Finally, the summary of this chapter will be presented.

CHAPTER FOUR RESULT AND DISCUSSION

In this chapter, the results obtained from the data analysis in the present study will be discussed. First, the overall findings will be presented. Next, I will examine the differences of linguistic behavior between men and women for a better understanding of men’s and women’s speech styles. Afterwards, men’s and women’s linguistic behavior will be discussed in terms of contexts and interlocutors.

Finally, the summary of this chapter will be presented.

4.1 General Findings

The total number of occurrences of different analytic categories is presented in Table 4.1. In the following, the average occurrences of each analytic category produced by the subjects are illustrated in Table 4.2. Because of the different numbers of male and female subjects in the study, the findings will be discussed according to the average number instead of the real number of occurrences.

The following are the linguistic categories to be observed: (1) hedges/fillers (wojuede“我覺得 ”, worenwei “ 我認為”, woxiang “我想 ”, nizhidao “你知道”, youdian “有點”, and youyidian “有一點”); (2) tag questions (duibudui “對不對”, shibushi “是不是”, houbuhou “好不好”, youmeiyou “有沒有”, bushima “不是嗎”, keyima “可以嗎”, bukeyima “不可以嗎”, haoma “好嗎”, shima “是嗎”, and youma

“有嗎”); (3) intensifiers (ruci “如此”, zheme “這麼”, and name “那麼”); (4)

joke-telling and humorous expressions; (5) polite forms (qing “請”, nikebukeyi “你可

不可以, ninengbuneng “你能不能, and nijiebujieyi “你介不介意); (6) empty

adjectives (keaide “可愛的”, xiyinrende “吸引人的”, mirende “迷人的”, youmeilide

Southern Min Dialect (Taiwanese).

Among the eight analytic categories, as shown in Table 4.2, men surpass women in six categories: tag questions, intensifiers, humorous expressions, polite forms, amount of talk and Southern Min Dialect. The only category which used more by females is hedges. In addition, males and females have no obvious differences in using empty adjectives. More details will be provided in the following sections.

Table 4.1 The total number of occurrences of each analytic category in male and female speech in the collected data (total number)

Analytic category Male Female Total

Hedges/Fillers (number) 219 245 464

Tag questions (number) 44 28 72

Intensifiers (number) 61 56 117

Humorous expressions (number) 35 17 52

Polite forms (number) 12 4 16

Empty adjectives (number) 3 3 6

Amount of talk (character) 56,612 43,833 100,445

Southern Min (character) 1,982 201 2,183

Table 4.2 The average frequency of each analytic category of speakers in the collected data (average number)

Male speaker (N=28) Female speaker (N=29) Analytic category

average occurrences average occurrences Hedges/Fillers

(number) 7.82 8.44

Tag questions

(number) 1.57 0.97

Intensifiers

(number) 2.18 1.93

Humorous expressions

(number) 1.25 0.58

Polite forms

(number) 0.43 0.14

Empty adjectives

(number) 0.11 0.10

Amount of talk

(characters) 2021.9 1511.5

Southern Min

(characters) 70.8 6.9

4.2 Hedges and Fillers

According to Lakoff (1975) and other previous studies (Bolinger, 1980; Holmes, 1992a; Coates, 1993), the expressions such as I think, you know, sort of, I guess, and I wonder are categorized as hedges and fillers. Lakoff suggests that hedges and fillers have the function of conveying uncertainty or of mitigating the degree of force in a statement. She further concludes that women use hedges and fillers more often because they believe that to assert themselves too strongly may be considered as

“unladylike” or “unfeminine.” Similar to Lakoff’s claim, Holmes (1986) has

observed that you know serves two major functions: one expresses the speaker’s

reflects uncertainty on both the speaker’s linguistic imprecision and the addressee’s attitude. In addition, Freed & Greenwood (1996:4) also addresses “the expression you know has often been described as a female hedging device, and interpreted as a marker of both insecurity and of powerless.” Therefore, in order to follow social values, women’s language tends to contain more hedges to avoid stating things too directly.

In Chinese, the counterparts of the hedges mentioned above are wojuede“我覺 得”, worenwei “我認為”, woxiang “我想”, nizhidao “你知道”, youdian “有點”/ and youyidian “有一點”.

In analyzing hedges, the finding of the present study confirms what Lakoff has proposed: everyone uses hedges in some situations but women use them more frequently. As shown in Table 4.3, totally 464 hedges or fillers are found. In addition to overall occurrences, Table 4.3 also indicates that on the average, each male speaker uses 7.42 hedges or fillers while each female speaker produces 8.45.

Nevertheless, the occurrences of the different types of hedges are not distributed evenly. Among the different kinds of hedges, the ones most frequently used by both sexes are in the category of I think/I wonder (wojuede “我覺得”/worenwei “我認為”

/woxiang “我想”). 90% of the Chinese hedges are wojuede “我覺得”/worenwei

“我認為” /woxiang “我想”, as in the following examples:

(1) TV Program 3, Female speaker 5:

但是 一般 的 女孩子 如果 說 要 用 模特兒 來 做

danshi yiban de nuhaizi ruquo shuo yao yong moteer lai zuo but general girl if say want use model come do

跳板 的 話, 我覺得 , 這 中間 陷阱 很 多.

‘But if general girls want to use model to be springboard, I think there are many traps in it.’

(2) TV Program 3, Male speaker 21:

這 點 我認為 馬 市長 有 責任 要 跟

zhe dian worenweil ma shizhang you zeren yao gen this point I think Ma mayor have responsibility shall to

大家 講 清楚 說 明白.

dajia jiang qingchu shuo mingbai everyone say explicit say clear

‘As for this point, I think Mayor Ma should have the responsibility to make clear to everyone.’

The next type is sort of (youdian “有點”/ youyidian “有一點”). But compared with I think/I wonder (wojuede “我覺得”/worenwei “我認為” /woxiang “我想”) type, this type is less frequent. Only 6% of hedges in the data are youdian “有點”

and youyidian “有一點”. Besides, Only 2% of hedges in the data are you know (nizhisao “你知道”). However, none of the speakers use the counterpart Chinese of I guess (wocai “我猜”). Among these three devices, the Chinese expression wojuede

“我覺得” stands out in the I think category; it is the top device used by both men and women, as Table 4.3 shows.

In brief, the present study has found that women use more hedges than men do,

as Lakoff (1975) has observed. However, the difference is not very huge. On the

average a man uses 7.82 hedges and a woman produces 8.45. Besides, among all the

hedges, the category of I think/I wonder occurs most frequently. Over 90% of the

hedges used by both men and women are in the category of I think/I wonder (wojuede

Table 4.3 The overall and average occurrences of hedges/fillers of each speaker Male speaker (N=28) Female speaker (N=29) Total Hedges Overall

occurrences

Average occurrences

Overall occurrences

Average occurrences

Overall occurrences

wojuede

“我覺得” ^{128 187}

worenwei

“我認為” ^{20 12}

woxiang

“我想”

I Think

/ I wonder

51

199 7.11

29

228 7.86 427

youdian

“有點”

youyidian

“有一點”

Sort

of 11 0.39 16 0.55 27

nizhidao

“你知道”

You

know 9 0.32 1 0.03 10

wocai

“我猜” I guess 0 0 0 0 0

Total 219 7.82 245 8.45 464

4.3 Tag Questions

According to Lakoff, tag questions such as isn’t it, don’t I, wasn’t there and so on are “used when the speaker is stating a claim but lacks full confidence in the truth of that claim” (1975:15). In Chinese, two general types of tag questions were observed by Hu (2002): (1) A not A tags (e.g. houbuhou “好不好”) and (2) particle tags (e.g.

keyima “可以嗎”). Based on these two basic types, the tag questions identified in

youmeiyou “有沒有”, bushima “不是嗎”, keyima “可以嗎”, bukeyima “不可以嗎”, haoma “好嗎”, shima “是嗎”, youma “有嗎”.

Lakoff (1975) asserts that tag questions occur more frequently in women’s speech than in men’s speech; however, the result of the present study seems to contradict her observation. That is, men use tag questions more frequently than women do. The result of the present study shows that regarding tag questions, each male produces an average number of 1.57 whereas each females 0.97, as shown in Table 4.4. Besides, as Table 4.4 shows, in not only soft talk shows but also socio-political talk shows, on the average men use more tag questions than women do.

On the average, each man uses 1.5 tag questions in soft talk shows and 1.64 in socio-political talk shows. By contrast, each woman uses such question to a much lesser degree than men do in both types of programs, 1.19 and 0.38, individually.

What’s worth noting is that when women’s frequency of tag questions in soft and

socio-political talk shows is compared, it is found that women use much fewer tag

questions in socio-political talk shows than in soft talk shows (3:1). Such ratio may

suggest that in socio-political talk shows, the topics under discussion are about social

and socio-political issues, which are masculine-oriented. Therefore, females tend to

avoid using too much “women’s language” in such programs. Instead, they change

their ways of talk to make their speech styles more similar to men’s.

Table 4.4 The overall and average occurrences of tag questions of each speaker in the collected data (total number/average number)

Male speaker (N=28) Female speaker (N=29) Program type Overall

occurrences

Average occurrences

Overall occurrences

Average occurrences Soft talk shows

(14 males; 21 females) 21 1.50 25 1.19

Socio-political talk shows

(14 males; 8 females) 23 1.64 3 0.38

Overall 44 1.57 28 0.97

Among all these tag question types, the present study asserts that A not A tag is the most popular form used in males’ as well as females’ speech, as Hu (2002) has suggested. In Hu’s (2002) corpus, 89.89% of the tags are in A not A forms, 7.76 % are particle tags and 4.35% are others. In the present study, as Table 4.5 shows, about 86% of the tag questions in males’ speech fall into A not A forms (duibudui “對 不對”, shibushi “是不是”, houbuhou “好不好”, youmeiyou “有沒有”). Similarly, 89% of the tag questions in females’ speech are in A not A forms, too, as the following examples show:

(3) TV Program 3, Female speaker 7:

所以 你 盡量 選 有 綠 燈 的 時候

Suoyi ni jinliang xuan you lu deng de shiho so you try best choose have green light time

才 開, 對不對 ？

Cai kai duibudui?

just drive, right no right

‘So you try your best to drive when the lights are green, don’t you?’

(4) TV Program 11, Male speaker 27:

後來 章孝嚴 走 的 時候, 是 好像 小 男生 做 錯

holai Zhangxiaoyen zuo de shiho, shi haoxiang xio nansheng zuo cuo later PN go time is like little boy do wrong

事情, 躲 在 媽媽 後面 喔, 媽媽 幫 他 開 道 擋住

shiqing, duo zai mama huomian mama bang ta kai dao dangzhu thing hide at Mom back Mom help he open road block

所有 的 記者, 那 時候 還 有 丁遠超 在 前面

suoyou de jizhe na shiho hai you Dingyuanchao zai qianmian all reporter that time still have PN at front

推 記者, 有沒有 ？

tuei jizhe youmeiyou push reporter have no have

‘Later when Zhangxiaoyen was leaving, he was like a little boy doing something wrong. It seemed that he hid behind his mom and his mom opened up a road for him, blocking all the reporters. At that time there was Dingyuanchao pushing the reporters in the front, wasn’t there?’

On the contrary, particle tags (bushima “不是嗎”, keyima “可以嗎”, bukeyima

“不可以嗎”, haoma “好嗎”, shima “是嗎”, youma “有嗎”) are occasionally used, nearly 14% in men’s speech and 11% in women’s. The following is an example of particle tag.

(5) TV Program 3, Female speaker 10:

這 是 美德, 不 是 嗎 ？

zhe shi meide, bushima this is virtue no yes PR

‘This is virtue, isn’t it?’

Table 4.5 The distribution of two tag question types in the collected data (total number v.s. percentage)

Male speaker (N=28) Female speaker (N=29) Tag question type

occurrence

A not A tags Particle tags A not A tags Particle tags overall

occurrences 38 6 25 3

Percentage 86.36% 13.64% 89.29% 10.71%

Total number

44 28

Average

number 1.57 0.97

Since A not A tags is the most popular form, among the four types of A not A tags (duibudui “對不對”, shibushi “是不是”, houbuhou “好不好”, youmeiyou “有沒有”), duibudui “對不對” is especially frequently used, in both men’s or in women’s discourse, as well as in both soft and socio-political talk shows, as Table 4.6 shows.

However, in Mandarin these tags such as duibudui “對不對”, shibushi “是不是”, and houbuhou “好不好” do not always mean “uncertain” or “lack of confidence.”

Instead, those tags are sometimes quite powerful and used to have people’s attention, as we often hear in election campaigns in Taiwan. It is the nature of the language which results in such differences between English tags and Chinese tags.

In conclusion, the present study has suggested a contrary finding to Lakoff’s

(1975) observation in terms of tag questions. Lakoff states that tag question is a

feature of “women’s language” and women use it more often. However, the present

study has found that on the average men use more tag questions than women do

(1.57:0.97). Besides, among all the tag questions types, A not A tags is used most

Table 4.6 The distribution of all tag question forms in the collected data (total number v.s. average number)

Number of occurrences Program

Type Types of tag questions Male speaker (N=28)

Female speaker (N=29)

duibudui “對不對” 11 9

shibushi “是不是” 2 7

houbuhou “好不好” 2 4

youmeiyou “有沒有” 5 2

bushima “不是嗎” 0 2

keyima “可以嗎” 0 0

bukeyima “不可以嗎” 0 1

haoma “好嗎” 1 0

shima “是嗎” 0 0

Soft talk shows (14 males;

21 females)

youma “有嗎” 0 0

duibudui “對不對” 14 3

shibushi “是不是” 2 0

houbuhou “好不好” 1 0

youmeiyou “有沒有” 1 0

bushima “不是嗎” 4 0

keyima “可以嗎” 0 0

bukeyima “不可以嗎” 1 0

haoma “好嗎” 0 0

shima “是嗎” 0 0

Socio-political talk shows (14 males;

8 females)

youma “有嗎” 0 0

Total number 44 28

Average number 1.57 0.97

4.4 Intensifiers

Lakoff (1975) has addressed that the intensive “so” is a linguistic device which can weasel on the intensity of people’s language when they feel it improper to reveal strong emotion or to make strong assertion. Take the sentence “I like him very much” for example, it states precisely I like him to a great extent. But the sentence

“I like him so much” weasels the intensity. Lakoff includes the intensives “so” and

“such” as “women’s language” and she holds the view that women use intensive “so”

more often than men to obscure the strength of emotion.

The Chinese counterparts of English “so” and “such” are ruci “如此”, zheme “這 麼”, and name “那麼”. In the collected data of the present study, totally 117 ruci

“如此”, zheme “這麼”, and name “那麼” are found. The result of the present study shows that the intensifier “so” appears as frequently in men’s as in women’s utterances, which is different from Lakoff’s claim,. As Table 4.7 shows, on the average, each male speaker uses 2.18 intensifiers and each female speaker uses 1.93.

Table 4.7 The overall and average occurrences of intensifiers (total number and average number)

Male speaker (N=28) Female speaker (N=29) Program type overall

occurrences

Average occurrences

overall occurrences

average occurrences Soft talk shows

(14 males; 21 females) 22 1.57 27 1.29

Socio-political talk shows

(14 males; 8 females) 39 2.79 29 3.63

Overall 61 2.18 56 1.93

However, the frequency differs in different types of programs. Both men and women use more intensifiers in socio-political talk shows than in soft talk shows.

Men use 2.79 intensifiers in socio-political talk shows but only 1.57 in soft talk shows.

Likewise, women use almost 3 times of intensifiers in socio-political talk shows than in soft talk shows (3.63 versus 1.29). More details will be discussed in Section 4.11.

In terms of the three Chinese counterparts of “so” (ruci “如此”, zheme “這麼”, and name “那麼”), men and women have similar frequency in using them. Men use zheme “這麼” most often, and so do women. Overall, zheme “這麼” has 31 occurrences in men’s speech and 32 occurrences in women’s. The next is name “那 麼”. Men use 26 name “那麼” and women 22. However, ruci “如此” is seldom used, as Table 4.8 shows. ruci “如此” rarely occurs in the data possibly because it is not a colloquial word but used more often in written language. Therefore, it does not occur frequently in males’ and females’ speech. The following are two examples of intensifier use in the data:

(6) TV Program 5, Male speaker 9:

你 那麼 討厭 當 人 嘛, 罰 你 生生世世

ni name taoyian dang ren ma fa ni shengshengshishi you so hate be human PR punish you lifetime

做 豬, 任 人 宰割.

zuo zhu ren ren zaige be pig allow people exploit

‘If you hate to be a human being so much, you will be punished to be a pig,

exploited by people.’

(7) TV Program 10, Female speaker 25:

我 是 覺得 說 很 可憐, 一個 直 航 這麼 重要 的

wo shi juede shuo hen kelian yige zhi hang zheme zhongyao de I be think say very poor a straight flight such important

政策, 到 最後 居然 被 解讀 為 總統 的

zhengce dao zueiho juran bei jiedu wei zongtong de policy to finally unexpectedly Past-tense interpret become president

重大 宣誓, 為 選舉 支票.

zhongda xuanshi wei xuanju zhipiao important declaration become election check

‘I think it is really a poor thing. Straight flight is such an important policy.

However, in the end it is interpreted as a very important declaration of the president, as an election check.’

Table 4.8 The distribution of all types of intensifiers (total number) Number of occurrences Program type

Male speaker (N=28) Female speaker (N=29)

zheme “這麼” 9 zheme “這麼” 19

name “那麼” 13 name “那麼” 8

Soft talk shows (14 males; 21 females)

ruci “如此” 0 ruci “如此” 0

zheme “這麼” 22 zheme “這麼” 13

name “那麼” 13 name “那麼” 14

Socio-political talk shows (14 males; 8 females)

ruci “如此” 4 ruci “如此” 2

Overall 61 56

In short, although the intensifier “so” is categorized as a feature of “women’s language” by Lakoff (1975), the present study has a contrary finding. On the average, a male uses “so” a little more often than a woman does (2.18:1.93).

Besides, both men and women use more intensifiers in socio-political talk shows than in soft talk shows.

4.5 Joke-telling and Humorous Expressions

In Lakoff’s (1975) work, she has observed that “women don’t tell jokes” and

“women have no sense of humor.” Parallel to Lakoff, Kuo (1995) also finds that in mixed-sex conversations, most humorous remarks are made by male speakers whereas females laughed more frequently. Since it is not easy to give “humorous expressions” a clear-cut definition, those remarks which make other speakers in the programs laugh are defined as “humorous” and were counted in the present study.

For example, the two following sentences are considered as humorous expressions.

(8) TV Program 3, Female speaker 7:

他 只是 一個 很 善良 的 海 蟑螂.

ta zhish i yige hen shanliang de hai zhanglang he just one very kindhearted sea cockroach

‘He is just a kindhearted sea cockroach.’

(9) TV Program 2, Male speaker 4:

是的, 為 您 介紹 的 產品 是 『別 想 高』,

shide wei ni jieshao de chanpin shi “biexianggao

yes for you introduce product is no want tall

只要 你 吃 了 它, 絕 對 長 不 高.

zhiyao ni chi le ta juieduei zhang bu gao as long as you eat ASP it absolutely grow no tall

‘Yes, the product being introduced now is Never Wish to Grow Tall.

As long as you eat this, you can never grow tall.’

The findings of the present study confirm Kuo’s observation that males express more sense of humor than females. As Table 4.9 and Table 4.10 demonstrate, males create about twice more humorous expressions than do females. On the average, each male speaker produces 1.25 humorous expressions. Women’s production of humorous expressions, however, is less, only 0.58 per person. Besides gender differences, the distribution of humorous expressions on program types is also different. Soft talk shows have about 4 times more of humorous expressions than do socio-political talk shows. This is predictable because the topics discussed in soft talk shows are not as serious as those in socio-political talk shows, and therefore the atmosphere in soft talk shows is more relaxing. Under such circumstances, people tend to be more humorous.

Table 4.9 The overall and average number of humorous expressions of each speaker in the collected data (total number/average number)

Male speaker (N=28) Female speaker (N=29) Program type overall

occurrences

average occurrences

Overall occurrences

average occurrences Soft talk shows

(14 males; 21 females) 28 2.00 14 0.67

Socio-political talk shows

(14 males; 8 females) 7 0.50 3 0.38

Overall 35 1.25 17 0.59

Table 4.10 The distribution of humorous expressions of each speaker in the collected data (total number)

Male speaker (N=28)

Female speaker (N=29) Program type and topic

Overall occurrences

Overall occurrences

Program 1 soft talk show (health) 1 0

Program 2 soft talk show (career) 9 3

Program 3 soft talk show (expense) 3 8

Program 4 soft talk show (horoscope) 0 2

Program 5 soft talk show (horoscope) 15 1

Program 6 soft talk show (education) 0 0

subtotal 28 14

Program 7 socio-political talk show (politics) 0 0 Program 8 socio-political talk show (politics) 1 1 Program 9 socio-political talk show (politics) 0 0 Program 10 socio-political talk show (politics) 0 0 Program 11 socio-political talk show (social issues) 2 1 Program 12 socio-political talk show (politics) 4 1

subtotal 7 3

Total 35 17

This gender difference in producing humorous remarks is explained by Lakoff (1975) and Mitchell (1985). Lakoff points out that in middle-class American society, customarily women cannot tell jokes nor understand jokes. Mitchell indicates that women tell fewer jokes because they gain little prestige for being good joke-tellers in formal contexts since men consider joke-telling improper for women. It is the difference in attitude that leads to this phenomenon. As a result, this study confirms what Lakoff and Mitchell have proposed, i.e. men are better joke-tellers.

However, among all male’s humorous expressions, a large proportion of them are generated by a single speaker in TV Program 5. Among 15 humorous remarks in TV Program 5, the male speaker produces 13 of them. This male speaker himself is a TV show host and he is famous for his comical character because he always says something funny when he shows up. He is good at making humorous remarks and he tries to do so in order to catch the audience’s eyes. If the 13 humorous expressions of this male speaker in TV Program 5 are not counted, only 2 humorous remarks are produced by the rest of the men in the same program. When the man’s speech in TV Program 5 is excluded, each male in the collected data on the average produces 0.81 humorous expressions. However, it does not change the result of the study because men’s average number (0.81) is still higher than women’s (0.59). The personal conversational style of the male in TV Program 5 does not have a great effect on the result of the study.

Besides, among the nine humorous remarks in TV Programs 2, seven of them are

produced by a male speaker to make fun of other female speakers in the same

program. The male speaker in this program treats teasing other female speakers as a

type of humor. Therefore, he keeps teasing others to make his speech funnier and

express solidarity and maintain social relationships. In a like manner, Tannen (1986) has argued that joking is one of the many ways of saying one thing and meaning another. Joking has its metamessages as well as indirect nature, and therefore, people cannot just depend on literal meaning to understand jokes. The performance of this male speaker in TV Program 2 confirms what Tannen (1986:60) has mentioned:

Joking is a kind of irony that has both rapport and defensive payoffs. The rapport benefit lies in the sensual pleasure of shared laughter as well as the evidence of rapport in having matching senses of humor. The defensive benefit is in the ability to retreat: “I was only joking.”

In conclusion, the present study has found that on the average men tell about two times of jokes than women do (1.25:0.59), as previous studies have suggested (Lakoff, 1975; Mitchell 1985; Kuo, 1995). Besides, in both soft and socio-political talk shows, men produce more humorous remarks than women do. Moreover, both male and female participants produce more humorous remarks in soft talk shows since they are not as serious as socio-political talk shows.

4.6 Polite Forms

Lakoff (1975) asserts that “women are the experts at euphemism” (p.55). They speak more properly, they talk more politely, and they are more careful to say words such as “please” and “thank you.” Besides, when making a request, they use “Will you”, “Won’t you”, “Do you mind” or “Please” to start the conversation rather than giving a direct order like “Close the door.”

In Chinese people also usually use qing “請” to exhibit politeness when asking

such as nikebukeyi “你可不可以 will you”, ninengbuneng “你能不能 will you”, nijiebujieyi “你介不介意 do you mind” to show politeness when asking for a favor or making a request. As Lakoff suggests, “Indirectness can function as a form of politeness.”

In the present study, polite forms do not occur frequently; only a total of 16

expressions are discovered. This phenomenon may be attributabled to the nature of

the data. The data were collected from TV talk shows and the speakers come to the

shows to discuss a certain topic or to express their own opinions. In that case, they

must be assertive enough to be considered experts of certain issues and therefore, they

will not be allowed the opportunities to “be too polite.” With this special domain in

our data, men use about 3 times more polite forms than women do. This finding

does not correspond to Lakoff’s observation, which states that women talk more

politely. Such difference in the findings may be due to the different types of data

collection. Lakoff’s data are from private and public contexts while the data in the

present study is only from public context. In the present study, men’s average use of

polite forms is 0.43 while each woman’s is 0.14, as presented in Table 4.11 and Table

4.12.

Table 4.11 The overall and average occurrences of different types of polite forms of each speaker in the collected data (total number/average number)

Male speaker (N=28) Female speaker (N=29) Total Polite forms _overall

occurrences

Average occurrences

overall occurrences

Average occurrences

overall occurrences qing “請”

(Please) 9 0.32 3 0.10 12

nikebukeyi

“你可不可以”

ninengbuneng

“你能不能”

(Will you / won’t you)

0 0 1 0.04 1

nijiebujieyi

“你介不介意”

(Do you mind)

3 0.11 0 0 3

Total 12 0.43 4 0.14 16

Table 4.12 The overall and average occurrences of polite forms of each speaker in the collected data (total number/average number)

Male speaker (N=28) Female speaker (N=29) Program type overall

occurrences

average occurrences

overall occurrences

average occurrences Soft talk shows

(14 males; 21 females) 7 0.50 2 0.10

Socio-political talk shows

(14 males; 8 females) 5 0.36 2 0.25

Overall 12 0.43 4 0.14

Among all the polite forms, the most frequent one is “please” (qing “請”), 12 out of 16 polite statements are in this form. Other forms of politeness such as nikebukeyi

“你可不可以 will you”, ninengbuneng “你能不能 will you”, nijiebujieyi “你介不介 意 are rarely used. Only four occurrences of them were found. The following are some examples of polite forms in the data.

(10) TV Program 3, Male speaker 7:

可不可以 請 問 一下, 為什麼 你 要 去 當 兵?

kebukeyi qing wen yixia weisheme ni yao qu dang bing would you please ask one time why you want go be soldier

‘Could you please tell us why you joined the army?’

(11) TV Program 11, Female speaker 26:

那 我 請 問, 她 為什麼 要 錄音？

na wo qing wen ta weisheme yao luyin Then I please ask she why want record

‘Please tell me why she recorded the speech?’

In summary, unlike Lakoff’s (1975) claim, in the present study polite forms

occur a little more often in men’s speech than in women’s speech. On the average, a

man uses 0.43 times of polite forms but a woman uses 0.14 times. Men use more

polite forms in both soft talk shows and socio-political talk shows. The differences

between Lakoff’s finding and the finding in the present study may result from

different languages and different contexts. However, the occurrences of polite forms

are so limited that the finding is not significant to generalize gender differences in

4.7 Empty Adjectives

“Empty adjectives” such as adorable, charming, sweet, lovely, divine and cute are believed to be used mostly by women. Lakoff (1975) points out that in English some adjectives are neutral so that both men and women may use them. However, certain adjectives are largely used by women. It is possible for a man to damage his reputation if he uses the words categorized in women’s language.

The Chinese counterparts of adorable, charming, sweet, lovely, divine and cute are: keaide “可愛的”, xiyinrende “吸引人的”, mirende “迷人的”, youmeilide “有魅 力的”, tianmeide “甜美的”, lingrenyukuaide “令人愉快的”, jibande “極棒的”, feifande “非凡的”. However, in the collected data, very few of them were found.

Only six occurrences of empty adjectives appear in the speakers’ speech. Among them, five are keaide “可愛的” and one is tianmeide “甜美的”. Other kinds do not occur at all.

As presented in Table 4.13, on the average men use empty adjectives almost as

frequent as women do. Each male and female uses about 0.1 empty adjective in

their speech. However, males’ and females’ use of empty adjectives are uneven in

different types of programs. In soft talk shows, women use twice more empty

adjectives than men do. Yet, in socio-political talk shows men use some of them but

women use none. The examples of empty adjectives are as follows.

Table 4.13 The overall and average occurrences of empty adjectives of each speaker in the collected data (total number/average number)

Male speaker (N=28) Female speaker (N=29) Program type overall

occurrences

average occurrences

overall occurrences

average occurrences Soft talk shows

(14 males; 21 females) 1 0.07 3 0.14

Socio-political talk shows

(14 males; 8 females) 2 0.14 0 0

Overall 3 0.11 3 0.10

(12) TV Program 3, Female speaker 8:

這 很 可愛 耶.

zhe hen keai yie this very lovely PR

‘This is very lovely.’

(13) TV Program 10, Male speaker 24

所以 一旦 掌握 了 權力 以後, 阿 覺得 權力 是

suoyi yidan zhangwo le quanli yiho a juede quanli shi so once seize ASP power after PR think power is

這麼 可愛的 , 有 這麼 多 的 好處, 你 知道 嗎,

zheme keaide you zheme duo de haochu ni zhidao ma so lovely have so many advantage you know PR

所以 他 其實 更 想 掌握 權力, 更 容易 腐化,

suoyi ta qishi geng xiang zhangwo quanli geng rongyi fuhua therefore he actually more want seize power more easy corrupt

更 容易 濫 權.

geng rongyi lan quan

more easy abuse power

‘So once he seizes the power, he thinks power is so lovely and has so many advantages, you know. Therefore, actually he wants more power and becomes corrupt easily and will abuse the power easily.’

For Lakoff (1975), empty adjectives are used more often by women and it is uncommon to hear men use those terms. However, the present study has found no obvious differences in empty adjectives. There are two possible reasons. For one thing, English and Chinese are two different languages and they are different in some aspects of language use. Perhaps empty adjectives could be one of those aspects.

For another, it may result from the nature of the data. The data in the present study were collected from public contexts and such adjectives may occur more frequently in private speech. Nevertheless, the occurrences of empty adjectives in the present study are so limited that the finding is not significant to generalize gender differences in language.

To sum up, despite the fact that empty adjectives are supposed to be used more frequently by women (Lakoff, 1975), the present study has found that empty adjectives are used as frequently by men as by women. Men do not use fewer empty adjectives than women do. However, the occurrences of empty adjectives are so limited and the finding may not be able to generalize to men’s and women’s linguistic patterns.

4.8 Amount of Talk

In both eastern and western cultures, it is a stereotypical image that women are

wordy and mouthy. Therefore, it has been generally believed that women always

talk more than men. However, some of the researcheres hold a different view. The

also supports the previous studies and contradicts such a stereotype. The present study indicates that men talk more than women in amount. As revealed in Table 4.14, the average amount of men’s talk is 2021.9 characters and women’s is 1511.5 characters. In other words, men speak 1.34 times more than women do. Therefore, it is obvious that a man speaks more characters than a woman.

Table 4.14 The average number of characters spoken by each individual in the collected data

Amount of talk

Male Speaker (N=28)

Female speaker (N=29)

Average number of characters taken by

each individual 2021.85 1511.48

To compare the amount of talk of males and females, the characters of each

speaker were counted in the collected data, which are listed in Table 4.15. In

addition, the average numbers of characters of each male and female speaker were

calculated, as Table 4.14 shows. In the present study a male on the average speaks

more characters than a female. Each man’s average number of characters is 2021.85

characters, while each woman’s is 1511.48 characters. In other words, on the

average a man talks more than a woman.

Table 4.15 The distribution of total characters uttered by each speaker in the collected data

Male speaker Number of

Characters Female speaker Number of characters

M1 3130 W1 802

M2 3486 W2 1602

M3 1404 W3 1185

M4 868 W4 2062

M5 1042 W5 1840

M6 173 W6 1319

M7 45 W7 1142

M8 4743 W8 884

M9 2012 W9 127

M10 1919 W10 1410

M11 640 W11 865

M12 980 W12 280

M13 928 W13 717

M14 834 W14 253

M15 2182 W15 1276

M16 2538 W16 2762

M17 3154 W17 1729

M18 2783 W18 1288

M19 1208 W19 1555

M20 1103 W20 230

M21 1753 W21 975

M22 1829 W22 3560

M23 1702 W23 3834

M24 2708 W24 1983

M25 2972 W25 2457

M26 3707 W26 1675

M27 3068 W27 1630

M28 3701 W28 2375

--- --- W29 2016

Total 56,612 Total 43,833

The finding of the present study does not correspond to the cultural stereotype, which believes that women talk more than men. However, it does confirm Tannen’s (1990) and Holmes’ (1992b) observations. Tannen (1990) suggests that men talk more in public contexts—at meetings, in mixed-group discussions, and in classrooms.

Besides, she argues that men speak for a longer time and take longer turns than women. Moreover, Holmes (1992a; 1992b) also indicates that in more formal contexts such as seminars, TV discussions, conferences and classrooms, men usually talk more and longer than women do. On the contrary, women feel more comfortable in private, informal contexts, and therefore they tend to contribute more in such circumstances. Holmes (1992b) has proposed that this phenomenon is related to different social orientations of the two genders. Males exhibit expertise, confirm status and power by dominating and participating more in the discussion, whereas females care much about solidarity when they have conversation with others.

However, since the contexts in the present study are TV programs, the result may be program-based. Besides, it may be due to personality factor, too.

To sum up, the data in the present study suggest that on the average, men talk in

a larger amount than women do in public domain. The finding of the present study

contradicts the stereotype which assumes that women are chattier and mouthier than

men are. However, it confirms Tannen’s (1990) and Nordenstam’s (1992) studies

which claim that men talk more than women, especially in public contexts.

4.9 Southern Min

In Taiwan, Southern Min is the most common vernacular language, compared with Hakka and other native languages, such as Amis and Atayal. Although at present, Mandarin has been made our official language, Southern Min still has a great number of speakers. Because of the change of Taiwan’s socio-political circumstance

, in recent years more and more people speak Southern Min in public domains. Oftentimes people switch between Mandarin and Southern Min in their daily conversations. Concerning the status of these two dialects

in Taiwan, Mandarin is considered the standard language while Southern Min is the

5. From 1945 to 2000, Taiwan was ruled by the Chinese Nationalists, or Kuomingtang (KMT), for over 50 years, which is a regime from Mainland China. KMT has been promoting Mandarin as the national language of Taiwan. Since 2000, Taiwan’s ruling party is changed to Democratic Progressive Party (DPP), which is mainly composed of local Taiwanese people. After winning the political power, DPP strongly proposes Taiwan’s localization and advocates the use of Southern Min, the language originally spoken in Taiwan. DPP argues that Taiwan’s primary language is not Mandarin, but Southern Min. Therefore, Southern Min is used by more and more people nowadays.

6. Holmes (1992a) remarks, “In order to define a language, it is important to look to its social and political functions, as well as its linguistic features. So a language can be thought of as a collection of dialects that are usually linguistically similar, used by different social groups who choose to say that they are speakers of one language, which functions to unite and represent them to other groups” (pp.141-142).

Similarly, according to Cipollone, Keiser and Vasishth (1998), “when a group of speakers of

a particular language differs noticeably in its speech from another group, we say that they are

speaking different dialects…. One criterion used to distinguish dialects from languages is mutual

intelligibility. If speakers of one language variety can understand speakers of another language

variety and vice versa, we say that these varieties are mutual intelligible…. Your variety of

speech and theirs are mutually intelligible but differ systematically, and are therefore dialects of

the same language” (p.368).

nonstandard

.

According to Cipollone, Keiser and Vasishth (1998), standard dialect relates to

“overt prestige”, which “defines how people should speak in order to gain status in the wider community.” Yet, nonstandard dialect signifies “covert prestige”, which defines how people should speak in order to be considered a member of some particular communities… In this sense, language becomes a marker of group identification (p.377). Therefore, when people make certain linguistic choices, they are affected by “social identity”. The language they choose to speak depends on how they identify themselves, how they want to be identified and show solidarity with the addressee (Cipollone, Keiser and Vasishth, 1998; Holmes, 1992a).

Holmes (1992a) remarks that vernacular language is considered “the most colloquial variety in a person’s linguistic repertoire. In a multilingual community, this variety will often be a nonstandardized ethnic or tribal language. The vernacular is the variety used for communication in the home and with close friends. It is the language of solidarity between people from the same ethnic group” (p. 80). What’s more, Holmes also points out that vernacular dialects “lack public or overt prestige, though they are generally valued by their users, especially as means of expressing solidarity and affective meaning” (1992a: 146).

In the same way, Hudson (1996) argues that the main reason for people to preserve more than one language is because of the social distinction which people want to be symbolized. For bilingual speakers, their language represents the

7. As Holmes (1992a) suggests, “To avoid the implication that nonstandard languages are

somehow deviations from the standard, some sociolinguists use the term vernacular as an

community and their language has social function which can’t be accomplished by other languages.

Holmes (1992a) suggests that women tend to use more of the standard forms than men do. In her point of view, “In every social class men use more vernacular forms than women... When men use a form more often than women, it is usually a vernacular form” (pp.168-170). Accordingly, women use more standard forms when they speak. According to Holmes’ (1992a) explanation, women use more standard forms than men because of the following reasons: (1) they are more status-conscious, (2) their roles are viewed as guardians of society’s values, and (3) they are subordinate. So they choose to speak as carefully and politely as they can. Besides, vernacular forms convey the meaning of masculinity. Thus, different genders prefer different forms of language.

Gumpers (1982) remarks that code switching is frequently found in informal speech; however, the present study has found that the speakers in TV talk shows switch from Mandarin to Southern Min repeatedly. Table 4.16 presents the average characters of Southern Min use of each male and female speaker. It shows that men speak more Southern Min than women do. In the total data, a man’s average use of Southern Min is 70.8 characters whereas a woman’s average is just 6.9. In other words, men speak approximately 10 times more Southern Min than women.

If Southern Min use is compared in soft or socio-political talk shows, men speak

more Southern Min than women do in both types of programs. As Table 4.16 shows,

in soft talk shows, a man on the average speaks 11 times more Southern Min than a

woman (11:1). Similarly, in socio-political talk shows, on the average a man speaks

5.8 times of Southern Min more than a woman. On the average a man speaks 130.6

Table 4.16 The overall and average characters switching to Southern Min (overall characters/ average characters)

Male speaker (N=28) Female speaker (N=29) Program type overall

characters

average characters

overall characters

average characters Soft talk shows

(14 males; 21 females) 154 11 22 1

Socio-political talk shows

(14 males; 8 females) 1828 130.6 179 22.4

Overall 1982 70.8 201 6.9

Nevertheless, Southern Min is unequally distributed in male and female speech

in the present study. Table 4.17 shows the distribution of total characters of

Southern Min in male and female speech.

Table 4.17 The distribution of Southern Min use between male and female speakers (total characters)

Male speaker (N=28)

Female speaker (N=29) Program type and topic

total characters of Southern Min

Program 1 soft talk show (health) 9 0

Program 2 soft talk show (career) 0 16

Program 3 soft talk show (expense) 0 0

Program 4 soft talk show (horoscope) 16 0

Program 5 soft talk show (horoscope) 85 0

Program 6 soft talk show (education) 44 6

Subtotal 154 22

Program 7 socio-political talk show (politics) 20 36 Program 8 socio-political talk show (politics) 164 75 Program 9 socio-political talk show (politics) 1472 65 Program 10 socio-political talk show (politics) 34 0 Program 11 socio-political talk show (social issues) 115 0 Program 12 socio-political talk show (politics) 23 3

Subtotal 1828 179

Total 1982 201

What’s more, the frequency of Southern Min use of males and females is also

unequally distributed in the programs. In half of all the programs (TV Programs 1, 3,

4, 5, 10 and 11), no women produce any Southern Min, as shown in Table 4.17. By

contrast, also as Table 4.17 presents, males do not produce any Southern Min only in

the 12 programs, the amount of Southern Min used by males even exceeds 100 characters. These three programs are programs about socio-political and social issues. Additionally, a male speaker in TV Program 9 keeps speaking Southern Min in all his turns without switching back to Mandarin because of his ethnic identity and political stance. This male speaker is a member of Democratic Progressive Party, which strongly proposes Taiwan’s localization and advocates the use of Southern Min.

Moreover, he works as National Policy Consultant and he speaks Southern Min to display his political stance and ethnic identity. Therefore, this program has a large amount of Southern Min use.

The unequal distribution of Southern Min is also obvious between political talk shows and soft talk shows. Compared with socio-political talk shows, very little Southern Min is spoken by males and females in soft talk shows. On the average, each male speaks 11.9 times more Southern Min in socio-political talk shows than in soft ones (130.6:11). Similarly, each female also speaks much more Southern Min in socio-political talk show. On the average, a female speaks 22.4 characters of Southern Min in socio-political talk shows whereas just one character in soft ones.

Please see Table 4.16.

As mentioned above, in the present study males use much more Southern Min

than females do. As previous studies suggest, males use more vernacular language

while females use more standard dialect. In addition, the degree of using Southern

Min depends on the speaker’s ethnic identity and political stance. Social and

political factors can strongly influence the attitudes to language. If the speakers

belong to DPP (Democratic Progressive Party), they tend to speak Southern Min in

the TV programs to exhibit their ethnic and political identity. Just as Holmes (1992a)

Besides, men and women speak Southern Min also in a different way. In the data of the present study, men use more code-switching but women use more code-mixing

. By Bokamba’s definition (1989), code-switching is intersentential and code-mixing is intrasentential. That means, in the present study men tend to continue speaking Southern Min for several sentences in a turn, but women only speak several characters within a sentence. Some male speakers in TV Program 9 (兩代電力公司 liang dai dian li gong si ‘Two-generation power company’) even keep speaking Southern Min for a whole turn when they switch to Southern Min, but none of the female speakers do so. Women usually only speak several vernacular words or may stop in a few sentences. The following are two examples of Southern Min use:

(14) TV Program 12, Male speaker 28:

你 去 看 三 場, 他們 都 贏 三 場, ni qu kan san chang tamen do ying san chang you go watch three period they all win three period 贏 兩 場 就 說, 哎 喲,

ying liang chang jiu shuo aiyou win two period say ouch

<

@ 阿扁 阿 真 正 很 厲害 喔 , 不是 帶賽 喔 @

>.

abin a jinjian jin lihai o msi daisai o PN PR really very great not unlucky PR

8. Bokamba (1989) defined code-switch as the mixing of words, phrases and sentences, from two distinct grammatical (sub)systems across sentence boundaries within the same speech event.

He defines code-mixing as the embedding of various linguistic units such as affixes, words,

phrases and clauses from two distinct grammatical (sub)systems within the same sentences and

speech event. In other words, code-switch is intersentential and code-mixing is intrasentential

但是, 但是 你 明明 就是…

danshi, danshi ni mingming jiushi…

but but you obvious are

‘You go to watch 3 periods and they win 3 periods, if they win 2 periods and say, ouch, A-Ban is really mighty, not out of luck but, but you obviously are... ’

(15) TV Program 12, Female speaker 29:

他 比較 講說, <

這 邊 選 -

>, 這 邊 要 加強, ta bijiao jiangshuo ji bing xuan zhe bian yao jiaqiang

he comparatively say this side elect this side need strengthen

這 邊 要 加強, 都 是 選票 的 考量…

zhe bian yao jiaqiang dou shi xuanpiao de kaoliang this side need strengthen all are ballot consideration

‘He tends to say, this side elect-, this side should be reinforced, and this side should be reinforced, which are all the consideration of ballots.’

Example (1) is a type of code-switching whereas example (2) is of code-mixing type. Example (1) is the pattern that most male speakers in the present study use when they are speaking Southern Min, which contains more than one Southern Min sentence. By contrast, example (2) is the pattern that more female speakers are apt to use, which includes just several Southern Min characters.

In brief, as previous studies (Holmes, 1992a; Hudson, 1996) have suggested, the

present study observes that males prefer vernacular language while females tend to

use standard forms. The males speak a greater amount of Southern Min than

females do though Southern Min is considered a nonstandard form in the public

domain, (for example, TV programs). As a whole, males still use more Southern

Min than females to show their political stance and represent their political and social

4.10 Context as an Influence on Speech Style

Language varies according to contexts and addressees (Giles & Smith, 1979;

Holmes, 1992a). The present study has found that the speakers change their ways of talk when the context is different. If men’s and women’s talk are examined in socio-political talk shows and soft talk shows, individually, it is found that their ways of speaking vary in different types of programs. First, in female speech, as demonstrated in Table 4.18, on the average female speakers talk much more in the total amount in socio-political talk shows than in soft ones (2236 characters to 1235 characters). Second, females use very few tag questions (0.38) and use no empty adjectives in socio-political talk shows. Third, female speakers speak much more Southern Min in socio-political talk shows than in soft talk shows (179 characters to 22 characters).

Since people adjust their speech styles according to different speech contexts and different participants, it is presumable that women may adjust their linguistic behavior in socio-political talk shows in order to have the same performance as male participants. It is asserted that men are more eloquent than women in public contexts (Tannen, 1990; Holmes, 1992b). Besides, because socio-political talk shows mainly discuss so-called “masculine-oriented topics” such as social and political issues, women need to change their ways of talk to be more like men’s to fit the context and topics. Therefore, in order to exhibit confidence as well as professionalism on TV programs, women choose to talk more than they normally do. They do so to reveal that they are as knowledgeable as men. Moreover, women avoid using too much

“women’s language”, such as tag questions and empty adjectives in socio-political

talk shows but choose manlike speaking manners. What’s more, in socio-political

Table 4.18 The average occurrences of each analytic category of each speaker in different types of programs in the collected data (average number)

Analytic

Programs categories

Tag questions (number)

Polite forms (number)

Amount of talk (characters)

Southern Min (characters) Soft

talk shows 1.57 0.50 1586 11

Male speaker

(N=28) Socio-political

talk shows 1.64 0.36 2457.1 130.6

Soft

talk shows 1.19 0.10 1235.47 1

Female speaker

(N=29) Socio-political

talk shows 0.38 0.25 2236 22.4

Male speaker

(N=28)

1.57 0.43 2021.9 70.8

Female speaker (N=29)

The total data

0.97 0.14 1511.5 6.9

In addition, when comparing men’s and women’s talk in the total data, it is found that in both socio-political talk shows and soft talk shows, women’s use of polite forms is less than men’s, though the occurrences are not many. In general, women are supposed to be more polite than men and use more polite language than men do.

However, in the present study, it is not the case. In public contexts such as TV talk

shows, women would not like to be too polite, gentle and submissive, which are the

As Lakoff claims (1990), “Yet women in public positions, in interviews on television, seem to be learning that they don’t have to be nice ladies and take it.” Thus, women in TV talk shows adjust their speaking styles, from ladylike to manlike.

Likewise, males’ speech styles vary in different types of programs. As Table 4.18 shows, male speakers talk in a less amount and use less Southern Min in soft talk shows than in socio-political ones. In soft talk shows, a man on the average speaks 1586 characters and 11 characters of Southern Min. Yet, in socio-political talk shows, a man on the average speak 2458 characters and 130.6 characters of Southern Min, which are 1.55 times more characters in amount and 11.9 times more Southern Min than in soft talk shows. Men talk less and use less Southern Min in soft talk shows is because soft talk shows are traditionally regarded as “women’s programs”.

In such programs topics are related to soft issues, such as horoscope and daily expense. To speak properly in “women’s programs”, men alter their ways of talk to be more like women’s. As a result, men not only use more women’s linguistic features such as polite forms than usual but also reduce masculine conversational style.

For example, they decrease the amount of Southern Min use. That’s why in soft talk shows they do not talk as much as they do in socio-political talk shows. Besides, they decrease the use of Southern Min because Southern Min is considered man’s language, which is not favorable in “women’s programs.”

In summary, the present study suggests that males’ or females’ speech styles are contexts-driven. Males and females shift their linguistic pattern as the contexts change. The male participants in the present study use more “women’s linguistic features” than usual and reduce “masculine conversational style” in soft talk shows.

Similarly, female participants change their ways of talk to be more like men’s in