3.2 The Results of the Oral Data in 2010
3.2.1.3 Increasing use of possessive pronouns
Following the discussions of subject pronouns and object pronouns, the Englishized possessive pronouns are explicated in terms of the referentiality, as shown in 3.2.1.3.1 and 3.2.1.3.2.
3.2.1.3.1 Referential possessive pronouns
For the distribution of referential possessive pronouns, just as the object position, third person singular is with the highest frequency (1.27‰), and followed by second person singular (0.60‰), first person singular (0.50‰), first person plural (0.29‰), third person plural (0.16‰) and second person plural (0.13‰).
As for the use of inanimate third person pronouns, the frequency of inanimate ta and tamen in the possessive position (0.53‰) is higher than the traditional object position (0.44‰), see (119). Therefore, again the Englishization of the neuter ta and tamen is supported.
(119) >蘆洲 目前 來 看 的 話 它 的 生活 機能 是 沒有 那麼 Luzhou muqian lai kan de hua ta de shenghuo jineng shi meiyou name 好
hau
‗So far Luzhou its vital function is not that good‘
It is found that in the oral data no matter which position, for referential use the plural forms of pronouns are less frequent than their singular counterparts. The reasons are as provided in 3.2.1.1.1: the one-to-one interactive nature, the replacement of plural form by singular form and the sense of respect and indirectness. Besides, comparing the results of the object and possessive position, the ranking of referential
90
pronouns in the oral data is the same in these two positions. Similar tendency can be obtained from the results of non-referential pronouns, as shown in 3.2.1.3.2.
3.2.1.3.2 Non-referential possessive pronouns
The most frequent non-referential pronoun in the oral data is second person singular (0.59‰) and followed by third person singular (0.18‰), first person plural (0.05‰), first person singular (0.03‰), and third person plural (0.02‰).
As in other two positions, second person singular is the most frequently used pronoun for non-referential use, as in (120) and (121). Given the above discussion, regardless of the language modes and the positions, for non-referential use the frequency of second person singular is the highest.
(120) 你 要 打開 你 的 耳朵 Ni yao dakai ni de erduo You want open you DE ear ‗You have to open up your ears.‘
(121) 你 要 聽 你 的 對手 在 講 什麼 Ni yao ting ni de duishou zai jiang sheme You want hear you DE rival at speak what
‗You have to listen to what your rival said.‘
The second highest pronoun for non-referential use is third person singular. The high frequency of third person singular can be associated with the nature of talk shows: the explication of events and arguments with indefinite referents, as in (122), where ta is used in a hypothetical context to refer to indefinite person.
(122) >有的 連 外面 有 孩子 都 沒 說 元配 在 他 的 那個 Youde lian waimian you haizi dou mei shuo yuanpei zai ta de nage Someone even outside have child all NEG say wife in he DE that 有沒有 靈堂 上 才 發現
youmeiyou lingtang shang cai faxian whether funeral up only find
‗Some even didn‘t say that he has an illegimate child, the wife just found it in his funeral‘
Overall, the above discussion shows that non-referential use of pronouns is not only significantly more frequent but also more various in the oral data than in the written data. Generally, just as Englishized subject and object pronouns, regardless of referentiality, the frequency of Englishized pronouns in the possessive position is significantly higher in the oral data (3.82‰) than in the 2010 written data (0.39‰).
To sum up, with regard to the distribution of non-referential pronuns in the oral data, its distribution is the same in all positions; the top three pronouns are: second person singular, third person singular and first person plural. The frequent use of second person singular and first person plural corresponds to the observation of the literature; these two types of pronouns can be used to engage the audience. The high frequency of non-referential third person singular may be attributed to the significantly more common use of third person singular in general, since the total frequency of third person singular (8.57‰) is only lower than second person singular (9.53‰).
As for the distribution of each referential pronoun, in the oral data regardless of positions all singular forms are more frequent than the plural forms due to the one-to-one interactive nature and the sense of respect and indirectness. For the ranking of the top three pronouns, in the subject position it is first person singular, third person singular and second person singular. However, in the object and possessive position, it is third person singular and then second person singular and first person singular. The reason why first person singular is significantly higher in the subject position is due to the fact that in the talk shows there are great opportunities of delivering personal opinions and attitudes, and self-mention ‗I‘ occurs in the subject position to express the subjective viewpoints.
To observe the distribution of each pronoun in speaking clearly, the results are
92
Table 11.5. The number and normalized frequency (per 1000 morphemes) of Englishized pronouns in 2010 oral data
Database Referentaility Person
Oral
Referential Non-referential Total
1st singular 416(6.76) 19(0.31) 435 (7.07)
1st plural 187(3.04) 27(0.44) 214(3.48)
2nd singular 327(5.32) 259(4.21) 586(9.53)
2nd plural 29(0.47) 5(0.08) 34(0.55)
3rd singular 477(7.75) 45(0.73) 522(8.49)
3rd plural 63(1.02) 10(0.16) 73(1.19)
Total 1499(24.37) 366(5.95) 1865(30.32)
Total of morphemes 61,517 61,517
In general, Table 11.5 shows that the high frequency of Englishized use demonstrates the pervasive influence of pronouns in oral Chinese. The widespread use of subject pronouns also suggests that Chinese may have moved from a topic-comment language to a subject-predicate language. Further studies are required, however, to investigate the changing typological feature of Chinese.
In addition, the results also indicate that no matter in which position or which person the pronoun occurs, the frequency of pronouns carrying the features of [+referential, +animate] is significantly higher than others. Since the main function of pronouns is for reference, and the subjects of sentences are usually animate, it follows that the pronouns with the features of [+referential, +animate] are the most frequently used.
Compared with the 2010 written data, the frequency of Englishized pronouns in the oral data (30.32‰) is significantly higher than the written data (1.69‰). This result is mainly due to different properties of writing and speaking, as observed in the literature. Most importantly, the widespread use of Englishized pronouns in the oral speech reveals that the Englishized usage of pronouns is indeed accepted as part of
daily use, and not merely limited in writing.
3.2.2 Long Pre-modifiers
The examples of long pre-modifiers in each subcategory from the spoken data are illustrated as below:
a. One clause or more, in the subject position
(123) >但是 這 兩位 柏林 派 來 的 這個 間諜 要求 說
b. One clause or more, in the object position
(124)>它 就是 全 世界 第四 個 擁有 美國 公債 最 高
‗It would be a unit which owns American government bond the fourth highest in the world.‘
c. Two phrases or more, in the subject position
(125) >所有 大 台北 地區 的 資產 股 都 在 那 條 台一線 上 Suoyou da Taipei diqu de zichan gu dou zai na tiao taiyixian shang
All big Taipei area DE assests stock all at that line Taiyixian up
‗All assests stocks in Taipei area are along Taiwan provincial road, route 1.
d. Two phrases or more, in the object position
‗Certainly I feel that you are a teacher with more avant-garde thoughts and
94
teaching methods.‘
Table 12. The number and normalized frequency (per 1000 morphemes) of long pre-modifiers in 2010 oral data
Data Number of phrases or clauses
Oral One clause or more Subject 13(0.21)
Object 20(0.33) Two phrases or more Subject 7(0.11)
Object 34(0.55)
Total 74(1.20)
Total of morphemes 61,517
Table 12 shows that the frequency of long pre-modifiers in the oral data is (1.20‰), which is significantly lower than the written data (6.95‰), since it is difficult to utter and also parse long sentences during the speech. It is suggested that under normal conditions a speaker cannot focus attention on more information than six words due to the limitation of short-term memory (cf. Pawley and Syder, 1976). In general, information is much less-tightly packed in conversation, and thus clauses tend to be shorter and less complex than the written registers (Biber et al., 1999). In other words, the syntax of a sentence should be kept fairly simple, or it would cause trouble to listeners and also speakers. However, for writing the readers can review the sentences at will, so the length of each sentence tends to be longer. Chafe and Danielwicz (1987) found that the number of the words per sentence is 6.2 in the conversation, 7.3 in the lectures, 8.4 in the letters and 9.3 in the academic papers. The language type in our oral data is closer to conversation than lectures, while the style of our written data is closer to academic papers than letters. Furthermore, Biber et al.
(1999) also found that premodifiers are more frequent in the news than in conversation, and longer pre-modifier sequences are more common in the news texts, corresponding to the present study.
In terms of the position of long pre-modifiers, the result in the spoken data is the same as the written data, namely, long pre-modifiers are more frequent in the object position than in the subject position. As for the types of long pre-modifiers, in the written data, the occurrences of one clause or more outnumber the tokens of two phrases or more; however, the distribution in the oral data is the opposite. Biber et al.
(1999) also found that the conversation texts contain significantly more non-clausal word forms, while the news texts are the opposite. In addition, in the conversation text, there are shorter clauses and a lower degree of embedding. Therefore, the results of the present study correspond to the findings of Biber et al (1999).
3.2.3.2 Insertion of yi and classifiers
The insertion of yi and classifiers is significantly higher in the oral data (2.58‰) than the 2010 written data (0.61‰), and it is found that the tendency of generalization of classifiers is more significant in the oral data. The majority of classifiers used is ge, and only twelve tokens out of 159 items of classifiers are not ge, as exemplified below.
Table 13. The number and normalized frequency (per 1000 morphemes) of Englishized insertion of yi and classifiers in 2010 oral data
Category Data Oral
Yi + 個 + abstract nouns 92(1.50)
+ concrete nouns 55(0.89)
Yi + 種 + abstract nouns 9(0.15)
+ concrete nouns 0 Yi + other classifiers + abstract nouns 0
+ concrete nouns 3(0.05)
Total 159(2.58)
Total of morphemes 61,517
96
a. Yi + ge + abstract nouns
(127) 那個 對 美國 來 講 是 多 大 的 一 個 衝擊 nage dui meiguo lai jiang shi duo da de yi ge chongji That to U.S come speak C/F many big NM one CL impact
‗For the U.S, what a huge impact it is.
b. Yi + ge + concrete nouns
(128) 你 怎麼 會 在 一 個 工人 所 坐 的 車廂 裡面 Ni zenme hui zai yi ge gongren suo zuo de chexiang limian You how can at one CL worker SUO sit DE car inside
‗How would you sit in a compartment for blue-collars?‘
The higher frequency of ‗yi + ge + abstract/concrete noun‘ in the oral data may be due to the fact that in writing the authors can always delete and edit unnecessary words whereas in speech, yige seems to provide the speaker with processing time before s/he thinks of a proper noun for the predication. Besides, it is suggested that through repeated uses in the daily usage, the construction ‗yi + ge + noun‘ is memorized and retrieved as a unit (Biq, 2004). As a result, in the real-time speech, such a ready phrase is more commonly adopted.
3.2.4 Passive Structure Bei
Just as the analysis of writing, we examined the four Englishized features in the passive bei sentences in speaking. As in (129), the bei sentence is with all Englishized features: non-negative, inanimate subject, without agent and non-past tense.
(129) 雖然 剛剛 講 說 刺激 這些 東西 是 要 被 打開 的 Suiran ganggang jiang shuo ciji zhexie dongxi shi yao bei dakai de Though just speak say stimulate these thing C/F want BEI open NM
‗Though we just said that stimuli these things should be opened up.‘
The frequency of Englishized bei structure is higher in the written data (0.61‰) than the oral data (0.23‰) (cf. Table 14). In the oral data, 14 tokens out of 31 bei sentences are Englishized; while in the written data, the frequency is even up to 25
tokens out of 27 items. The results reveal that the use of Englishized bei structure is widely adopted in modern Chinese, especially in writing. Similarly, in English the passives are relatively infrequent in the spoken language, except for the lexicalized phrases like ‗I got hit.‘ (Chafe, 1990). Chafe and Danielwicz (1987) found that there were only three passives in conversations, but 22 occurrences in academic writing. In fact, plenty of studies of English speaking and writing reveal that the passives occur more frequently in writing (O‘Donnell, 1974; Blass and Siegman, 1975; Bennett, 1977; Biber et al., 1999).
As for the frequency of each Englishized feature, the distribution is relatively even among four Englishized features in the oral data. That is, unlike the significantly higher frequency of agentless passives in the written data, while in speaking the number of agentless passives is close to that of other features. It may be associated with the spontaneous feature of spoken language, leading to more redundant use of words. Take (130) as an example, the agent renjia is unimportant and omissible information; however, it is expressed overtly in the spoken data due to the real-time feature of speaking. On the other hand, as mentioned in 3.1.4, no similar example is found in our written data, since it can be deleted in writing.
(130) 那邊 萬一 等一下 被 人家 拿走 或 幹嘛 Nabian wanyi dengyixia bei renjia nazou huo ganma That in case wait a minute BEI people take away or what ‗What if later it was taken away by people‘
98
Table 14. The number and normalized frequency (per 1000 morphemes) of Englishized passive bei structure in 2010 oral data
Category Data Oral (2010) Non-negative (feature 1) 1(0.02) Inanimate subject (feature 2) 2(0.03)
w/o agent (feature 3) 0
non-past (feature 4) 1(0.02)
1+2 0
1+3 0
1+4 3(0.05)
2+3 3(0.05)
2+4 1(0.02)
3+4 1(0.02)
1+2+3 0
1+2+4 0
2+3+4 0
1+3+4 1(0.02)
1+2+3+4 1(0.02)
Total of Englishized uses 14(0.23)
Total of morphemes 61,517
3.2.5 Lexical Verbal Nominalization
The examples of each subcategory of lexical nominalization in the oral data are illustrated below:
a. Nominalized subject
(131) 問卷 調查 都 做 好 了 Wenjuan diaozha dou zuo hao le Questionnaire survey all do good PT
‗Questionnaire surveys are all finished.‘
b. Nominalized object
(132) 之後 她 就 隨著 工程 的 進行 她 會 一直 跟 你
Zhihou ta jiu suizhe gongcheng de jinxing ta hui yizhi gen ni After she LK with engineer NM process she can always with you
‗Afterward following the proceeding of construction she kept tagging along after you.‘
(133) 他 說 有 2 3 秒 的 遲延 所以 剛剛 122 是 對 的 Ta shou you 2 3 miao de chiyan suoyi ganggang 122 shi dui de He say have 23 second NM delay so just 122 C/F right NM ‗He said that there is two to three seconds delay so the number 122 was right.‘
c. Weak verb + abstract noun
(134) 如果 我們 給 它 做 設計 之後 Ruguo women gei ta zuo sheji zhihou If we give it do design after ‗If after we make a design for it‘
(135) 去 做 這樣 的 一 個 交換 qu zuo zheyang de yi ge jiaohuan Go do such NM one CL exchange ‗to do this kind of exchange‘
Table 15. The number and normalized frequency (per 1000 morphemes) of lexical verbal nominalization in 2010 oral data
Category Data Oral
Nominalized subject 4(0.07)
Nominalized object 12(0.20)
Weak verb + abstract noun 15(0.24)
Total 31(0.50)
Total of morphemes 61,517
Compared with the written data, the frequency of lexical verbal nominalization is lower in the oral data (4.16 v.s. 0.50 per 1000 words), which can be associated with the result of long pre-modifiers. Chafe and Danielwicz (1987) also found that the occurrences of nominalizations per thousand words are up to 92 in academic writing, but only 27 cases in conversations. The result of the present study corresponds to their findings. As discussed in 3.1.5, nominalization permits the concentration of information and the construction of an argument (Banks, 2005), while in the speech information should not be too dense, or it would increase the processing load of the
100
hearers. Besides, nominalization is associated with detachment feature (Chafe, 1990);
however, in the talk shows the speakers basically aim at engaging the audience.
As for the frequency of each subcategory, another similarity between the result of long pre-modifiers and lexical verbal nominalization is the frequency of lengthening objects is higher than lengthening subjects in both written and oral data.
Noticeably, in the spoken data, the occurrence of ‗weak verb + abstract noun‘ is the highest due to the fact that the use of weak verb 做 zuo occurs much more frequently in speaking, as shown in (133) and (134). As it is a rather colloquial usage, the frequency is much lower in writing. Besides, it is found that in conversation participants tend to repeatedly use some common verbs (mostly weak verbs, i.e. get, make, take, etc.), which may be due to the fact that time limitation leads speakers to rely heavily on those common verbs while the writers have more time to select different lexical items (Biber et al., 1999).
3.2.6 Concessive Clauses in the Final Position
There is no such example in the present data. Just as suggested in section 3.1.6, it may be due to the general infrequent use of concessive clauses in Chinese.
3.2.7 Other Englishized Patterns
Other Englishized patterns are new expressions translated from English, and originally their English counterparts are mainly used in formal writing or speech, for example, ‗concerning‘, ‗with respect to‘, ‗one of‘, and ‗the former… the latter‘.
Therefore, it is expected that these Englishized patterns would be less frequent in the oral data, as illustrated below.
3.2.7.1 Other Prepositions
Table 16. The number and normalized frequency (per 1000 morphemes) of other prepositions in 2010 oral data
Data Category
Oral 有關(相關) youguan/xiangguan 5(0.08)
對於 duiyu 6(0.10)
針對 zhendui 0
由於 youyu 0
Total 11(0.18)
Total of morphemes 61,517
In congruence with our expectation, the frequency of this usage is significantly higher in the written data (1.04‰) than the oral data (0.18‰), corresponding to Hsu‘s observation that they are used exactly and extensively in the news reports and professional registers. Besides, it also fits into the study of Chafe and Danielwicz (1987); that is, the occurrences of the prepositional phrases per thousand words in academic writing are higher in number than those in conversation. Most importantly, just as suggested above, the English counterparts of these Englishized prepositions are also more commonly used in the formal registers. Thus, it is normal that its frequency is low in our conversation database.
In congruence with our expectation, the frequency of this usage is significantly higher in the written data (1.04‰) than the oral data (0.18‰), corresponding to Hsu‘s observation that they are used exactly and extensively in the news reports and professional registers. Besides, it also fits into the study of Chafe and Danielwicz (1987); that is, the occurrences of the prepositional phrases per thousand words in academic writing are higher in number than those in conversation. Most importantly, just as suggested above, the English counterparts of these Englishized prepositions are also more commonly used in the formal registers. Thus, it is normal that its frequency is low in our conversation database.