Theoretical Framework

To tackle the first research question: what is unique about the novel usage of YONG in terms of its grammatical function and semantic properties? three theoretical frameworks are adopted in the current study: Construction Grammar, Frame Semantics and Qualia Structure.

Construction Grammar is used to define the form-meaning pairing of YONG and its associated pattern in which YONG is used. Frame Semantics is utilized to describe the semantics of YONG with the representation of the possible scenario in which various activities denoted by YONG take place. Qualia Structure is exploited to account for and differentiate the potential ambiguity of YONG in context by semantic coercion from the complement of YONG. To deal with the second research question (i.e. why, how and by what mechanism does this YONG come into being in Taiwan Mandarin?), constructional approach is also used to account for the emergence of the novel use of YONG in Taiwan Mandarin. By adopting constructional approach, the lexical semantics of YONG and its related grammatical change can thus be explained with a unified theoretical framework.

1.3.1 Construction Grammar

According to Goldberg (1995), the definition of a construction is that “C is a CONSTRUCTION iffdef C is a form-meaning pair <Fi, Si> such that some aspect of Fi or some aspect of S_i is not strictly predictable from C’s component parts or from other previously established constructions.” In other words, constructions are defined as form-meaning pairings which serve as the basic units of language. Namely, a construction functions in the same way as a lexical item functions in language. Like lexical items, one construction thus has its own meaning and can endow a particular interpretation for its component parts. The English Ditransitive (double-object) Construction would be a

significant example (Goldberg, 1995): the form of the construction is [Subj V Obj1 Obj2], and the meaning of the construction is [X CAUSE Y to RECEIVE Z] (intended or actual transfer). For instance, the word “bake” is not essentially a verb coding actions of transfer.

Nevertheless, in a sentence like “She baked him a cake”, the Ditransitive Construction would coerce its meaning on the verb “bake”. “Bake” here thus obtains the transfer sense from the Ditransitive Construction.

Taking a similar viewpoint, in Radical Construction Grammar, Croft (2001) defined constructions as “pairings of form and meaning that are at least partially arbitrary”.

Constructions are thus symbolic units conceived as the model in the following:

Fig. 1 Model of the symbolic structure of a construction in Radical Construction Grammar (Croft, 2001)

For this reason, a construction can be a language unit of any size from individual words to larger phrasal pattern or even a sentence. Following this perspective, the lexical item YONG itself is definitely a form-meaning pairing and thus should be viewed as a construction. On the other hand, as shown in (6), the actual event in the [YONG + NP] pattern (e.g. 用早餐 yong zaocan ‘to prepare/cook breakfast’) is underspecified and the meaning of this pattern is not strictly predictable from its internal components. Due to its opaque semantic interpretation, this pattern surely could be regarded as a construction. Besides, as shown in

(6), YONG may have various meanings like ‘to cook’, ‘to write’ and ‘to probe’. However, it is cognitively uneconomical to list all these meanings as polysemous senses of YONG because it would neglect the creative use of YONG in context and because “an elaborate construction should be able to economically predicate the coerced interpretation without burdening to list extra senses or rules in the lexicon” (Chang, 2005). The seemingly irrelevant contextual interpretations of YONG (i.e. ‘to cook’, ‘to write’ and ‘to probe’ in (6)) may share some semantic properties endowed by the [YONG + NP] pattern. Accordingly, the constructional approach is adequate to be exploited here to account for the meaning encoded in the novel use of YONG in its related patterns. In sum, Construction Grammar is used to define the form-meaning pairing of the novel use of YONG and its associated pattern in the current study.

Further more, since the emergence of this novel use of YONG in Taiwan Mandarin is essentially a grammatical change involving re-association of an existing form with a new meaning, the construction-based approach thus would be adequate to study the development of this novel YONG in Taiwan Mandarin.

1.3.2 Frame Semantics

Fillmore and Atkins (1992) have proposed that verbal semantics can only be acquired in reference to the associated semantic frame. A “semantic frame” is defined as “a structure background of experiences, beliefs or practices, constituting a kind of conceptual prerequisite for understanding the meaning” by Fillmore and Atkins (1992). In other words, a “semantic frame” is the conceptual structure, or more precisely, the image schema, which supports and motivates a word’s meaning. For instance, the “commercial transaction” frame contains four core participants (i.e. frame elements) involved in a commercial transaction event: BUYER, SELLER, MONEY and GOODS. The lexical semantics of a verb belonging

to this frame (e.g. buy, sell, spend and pay etc.) is thus characterized by virtue of its profiled frame elements.

Similar to the novel use of YONG in the current study, Liu (2005) has noted that a group of Mandarin verbs only “denote a manner, rather than naming a specific activity”.

These verbs are “used to ‘set a frame’ for various activities to take place”. However, Liu (2005) also noted that these “frame-setting” verbs are unique in that “the frame is lexically inferred rather than syntactically realized with an array of arguments (‘frame elements’)”.

She takes 趕 GAN for example, the semantic frame set by GAN may be spelled out with its four properties: Presupposition, Manner, Telicity and Agent-control. Since YONG and GAN share similar “frame-setting” functions, Frame Semantics surely could be adequately utilized to probe into its lexical semantics in the similar fashion. Therefore, in the current study, the frame-based approach will be exploited to depict the verbal semantics of the novel use of YONG.

1.3.3 Qualia Structure

According to Pustejovsky (1995), Qualia Structure specifies four essential dimensions of the meaning of a word:

 CONSTITUTIVE: the relation between an object and its constituents, or proper parts.

 FORMAL: factors which distinguish the object within a larger domain.

 TELIC: purpose and function of the object.

 AGENTIVE: factors involved in the origin or “bring about” of an object.

As Pustejovsky (1995) has clarified, Qualia Structure not only characterizes our knowledge of words, but also suggests interpretations of words in context. Take the English verb “enjoy”

for example:

(8) a. Mary enjoyed the movie last night. (watching) b. John quite enjoys his morning coffee. (drinking)

The different contextual meanings of “enjoy” are rendered from information of the complement. More specifically, it is the Telic role of “movie” and “coffee” that project the activities of “watching the movie” and “drinking coffee” respectively to the interpretation of the VP. In other words, “Qualia Structure provides a compositional means for meaning coercion based on characterizations of the four different roles: Constitutive, Formal, Telic, and Agentive.” (Liu, 2005).

Further more, another advantage of Qualia Structure is that it can solve the potential ambiguity in the interpretation of the predication in a sentence. For example, as illustrated in Pustejovsky (1991), the English sentence “John began a novel.” may have two possible readings:

(9) a. John began to read a novel.

b. John began to write a novel.

The two distinct interpretations can be viewed as deriving from two distinct roles of the NP complement. More specifically, the Telic role [Telic = read] and the Agentive role [Agentive

= write] of “novel” project the two interpretations ‘read’ and ‘write’ to (9a) and (9b) respectively.

As illustrated previously, the novel use of YONG does not specify the actual event undertaken. Instead, it only denotes the manner by which the activity is fulfilled. The crucial information about the “ellipsed” event is supplied by YONG’s complement object NP. Therefore, the application of Qualia Structure would be profitable to explain how contextual meanings of the predication in the associated pattern of YONG can be obtained

by semantic coercion of the complement’s pre-defined Qualia role. As a result, how the various contextual interpretation of YONG in (6) as ‘to cook’, ‘to write’ and ‘to probe’ in the [YONG + NP] construction are obtained can be efficiently accounted for via Qualia Structure. Thus, in this vein, by incorporating Construction Grammar and Qualia Structure, the potential semantic ambiguity of YONG in the same context can be differentiated and accounted for in a systematic way.