III. Semantics-Pragmatics Analysis of Polysemous Modal Verb
3.3 Summary
國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
3.3 Summary
In this chapter, we have identified all the possible modal interpretations of each
polysemous modal verb used in utterances in Mandarin Chinese by applying the
alternative methodology of context-derived approach. Based on our present
framework suggested by Klinge (1993), these modal expressions are not encoded in
modal lexemes as the traditional descriptive-oriented approach argues but induced by
the inference process in which the modal verbs which share the same semantic field of
POTENTIALITY specify the correspondence between the unverified SITUATION
REPRESENTATION described by different concepts of situation type, i.e.
WORLD-EVENT and ANGENT-EVENT and assumptions about the referential situation, i.e. WORLD SITUATION. In the next chapter, based on the principled account for the
different modal interpretations derived by the utterances of sentences containing each
Chinese polysemous verb in contexts, we will aim to elaborate how our present
methodology can best explicate modal co-occurrence in Mandarin Chinese.
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
71
Modal Co-occurrence in Mandarin Chinese
It is well-known that the adjacency of modal verbs in Chinese is well developed
and used frequently, while not all languages have such phenomenon, such as Standard
English and Korean. Generally speaking, the order of words is a factor that dominates
the meaning of sentence. Therefore, many linguists pay much attention to the
multiple-modal construction. In what follows we will use the traditional
descriptive-oriented approach to generate the initial observation on the
multiple-modal construction of polysemous modal verbs in Mandarin Chinese and
then apply our alternative semantics-pragmatics interface approach to explicate the
regularity of distribution of Chinese modal co-occurrence.
4.1 Initial Observation within Descriptive-oriented Approach
As we have discussed in earlier section, traditional studies identify that modality
system composes of three-layer modal interpretations cross-linguistically: epistemic,
deontic and dynamic. When these three-layer modal expressions are assigned in the
sentences containing two or more polysemous modal verbs in Mandarin Chinese, their
ordering restriction is not arbitrary but accords with a severe regulation. First, we can
find the same type of modal expression can co-occur in one sentence. See the
following examples:
‧
“He should be able to finish this homework.”
In (84) and (85), YINGGAI and HUI both have identical epistemic modal expression
because they specify speaker’s judgment of probability to the propositions ‘he hearing
this news’ and ‘he finishing this homework’ in both sentences respectively. So the
same kind of modal expression can be identified in one utterance. However, it does
not imply that these two modal verbs with identical modal expression can freely
interchange to each other in one utterance. Consider the following examples:
(86) a. 他 可能 應該 知道這件事
Ta keneng yinggai zhidao zhe-jian shi.
He might should know this matter
“It is possible that he should know this matter.”
b. 他 應該 可能 知道這件事
Ta yinggai keneng zhidao zhe-jian shi.
He should might know this matter
“It is possible that he should know this matter.”
(87) a. 他 可能 會 去美國
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
Ta keneng hui qu meiguo He might will go America
“It is possible that he will go to America.”
b. *他 會 可能 去美國
Ta hui keneng qu meiguo He will might go America
“It is possible that he will go to America.”
In sentence (86), both YINGGAI and KENENG are interpreted as epistemic
interpretation co-occurring before common verbs and they can also interchange with
each other. However, (87b) is ungrammatical even though both KENENG and HUI
have epistemic interpretation. In fact, they can change each other in theory but they
are contradictory in semantics because KENENG expresses that something may
happen or may not happen, i.e. possibility while HUI denotes that something will
happen, i.e. certainty. When KENENG occurs before HUI, KENENG+HUI expresses
that it is possible that something will happen. On the contrary, HUI+KENENG
expresses the contradictory meaning, that is to say, it cannot express the certainty for
the uncertain things. So sentence (87b) is ungrammatical. On the other hand, YINGGAI
and KENENG in (86) can interchange with each other because they both interpret
possibility. Therefore, here we may temporally conclude that the order of modal verbs
in the same modal expression can interchange with each other only if their semantic
property is compatible.
‧
When different types of modal verbs occur with each other, they should obey the
ordering restriction of multiple-modal construction. The order of co-occurrence must
be epistemic > deontic > dynamic and they cannot interchange. Both (88) and (89)
can support this proposition:
(88) a. 你 應該 可以 能 做那件事
“You should be able to do that matter.”
c. *你 能 應該 可以 做那件事
“He should be willing to come.”
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
c. *他 願意 會 應該 來
Ta yuanyi hui yinggai lai He be willing to will should come “He should be willing to come.”
The above analysis reveals that not only can the different modal expression but the
same one be identified in one sentence. Meanwhile, all the evidences so far seem to
yield the implication that the order of the same type of modal verbs can interchange if
they have similar semantic property while the order of different types of modal verbs
must obey the ordering restriction. Nevertheless, the modal co-occurrence is still not
allowed in some cases even though these two polysemous modal verbs constitute
correct variants of multiple-modal expressions. Back to (21) as repeated in (90)
below:
(90) ‘You will be able to earn much money.’
a. 你 會 能 賺大錢
ni hui neng zhuan da-qian you will can earn much-money b. *你 能 會 賺大錢
ni neng hui zhuan da-qian you can able to earn much-money
(91) a. 你 會 可以 辦到的
ni hui keyi bandao-de you will able accomplish
‧
In (90b) NENG has deontic interpretation denoting personal promise while HUI has
dynamic interpretation denoting ability, in which they should be able to co-occur in
theory but result in ungrammatical pattern. The multiple-modal construction HUI +
KEYI in (91) has similar situation but they can interchange if they are both negated as
in (91b). Additionally, we may also confront another case in which the same form of
modal co-occurrence can be analyzed as different variants of multiple-modal
expressions. Consider the following examples extracted from Huang (1999):
(92) a. 他 應該 要 來
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
be I owe Yanji I necessary should give back ‘It’s me that owe Yanji so it’s necessary that I should give back.’
d. 戀愛 應該 要 衡量彼此的身分 (Huang, 1999)
lianai yinggai yao hengliang bici de shenfen To love necessary should judge each other POSS status ‘Before falling in love it is necessary that we should judge the status with each other.’
(92a) and (92b) have the same pattern of modal co-occurrence but the former
constitute an [epistemic-judgment + dynamic-volition] variant while the latter has
identical ‘deontic-obligation’ interpretation. However, with the help of context
information such as ‘Owing money’ in (92c) and ‘falling in love’ in (92d) we can see
YINGGAI+YAO in each sentence only has identical deontic expression because both
context information provide an obligation force.
To sum up, all of these problems raised so far can be generalized as follows: since
we have identified a three-layer modal construction in Mandarin Chinese, we expect
every Chinese polysemous modal verb should be able to co-occur in any position of
multiple-modal construction if they are semantically compatible. However, sentence
(90b) reveals that we cannot predict ungrammatical modal co-occurrence only
depending on the investigation of various senses of lexical meaning upon polysemous
modal verb itself. On the contrary, we should consider other constraints involved in
the restriction of modal co-occurrence, e.g. the elements surrounding polysemous
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
modal verb or the elements beyond sentence level, i.e. contextual information, as
revealed in sentences (92c) and (92d). Additionally, we cannot also generate all
variants of multiple-modal expressions of identical modal co-occurrence under the
framework of traditional descriptive-oriented approach as demonstrated in sentences
(92a) and (92b).
4.2 Modal co-occurrence within Semantics-Pragmatics Interface Approach
4.2.1 Hypothesis
Since the discussion so far suggests that modal interpretation is determined by the
inference process in which each polysemous modal verb possesses different core
meaning in their lexical semantics and thereby assigns different constellations of
correspondence and non-correspondence between the SITUATION REPRESENTATION
described by different concepts of PROPOSITIONAL CONTENT and the referential
WORLD SITUATION, then we will reasonably speculate whether these different core meanings encoded in each polysemous modal verb become a crucial point to motivate
their ordering restriction in multiple-modal construction. In the previous sections, as
Klinge (1993) suggests, we assume that the five Chinese polysemous modal verbs in
our study also share the same semantic field of POTENTIALITY and thereby provide
different degrees of procedural information about the POTENTIAL correspondence
between SITUATION REPRESENTATION and WORLD SITUATION. In fact, three different
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
degrees of procedural information have been identified in the last chapter, in which
each degree encompasses its own representative polysemous modal verbs and they are
depicted as follows:
I. YINGGAI generates the only outcome of correspondence between SITUATION
REPRESENTATION and WORLD SITUATION by eliminating the possibility of non-correspondence between SITUATION REPRESENTATION and WORLD
SITUATION.
II. YAO and HUI directly specify the correspondence between SITUATION
REPRESENTATION and WORLD SITUATION.
III. KEYI and NENG hold the co-existence of correspondence/non-correspondence
between SITUATION REPRESENTATION and WORLD SITUATION.
From the initial observation of these three degrees of procedural information, the first
degree denoted by YINGGAI has the highest assurance of correspondence between
SITUATION REPRESENTATION and WORLD SITUATION than the other two degrees because the speaker first rules out the possibility of non-correspondence when s/he
has full evidence in which it is impossible that SITUATION REPRESENTATION
eventually is not a true description of WORLD SITUATION so that the current
SITUATION REPRESENTATION nearly corresponds hundred percents with the referential
WORLD SITUATION. Compared with YINGGAI, though YAO and HUI also directly point
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
out the outcome of correspondence between SITUATION REPRESENTATION and WORLD
SITUATION, it still has the chance that SITUATION REPRESENTATION is not a true description of WORLD SITUATION. Nevertheless, YAO and HUI still have higher
assurance than KEYI and NENG because the later two modals only hold the relation
between SITUATION REPRESENTATION and WORLD SITUATION on the neutral position
without assuring the correspondence or non-correspondence between SITUATION
REPRESENTATION and WORLD SITUATION.
Additionally, although the utterances of sentences containing three modal verbs
occur rarely in corpus data, these three levels of procedural information mentioned
above themselves formulate a systematic pattern in the examples below:
(93) 地方首長 應該 要 能 傾聽市民的心聲 (GigaWord Corpus)
defang shouzhang yinggai yao neng qingting shimin de xinsheng local leader must should can listen citizen of heartfelt wish ‘Local leaders should be able to listen to citizens’ heartfelt wishes.’
(94) 國軍 應該 要 隨時 可以 進入戰備狀態 (GigaWord Corpus)
guojun yinggai yao suishi keyi jinru zhanbei zhuangtai military must should anytime can into combat readiness situation ‘Military should be able to get into combat readiness in anytime.’
(95) 小孩子 應該 會 能 了解父母的用心 (GigaWord Corpus)
xiaohaizi yinggai hui neng liaojie fumu de yongxin child probably will can realize parents of concern
‧
‘In terms of the promotion of domestic democracy, it is obligatory for political party to be able to set an ideal model.’
(97) 稅制結構 應該 要 能 與經濟同步正向成長 (GigaWord Corpus)
shuizhi jiegou yinggai yao neng yu jingji tongbu tax structure should will can with economic
zhengxiang chengzhang synchronous growing
‘Tax structure should be able to have synchronous growing with economic.’
(98) 對方 應該 會 可以 接受和解的條件 (GigaWord Corpus)
duifang yinggai hui keyi jieshou hejie de tiaojian other side probably will can accept reconciliation of condition
‘The other person involved probably will be able to accept the condition of reconciliation.’
Among all these examples, any other ordering variants in each sample will result
in semantically unacceptable sentences. Hence, we can see YINGGAI always occupies
the top position and then YAO & HUI and KEYI & NENG follow in that order. Although
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
this has the same finding with descriptive-oriented approach, the advantage of our
alternative analysis is that we ensure all the five polysemous modal verbs in Mandarin
Chinese are regarded as an unambiguous and consistent element with their intrinsic
lexical semantics because the three categories of modality - epistemic, deontic and
dynamic - where descriptive-oriented approach argues that modals encode these three
modal interpretations are actually derived from our assumptions described by
different SITUATION REPRESENTATIONS, i.e. WORLD-EVENT (POSSIBILITY,
PREDICTION and GENERIC) vs. AGENT-EVENT (OBLIGATION, PERMISSION, VOLITION,
ABILITY) about referential WORLD SITUATION. Thus, we assume that the STRENGTH of prediction to the POTENTIAL correspondence between SITUATION REPRESENTATION
and WORLD SITUATION represented by modal verbs plays an important role in the
ordering restriction in multiple-modal construction. The more STRENGTH about the
outcome of correspondence/non-correspondence the modal verb possesses, the higher
position in multiple-modal construction it occupies. In what follows, we will see how
the examples below found in the corpus data accord with our prediction.
4.2.2 Data Distribution
4.2.2.1 The Co-occurrence of Different Category
I. The Co-occurrence with YINGGAI
First, according to our hypothesis in the last section, YINGGAI can adjoin all the
‧
other four polysemous modal verbs. See the following examples:
YINGGAI + NENG
(99) 鳥林投居民要求合理，政府 應該 都 能 接受 (GigaWord Corpus)
Niaolintou jumin yaoqiu heli, zhengfu yinggai Niaolntou resident requirement reasonable government probably dou neng jieshou
all can accept
‘The requests of the residents in Niaolintou are reasonable so the government should probably be able to accept it all.
(100) 大學法部分條文修正案 應該 能 在上午順利通過 (GigaWord Corpus)
Daxuefa bufen tiaowen xiuzhengan yinggai neng zai University Law partial clause amendment propably can in shangwu shunli tongguo
morning successfully pass
‘The partial amendment of University Law probably will successfully be able to pass in the morning.’
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
In the above examples, NENG first has the SITUATION REPRESENTATIONS of
WORLD-EVENT as its scope to denote the correspondence or non-correspondence to be verified by referential WORLD SITUATION. Next, YINGGAI occupies before NENG to
reject the possibility of non-correspondence between SITUATION REPRESENTATION
and referential WORLD SITUATION in ensuring this unverified WORLD-EVENT is the
true description of WORLD SITUATION. As we have discussed previously, NENG only
holds the relation between SITUATION REPRESENTATION and WORLD SITUATION on
the neutral position. In other words, it has the lowest STRENGTH of prediction to the
potential correspondence between SITUATION REPRESENTATION and WORLD
SITUATION so that it always occurs after YINGGAI which has the strongest prediction to the potential correspondence between SITUATION REPRESENTATION and WORLD
SITUATION. For instance, the proposition ‘the government accepting all requirements of Niaolintou residents’ in (99) describes the SITUATION REPRESENTATION of future
unverified WORLD-EVENT. NENG here first denotes both possibilities of the
government accepting or not accepting the requests of Niaolintou residents. Then
YINGGAI rules out the possibility of which the government won’t accept the residents’
requests so that it specifies the proposition ‘the government accepts all the requests of
Niaolintou residents’ eventually truly describes the referential WORLD SITUATION. The
rest two utterances of (100) and (101) also accord with this mechanism. On the other
‧
hand, the inferential process of utterances encompassing YINGGAI + KEYI also has the
same mechanism with YINGGAI + NENG. See the following examples:
YINGGAI + KEYI
‘Buying the excessively-produced vegetable of each place probably can gradually increase the price of vegetable.’
(103) 相信明年經濟成長率 應該 還是 可以 保持在百分之六左右 be kept at around six percents.’
(104) 建立貨物直航構想 應該 可以 避免兩岸主權爭議 (GigaWord Corpus)
jianli huowu zhihang gouxiang yinggai keyi bimian liangan establish goods direct delivery idea probably can avoid both side zhuquan zhengyi
sovereignty dispute
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
‘The idea of establishing direct delivery of goods between the two sides (of Taiwan and Mainland China) should be able to avoid sovereignty dispute.’
Although KEYI represents the speaker’s expectation in that the SITUATION
REPRESENTATION corresponds with WORLD SITUATION or not, it still only specifies the POTENTIAL of correspondence between SITUATION REPRESENTATION and WORLD
SITUATION. That is to say, KEYI still has lower STRENGTH of prediction than YINGGAI. Thus, KEYI, like NENG, always occurs after YINGGAI in the utterance. Hence, taking
the utterance of (102) for instance, KEYI first has the SITUATION REPRESENTATIONS
of WORLD-EVENT described by ‘buying the excessively-produced vegetable of each
place gradually increasing the price of vegetable’ as their scope to denote the
correspondence or non-correspondence to be verified by referential WORLD
SITUATION and then YINGGAI occupies the higher position and thereby immediately rejects the possibility of non-correspondence to ensure this SITUATION
REPRESENTATION is the true description of WORLD SITUATION. The same mechanism can also apply in the utterances of (103) and (104). Next, we will see the utterances
encompassing YINGGAI + HUI and YINGGAI + YAO.
YINGGAI + HUI
(105) 未來政府在調整薪資時 應該 會 考慮到企業界的生產成本
(GigaWord Corpus) weilai zhengfu zai tiaozheng xinzi shi yinggai hui kaoludao qiyeje
‧
future government on adjust salary when probably will think about industry de shengchan chengben
of production cost
‘In the future, it is very likely that the government will take the production costs of industry into consideration when adjusting salary scales.’
(106) 經濟部認為下半年經濟情勢 應該 會 較樂觀 (GigaWord Corpus)
Jingjibu renwei xiabannian jingji-qingshi Ministry of Economic Affairs think second half year economic situation yinggai hui jiao leguan
probably will more positive
‘Ministry of Economic Affairs thinks that the economic situation probably will be more positive in the second half of the year.’
(107) 美國方面只要達到目的 應該 會 立刻停止軍事攻擊行動
(GigaWord Corpus) Meiguo fangmian zhiyao dadao mudi yinggai hui like
America side as long as achieve destination probably will immediately tingzhi junshi gongji xingdong
stop military attack action
‘In the perspective of America (government), as long as they achieve their goal, military attack is very likely to be immediately stopped.’
As we have discussed in the last chapter, the speaker uses HUI to directly predict the
highly possible correspondence between SITUATION REPRESENTATION and WORLD
SITUATION. Nevertheless, it still leaves the chance that the SITUATION
‧ 國
立 政 治 大 學
‧
N a tio na
l C h engchi U ni ve rs it y
REPRESENTATION of unverified WORLD-EVENT eventually may not truly describe the actual statement of affairs. Thus, even though HUI has higher assurance than NENG
and KEYI, HUI still has lower STRENGTH of prediction than YINGGAI so that HUI
generally follows after YINGGAI and precedes NENG and KEYI in one utterance. For
example, in Sentence (105), the SITUATION REPRESENTATIONS of WORLD-EVENT
described by ‘the government thinking about the production cost of industry when
adjusting salary’ is first specified by HUI to denote the highly possible
correspondence between this unverified WORLD-EVENT and referential WORLD
SITUATION. Then YINGGAI occupies the higher position to further strengthen the only
SITUATION. Then YINGGAI occupies the higher position to further strengthen the only