘to vomit’. Palatal stops can be followed by
-ɣ-, -r- or -l- as in ɲcʰɣaʁ ‘birch bark’.
Third, Zbu has a very complex vowel system,
with a series of velarized vs. plain vowels (Sun
2000, 2004:272).
3. M orp h osy nt ax
Rgyalrong languages have a complex verbal
morphology, unlike anything else found in
the Sino-Tibetan family. They present all the
characteristics of polysynthetic languages: 1)
A high word to morpheme ratio (most verbal
forms include at least three morphemes, and
can have up to eight); 2) Head-marking typology (e.g., grammatical relationships are marked
mainly on the verb, and within the noun phrase
possession is marked on the possessed rather
than on the possessor); 3) Presence of nominal
incorporation.
For want of space, we restrict this section to
a description of four of the typologically most
unusual features of Rgyalrong languages: stem
alternation, inverse marking, incorporation, and
word-order typology.
3.1 Verb stem alternations
Verb stem alternations in Rgyalrong languages
were ijirst discovered by Sun (2000, 2004). All
four Rgyalrong languages have alternations of
some kind, with Zbu presenting by far the most
complex system.
Three stems are found across the four languages: stem 1 (basic stem), stem 2 (aorist),
584
stem 3 (non-past transitive, singular non-inverse
agent).
In Zbu, the alternations are largely unpredictable (as with strong verbs in Germanic). Some of
these segmental and tonal alternations present
polarity:
Table 1. Stem alternations with polarity in Zbu
(Sun 2004:276)
alternation stem 1
stem 2
meaning
i-ə
ə-i
i-e
e-i
smə̄
vī
ʁlē
qʰjîv
to be cooked
to come
to speak ill of
to be bitter
smî
və̄
ʁlî
qʰjêv
Some alternations are partially productive,
as demonstrated by the fact that they apply
to Tibetan (or even Chinese) loanwords. For
instance, the Zbu verb nkrós ‘to discuss’, a
denominal verb derived (by the preijix n-) from
the borrowed Tibetan noun gros ‘discussion’, has
the following three stems: nkrós, nkhrús, nkrə́m
(personal ijieldwork).
In some cases, it can be proven that some
stem alternations are secondary from a diachronic point of view. For instance, the Japhug
a/e, o/e, ɯ/i vowel alternations between stem 1
and stem 3 have been shown to originate in the
fusion of a sufijix *-jə (still attested in Tshobdun,
see Sun 2003) with the verbal stem (see Jacques
2008:234–235). In most cases, the irregular alternations go back to proto-Rgyalrong and possibly
to an even earlier stage.
3.2 Inverse marking
Direct/Inverse alignment is a type of agreement
system in which a marker (the inverse marker)
indicates that the agent is lower than the patient
on the empathy hierarchy:
1. SAP (Speech Act Participant, 1st or 2nd person) >> human >> animal >> inanimate >>
generic
Verbal forms where the agent is ijirst or second
person and the patient is third person (SAP>3)
are called direct, as are forms with two third
For use by the Author only | © 2017 Koninklijke Brill NV
585
Rgyalrong Language
person arguments (3>3) when the agent is higher
than the patient on hierarchy (1). Direct forms
never have the inverse preijix in Japhug.
The existence of inverse marking in Rgyalrong languages was ijirst noticed by DeLancey
(1981) and described by Sun and Shi (2002) and
Jacques (2010). The following table presents the
paradigms of regular transitive and intransitive
verbs in the non-past (the only ijinite verbal form
without tense/aspect/mood preijixes) in Japhug
Rgyalrong, only taking into account verbal forms
where both arguments are singular. We only
present here a restricted account of Japhug verbal agreement.
The -Ø sufijix marks non-ijirst person singular;
it is replaced by other sufijixes in the dual and
plural. The symbols Σ1 and Σ3 represent verbal
stems 1 and 3, respectively. The rows here refer
to person of the singular agent of the transitive
verbs (A) or the sole argument of the intransitive
ones (S), and the columns to the person of the
singular patient (P). For example, the form for
a transitive verb with second-person agent and
third-person patient (2>3) is found in the second
row, third column (tɯ-Σ3-Ø) (See Table 2).
The “local scenario” forms where both arguments are SAP, 1>2 and 2>1, are marked by
special portmanteau preijixes ta- and kɯ- respectively which are synchronically unanalyzable;
the sufijixes in these forms are coreferent with
the patient (non-ijirst person singular -Ø for 1>2
and ijirst person singular -a for 2>1).
In “mixed scenarios” when one argument is
SAP and the other is third person, we ijind two
situations. First, when the SAP is the agent, the
verbal form is similar to that of an intransitive
verb save for the fact that stem 3 appears instead
of stem 1. When the SAP is patient, sufijixes and
preijixes are still the same as that of intransitive
verb forms, but the inverse preijix wɣɯ́ - or wɣ- is
placed immediately before the verbal stem. The
following examples illustrate 1>3 and 3>1 forms:
2. kɯki mbro ki
ju-nɯtsɯm-a
this horse this ipf-take.away-1sg
I will take away this horse. (Japhug; The Fox, 59)
3. a-rɣa
nɯ-mbala
kɯ
1sg.poss-neighbor 3pl.poss-bull erg
wɣɯ́ -nɯtsɯm-a
ɲɯ-ŋu
inv-take.away-1sg ipf-be
The neighbors’ bull is about to take me away!
(Japhug; The Demon, 8)
When both arguments are third person (“nonlocal scenario”), the transitive verb has two possible forms, either with the inverse preijix wɣɯ́ - /
wɣ- and with stem 1, or without this preijix and
with stem 3. The choice between the two forms
is dictated by both semantic and pragmatic
factors.
In the form Σ3-Ø the singular sufijix -Ø marks
the agent, while in wɣɯ́ -Σ1-Ø, the same sufijix
marks the patient. With an animal agent and
human patient, direct forms are possible in some
cases; inverse is obligatory with an inanimate
agent and animate patient, and with a generic
agent (even a human agent) in Japhug; human
generic arguments are lower than inanimates on
the empathy hierarchy in this language.
Inverse morphology is relatively widespread
in American languages (see Zúñiga’s 2006 for
a recent survey) but rare on other continents.
Rgyalrong is the only group of languages in Eurasia with a consistent inverse system, where the
same morpheme is used in both mixed and nonlocal forms (and even in 2>1 local forms in Situ
and Tshobdun).
Table 2. A restricted Japhug verbal paradigm (non-past, both arguments singular)
1P
1 A/S
2 A/S
kɯ-Σ1-a
3 A/S
wɣɯ-Σ1-a
2P
3P
Intr
ta-Σ1-Ø
Σ3-a
Σ1-a
tɯ-Σ3-Ø
tɯ-Σ1-Ø
tɯ-wɣ-Σ1-Ø
Σ3-Ø
Σ1-Ø
wɣɯ-Σ1-Ø
For use by the Author only | © 2017 Koninklijke Brill NV
Rgyalrong Language
3.3 Incorporation
In Japhug and other Rgyalrong languages, we ijind
a peculiar incorporation-like construction called
indirect incorporation (Jacques 2012c). Verbs
with indirect incorporation are compounds comprising a nominal stem with vowel changes (o, a,
u > ɤ/a, i > ɯ) and a verbal stem. Unlike prototypical incorporation, indirect incorporation is
built not by coalescence of noun and verb, but by
denominal derivation of a nominal compound,
made up of a nominal root and a verb root, by
means of a derivation preijix nɯ-, nɤ-, ɣɯ- or sɯ-;
most (but not all) verbs with indirect incorporation still have corresponding compound action
noun in Japhug, and this derivation is still synchronically transparent (see Table 3).
Verbs derived by indirect incorporation are
not numerous—only about thirty in Japhug
have been found—but the process is still marginally productive, as some verbs have incorporated a noun borrowed from Chinese (‘earn
money’ above, from Chinese piàozi 票子, for
instance).
The incorporated noun can be either the only
argument of an intransitive verb (as in ‘gallop’),
the patient of a transitive verb (‘to fell trees’) or
an adjunct (the other examples); it is never the
agent of a transitive verb. The derived verb is
generally intransitive, except in some rare cases
like ‘to hit with the head’ when the original verb
was transitive and the incorporated noun was
an adjunct. The following pair of examples taken
from the same story illustrates the object-verb
vs. indirect incorporation (Jacques 2012a):
586
4. ɬasa ju-kɯ-ɕe
tɕe, nɯ tɕu
Lhasa ipfv-genr:S/O-go conj dem conj
pɕawtsɯ kɤ-fsoʁ
ɲɯ-mbat
money inf-earn const-easy
‘If one goes to Lhasa, money is easy to earn
there.’
(Lobzang, 22)
5. nɤ-mbro
nɤ-rŋɯl
2sg.poss-horse 2sg.poss-silver
tu-rke-a
tɕe
ipfv-put_in[3]-1sg conj
kɯ-ɣɯ-pɕawtsɯ-fsoʁ
jɤ-ɕe
nmlz:S/A-derivation-money-earn imp-go
tɕe
conj
‘I will prepare a horse and some silver for you,
go to earn money.’
(Lobzang, 17)
Although many languages around the world present more extensive incorporation than Rgyalrong languages, this phenomenon is exceedingly
rare in languages of China and Asia in general
(excluding the polar and subpolar regions).
3.4 Word order typology
Rgyalrong languages have a highly unusual word
order typology, as they are strictly verb ijinal
but have mainly preijixing morphology, both in
the nominal and in the verbal domain. Below a
simpliijied verbal template for Japhug is given,
where we ijind fourteen preijixal slots and only
three sufijixal slots.
Table 3. Examples of indirect incorporation in Japhug
noun
verb
compound action
noun
derived verb
meaning
mbro ‘horse’
qʰu ‘back’
si ‘timber’
tɯ-ku ‘head’
pɕawtsɯ ‘money’
rɟɯɣ ‘to run’ it.
ru ‘to look’ it.
pʰɯt ‘to chop’ tr.
tɕʰɯ ‘to gore’ tr.
fsoʁ ‘earn’ tr.
mbrɤ-rɟɯɣ ‘horse race’
qha-ru ‘a look back’
sɯ-pʰɯt ‘tree felling’
—
pɕawtsɯ-fsoʁ ‘earning
money’
nɯ-mbrɤ-rɟɯɣ
nɤ-qʰa-ru
ɣɯ-sɯ-pʰɯt
nɤ-kɤ-tɕʰɯ
ɣɯ-pɕawtsɯ-fsoʁ
‘to run’
‘to look back’
‘to fell trees’
‘to hit with the head’
‘to earn money’
For use by the Author only | © 2017 Koninklijke Brill NV
587
rgyalrong language
1 2
3
4
a- mɯ- ɕɯ- tɤ/mɤ- ɣɯ- pɯetc.
trnsl
cisl
irr neg
dir
5 6
7
8
9
10
tɯ- wɣ-, ʑɣɤ- sɯ- rɤ- nɤkɯ- asɯ
prog
inv
anti
2
refl caus pass appl
11 12 13 14
15 16 17 18
a- / nɯ- ɣɯ-, incor- verb -t -a -nɯ
ɤrɤ- porated stem
noun
other
pst 1sg agr
pass auto
Complex morpho-phonological rules govern the
interaction between preijixes: the morphology is
not plainly concatenative.
entire world that share strict verb-ijinal order
and mainly preijixing morphology with Rgyalrong languages are Yenissean and Athabaskan.
6. ɲɯ-tɯ-ɤ́<wɣ>sɯ-zgroʁ
const-2-prog<inv>-attach
He is attaching you. (elicitation)
4. H i s t o r i c a l L i n g ui s t i c s
Although some combinations are not possible
(for instance the inverse preijix cannot appear
with the reflexive, which causes the verb to
become intransitive), it is quite common to ijind
examples with ijive preijixes in a row in traditional stories:
7. a-ɣɯ-lɤ-kɯ-sɯ-mtshám-a
irr-cisl-prf:upstream-2>1-caus-hear-1sg
You will come here to tell me. (Japhug; The
three sisters, 132)
With elicited examples it could be possible to
obtain eight preijixes or even more. It may be
noted that while the sufijixes have no or few
irregular forms, the preijixes present many irregularities, which suggests that they underwent
grammaticalization signiijicantly earlier.
This pattern of preijixation is unusual given
the strict verb-ijinal word order, which usually
correlates with mainly or exclusively sufijixing
morphology. The only language families in the
The importance of Rgyalrong languages for
the study of comparative Sino-Tibetan linguistics has long been noticed (see for instance
DeLancey 1981).
In phonology, Rgyalrong languages appear to
be extremely conservative: they preserve archaic
clusters only found in a few other languages or
in reconstructed languages such as Proto-LoloBurmese or Old Chinese (see Table 4).
The fact that loanwords from Tibetan into
Rgyalrong languages preserve many features lost
in all Tibetan languages of China (such as ijinal -s,
clusters of the type bsC- or brC-, etc.) is a further
conijirmation that these languages have a generally conservative phonology.
The phonological conservatism of Rgyalrong
languages makes them a useful model for the
reconstruction of other subgroups of the SinoTibetan family (see in particular the reconstruction of the Naish group, Jacques and Michaud
2011).
In derivational morphology (see for instance
Sun 2006; Jacques 2012a), Rgyalrong languages
preserve many afijixes and morphological
Table 4. Japhug Rgyalrong compared to other ST languages
Meaning
Japhug
Tibetan
louse
shame
body
to steal
zrɯɣ
-zraʁ
-skʰrɯ
mɯrkɯ
shig
Proto-Lolo-Burmese
(Bradley/Matisoff)
*s-rakL
sku
rku
*ko²
Old Chinese
(Baxter and Sagart)
*srik > shī 蝨
*srˁɨk > sè 色
*qʰ(r)ˁo > qū 軀
*kʰˁ(r)o-s > kòu 寇
For use by the Author only | © 2017 Koninklijke Brill NV
rgyalrong language
processes which may be reconstructible to
Proto-Sino-Tibetan, such as the causative sɯ-,
sɯɣ-, z- (with three regular and four irregular
allomorphs), anticausative prenasalization and
the oblique nominalization preijix sɤ-, agent or
patient nominalization kɯ-/kɤ-, action noun
nominalization in tɯ-, passive a- < *ŋa-, etc.
In inflectional morphology, Rgyalrong languages constitute an important piece of evidence in the debate regarding the antiquity of
verbal agreement in the Sino-Tibetan family.
The Rgyalrong agreement system presents many
similarities with that of Kiranti, a group of SinoTibetan languages spoken in Nepal, and most
scholars accept the fact that at least some of
the verbal morphology found in Rgyalrong and
Kiranti is cognate (Jacques 2012b).
5. Conclusion
Given the considerable importance of these
languages for both typology and comparative
linguistics, it is surprising that so few scholars engage in the description of these highly
endangered languages, whose importance for
the study of the Sino-Tibetan family as a whole
can be aptly compared to the role of Sanskrit in
Indo-European. In addition to their importance
for understanding the history of Sino-Tibetan
as a whole, they open radically new perspectives for Old Chinese historical phonology and
morphology.
A c k nowle dgment
I wish to thank Lin Youjing, Zev Handel, Waltraud Paul, and Jackson T.S. Sun for comments
on this article; I am responsible for any remaining mistakes.
Bibliography
Btsan.lha Ngag.dbang Tshul.khrims, Rgyal.rong dmangs.khrod gtam.tshogs [Popular Rgyalrong texts],
Běijīng 北京: Mi.rigs dpe.skrun khang, 2010.
DeLancey, Scott, “The Category of Direction in TibetoBurman”, Linguistics of the Tibeto-Burman Area 6/1,
1981, 83–101.
Huáng Liángróng 黄良荣 and Sūn Hóngkāi 孙红开,
Hàn-Jiāróng cídiǎn 汉嘉戎语词典 [A Chinese-
588
Rgyalrong dictionary], Běijīng 北京: Mínzú 民族
出版社, 2002.
Jacques, Guillaume, “Phonologie et morphologie du
Japhug (rGyalrong)” [Phonology and morphology
of Rgyalrong], dissertation, Université Paris VIIDenis Diderot, 2004.
Jacques, Guillaume, Jiāróngyǔ yánjiū 嘉絨語研究
[Study on the Rgyalrong language], Běijīng 北京:
Mínzú 民族出版社, 2008.
Jacques, Guillaume, “The Inverse in Japhug Rgyalrong”, Language and Linguistics 11/1, 2010, 127–157.
Jacques, Guillaume, “Argument Demotion in Japhug
Rgyalrong”, in: Gilles Authier and Katharina Haude,
eds., Ergativity, Valency and Voice, Berlin: Mouton
de Gruyter, 2012a, 199–225.
Jacques, Guillaume, “Agreement Morphology: The
Case of Rgyalrongic and Kiranti”, Language and
Linguistics 13/1, 2012b, 83–116.
Jacques, Guillaume, “From Denominal Derivation to
Incorporation”, Lingua 122/11, 2012c, 1207–1231.
Jacques, Guillaume and Chen Zhen, Une version rgyalrong de l’épopée de Gesar [A Rgyalrong version of
the epic of Gesar], Gyalrong Studies 1, Senri Ethnological Reports 93, Osaka: National Museum of
Ethnology, 2010.
Jacques, Guillaume and Alexis Michaud, “Approaching the Historical Phonology of Three Highly
Eroded Sino-Tibetan Languages: Naxi, Na and
Laze”, Diachronica 24/4, 2011, 468–498.
Lín Xiàngróng 林向荣, Jiāróngyǔ yánjiū 嘉戎語研究
[Study on the Rgyalrong language], Chéngdū 成都:
Mínzú 民族出版社, 1993.
Lin Youjing, “Tense and Aspect Morphology in the
Zhuokeji rGyalrong Verb”, Cahiers de LinguistiqueAsie Orientale 32/2, 2003, 245–286.
Lin Youjing, “Units in Zhuokeji rGyalrong Discourse:
Prosody and Grammar”, dissertation, University of
California at Santa Barbara, 2009.
Sun, Jackson T.-S., “Parallelisms in the Verb Morphology of Sidaba rGyalrong and Lavrung in rGyalrongic”, Language and Linguistics 1/1, 2000, 161–190.
Sun, Jackson T.-S., “Caodeng rGyalrong”, in: Graham
Thurgood and Randy J. LaPolla, eds., Sino-Tibetan
Languages, London: Routledge, 2003, 490–502.
Sun, Jackson T.-S., “Verb-Stem Variations in Showu
rGyalrong”, in: Ying-chin Lin et al., eds., Studies on
Sino-Tibetan Languages: Papers in Honor of Professor Hwang-Cherng Gong on His Seventieth Birthday:
Language and Linguistics, Taipei: Institute of Linguistics, Academia Sinica, 2004, 269–296.
Sun, Jackson T.-S., “Jiāróng yǔzǔ yǔyán de yīngāo:
liǎnggè gè’àn yánjiū 嘉戎语组语言的音高:两个
个案研究” [On pitch in the rGyalrongic languages:
Two case studies], Yǔyán Yánjiū 语言研究 25/1,
2005, 50–59.
Sun, Jackson T.-S., “Jiāróngyǔ dòngcí de pàishēng
xíngtài 嘉戎语动词的派生形态” [Derivational
morphology in the Rgyalrong verb], Mínzú
Yǔwén 民族语文 4, 2006, 3–14.
For use by the Author only | © 2017 Koninklijke Brill NV
589
Rhetoric
Sun, Jackson T.-S., “The Irrealis Category in rGyalrong”, Language and Linguistics 8/3, 2007, 797–819.
Sun, Jackson T.-S. and Shí Dānluó 石丹羅, “Cǎodēng
Jiāróngyǔ yǔ ‘rèntóng děngdì’ xiāngguān de yǔfǎ
xiànxiàng 草登嘉絨語與 ‘認同等第’ 相關的語法
現象” [Empathy hierarchy in Cǎodēng rGyalrong
grammar], Language and Linguistics 3/1, 2002,
79–99.
Zúñiga, Fernando, Deixis and Alignment, Inverse
Systems in Indigenous Languages of the Americas,
Amsterdam: Benjamins, 2006.
Guillaume Jacques
Rhetoric
There are many deijinitions of rhetoric. In her
book, Rhetoric in Ancient China, Xing Lu deijines
it as “the art of persuasion, the artistic use of
oral and written expressions, for the purpose of
changing thought and action at social, political
and individual levels” (1998:2). This deijinition
will act as our guide in this entry.
In Chinese, persuasion was known as shuì
說, explanation as míng 明, and argumentation
as biàn 辯. Although there is overlap between
these terms (and others), Lu argues that each
word has a particular function in conceptualizing and contextualizing persuasive discourse.
For example, shuì is associated with face-toface persuasion and míng deals with the use of
symbols in social and epistemological contexts.
Lu suggests that the term míngbiànxué 明辨學
is comparable to the Western study of rhetoric,
with míng aiming to seek truth and justice and
biàn concerning the art of persuasion. This term
also captures the contradiction inherent in the
two key concepts of Western rhetoric, viz. truth
and/or persuasion.
Distinctions between “Western” and Chinese concepts of rhetoric have been frequently
drawn (Harbsmeier 1999; Shankmann and Durrant 2002; Jullien 2004). The distinctions include
the origins of rhetoric, with the law courts of
Classical Greece providing the impetus for the
use of forensic rhetoric while, in China, rhetoric
was used to persuade rulers (Graham 1989). This
leads to a further distinction of Classical Greek
rhetoric, namely its use in Greece among equals,
while Chinese rhetoric was more “bottom-up”.
This, in turn, meant that the Chinese persuader
had to be careful. The Chinese philosopher
Guǐgǔzi 鬼谷子 (more or less a contemporary
of Aristotle) was clearly aware of this and the
influence of the audience upon rhetorical style.
He considered persuasion from below (yīn 陰)
to above (yáng 陽) to be a disturbance of the
natural order of things. Persuasion from below
to above or from an inferior to a superior was
yīn and required special effort. Persuading from
above to below was yáng and required less effort.
“Yáng . . . encourages straightforward speaking.
Yīn . . . encourages speaking in forked tongue”
(Tsao 1985:103).
A third distinction which developed as a consequence of the different Chinese and Western
rhetorical tradition is that Western rhetoric is
considered more “agonistic”, while Chinese is
considered more “irenic” (Durrant 2002:283).
But, as Durrant also points out, Chinese could be
agonistic, giving as examples Wáng Chōng 王充
(27–110 CE) and others who criticized the great
historian Sīmǎ Qiān 司馬遷 (c. 145–90 BCE).
This is a salutary reminder that there were (and
remain) many different ways of “doing” rhetoric
in both these great traditions. In this context,
while it is true that Chinese rhetoric is more
associated with writing than with speech, while
Western rhetoric is more associated with speech
than with writing, it is certainly not true to say
that this was ever at the expense of the other.
One reason why it is difijicult to be accurate
about the origins of Chinese rhetoric is that
rhetoric did not constitute a discrete discipline
in Classical China. As illustrated above, different
terms described different aspects of rhetoric.
There were others, including: yán 言, ‘speech
and the use of language’; cí 辭, ‘modes of speech,
types of discourse, eloquence, style’; and jiàn
諫, ‘giving advice, persuasion’ (Lu 1998:5). The
current Chinese term for rhetoric is xiūcí 修辭,
which literally means something like ‘reijining
words’. Its ijirst recorded use is the Confucian
classic, the Book of Changes (Yìjīng 易經).
The term—to mean rhetoric as an independent discipline—is, however, recent. It is ijirst
commonly used when Chinese scholars started
For use by the Author only | © 2017 Koninklijke Brill NV