O RGANIZATION - 學習泰雅語動態事件編碼:以新竹某國小族語學習為例

CHAPTER 1 INTRODUCTION

1.3 O RGANIZATION

The next chapter begins with a general review on the typological studies on motion events. And then it moves on to a more language-specific aspect, describing the motion event encoding in Chinese and in Atayal. This chapter also reviews the studies on the acquisition of motion event encoding, and introduces the Heritage Language Learning Program in Taiwan. In Chapter 3, a more detailed description of the children

participating in this study will be provided, in particular their language background.

Chapter 4 will cover the methodology of the narration task, testing children's production of motion events. Details would be provided concerning the materials and the procedure of the narration task. The results of the task would be presented and followed by a brief summary and discussion. Chapter 5 will provide the implementation of two

comprehension tasks, namely, listening comprehension task and act-out task.

Methodology of the two tasks and the results would be provided in detail. Influences from language experience will be taken into consideration in Chapter 6 and the concluding remarks are given at the end.

Chapter 2 Literature Review

2.1 Linguistic Typology of Motion Event

A motion event typically involves an entity moving from one place to another.

According to Talmy (1985), a motion event consists of four basic semantic components, which are motion, figure (one object), ground (the reference with respect to which the figure moves), and path (the course followed by the figure with respect to the ground). An example of motion event in English is given below:

(1) The girl

RAN OUT of the house.

Figure Motion+Manner Path Ground

(Example from Naigles et al., 1998)

Path is paid special attention to and is viewed as the core schema to determine typological differences in motion event encoding. According to Talmy (1985, 1991, 2000), languages can be classified into two types, the satellite-framed languages

(S-languages) and the verb-framed languages (V-languages), depending on how the core schema ‘path’ is encoded in the motion events. In an S-language, the core schema is encoded in satellites, or particles. For example, in the English sentences I blew the ant off

my plate and the rock rolled down the hill, the path components off and down, are encoded

in a peripheral position by a particle. English, German, and Chinese all belong to the S-languages. In a V-language, the core schema, path, is encoded in the main verbs, and if there is a necessity of adding manner of the motion, then manner will be placed in a peripheral position. Take Spanish as an example, in this sentence la botella salió de la

cueva (flotando) “the bottle move-out the cave”, the path element is expressed in the main

verb and no manner element is specified. Languages like Spanish, Italian, and French fall into this category. Following Talmy’s classification, a number of studies were conducted to further examine the lexicalization patterns of these two different types of

languages.While the dichotomous classification can distinguish most of languages, some languages appear to fall out of these two categories, which challenge Talmy’s

classification. Thus, Slobin (2004) revised Talmy’s dichotomous typology by adding a third category: the equipollently-framed language, in which both path and manner are expressed with equivalent grammatical forms. The serial verb construction in Chinese is a typical example, as is shown below.

(2) qing1-wa1

tiao4 chu1 ping2-zi

frog jump exit jar

‘the frog jump out of the jar.’

Both the manner verb tiao4 and the path verb chu1 are equally important, since it is not correct to say qing1-wa1 tiao4 ping2-zi “the frog jump the jar”. In this study, we mainly adopted the three-way typology (Slobin, 2004) and classified Chinese as an equipollently-framed language, in which manner and path are equally important though whether a dichotomous typology or a tripartite typology is more suitable for the

classification of languages all over the world is still under debate and needs further investigation according to previous studies (Slobin, 2004; Talmy, 1985, 1991, 2000).

2.2 Motion Events in Chinese

Chinese is classififed as a satellite-framed language in Talmy’s two-way motion

event typology because the path component is viewed as a particle when used with the manner verb to form a motion event construction in Chinese. For example, tiao3 jin4

ping2-zi “jump enter the jar”. In Talmy’s classification, tiao4 “to jump” is the main verb,

and jin4 “enter” functions more like a particle or a preposition, as the preposition into in English. However, Tai (2003) held a different view toward the classification. He argued that Chinese is not so much a satellite-framed language as a verb-framed language. Given the same example, it is grammatical to say jin4 ping2-zi “enter the jar”, but the expression

tiao4 ping2-zi “jump the jar” is ungrammatical. Tai pointed out that the path element in

Chinese is more crucial and essential in a motion event construction since it can be used alone to express a motion event, while manner by itself cannot form a correct expression of motion event. Therefore, the path element should not be taken as a particle; instead, it should also be viewed as a main verb. Based on this argumentation, Chinese encodes a motion event by using a combination of a manner verb, a path verb, and a deixis, such as

tiao4 chu1 lai2 “jump exit come”. Manner and path all belong to verb, while deixis

signifies a reference point of direction from the perspective of speakers. Because of the controversial nature of the serial verb construction, Slobin (2004) proposed a third type of languages, namely the equipollently-framed languages, in which both the manner and the path elements are encoded with equivalent grammatical forms.

Chen (2005) examined the motion event encoding across different age groups of Mandarin-speaking children with the story Frog, where are you?. Both data from children and adults supported the view that Chinese should be an equipollently-framed language.

Chen conducted the analyses following Slobin’s study (1996), which considered various elements in the encoding of motion events, including types of motion verbs used and the presence of the ground element. Motion verbs were classified into three types: the manner

of motion verb, the path verb, and the non-motion neutral verb.³ With regard to the ground information, if it was specified in the motion event, then the clause was marked as

“plus ground”, otherwise, “minus ground”. It was found that Chinese possesses the characteristics of both the S-languages and the V-languages. More manner verbs (72% of the total) were found in Chinese as was shown in English (74% of the total), while less ground information (minus ground 48% vs. plus ground 52%) was specified, as was shown in Spanish (minus ground 37% vs. plus ground 63%). Moreover, frequent uses of the two deictics lai2 ‘come’ and qu4 ‘go’ were found in many motion event descriptions.

The Manner+Path+Deixis construction and the Path+Deixis construction were the most common constructions observed in the elicitation.

2.3 Motion Events in the Austronesian Languages

Not much attention was paid to the expression of motion events in Austronesian languages until Huang and Tanangkingsing (2005), in which motion events were examined in six Austronesian languages, including Cebuano, Tagalog, Saisiyat, Squliq Atayal, Malay, and Tsou. Though these languages show minor differences in their constructions of motion events, they are similar in the dominate use of path verbs and the absence of ground information in motion event encoding. It is concluded that Cebuano and Tagalog are most strongly Verb-framed languages. Squliq Atayal is still more of a Verb-framed language than a Satellite-framed language. Saisiyat shows an incipient characteristic of macro-event language, like Tsou. Huang and Tanangkingsing (2005) further proposed to conceptuali e Talmy’s model as a grid, with the vertical axis

representing path salience, and horizontal representing manner salience, so that the exact

3 Examples of the manner of motion verbs are like pa2 ‘climb,’ pao3 ‘fall down,’ and fei ‘fly,’ and examples of the path verbs are like dao4 ‘arrive,’ luo4 ‘drop,’ and hui2 ‘return.’ The non-motion neutral verbs are like pa1 ‘bend over,’ tang3 ‘lie,’ and na2 ‘take.’

position of each language relative to other languages could be clearly plotted, as shown in Figure 1 below.

Figure 1 Path and Manner Salience of Six Austronesian Languages (Adopted from

Huang & Tanangkingsing, 2005)

2.4 The Acquisition of Motion Events

In addition to the focus on the typological classification of motion events, some researchers have examined children’s acquisition of motion events encoding. Motion events for children seem to be fundamental since children can associate them to similar concepts in cognition.

alişkan and Slobin (1999) examined how 3 to 11 year-old children speaking different languages, including English, Spanish, and Turkish, encoded motion events by asking them to narrate the frog story. They found that children of different language backgrounds showed the development of manner and path verbs at divergent ages.

Comparing with Spanish children and Turkish children, there was a higher tendency for English children to encode the manner element in the main verb and as children grew older, they would form a preferred motion event construction.

Hickmann and Hendriks’ finding (2010) also supported this view. They studied the

acquisition of motion event in children aged from three to ten in two languages, English and French. Cross-linguistic differences were observed. First, the English speakers expressed motion events in a compact way by using multiple types of ground information.

Also, compared with the French speaker, they tended to encode manner in the verb root and path was encoded in other devices (not in the verb root).

Moreover, Slobin (2003) and Hohenstein (2005) also found that children learning V-language or S-language encoded motion events in different ways. Hohenstein (2005) investigated the lexical bias in the 3.5 and 7 year-old English and Spanish children. He found that only the older age group showed a preference toward different lexicalization patterns, which implied that when children grow older, they form a preferred construction.

English-speaking children at the age of 7 tended to shift to a more manner-oriented perspective in their similarity judgments of motion events, while Spanish-speaking children at the same age showed no preference. Slobin (2003) examined the encoding of visual path, such as look into/through, and physical path, such as walk into/through in children speaking English, Russian, Spanish and Turkish. It was found that children of the verb-framed languages, such as Spanish and Turkish, analyzed the visual paths into fewer components. However, unlike Spanish- and Turkish-speaking children, English- and Russian-speaking children often elaborated path by adding adverbs of directionality, such as down and around.

So far, we have reviewed the studies on the acquisition of motion event encoding in the V-languages and the S-languages. The results have revealed that children are sensitive to the language-specific difference in motion encoding. Recently, some results are coming out regarding children's acquisition of motion events in the third type, the equipollently-framed languages.

The study of Chinese motion event acquisition has been prosperous in recent years.

Chen (2005) analy ed the developmental course of Chinese children’s motion event encoding, from 3 to 9 years old. The result revealed that children had not reached the adult-like performance at the age of 9. Lin (2006) in her thesis investigated Chinese children’s motion event constructions and found that young children tended to omit the manner element and simply encoded motion events into the Path + Deixis construction, such as chu1 lai2 ‘exit come’. However, the older children showed a preference for the use of the Manner + Path + Deixis (M+P+D) constructions containing, such as pao3 chu1

lai2 ‘run exit come’. Therefore, the applications of the M+P+D construction increased

with age. Ku (2007) also examined Chinese children’s development of manner-of-motion verbs in motion event encoding by asking children to narrate the frog story. The results echoed Lin (2006)’s finding. An increase use of manner verbs was observed and Chinese children attended to treat motion events in the way as adults, which was to encode both manner and path equally.

While studies have been carried out to investigate how monolingual children learn to encode motion events in their ambient language, some researchers have extended this acquisition issue to bilinguals or second-language learners, especially those who are learning two languages that are typologically different in motion event encoding.

Nicoladis and Brisard (2002) examined the English-French bilinguals’ encoding of motion events. They found that the English-French bilinguals tended to encode the path element in gesture and speech, and producing comparable number of manner verbs, either in English or in French. The result was counter to the expectation because it is expected more manner verbs should be used in English narrative since English is classified as an S-language according to Slobin’s three-way typology.

Some studies are interested in the acquisition of motion events in second language learning. Navarro and Nicoladis (2005) studied the advanced Spanish learner of English

on motion event encoding, and found that L2 (Spanish) speakers exhibited a certain influence from their L1 (English). Hohenstein and colleagues (2006) studied

Spanish-English bilinguals and found bilingual speakers behave in between the two languages they had been exposed to in terms of motion events.

Studies on the Chinese-English language learners also revealed interference from the other language. Wu (2008) investigated the influence of L2 (English) on the expression of motion events in L1 (Chinese) in a group of English learners of Chinese with different L2 proficiencies. She found that the advanced learners were more likely to be affected by the motion event encoding system of L2, which could result from a stronger conceptual link between L1 and L2. The study also indicated that Chinese children at old as 10 had not completely mastered the encoding of motion events in their L1. Wu (2011) investigated how the Chinese learners of English learned to express the target-like manner, for example, the directional complement (DC),⁴ such as dao4 ‘to’,

jin4 ‘into’, lai2 ‘come’, and hui2 ‘return’. DCs can function as a main verb or a path

satellite in Chinese. The dual functions of DCs posed considerable challenges for learners, whose L1 (i.e. English) does not have this distinction. The difficulty increased along with the complexity of the DC construction. When a more complex DC construction, such as

ta1 zou3 jin4 lai2 le ‘he walk into hither

PERF’ it becomes more challenging for these Chinese learners of English, since no such construction is used in their L1 (English). In Wu (2011), it was also shown that the Heritage Language Learners (HLL) performed better when compared with the Foreign Language Learners (FLL) since HLLs have more access to the natural context of the target language. In our study, the subjects included were all Heritage Language Learners (Valdés, 2000). Therefore, in section 5 of this chapter, the definition of Heritage Language Learner and a more detailed description of

4 DCs in Wu (2011) consist of path elements, such as jin4 ‘in,’ chu1 ‘out,’ and deixis elements, lai2

‘hither,’ and qu4 ‘thither.’

the HLLs in our study will be introduced together with the language policy implemented.

2.5 Heritage Language Learning Program in Taiwan

Taiwan, with a total population of around 21 million, is a multi-cultural and

multi-lingual society due to its historical background and language policy. Huang (2007) pointed out the difficulty of revitalizing indigenous language lies in the fact that over the past fifty years, people in Taiwan were forced to use the national language, Mandarin Chinese, to communicate with each other. Peoples of minority languages, such as the Austronesians, were deprived of the right to receive education in their own languages. In consequence, many young people nowadays mainly communicate in Chinese in their daily lives, and these indigenous languages are facing extinction.

According to Chen (2010), language policy influences the multilingual

evolution in Taiwan. Among the three language policies that have been implemented over the past fifty years, namely the national language policy, the mother tongue language policy, and the new English language policy, the implementation of the mother tongue language policy helped preserve the indigenous languages in Taiwan. According to the mother tongue language policy, 1-2 periods (40 minutes per period) of local language teaching per week should be included in the school curriculum. The aboriginal languages have been included in the local language teaching program since September 2001. Until now, the indigenous languages have been taught in school for ten years. These indigenous children, having some exposure to the indigenous language at home or in the community, start to learn the indigenous language via the formal school education. This group of children who have some exposure to the indigenous language at home or in the community therefore can be viewed as heritage language learners (HLL).

Valdés (2000) defines the heritage language learners as “individuals raised in homes

where a language other than English is spoken and who are to some degree bilingual in English and the heritage language.” The original definition illustrates the heritage

language learners in the US, and therefore is more English-centered. English, in Valdés’s definition, can be substituted by any other dominant language for heritage language learners in other areas around the world. A characteristic of heritage language learning is that the heritage language was first acquired at home but was not completely acquired because of the switch to another dominant language. Therefore, different from second language acquisition (SLA), heritage language learners have had some exposure to the heritage language at home or in the community. Also, unlike first language acquisition, the language used outside their home or community is a dominant language. Heritage language learners behave like neither L1 nor L2 speakers of the heritage language (HL) because of the curtailed acquisition during childhood, according to Lynch (2003). Valdés (2005) indicates that a tremendous variation of language proficiency is observed in heritage language learners, which could result from the diverse linguistic experience at home (Carreira, 2004). Due to the diverse linguistic proficiency of heritage language learners, Kondo-Brown (2010) points out the importance to identify the heritage language learners in order to serve the needs and interests of HLLs and help them advance their competence in the heritage language from curriculum planning.

In Taiwan, the indigenous children learning their indigenous language, including the subjects in our study, can be identified as heritage language learners. The subjects in our study all have had some exposures to the heritage language at home or in the community, because the elder still use the heritage language while communicating with each other.

Nevertheless, these children’s acquisition of the heritage language is not complete because they receive their education in Chinese, the dominant language in Taiwan.

As indicated by Huang (2007), the government has put a lot of efforts to revitalize

the indigenous languages with some strategies, such as the development of the indigenous language textbooks and training the indigenous language teachers. It is important to identify the group of heritage language learners and provide them with proper instruction in the indigenous language learning. However, the implementation of the new English policy worsens the situation of the heritage language learning. Chen (2006) points out that English is usually viewed as providing personal benefits by means of an international outlook and socioeconomic advantage. Huang (2007) also indicates that parents hold a more positive attitude toward English learning than the indigenous language learning.

Therefore, it is likely that the promotion of English may suppress the learning of the indigenous language, because the HLL will be encouraged to devote efforts to studying English, rather than their indigenous language.

As the heritage language learning program has been implemented for 10 years, we think it is time for us to evaluate whether this program has helped preserve indigenous languages. Moreover, it may be interesting to see if the promotion of English learning has influenced the HLL’s learning of their indigenous language.

2.6 Summary

Based on the typological studies on motion events in Chinese (Slobin, 2004), Chinese is classified as an equipollently-framed language, in which both manner and path are encoded in equivalent grammatical forms. The Manner+Path+Deixis construction is the most frequent pattern used among adults (Chen, 2005). On the other hand, the target language in this study, Squliq Atayal, is proposed to be more like a verb-framed language (Huang & Tanangkingsing, 2005). According to Huang and Tanankingsing (2005), more

在文檔中學習泰雅語動態事件編碼:以新竹某國小族語學習為例 (頁 14-0)