Chapter 1 Introduction
1.4 Organization of the Thesis
The thesis is organized as follows. Chapter one is a general introduction of the thesis. Chapter two reviews previous studi es related to emotion verbs. Chapter three briefly describes the database and the framework adopted in the study . Chapter four presents the findings motivating this research. Based on the findings, chapter five proposes a frame-based analysis of Mandarin emotion verbs. Finally, chapter six concludes the study and suggests future research topics. Finally, the conclusion of this study and suggestions of future research topics are given in chapter seven.
Chapter 2 Literature Review
This chapter reviews previous studies on emotion verbs as a foundation of the research. A variety of studies have looked at emotion verbs. The verbs discussed, however, are not always referred to as ‘emotion verbs’ in the literature. Therefore, the terms used in pervious works will be briefly listed in section 2.1 in advance. Researches on emotion verbs from different perspectives with diverse approaches will be introduced in the following sections. Section 2.2 introduces the studies on the typology or categorization of emotion verbs in different approaches (Levin 1993, Tsai et al 1996, Chang et al 2000, Liu 2002, and Lai 2004, Berkeley FrameNet Project). Section 2.3 summarizes the chapter and points out the direction of this study.
2.1 Terms Used in the Previous Studies
‘Emotion verbs’ are frequently viewed as a sub-class or sub-type of ‘mental verbs’, or ‘psych verbs’ by some researchers. For example, Croft mentioned that “the class of mental verbs (also known as ‘psych verbs’) includes verbs of perception, cognition and emotion” (1993: 55). Additionally, FrameNet defined the Mental_activity frame : ‘In this frame, a Sentient_entity has some activity of the mind op erating on a particular Content or about a particular Topic. The particular activity may be perceptual, emotional, or more generally cognitive. ’ As to Mandarin, Mei et al (1993) classified mental predicates into three sub-classes: cognition verbs, affectio n or emotion verbs, and will verbs. The related terms used in previous studies are briefly listed: ‘mental verbs/predicate ’ (Vendler 1972; Huang 1982; Sweetser 1990; Croft 1991, 1993; Wierzbicka 1996;
Goddard and Wierzbicka 2002; Su 2002, 2004 ), ‘psychological or psych
Chen 1994, Hana 1996, Yang 2000, Lai 2004). The terms ‘emotion verbs/words/domain’ used in this study are also used in Chang et al (2000), Liu (2002), Les Bruce (2003), Lin (2006), Hsiao (2006, 2007) , and Berkeley FrameNet Project. In addition to studies explicitly on the whole class of verbs in question, there are also studies focusing on only one or two emotion verbs, for example, Tsai et al (1996).
2.2 Typological Studies of Emotion Verbs
Recently, some researchers tried to investigate the typology or categorizations of emotion verbs in different approaches. To introduce these approached in a clear way, the approaches will be organized in divided part s.
2.2.1 Alternation-based Approach
Levin (1993) is a pioneering work on English verb classes and diathesis alternations. Diathesis alternations, by definition, refer to alternations in the expression of arguments, sometimes accompanied by changes of mea ning (Levin 1993:2). It is believed that the grammatical construction is encoded in the lexicon, i.e. the meaning of a verb. Therefore, Levin divided English Psych-verbs, verbs of psychological state , into four sub-classes in terms of transitivity and their expressions of arguments.
Amuse verbs: transitive verbs whose object is the experiencer and whose subject is the cause of the change.
Admire verb: transitive verbs with experiencer -subject.
Marvel verb: intransitive, experiencer as subject, express the stimulus/object of emotion in a PP headed by one of a variety of prepositions.
Appeal verb: the least in four subclasses; intransitive, taking the stimulus as subject and expressing the experience in a PP headed by one of a variety of prepositions.
In addition to verbs classes, Levin also presented detailed alternation patterns to test the properties of each class of verbs. Take the Amuse type verbs for example. (Here only part of the used alternations in her study is listed.)
(5) *Causative Alternation (most verbs):
a. The clown amused the children.
b. * The children amused (at the clown).
(6) Middle Alternation:
a. The clown amused the little children.
b. Little children amuse easily.
(7) PRO-Arb Object Alternation1:
a. That joke never fails to amu se little children.
b. That joke never fails to amuse.
This alternation-based framework provided an important criterion to distinguish the properties of verbs. However, some of these verbs included in this class should be further distinguished, and Levin herself also agreed with the argumentation (Levin 1993:191). She further stated Grimshaw ’s (1990) argumentation:
For instance, Grimshaw (1990) argues that some of these verbs, such as amuse, allow the subject/stimulus argument to receive an agentive inter pretation, while others, such as concern, do not; this distinction could be the basis for further subdivision of these verbs. (Levin 1993:191)
1 Levin (1993:38) explained the PRO -arb object alternation:
“In this alternation, the unexpressed object in the intransitive variant receives what has been called an
Moreover, Liu (1996) mentioned that “a purely alternation-based approach may not be adequate for categorizing and describing Mandarin verbs because of the typological and parametric variations among languages. ”
2.2.2 Frame-based Approach
In FrameNet (http://framenet.icsi.berkeley.edu/) , an on-line English lexical resource based on frame semantics and supported by corpus evidence, shows the hierarchical relationship ( i.e. ‘generation’ in FrameNet) between the parent frame Emotions with its children frames. The Emotion frame has 12 children frames with
“Using2” relationship: Predicament frame, Judgments frame, Feeling frame, Forgiveness frame, Contrition frame, Desiring frame, Subject_stimulus frame, Expereience_subject frame, Experience_object frame, Emotion_heat frame, Emotion_directed frame, and Emotion_active frame. Figure 1 illustrates the relationship schema of the Emotion frame with the twelve children frames; Table 1 presents some lexical units in each child frames.
Figure 1. The Twelve Frames and Their Relations under Emotions Frame in Framenet
2‘Using’ is one of the ‘frame-to-frame relations’ given in FrameNet. It is a presupposition relation. ‘A using B’ means ‘A presuppos es B as background.’ When a frame uses part of background information (E.g. some core semantic frame elements ) of another frame, the two frames are having a using relationship.
Table 1: 12 Children Frames of the Emotions Frame and Lexical Units in Each Child Frame
No. Frame Name Lexical Units
1 Predicament misfortune, problem, trouble, fix, mess, jam…
2 Judgment admire, disapprove, blame, mock, scorn…
3 Feeling feel, feelings, emotion, experience … 4 Forgiveness forgive, forgiveness, pardon, excuse…
5 Contrition contrition, guilt, guilty, repent, remorse, sorry…
6 Desiring desire, ambition, eager, hope, long, lust, want…
7 Subject_stimulus amazing, amusing, annoying, boring, delight, disgusting, dull, enjoyable, sad, scary, t iring, worrying…
8 Experiencer_subj adore, afraid, dislike, empathy, envy, fear, hate, like, pity, regret, nervous, love…
9 Experiencer_obj amaze, anger, annoy, bore, cheer, please, sicken, upset, trouble, surprise, shock, shame…
10 Emotion_heat Boil, burn, simmer… (e.g. I’m boiling with anger)
11 Emotion_directed Agony, anxious, blue, concern, discouraged, distress, furious, grief, mad, sad, upset…
12 Emotion_active Worry, obsess, agonize, fret…
FrameNet proposed a significant analysis to present the range of semantic and syntactic combinatory possibilities of each word in each of its senses . Moreover, they organized clear frame -to-frame relationships. However, FrameNet did not clarify a explicit hierarchy of the frames. In addition, the analysis based on English lexicon may not be adequate or equivalent to Mandarin Chinese.
2.2.3 Corpus-based Approach
With the development of corpus and data, a number of studies attempted to investigate the properties of Mandarin emotion verbs and categorizing t he verbs from a corpus-based approach.
2.2.3.1 Tsai et al (1996)
In response to the need of fine-tuning verbal semantics, Tsai et al (1996) observed and differentiated a near-synonyms pair to manifest the argument that syntactic contrasts can be predicated from lexical semantics and underlines the importance of interactions between semantics (lexical semantic properties) and syntax (syntactic behaviors).
The near-synonyms pair of emotional state verbs , gaoxing 高興 and kuaile 快樂, are semantically similar. Observing the corpus distribution of these two verbs , however, Tsai et al found gaoxing and kuaile are syntactically distinct in many ways and then proposed that the verbs can be differentiated according to their syntactic distribution:
Table 2: The Syntactic Distribution of Gaoxing and Kuaile Verb
Function
GAOXING 高興 (280)
‘be glad’
KUAILE 快樂 (365)
‘be happy’
Predicate 224 (80%) 119 (32%)
Complement 8 (3%) 17 (5%)
Adverbial 47 (17%) 30 (8%)
Adjective 116 (31.7%)
Nominalized 1 (0.3%) 83 (23%)
Gaoxing is typically used as a predicate ( gaoxing takes sentential objects while kuaile cannot.), but barely or never used as adjective (0%) or a noun or nominal modifier (0.3%); contrarily, kuaile shows a much higher frequency of nominalization (23%) and adjectival using (32%). As examples below:
(8) Transitive predicate:
他們都很高興/*快樂創刊號終於出來了。
tamen tu hen gaoxing/*kuaile chuangkanhao zhongyu chu lai le.
they all very glad/*happy initial issue finally publish PERF/CRS
‘They are all very glad/*happy that the initial issue is published finally. ’
(9) Adjective:
如何做個*高興/快樂的上班族。
ruhe zuo ge *gaoxing/kuaile de shangbanzu how do CL *glad/happy DE worker
‘How to be a *glad/happy worker ’
(10) Nominalized:
人有追求*高興/快樂，逃避痛苦的本能。
ren you zhuiqiu *gaoxing/kuaile, taobi tongku de benneng people have seek *glad/happy, evade pain DE instinct
‘People have instinct to seek *gladness/happiness and evade pain ’
Tsai et al proposed that these syntactic contrasts are systematically accounted for with lexical features <change of state> and <control>.
(11) Change-of-state:
a. 客人高興/*快樂了會賞你錢。
keren gaoxing/*kuaile le, hui shang ni qian customers glad/*happy PERF will give you money
‘Customers would give you money if they have become glad/*happy. ’ b. 他們談得正高興/*快樂。
tamen tande zheng gaoxing/*kuaile le they chat PROG glad/*happy
‘They are chatting gladly/*happily. ’ c. 從此不再快樂/*高興。
congci buzai kuaile/*gaoxing frome-now-on no-longer happy/*glad
‘No longer being happy/*glad from now on. ’
(12) Control:
a. 生活快樂/*高興最重要。
shenghuo kuaile/*gaoxing zui zhongyao life happy/*glad most important
‘Life be happy/*glad is what really matter. ’ b. 別高興/*快樂!
bei gaoxing/* kuaile don’t glad/*happy
‘Don’t be glad/*happy!’
c. 你應該高興/*快樂才對。
ni yingai gaoxing/* kuaile cai dui you should glad/*happy only:then right
‘You should be glad/*happy and only then it is right. ’ d. 為他高興/*快樂。
wei ta gaoxing/* kuaile for he glad/*happy
‘Being glad/*happy for him.’
Hence, they divided emotional stative verbs semantically into two groups:
homogeneous state verbs which are characterized as <-change of state, -control> and result state verbs characterized as <+change of state, +control> .
Tsai et al (1996) offered convincing evidence that syntactic behaviors can be predicted based on lexical semantics, but it seems that the dichotomous distinction cannot account for the whole semantic field of emotional predicates by only looking at two verbs. Moreover, the semantic features proposed above are general factors which pertain to eventivity. They are too broad to explain the unique b ehavior of the particular class of emotional verbs (Liu, 2002:79).
2.2.3.2 Chang et al (2000)
Chang et al (2000) followed Tsai et al’s dichotomous study mentioned above, and then re-examined a succinct difference over a broader range — the whole semantic field of verbs of emotion. In the study, they first categorized emotional verbs into seven subfields, i.e. Happiness, Depression, Sadness, Regret, Anger, Fear, and Worry; then they revised Grandy’s (1992) definition of a semantic field to propose that there are two covering terms that form a Contrast Pair (Type A and B) that define each semantic field.
And then they thoroughly examined the seven contrast pairs of subfields of emotional verbs and predicted that the other verbs being members of the field will behave like
Chang et al examined the contrasts between the two types based on five distributional syntactic criteria. They found that type A and type B verbs can be characterized as below:
Table 3: The Contrastive Distribution of Type A and Type B Verbs
Criteria Verb types Type A verbs (e.g. gaoxing 高興) Type B verbs (e.g. kuaile 快樂) Grammatical
function
Mostly as predicates Mostly as nominal uses
Co-occurrence restriction
Strict selectional restrictions on the head when the function as adjuncts
Looser selectional restrictions on the head when the function as adjuncts
The appropriateness in the imperative and evaluative constructions
Most verbs can appear in either type of construction
Few verbs cannot appear in either type of construction
Verbal aspect Predominantly represent inchoative state
Rarely represent inchoative state
Transitivity Can take causes or goals as their direct objects
Seldom take causes or goals as their direct objects
From the above contrast, Chang et al generalized that type A verbs are preferred for indicate transition while type B verbs are preferred for homogeneity. What kind of reason drives the grammatical differences? They conclude that the grammatical contrasts derive from their morphological structures. A VV compound is double-headed and does not elaborate. Therefore, in VV compounding, the concept of an event is not so
clear. It is common morpho-lexical strategy in Mandarin to link two antonyms or synonyms to form the conc ept of ‘kind’ or ‘property’, and therefore it is natural for VV compounds to be chosen as the representation of homogeneity.
Hence, verbs of emotion are morphologically separated by Chang et al into Non-VV compound and VV compound. Type A verbs are Non-VV compound, while type B verbs are VV compound.
Type A: gao xing 高興(non-VV), nan guo 難過(non-VV), hou hui 後悔(non-VV), shang xin 傷心(non-VV), sheng qi 生氣(non-VV), hai pa 害怕(non-VV), dan xin 擔心 (non-VV)
Type B: kuai le 快樂(VV), tong ku 痛苦(VV), yi han 遺憾(AN or VO), bei shang 悲傷 (VV), fen nu 憤怒(VV), kong ju 恐懼(VV), fan nao 煩惱(VV)
Chang et al (2000) extended range of the study to seven subfields of emotion verbs, and further proposed a morphological -makeup explanation, but these seven types of verbs were still not enough or adequate to present the heterogeneous properties of Mandarin emotion verbs (see Chapter 4). Besides, there is a flaw in their analysis: the verb yi han 遺憾 judged as a AN or VO compound is located in the type B, i.e. VV compound verbs. Liu (2002:79) pointed out “the morphological account is at best an observation associated with and resulted from a deeper semantic relation. It fails to explain the semantic driving force for the paradigmatic variation.”
2.2.3.3 Liu (2002)
Different from Tsai et al (1996) and Chang et al (2000), Liu (2002) looked at near-synonyms pairs of emotion activity verbs, XIANMU 羨慕‘envy’ and JIDU 嫉妒‘be jealous of’, and found that CAUSE is a crucial parameter in the event structure of
conceptualizing emotional activities. The event can be perceived as a self -motivated action with the Expeirencer coded as an actor projecting an internal state towards a Target or Stimulus. (e.g. 我羨慕他。wo xianmu ta ‘I envy him.’) It can also be perceived as a caused state with the Experiencer being coded as a Causee undergoing a change affected by a Causer. (e.g. 他讓我羨慕。ta rang wo xianmu ‘He makes me envious.’)
Figure 2: Two Conceptual Frames used in describing emotional activities.
a) SELF-MOTIVATED ACTION b) CAUSED STATE
Liu mentioned that these two conceptual frameworks may be conflated into a three-stage causal chain: Causer arouses an emotional state in the Experiencer toward a Target.
Figure 3: The Causal Chain of Emotional Activities.
Based on the causal frameworks, Liu highlighted the crucial role of CAUSE as a semantically essential variable . Liu found that the CAUSE may be encoded as an NP-object, an NP-subject, a preposed NP-Goal, a clausal object, or a preposed clausal-subject and found the CAUSE form is often overtly marked in a sentence (1 3).
However, there are also cases where the role CAUSE is not overtly mentioned (1 4).
(13) a.我 非常 嫉妒[NP 他的成功]。
wo fenchang jidu ta de chenggong I unusually JIDU he De success
‘I am very jealous of his success. ’ b.對[NP他的成功] 我 非常 嫉妒。
dui ta de chenggong wo fenchang jidu to he De success I unusually JIDU
‘Of his success, I am very jealous. ’
c. 我 羨慕 [CP他會說不同的語言]。
wo xianmu ta hui shui butong de yuyan
I XIANMU he can speak not same De Language
‘I envy that he can speak different languages.’
(14) a. 你 只會 嫉妒。
ni zhi hui jidu you only can JIDU
‘You are only capable of bei ng jealous.’
b. 被 貪婪 嫉妒所 緊縛
bei tanlan jidu suo jinfu
‘Tightly bound by greed and jealousy. ’
Examining distribution of CAUSE between the verbs, XIANMU 羨慕‘envy’ and JIDU 嫉妒‘be jealous of’, Liu found that an obvious skewing in the frequency of overtly marked CAUSE.
Table 4: Distribution of CAUSE: XIANMU 羨慕 vs. JIDU 嫉妒 XIANMU 羨慕 (112)
‘envy’
JIDU 嫉妒 (38)
‘be jealous’
Mention (NP or S) 100% (112) 45% (17)
No Mention 0% (0) 55% (21)
Hence, Liu proposed that there are two types of Cause, External vs. Internal CAUSE, and classified emotional activity verbs into two groups based on the distinctive role CAUSE. Externally caused verbs are more controllable and more justifiable;
internally caused verbs are less control and less justifiable.
Liu also showed that other sets of emotional verbs also support her finding, such as verbs of anger (SHEGQI 生氣 vs. FENNU 憤怒), verbs of fear (HAIPA 害怕 vs.
KONGJU 恐懼), verbs of sadness (SHANGXIN 傷心 vs. BEISHANF 悲傷), and verbs of depression (NANGUO 難過 vs. TONGKU 痛苦). Finally, Liu utilized MARVS (the Module-Attribute Representation of Verb Semantics) (Huang et al. 2000, Chang et al.
2000) to specify the event module and role module of externally and externally caused emotional activity verbs.
Liu (2002) clearly presented the interaction between syntactic behavior and semantic properties. The three-stage causal chain proposed in her study provided a simple and useful cognitive schema of conceptualization of emotion events. Therefore,
following Liu’s study, we will propose an extended and elaborated causal schema based on the causal relation established in the study (see section 5.2).
2.2.3.4 Lai (2004)
In this paper, Lai attentively observed the syntactic realizations or constructions of three set of verbs, trying to explore the subtle distinctive features and meaning in Chinese emotion (psychological) predicates. In order to single out representative verbs for convincing investigation , Lai selected verbs with four criteria: the argument structure, the meaning of verbs, the frequency of occurrence, and the morphological structures. Three most frequently used monosyllabic psych verbs and their most-used bi-syllabic counterparts were finally selected for further examination . The selected emotion verbs are愛 ai ‘love’, 喜歡 xihuan ‘like’, 氣 qi ‘angry or get angry’, 生氣 shengqi ‘get angry’, 怕pa ‘fear or be afraid’ and 害怕haipa ‘be afraid’. Although did not be selected, the verb 嚇 xia ‘frighten’ is put into the following discussion in the study.
Table 5: Interaction between Criteria and Psych Predicates
Predicate A-structure Meaning Frequency Morpho-structure3 (Number of variants) a qi 氣 (Th, Exp; Exp, Th) negative 617 3 (V#, #V, mor-#) b fan 煩 (Th, Exp; Exp, Th) negative 54 3 (V#, #V, mor-#) c pa 怕 (Th, Exp; Exp, Th) negative 928 2 (V#, mor-#)
d xia 嚇 (Th, Exp) negative 147 2 (V#, #V)
3 According to Lai (2004:50), “Morph -structure” is “morphological structure,” referring to components of a compound. The symbol “#” stands for the psych predicates listed in the first column. “V” stands for a
e nu 怒 (Exp) negative 68 1 (V#)
f hen 恨 (Exp, Th) negative 140 2 (V#, mor-#)
g ai 愛 (Exp, Th) positive 1925 3 (V#, #V, mor-#)
h teng 疼 (Exp, Th) positive 85 2 (#V, mor-#)
Considering a sentence like (15), Lai found that there are two readings deduced from the same clause, as (16) illustr ates.
(15) 他氣他媽媽 ta qi ta mama he anger he mother
(16) a. ‘He was angry with his mother. ’(Exp, Th) b. ‘He made his mother angry.’ (Th, Exp)
The meaning coded in (15) can be either ‘he was angry with his mother ’ or ‘he made his
The meaning coded in (15) can be either ‘he was angry with his mother ’ or ‘he made his