2.4 Studies about the Three Public Discourses
2.4.2 Discursive approaches to language policy and planning
This second analysis of the study, in Chapter 5, draws from how English in language planning is talked about. The purpose is to investigate how layers of language ideologies are unfolded and interacting in such metadiscourse. Language planning, according to Weinstein
(1980), describes ―a government authorized, long-term, sustained, and conscious effort to alter a language‘s function in a society for the purpose of solving communication problems‖
(56). Language policy, on the other hand, acts as an umbrella term for a wider range of
47
language management. Spolsky (2004) provides a broad generalization to include ―all the language practices, beliefs and management decisions of a community or polity‖ (9). With
this definition, language policy does not necessarily involve institution-directed
implementation. Rather, any conventions about language use could be a form of language policy. It is also necessary to make distinctions between ‗language policies‘ and ‗language policy.‘ Policies, according to Spolsky (2004), are usually explicitly written plans about
language use. Yet, several complexities are also addressed by Spolsky (2004). First, he clarifies that a language policy does not always lead to its actual implementation.
Implementation would not necessarily promise successful stipulation to impact on existing language use. Second, the interaction between a social factor in favor of certain language policies and the policies themselves to be probable, instead of causal. Besides, as far as national languages are concerned, they are not always specified in constitutions. These observations lead Spolsky (2004) to propose that language use develops even without making
references to explicitly written interventions. By the same token, language use is never free from constraints even without written regulations. Spolsky‘s (2004) language policy
addresses all the regulations and management on language use. Language policy is understood as rules, either written or unwritten, of language use which is imposed on a social unit of people. Such rules need not be imposed by institutional power, though discussion on institutional influences on language policy prevails.
48
Discourse contributes tremendously to meaning making of language policy. Mortimer
(2016) suggests two ways of understanding policy. Policy could be viewed as representations which are open for interpretation, a property Mortimer assimilates to Gee‘s little-d discourse (1999). Policy could also be seen as Discourse in Gee‘s terminology (1999) because it
embodies socially shared decorum of using and viewing language. Scholarly attention on how language policy is talked about notes a gradual emphasis on addressing socially constituted understanding of language management (Barakos 2016). This trend called for discursive approaches to language policy (Johnson 2011; 2016; Savski 2016). Discursive approaches to language policy center on reflexivity and account for shared assumptions about language through metapragmatic discourse (Mortimer 2016). As Savski (2016) notes, ―policy
meaning is discursively constructed and that ‗discourse about policy‘ can thus be considered constitutive of policy meaning and constituted by it‖ (55). Through talking about language
policy, social actors interpret language policy and reveal their beliefs about languages.
Discourse about language policy thus serves as a site where language ideologies are manifested, evaluated, and contested (Lawton 2016; van Splunder 2016; Weber 2016).
The relationship between language ideology and language policy is summarized in Spolsky (2004), who notes ―language ideology is language policy with the manager left out‖
(14). Language ideology (Silverstein 1979; Rumsey 1990; Irvine 1998; Schieffelin et al.
1998), which refers to beliefs about language speakers accumulate along their social
49
experiences, could be both a motive and a result of policy enactment (Shohamy 2006).
Rumsey defines language ideology as ―shared bodies of commonsense notions about the nature of language in the world‖ (1990: 346). Rumsey‘s definition focuses on the
homogeneity of language ideology and addresses less about dynamics and multiplicity of language ideologies (Kroskrity 2004). Metapragmatic discourse serves as a site for investigation into language ideologies (Woolard 1998). This study takes this a step further and suggests that metapragmatic discourse about language policy could be molded by prevailing language ideologies which serve as shared knowledge for communication, and which are also further evaluated by speakers.
The study thereby adopts a discursive analysis to metadiscourse on English in language planning and aims to supplement the ideologizing process of English with the analysis. The analysis draws on Blommaert‘s sociolinguistic scales (2007; 2010), reviewed previously in Section 2.3, and also Vandenbroucke‘s (2015) analysis scheme to relate the contesting nature of language ideologies in the Taiwanese context to the multiple spatial identities Taiwan has.
Sociolinguistic scales (Blommaert 2007; 2010) metaphorically see geographical movement as vertically layered and power-invested. The framework of sociolinguistic scales can account for the relation between spatial identities and linguistic practice. A locale (Giddens 1991) could function multiple spatial identities (Vandenbroucke 2015) which semiotic resources are seen to lock intimately with (Blommaert et al. 2005; Blommaert 2007; 2010). Semiotic
50
practices can therefore conform to multiple references ranging from homogenous and global to situated and local, depending on the associations between the spatial identities and semiotic norms (Blommaert 2010). Vandenbroucke (2015) discusses how Brussels projects distinctive identities of Brussels as a capital, a non-official unified European city and a global city in neoliberalism with shop signs featuring French/Dutch bilingualism, Dutch monolingualism and English. As Taiwan positions itself with multiple identities, Chapter 5 discusses that discourse concerning language planning potentially invokes discussion about languages at different scale-levels.
2.4.3 Intertextuality
The third analysis, in Chapter 6, focuses on how the debate regarding a phonetic system for Taiwan Mandarin involves metadiscourse about English. In addition to the frameworks of language ideology and discursive approaches to language policy, the study also draws from the theoretical frameworks of intertextuality and indexicality to account for the observations. Metalinguistic discourse (Silverstein 1979; 1985; 1993) is one of the sites where the three frameworks, language ideologies, indexicality, and intertextuality, intersect.1 Intertextuality describes how current discourse is a combined product of both elements from other speakers in other contexts as well as elements newly created in the current context
1 An earlier version of the review in this first appe section ars in Lee, Wan-Hsin. accepted 2020. What does guojijiegui mean?: Language ideology, intertextuality and indexicality in online metapragmatic discourse.
Monumenta Taiwanica 19.
51
(Bakhtin 1986; Fairclough 1992b; Hodges 2015). A text always both links to prior utterances and feeds later communication (Bakhtin 1986). Intertextuality could be understood as cohesion across texts (Bauman 2005; Hodges 2015). The term ‗intertextuality‘ was first
coined by Kristeva (1980), who developed Bakhtin‘s idea where a text is seen as ―a mosaic of quotations‖ (Kristeva 1980, 66). The act of quoting and repeating facilitates social interaction
through discursive practice (Tannen 2006). For example, Tovares (2005) discusses how utterances from TV programs are borrowed, repeated, and readjusted in family interaction.
Kristeva (1980) further identifies the concept of horizontal and vertical intertextuality.
Horizontal intertextuality addresses the phenomenon of how speakers select utterances from previous turns and rephrase them to sustain the current discourse (Du Bois 2014). Such a conceptualization is also discussed in different terminology (Hodges 2015). For instance, the act of reintroducing prior utterances is termed ‗recycling‘ by Tannen (2006). Vertical intertextuality refers to the way that speakers see a text as belonging to a certain genre by identifying linguistic cues to salient characteristics of various types of texts (Kristeva 1980;
Briggs & Bauman 1992; Johnstone 2008). Briggs and Bauman (1992) adopt the concepts of intertextuality to yield an account of how a genre is recognized and how the knowledge about various genres is circulated through discourse. They propose that linguistic features are foregrounded by speakers and picked up by recipients who link the current text to prior understanding. However, Briggs and Bauman (1992) also state that the idea of generic
52
intertextuality offers only a sketch, and that pragmatics and metapragmatics still play a crucial role in shaping and understanding discourse.
Silverstein (2005) adopts a related concept termed interdiscursivity. According to Silverstein (2005), intertextuality describes ―a directionally neutral state of comparability of texts in one or another respect‖ (7), and thus differs from the inquiry of linguistic
anthropology, which emphasizes the communication process where a stretch of discourse is at participants‘ disposal to meet various social purposes (Silverstein 2005). Therefore,
Silverstein distinguishes intertextuality from what he terms ‗interdiscursivity.‘
Interdiscursivity deals with ―how intertexts are created, that is, how they are generated in
events of communication through techniques of interdiscursivity deployable as role strategies of the participants‖ (Silverstein 2005, 7). Bauman (2005) states that interdiscursivity mostly refers to discursive practice while intertextuality refers to the cohesive connections among
texts. This study does not distinguish between intertextuality and interdiscursivity. The term
‗intertextuality‘ is used throughout.
A discussion of intertextuality and the meaning-making process draws attention to the concept of indexicality (Hill 2005; Eckert 2008; Jaffe 2016). The fact that speakers understand and choose to use recurrent utterances shows that indexicality is constructed on a somewhat shared knowledge regarding the intended meanings of a phrase. The experience of intertextuality is also a process of indexicalization (Hill 2005). Speakers interpret linguistic
53
practice by linking it to social contexts where it is situated. Linguistic practice which
conforms to social expectations usually go unnoticed whereas unexpected linguistic practice invites language users‘ attention to the practice itself (Hübler & Bublitz 2007). Anton (1998)
notes the distinction between attending ‗from‘ speech, which speakers mostly do in communication, and attending ‗to‘ speech, which speakers do when expectations fail (199).
Expectedness can be also deployed to achieve other goals. In their study on mock language in films, Bucholtz and Lopez (2011) discuss how European-American characters‘ portrayal of stereotypical African American English is interpreted as parodic. The European-American performers deploy linguistic features which are stereotypically considered to belong to AAE to perform inauthentic blackness (Bucholtz & Lopez 2011). In indexicalization, linguistic practice usually points to a number of potential meanings (Silverstein 2003; Eckert 2008).
This suggests indeterminacy and multiplicity of indexical values (Silverstein 2003; Hill 2005;
Jaffe 2016). As linguistic practice indexes multiple meanings, different indexical values coexist (Silverstein, 2003). Eckert (2008) notes that indexical meanings are not discrete from one another but are ideologically associated (Eckert, 2008). In her study on the use of the
Spanish word mañana by Anglophone speakers, Hill (2005) states that intertextuality should be seen as both a source and evidence of indirect indexical meanings. She notes, ―The
apparent opacity indirect indexicality has is not inherent in its semiosis but resides in the particular context of ideological commitments and understandings of the world held by
54
speakers‖ (2005, 115). Jaffe (2016) discusses how English-speaking mothers get confused
and even irritated when their Spanish-speaking mothers-in-law address their baby boys as pobrecito, literally ‗poor little boy.‘ The address term is interpreted by the English-speaking
mothers as the mothers-in-law‘s accusations that they are not looking after their babies well.
Jaffe (2016) elaborates that indexical meanings are constantly examined and reinterpreted.
Taking a slightly different approach, the study discusses how a fixed expression invokes extra-textual knowledge (John Gumperz 1996) that the viewers deploy as a criterion for judgments.
Chapter 6 is inspired by the observations that the metadiscourse of English is introduced in the debate of Taiwan Mandarin‘s phonetic systems and that the public treats the
value guojijiegui as given, shared and needless to define. The issue is worth exploring for its theoretical and empirical implications. Theoretically, the chapter hopes to complement past studies on intertextuality by analyzing how a fixed expression such as guojijiegui, which is also an indexical value without a specified definition, invokes intertextual interpretations.
Studies on intertextuality have focused on address terms (Hill 2005), identity labels (Wong 2005), and evaluative expressions (Adachi 2016). Studies on intertextuality have discussed how a term, when used in a new context, carries its indexical meanings from prior texts with it (e.g. Hill 1992; Wong 2005), and how indexicalization, the link between language use to social categories, is deployed to serve other sociopragmatic purposes such as scripting
55
parodies (Bucholtz & Lopez 2011). Furthermore, this section presents how the metadiscourse about phonetic systems can be a full-fledged ideology-laden and indexicality-fused debate.
Empirically, to the best of my knowledge, past literature on intertextualiy with Mandarin Chinese data has been found to focus on literature and art (H.-l. Chang 2007; Fischer 2009), information transmit in media (W. Wang 2008; C. Wang 2017) and courtroom discourse (Shi 2014). Studies on intertextuality in the Taiwanese context feature in research on drama and film (Wood 2007; W.M. Wang 2016), and EFL studies (Chi 1995). Less attention has been paid to intertextuality in everyday discourse. How intertextuality works and affects speakers‘
everyday discourse is thus worth exploring.
The study draws from several popular research inquiries respectively supported by great bodies of literature. The study reviews directly relevant framework to build the theoretical sketch for data observation and discussion.
56
C
HAPTER3
M
ETHODOLOGYA study about the ideologizing process of English starts with a definition of English. As the use of English is indigenized with local features (Mufwene 2010), English in Taiwan is
also imbued with local meanings. Curtin (2007) discusses the phenomenon that some English lexical items have their unique meanings used solely in Taiwan. For example, the letter ‗Q‘
can refer to chewiness. Wasserfall (in press 2021) also suggests that English lexical insertion in Taiwanese Mandarin contexts is adapted and conventionalized. For example, the noun ‗line‘
can be used as a verb to mean to contact a person via line (Wasserfall in press 2021).
Sometimes it becomes challenging to argue whether these indigenous features are English.
Consequently, what practice can be counted as ‗using English‘ is itself worth exploring. As this study addresses the public‘s attitudes and language ideologies concerning English, the
study approaches the issues of ‗English‘ in accord with lay conceptualizations. When linguistic practice is referred to as ‗using English,‘ the study considers it potential for further discussion.
The study investigates how English is ideologized through examining how it is represented in three public discourses, in crosslinguistic humorous discourse in TV programs, in directly elicited online metadiscourse about officializing English, and in indirectly elicited online metadiscourse about English in a debate regarding phonetic systems for Taiwan
57
Mandarin. These three public discourses contribute to respective dimensions of profiling English in Taiwan—to inform and to entertain with English, to debate about English, and to debate around English. The goals are double-folded: to yield an understanding of sociolinguistic vitality of English in Taiwan and to pinpoint the dynamics of the ideologizing process of English in the Taiwanese context.
This study proposes that language representation is a context-specific product of ideological interpretations. Speakers‘ discourse can be seen as an embodiment of reflexivity
(Lucy 1993; Verschueren 2004; Caffi 2006; Hübler & Bublitz 2007) on their social knowledge about language. Namely, discourse is expected to be layered with different levels of ideologizing work. It‘s a combined product of anticipations and opinions on these expectations. Martin and White (2005) also address ―the anticipatory aspect‖ (93) of utterances with the dialogistic perspective. They note,
all verbal communication, whether written or spoken, is ‗dialogic‘ in that to speak or write is always to reveal the influence of, refer to, or to take up in some way, what has been said/written before, and simultaneously to anticipate the responses of actual, potential or imagined readers/listeners (92).
At least two levels of ideological process are identified in a given discourse, one established from long-term social interaction and one emergent in current interaction. The levels are
detectable from discursive arrangement of information. When speakers react to others‘
discourse, the speakers‘ discourse reveals their values, attitudes, and beliefs concerning not others‘ prior discourse, but the ideological understanding of such discourse. The study terms
58
the outperformed attitudes, beliefs and values ‗ideological stances,‘ following Jaffe‘s (2016)
terminology. Ideological stances tend to share the discursive structure for new information.
They are the focus of a complex sentence, they follow a transition, and they contain explicit
evaluative terms. The study terms the backgrounded and implicit social experiences about languages ‗established language ideologies‘ because they are pieces of available knowledge with which speakers make sense of others‘ discourse and proceed to evaluate. Established
information tends to be treated as old information. It occupies sentence-initial position, in subordinate clauses as a subject or a topic. They can also be the preferred answers in rhetorical questions because they are the expected information. An established language
ideology does not have to be embraced by the public. The study adopts Martin and White‘s (2005) view that all linguistic performances reveal speakers‘ attitudes toward prior discourse.
Martin and White (2005) specify the evaluative nature of utterances. The use of conjunctions and connectives, for instance, introduces two opposite propositions and prefer one to the other. It justifies how utterances are evaluative to an expected, known assertion. The relations
among others‘ discourse, established language ideologies and ideological stances sketch what the study calls an ‗ideologization web,‘ to emphasize the entanglement of language ideologies
involved in interpreting linguistic practice. The sketch of the ideologization web is presented in Figure 2.
59
Figure 2. An ideologization web sketches the connection among others’ discourse, established language ideologies and ideological stances.
The study presents others‘ discourse and ideological stances in concrete lines and frames as
they are discursively overt. Established language ideologies mediate in-between are presented in dotted frames and lines because they are often backgrounded and remain relatively implicit.
This model captures language ideologization as progressive, layered and context-dependent when communication proceeds. As interaction continues, the ideologization process could be
extended. Others‘ discourse is also a product of earlier multilayered ideologization. Speakers‘
ideological stances will also act as others‘ discourse that invokes further ideologization. The
model can be as complicated as Figure 3 presents.
others‘ discourse
speakers‘ expectations and screening established language ideologies
speakers‘ discourse:
ideological stances
60 Figure 3. An extended ideologization web.
This study will analyze the three public discourses by identifying the duel ideological focuses and discuss their interaction.
The outline of the Method is as follows. Section 3.1 describes how the study collects and analyzes discourse about crosslinguistic conversational joking in variety shows. Section 3.2 outlines how the study draws from online metadiscourse regarding English in language planning in Taiwan to discuss the interaction among language ideologies. Section 3.3 describes how metadiscourse concerning English is elicited due to its social imagination in the Taiwan context.
established language ideologies
Speaker C‘s ideological stances Speaker B‘s discourse
established language ideologies
Speaker B‘s ideological stances Speaker A‘s discourse
61