• 沒有找到結果。

Tiau11(條) is used for long and flexible entities such as towels

N/A
N/A
Protected

Academic year: 2021

Share "Tiau11(條) is used for long and flexible entities such as towels"

Copied!
14
0
0

加載中.... (立即查看全文)

全文

(1)

Chapter 5

The Categorical Structure of Classifiers Gien24, Dung5, and Zak2

5.1 Introduction

The majority of Hakka individual classifiers collocate with inanimate nouns, as shown in Table 3.5. Among them, the shape classifiers have received much attention because the physical properties of the objects they co-occur with are more easily perceived such as longness, flatness, and roundness. It is also easier for us to construct the prototypical structure of them. For example, Wu (2001) has done a research on shape classifiers for long objects in Hakka, including tiau11(條), ki24 (枝), and moi11 (枚). In her study, the membership in each classifier category is clearly manifested. Tiau11(條) is used for long and flexible entities such as towels; ki24 (枝) is used for long and rigid objects like pens; moi11 (枚) is used for long, rigid, and especially small objects like needles.

However, a great number of inanimate classifiers are defined mainly by the functional properties of the objects they associate, since the functional properties of their associated objects are more easily perceived than their physical characteristics.

For instance, tai (臺), pun (本), king (間) in Taiwan Min are only labeled as functional classifiers in Wang’s study (2001:46), and are treated as “a classifier for transport,”“a classifier for written materials,” and “a classifier for buildings and shops,”

respectively. This phenomenon triggers an interesting question: Can inanimate classifiers, especially those which are more functionally-based, be semantically analyzed in terms of more specific features, instead of being defined with a simple definition--a classifier for a certain function?

In the present chapter, we intend to analyze the so-called functional classifiers gien24 (間), dung5(棟), zak2(座) in Hakka in ways like what the previous studies

(2)

have done for shape classifiers (e.g. Tai and Wang 1990, Tai and Chao 1994). The prototype theory will serve as the analysis foundation for these classifiers. At the same time, metaphorical extension or even cultural factors will be used to explain the extended members of the prototypical ones in each classifier category when necessary.

We wish to argue that the three classifiers are also definable to some extent in terms of features like shape, size, and function. In doing so, we will address the following questions.

1. What are the prototypical members and peripheral members of gien24 (間), dung5 (棟), and zak2 (座)?

2. How do the semantic properties such as shape, size, and function have influence on the selection of an appropriate classifier? Which semantic feature is primary and which is secondary in each classifier category?

This chapter is organized as follows. Section 5.2 presents Matsumoto’s (1993) study on classifiers for buildings in Japanese. The result shows that semantic features such as shape, size, and function decide the selection of an appropriate building classifier. In the present work, we are going to find out how these features can apply to the case in Hakka. Section 5.3 examines the semantic dimensions of gien24 (間), dung5 (棟), and zak2 (座) in Hakka respectively. Section5.4 concludes the chapter.

5.2 Matsumoto (1993)

Matsumoto’s(1993)study on –ken and –mune, which are classifiers for houses and buildings in Japanese, intends to show that functional classifiers can also be analyzed as shape classifiers do, and are definable to some extent. He designs an experiment on the acceptability of –ken and –mune for various photographs of buildings. After looking at each photo of the buildings, the subjects judge whether the collocation between the building classifiers and the entities is appropriate or not. 20

(3)

native speakers of Japanese participated in the experiment, and the acceptability values for each classifier selection range from 0 (completely unacceptable) to 6 (completely acceptable). The analysis of –ken and –mune is conducted in the framework of prototype theory, and the result shows that each of them displays different prototype conditions.

Matsumoto found that the acceptability of –ken varies according to the function and size of the buildings. The result indicates that houses for dwelling, shops and stores for commercial activities especially drew high acceptability of –ken, while buildings that are used for public purposes such as post offices and churches and those for miscellaneous purposes like office buildings drew relatively low acceptability, and buildings that are not primarily for human use such as store houses drew the lowest acceptability. The semantics of –ken can be defined by a more precise statement about the function of its associated buildings. The primary feature of -ken is [+ human use:

dwelling or commercial purposes]. For example, the Shinto shrine is used for a public purpose, but its interior is not used by human beings at all. Thus, the use of –ken for the Shinto shrine drew completely unacceptability in the test. In conclusion, Matsumoto (1993) contends that –ken, as a classifier for the buildings used by human beings has the prototype condition--used for dwelling or commercial purposes.

The effect of size difference is clearly observed when buildings with the same function are compared, so the size feature serves as the secondary feature in the –ken family. The closer the size is to that of the ordinary house, the higher the acceptability is. For example, a shop in a medium size got 5.8 points in the experiment with 6 points as the full points, while a large grocery store got only 3.1 points.

-Mune is another Japanese classifier for buildings studied by Matsumoto. Its prototype structure is found to be different from that of –ken. The use of –mune for the same set of building photos tested by –ken shows that the acceptability is the

(4)

highest for relatively large buildings. As for the function of buildings, it does not affect the acceptability ratings. Another semantic property that affects the acceptability of –mune is shape. Many speakers find that –mune is more acceptable for those buildings that are flat in shape like the chicken house and the large storehouse since both of them are nottallerthan one floor.Matsumoto’sfindings suggest that the prototype condition of –mune is: moderately large and relatively flat in shape. They are equally important in the –mune category.

Matsumoto (1993) proves that functional classifiers can also be semantically analyzed with semantic features. The selection of –ken and –mune may not just depend on the simple statement such as classifiers for buildings, but on some more specific semantic features of the associated buildings. Based on Matsumoto’sstudy, we turn to investigate the prototypical structure of the classifiers for buildings in Hakka. We aim to find that classifiers for buildings in Hakka are also of high explanatory value in that their semantic meanings can be further decided by specific semantic features.

5.3 The Semantic Dimensions of Gien24(間), Dung5 (棟), and Zak2 (座) in Hakka In this section, we examine the categorical structure of gien24 (間), dung5 (棟), and zak2 ( 座 ) respectively. Based on the data, we are going to find out the prototypical members of them. Also, we will investigate how the semantic properties such as shape, size, and function influence the proper selection of the classifiers for a particular building.

5.3.1 The Semantic Dimensions of Gien24 (間)

The original meaning of gien24 is a separate space divided by building construction (Syu 2003). An enclosed space with interior may be viewed as a single

(5)

room. Afterwards, gien24 is used as a classifier for counting rooms or houses (with a number of rooms in it) in Hakka.

Buildings which collocate with gien24 are presented in Table 5.1. The so-called ordinary building refers to houses or rooms in average size such as restrooms or apartments.

Table 5.1 Nouns collocate with gien24

Feature Category Group Examples

1. Houses or rooms for dwelling

fong11gien24 房間 ‘room,’

gun24 ngi5 公寓 ’apartment,’

li24 sa5 旅館 ‘hotel’

2. Rooms, houses, or buildings for public purposes

gau5 siit11 教室 ‘classroom,’

pan5 gung24 siit11 辦公室

‘office’ngiun11 hing11 銀行

‘bank,’hok5 gau31 學校

‘school,’yi24 yen5 醫院

‘hospital,’ziin31 so31 診所

‘clinic,’meu5 廟 ‘temple,’

yiu11 kiuk5 郵局 ‘post office,’

bu31 sip5 ban5 補習班 ‘cram school,’lo11 ngin11 yen5 老人 院 ‘old people caring center,’

bok11 vut5 gon31 博物館

‘museum, [ordinary

size]

[+human use]

3. Shops and stores for ’cap5 fo5 diam5 雜貨店

(6)

commercial activities ‘grocery store,’con24 tang24 餐 廳 ‘restaurant, ’ga24 bi24 diam5 咖啡廳 ‘coffee shop,’hi5 yen5 電影院 ’theater’, bak11 fo5 gung24 sii24 百貨公司

‘department store’

Gien24 is a classifier used to enumerate various buildings or rooms in ordinary size such as apartments which are not as tall as office buildings. Thus, the size feature [+ordinary] decides the central members in the gien24 family. Gien24 (間) collocates with ordinary houses or rooms for dwelling in average size as in yit2 gien24 fong11 gien24 (一間房間)‘a room,’yit2 gien24 gun24 ngi5 (一間公寓) ‘an apartment.’

Both rooms and apartments can be considered as possessing some kind of enclosed environment.

Different human purposes are not significant in the selection of gien24. We can see that gien24 also collocates with houses or rooms for public purposes as in yit2 gien24 gau5 siit11 (一間教室)‘a classroom,’and yit2 gien24 pan5 gung24 siit11 (一 間辦公室)‘an office.’Thesizeofaclassroom iscloseto thatofaliving room.

Therefore, gien24 can be used to classify classrooms.

Next, we are going to discuss more examples. In Table 5.1, we can see gien24 is also used for public purposes as in yit2 gien24 ngiun11 hing11 一間銀行‘a bank’;

yit2 gien24 hok5 gau31 一間學校‘a school’; yit2 gien24 yi24 yen5 一間醫院‘a hospital’; yit2 gien24 ziin31 so31 一間診所‘a clinic’; yit2 gien24 meu5 一間廟

‘a temple’; yit2 gien24 yiu11 kiuk5 一間郵局‘a post office’; yit2 gien24 bu31 sip5 ban5 一間補習班‘a cram school’; yit2 gien24 lo11 ngin11 yen5 一間老人院

‘ an old people caring center’; yit2 gien24 bok11 vut5 gon31 一間 博 物 館 ’a

(7)

museum.’ Members here have two characteristics. First, they are used for public service, functioning as an institution such as a school, a temple, a bank, or a post office. Second, the activities taken in these institutions are more serious than daily routines, requiring more regulations or formality. Third, the space is larger.

Gien24 is also used for shops and stores for commercial activities as in yit2 gien24 cap5 fo5 diam5 一間雜貨店‘ a grocery store’; yit2 gien24 con24 tang24 一間餐廳‘ a restaurant’; yit2 gien24 ga24 bi24 diam5 一間咖啡廳‘a coffee shop’; yit2 gien24 hi5 yen5 一間電影院‘a theater’; yit2 gien24 bak11 fo5 gung24 sii24 一間百貨公司‘a department store.’Objects here differ from those above. First, they are not used for dwelling but for commercial activities; in other words, the function of the buildings changes. However, they are still used for human beings. The size of the objects here are bigger since there are a lot more people involved in commercial activities, and a larger space is definitely needed.

From above, we conclude that gien24 is a classifier especially used to classify buildings for human use. The prototypical property of gien24 is: [+ ordinary size].

Gien24 can be extended for classifying buildings with bigger space such as a house or an apartment. Both of them are still buildings for human use and possess some kind of enclosed space at the same time. One thing we should note is that only buildings in which human beings dwell or hold activities collocate with gien24. Therefore, the collocation between gien24 and buildings for non-human beings such as animals or objects is odd. Consider the following examples, *yit2 gien24 diau24 e31 lung11 (一 間鳥籠)‘a bird cage,’*yit2 gien24 cong24 ku5 (一間倉庫)‘a storehouse.’

5.3.2 The Semantic Dimensions of Dung5 (棟)

According to the Hakka Dictionary of Taiwan (Syu 2003), dung5 (棟) originally refers to the central pillars of the house, which is the most essential part of a building.

(8)

Dung5 (棟) is then used as a classifier for buildings, especially large ones, and these large buildings also possess some kind of enclosed space. Consider the nouns collocating with dung5 in Table 5.2.

Table 5.2 Nouns collocate with dung5

Feature Category Group Examples

1. Buildings for dwelling gun24 ngi5 公寓 ’apartment,’

vuk11 e31 屋仔 ’house,’tai5 ha5 大廈 ‘mansion’ pet5 zong24 別莊 ‘villa’

2. Buildings for

miscellaneous purposes

pan5 gung24 tai5 leu11 辦公大樓

‘office building’

[+human use]

3. Shops and stores for commercial activities

bak11 fo5 gung24 sii24 百貨公司

‘department store’

+ large or +tall

[- human use]

1. Storage gung24 cong31 工廠 ‘factory’

The collocation of dung5 is rather restrictive, and different human purposes do not seem to be influential in the selection of dung5. Dung5 is typically used as a classifier for a large building such as a mansion or a villa. It is also extended to collocate with large buildings for miscellaneous purposes, especially those which are very tall in height, as in yit2 dung5 pan5 gung24 tai5 leu11 一棟辦公大樓 ‘an office building,’and for commercial activities such as a department store, which is usually very large and tall. Based on the resemblance in large size, dung5 is naturally

(9)

extended for buildings in large size for non-human purposes as in yit2 dung5 gung24 cong31 一棟工廠 ‘a factory.’Dung5 can be defined by the following semantic features: [+large] or [+tall]. Most members in the dung5 family possess the feature [+large], and the feature in height [+ tall] is equally important since we also use dung5 for buildings which are very tall but not large in the base, i.e. they do not cover much land; for example, yit2 dung5 101 tai5 leu11 (一棟台北 101 大樓) ‘the Taipei 101 building.’

5.3.3 The Semantic Dimensions of Zak2 (座)

Zak2, as a noun meaning a base originally (Lu 1988), typically collocates with constructions or buildings with a wide base such as a bridge or a dam. The examples are illustrated in Table 5.3.

Table 5.3 Nouns collocate with zak2

Feature Category Group Examples

1. constructions with flat base

san24 山 ‘a mountain’kieu11 橋 ‘a bridge’ho11 ba5 河壩 ‘a dam’

2. buildings with wide space

fa24 yen11 花園 ‘garden,’

tung24 vut5 yen11 動物園 ‘zoo’

[+ wide base] [- human dwelling]

3. for animals or objects dwelling

gung24 cong31 工廠 ‘factory,’

cong24 ku5 倉‘storehouse’

gieu31 deu5 狗屋 ‘dog house,’

diau24e31 lung11鳥籠 ‘bird cage’

The prototypical condition of zak2 is: [+ wide base]--a building possesses a

(10)

wide base. For example, yit2 zak2 fa24 yen11 一座花園 ‘a garden,’yit2 zak2 tung24 vut5 yen11 一座動物園 ‘a zoo.’Both of them cover wide space.

[- Human dwelling] explains the reason why zak2 can be extended to classify a dog house or a birdcage once in a while as in yit2 zak2 gieu31 deu5 一座狗屋 ‘a dog house.’We may see the effect [+ wide base] does not work here anymore since a dog house is small and does not possess a wide base. Zak2 is also found to classify a factory or a storehouse since both of them are not typically used for human dwelling, and are usually not very tall in height.

One thing we should note is that zak2 in Hakka does not collocate with buildings or construction for human dwelling; therefore, people may tend to say yit2 zak2 gim24 sii5 tap11 一座金字塔 ‘a pyramid, rather than *yit2 gien24 gim24 sii5 tap11 一間金字塔, or *yit2 dung5 gim24 sii5 tap11 一棟金字塔, and yit2 gien24 gun24 ngi5 一間公寓 ‘an apartment,’instead of *yit2 zak2 gun24 ngi5 一座公寓

‘an apartment.’

5.3.4 Summary

This section demonstrates that the classifiers gien24, dung5 and zak2 in Hakka can be semantically analyzed with semantic features instead of just being dubbed as classifiers for buildings.

We have found that gien24 and dung5 are both classifiers for buildings with enclosed space for human dwelling use. The prototypical members of gien24 are relatively close the size of an ordinary house. In identifying large buildings, sometimes tall in height, dung5 is preferred as in yit2 dung5 tai5 ha 一棟大廈 ‘a mansion’yit2 dung5 pet5 zong24 一棟別莊 ‘a villa.’Sometimes, dung5 is extended to be used for large buildings for objects storage such as a factory.

(11)

Zak2 in Hakka is typically used for constructions with a wide base, as in yit2 zak2 san24 一座山 ‘a mountain’ yit2 zak2 kieu11 一座橋 ‘a bridge,’yit2 zak2 ho11 ba5 一座河壩 ‘a dam.’Based on the meaning of zak2, a wide base, all of the prototypical objects seem to have a wide base covering much land. Then, based on the shape similarity, zak2 can be used for a garden and a zoo which are usually not taller than one or two floors and cover wide space at the same time. In the extension category, zak2 even collocates with entities such as a birdcage or a doghouse. Under this condition, the semantic property of shape, [+ wide space], is not taken into consideration. It is the category, [-human dwelling], that has influence on the selection of zak2 here.

In sum, The prototype structure of gien24(間) , dung5 (棟), zak2 (座) can be simply presented in figure 5.1

Figure 5.1 The prototype structure of gien24 (間), dung5 (棟), and zak2 (座)

5.4 Conclusion

The semantic study of gien24, dung5, and zak2 cast some insights on the categorical structure of building classifiers. We have proved that they are definable to

Buildings

[+enclosed space where human dwell] [-enclosed space where human dwell]

[+ordinary size] [+large] or [+tall] [+wide base]

gien24 dung5 zak2

(12)

some extent. They even possess some specific physical properties for classification such as size and height. According to figure 5.1, there are two major groups in the Building category. Gien24 and dung5 are used for classifying buildings for human dwelling, and zak2 is for those which are [-human dwelling].

According to the result of the study, we may see some overlap between gien24 and dung5, but there will not be any overlap between gien24 and zak2 or dung5 and zak2 because both of gien24 and dung5 are for human use and zak2 is for non-human purposes.

Generally speaking, with respect to large buildings for human dwelling, dung5 is preferred over gien24 and zak2. Dung5 is acceptable in collocation with buildings for commercial activities or public purposes which are generally collocated with gein24 because they are usually large in size, as shown in (1).

(1) yit2 dung5 tai5 ha 一棟大廈 ‘a mansion’

yit2 dung5 pan5 gung24 tai5 leu11 一棟辦公大樓 ‘an office building’

yit2 dung5 bu31 sip5 ban5 一棟補習班 ‘a cram school yit2 dung5 hi5 yen5 一棟電影院 ’a theater’,

yit2 dung5 bak11 fo5 gung24 sii24 一棟百貨公司 ‘a department store’

yit2 dung5 yi24 yen24 一棟醫院 ‘a hospital’

We should note that buildings in (1) all possess enclosed space in which human beings dwell or hold activities. People prefer dung5 to gien24 for classifying these large buildings in (1). When speakers use dung5 to classify large buildings, they intend to focus on their appearance. However, it is also possible to say yit2 gien24 tai5 ha5 一間大廈 ‘a mansion,’or yit2 gien24 tai5 leu11 一間大樓 ‘an office building’because gien24 is also used to classify buildings with enclosed space. Dung5

(13)

does not collocate with buildings in small size or without enclosed space as shown in (2).

(2) *yit2 dung5 gieu31 deu5 一棟狗屋 ‘a dog house’

*yit2 dung5 fa24 yen11 一棟花園 ‘a garden’

*yit2 dung5 tung24 vut5 yen11 一棟動物園 ‘a zoo’

However, there are two exceptions in (3) because factories and storehouse also have enclosed space.

(3)

yit2 dung5 gung24 cong31 一棟工廠 ‘a factory’

yit2 dung5 cong24 ku5 一棟倉庫 ‘a storehouse’

From above, we may state that dung5 and gien24 are interchangeable, but dung5 and zak2 are not.

Zak2 is typically used for big, magnificent constructions or buildings with wide flat base in a sense. Zak2 can also collocate with entities for non-human purposes, as shown is (4).

(4) yit2 zak2 san24 一座山 ‘a mountain’

yit2 zak2 kieu11 一座橋 ‘a bridge’

yit2 zak2 gung24 cong31 一座工廠 ‘a factory’

yit2 zak2 cong24 ku5 一座倉庫 ‘a storehouse’

yit2 zak2 gieu31 deu5 一座狗屋 ‘a dog house’

Zak2 does not collocate with houses or rooms in ordinary size or buildings for human activities as shown in (5). Moreover, all building objects here have enclosed space.

(14)

(5) *yit2 zak2 fong11 gien24 一座房間 ‘a room’

*yit2 zak2 gau5 siit11 一座教室 ‘a classroom’

*yit2 zak2 pan5 gung24 siit11 一座辦公室 ‘an office’

*yit2 zak2 yi24 yen5 一座醫院 ‘a hospital’

*yit2 zak2 hok5 gau31 一座學校 ‘a school’

*yit2 zak2 hi5 yen5 一座電影院 ‘a theater’

*yit2 zak2 pan5 gung24 tai5 leu11 一座辦公大樓 ‘an office building’

There is an exceptional case in the gien24 category, that is, *yit2 gien24 fo31 fong11 (一間夥房) ‘a traditional house.’Fo31 fong11 (夥房) is a traditional house in Hakka culture. Though it is also an ordinary house with a bit larger space, Hakka people tend to use zak2 (座) for counting it metaphorically. It may be due to the Hakka culture, in which Hakka people consider their home as very significant, like a base of their life. Therefore, they use the classifier zak2 (座), which is usually used for objects with a wide firm base, to count the traditional house. Another possible reason is that the traditional house, which consists of several houses, covers wide space;

therefore, Hakka people use zak2 for classification.

The interchangeability phenomenon of gien24, dung5, and zak2 supports the idea that the human conceptual mechanism can be revealed by classifiers. We find that in every interchangeable phrase, the semantic features of the classifier in question resemble the expected classifier. In other words, when a speaker chooses another classifier to replace the expected one, he does not do it as he pleases but based on the semantic feature of the entities.

數據

Table 5.2 Nouns collocate with dung5
Table 5.3 Nouns collocate with zak2
Figure 5.1 The prototype structure of gien24 (間), dung5 (棟), and zak2 (座)

參考文獻

相關文件

Besides adopting effective strategies, such as flexible grouping, differentiating learning tasks and promoting self-access learning, to cater for learners’ diverse needs, due

How would this task help students see how to adjust their learning practices in order to improve?..

The long-term solution may be to have adequate training for local teachers, however, before an adequate number of local teachers are trained it is expedient to recruit large numbers

Consequently, Technology Education is characterized by learning activities which provide students with authentic experiences in various technological areas such as

This elective is for those students with a strong interest in weather and climate. It aims at providing a more academic and systematic foundation for students’ further study pursuit

obtained by the Disk (Cylinder ) topology solutions. When there are blue and red S finite with same R, we choose the larger one. For large R, it obeys volume law which is same

For pedagogical purposes, let us start consideration from a simple one-dimensional (1D) system, where electrons are confined to a chain parallel to the x axis. As it is well known

To improve the convergence of difference methods, one way is selected difference-equations in such that their local truncation errors are O(h p ) for as large a value of p as