Comprehension of conversational implicatures in L3 Mandarin

1 Introduction

Learning a second language (L2) does not simply mean having an understanding of the grammar and the vocabulary. It also entails being familiar with the sociocultural conventions as well as norms of language use. That is, a language learner needs to know what to say or not to say in a given situation. Part of becoming a competent speaker in an L2 requires a language learner to be adept at pragmatics (Taguchi 2019). Crystal (1997: 301) defines pragmatics as “the study of language from the point of view of users, especially of the choices they make, the constraints they encounter in using language in social interaction and the effects their use of language has on other participants in the act of communication”. In second language acquisition (SLA) research, L2 pragmatics investigates how nonnative speakers acquire pragmatics and pragmatic competence. Taguchi (2019: 4) maintains that pragmatic competence is a “multi-dimensional and multilayered construct that involves (1) linguistic and sociocultural knowledge of what forms to use in what context; (2) interactional abilities to use the knowledge in a flexible, adaptive manner corresponding to changing context; and (3) agency to make an informed decision on whether or not to implement the knowledge in the community”.

Given the complexity of this definition, achieving pragmatic competence for an L2 learner is an arduous feat. Taguchi (2019) breaks down the specific challenges in achieving pragmatic competence for an L2 learner. The first challenge is the first language (L1) or any other language that the learner previously acquired. That is because such previously acquired languages not only require the L2 learner to maintain control over preexisting pragmatic representations but also to relearn new connections between linguistic forms and the social contexts in which those linguistic forms appear (Bialystok 1993). Another challenge in achieving L2 pragmatic competence is the lack of saliency in expressing social norms and conventions of communication, which makes it difficult for L2 learners to gauge what linguistic means are used in different social encounters. Finally, according to Thomas (1983), pragmatic competence, due to its multifaceted nature, necessitates knowledge of pragmalinguistics (i.e., knowledge of linguistic resources for performing a communicative act) and sociopragmatics (i.e., knowledge of sociocultural norms and conventions associated with the act). Bardovi-Harlig (2000) argues that grammar and pragmatics are distinct but interdependent in L2 learning. Going a step further, Taguchi (2019) maintains that while knowledge of vocabulary or grammatical rules does not readily lead to better pragmatic performance, pragmatic competence cannot be achieved without knowledge of the formal aspects of the language. In that respect, Taguchi (2019) argues for a threshold linguistic knowledge that is prerequisite for pragmatic performance.

2 Models of morphosyntactic transfer in L3 acquisition

There have been a plethora of studies investigating when, how, and to what extent prior language experience, be it from the first, second, or both languages, affects the early phases and subsequent development of adult third language (L3) acquisition (Puig-Mayenco et al. 2020). One hypothesis attributes a privileged role to the L1 in L3 acquisition (Hermas 2010, 2015; Jin 2009; Na Ranong and Leung 2009). Another hypothesis, the L2 status factor hypothesis (Bardel and Falk 2007; Bardel and Sánchez 2017; Falk and Bardel 2011), maintains that a second language learned in adulthood holds a special position as a source of morphosyntactic transfer. The Cumulative-Enhancement Model (Berkes and Flynn 2012; Flynn et al. 2004) suggests that both previously learned languages can be utilized for transfer at any stage of L3 acquisition. According to the Typological Primacy Model (Rothman 2010, 2011, 2015), all grammars from previously learned languages are available for wholesale transfer that may be facilitative as well as non-facilitative. Finally, the Scalpel Model posits that there is property-by-property transfer from either the L1 or the L2 to the L3, and such transfer may be facilitative or non-facilitative. Although these models say nothing about whether such transfer would extend to L3 pragmatics, I hypothesize that in the present study the typological proximity between the participants’ L1 (Cantonese) and their L3 (Mandarin) would provide support for the Typological Primacy Model (Rothman 2010, 2011, 2015) as the participants could potentially be inclined to transfer the processes by which they decode conversational implicatures from L1 Cantonese into L3 Mandarin. Such a finding would also substantiate the notion of a privileged role of the L1 in L3 acquisition.

3 Previous research on comprehension of conversational implicatures

Taguchi (2018) argues that indirect communication such as sarcasm, irony, humor, and understatement are so prevalent in everyday interactions that it is impossible to fathom individuals going about their days without using them. A conversational implicature, or simply implied meaning depending on a context (Grice 1975), is an example of indirect communication and decoding it is a core element of pragmatic competence. According to Blome-Tillmann (2013: 174–175), “conversationally implicated contents are utterance contents that are only loosely related to what is said by an utterance – they are not part of the conventional meaning of an utterance”.

Over the last three decades, L2 researchers have employed various theoretical frameworks to study the acquisition of conversational implicatures. One of the theoretical frameworks is Grice’s (1975) Cooperative Principle, and the four Gricean conversational maxims. Grice (1975) asserts that understanding the meaning of a statement assumes that the individuals involved in the conversation adhere to the Cooperative Principle, which states: “Make your contribution such as is required, at the stage at which it occurs, by the accepted purpose or direction of the talk exchange in which you are engaged” (Grice 1989: 26). In addition, Grice (1989) posits that conversational participants also adhere to four overarching conversational maxims: Quantity, Quality, Relation, and Manner (Grice 1975: 45–47). Sperber and Wilson (1995) propose that the four Gricean maxims can be condensed into a single principle known as the Communicative Principle of Relevance. According to this principle, human cognitive processes are structured to obtain the maximum cognitive impact while minimizing processing effort. To achieve this optimal outcome, individuals should focus exclusively on what is most relevant during communication.

L2 researchers following a Gricean paradigm have usually investigated whether L2 learners can interpret implicatures stemming from flouting certain implicatures (see, among others, Bouton 1992, 1994b, 1999; Roever 2005; Shively et al. 2008). Some other researchers, working within the paradigm of Relevance Theory, according to which comprehension of implicatures is due to our relevance-seeking cognition (Taguchi and Yamaguchi 2019), have studied factors such as contextual cues, processing load, and conventionality (Taguchi 2005, 2007, 2008, 2011, 2012; Taguchi et al. 2013a).

To date, different variables have been investigated in studies on implicature comprehension. The most widely researched variables have been the effect of L2 proficiency, individual characteristics, study contexts (particularly the effect of study abroad), and finally the effect of instruction on successful implicature comprehension. Previous literature indicates that L2 proficiency usually leads to a better performance in comprehension of conversational implicatures (Bouton 1988, 1992, 1994a, 1994b; Köylü 2018; Shively et al. 2008; Taguchi 2005; Taguchi et al. 2013a; Yamanaka 2003).

The methods used to investigate how learners acquire conversational implicatures have evolved significantly in the last four decades. Initial studies, such as Bouton’s (1988) work, utilized multiple-choice tasks administered to large groups of English language learners to examine the developmental process of acquiring conversational implicatures. Building on this foundation, later researchers, like Taguchi (2005), began employing aural tasks where learners listened to audio stimuli and selected the correct interpretation of a conversational implicature from written multiple-choice options. The approach further advanced when Shively et al. (2008) used an audiovisual task followed by a written production component to investigate conversational implicatures. Finally, Köylü (2018) implemented a methodologically innovative approach to investigate conversational implicatures, which involved an audiovisual task where participants viewed a video, followed by an oral production component that required the participants to verbally report their interpretations of the conversational implicatures.

SLA researchers investigating conversational implicatures have usually studied whether L2 learners of English coming from different L1 backgrounds were on a par with native speakers of English in their correct interpretations of those implicatures. That is, the target language examined has usually been English. However, some earlier studies such as Hong (1997, 2011) and Sun and Zhang (2008) looked into L2 Chinese learners’ use of speech acts. More recently, Taguchi et al. (2013b) investigated how 31 intermediate-level American learners of Mandarin used formulaic expressions during their 10-week stay in China. The researchers administered a computerized oral discourse completion task to these learners at the beginning and end of their study abroad experience. This allowed the researchers to measure the learners’ production of formulaic expressions, and their appropriateness, fluency, and frequency. The comparison of the pre- and posttest data showed that the learners made significant improvements in all three of these measures, suggesting that the study abroad context had a positive impact on the learners’ use of formulaic language. Finally, Wen (2014) used a written discourse completion task to investigate the pragmatic development of 48 learners of Chinese as a foreign language (CFL). The focus was on how these learners used request strategies and modification devices in different contexts. The findings showed that the advanced-level CFL learners exhibited greater pragmalinguistic and sociopragmatic development compared to their lower-proficiency counterparts.

Previous research on the acquisition of pragmatics in L3 Mandarin is almost nonexistent. Most of the previous literature and books published look into the acquisition of pragmatic features such as speech acts, pragmatic routines, and conversational implicatures in L2 English (for a review, see Xiao 2018). The only studies that have looked into the acquisition of pragmatics in Mandarin are L2 studies (see, among others, Hong 2011; Jin 2012, 2015; Li 2012, 2013, 2014; Li and Taguchi 2014; Taguchi et al. 2013a; Taguchi et al. 2013b; Wen 2014; Yang 2014, 2016; Yang and Zhu 2016; Zhang and Yu 2008). Thus, Yang (2018: 275) maintains “Chinese L2 pragmatics has been given far less attention in relation to other areas in Chinese SLA”. This absence of acquisition studies in pragmatics in L2/L3 Mandarin demonstrates the dire need for experimental research in this domain. In that respect, the L1 Cantonese L2 English language background of the participants in the current study creates a unique opportunity to test the comprehension of conversational implicatures in L3 Mandarin.

In addition, in their discussion of future directions in assessing implicature comprehension, Taguchi and Yamaguchi (2019) underline the significance of innovation in instrumentation. They state:

Future research should expand the scope of instrumentation in assessing L2 implicature comprehension. Previous studies mainly used a highly controlled, decontextualized listening or reading test with researcher-made dialogues, limiting the generalizability of the findings in real-life situations. This limitation can be solved by developing instruments that reflect real-life inferential practices. Particularly useful in this direction is the use of multimodal input combining visual, auditory, and textual information (Taguchi and Yamaguchi 2019: 42).

Moreover, Yang (2018) highlights the urgent need for more empirical efforts to examine different subareas of L2 Chinese pragmatics, such as the development of pragmatic competence. Taking into account these gaps in the literature, and adopting the design in Köylü (2018), the current study investigates whether L1 Cantonese L2 English L3 Mandarin learners can successfully interpret conversational implicatures in an authentic audiovisual interpretation task in their L3 Mandarin.

4 The current study

This study investigated the following research questions:

1. Can L1 Cantonese L2 English L3 Mandarin learners successfully interpret conversational implicatures in an authentic audiovisual interpretation task in their L3 Mandarin?

2. What does the variation in learner responses reveal about L1 Cantonese L2 English L3 Mandarin learners’ interlanguage development in conversational implicatures in their L3 Mandarin?

5 Results

An independent samples t test was conducted to compare the mean scores of successful implicature calculation. That is, the two groups (L3 and native speakers of Mandarin) were compared to see if they could arrive at the intended implicatures for each context. The prerequisites for the t test were met. Normality was assessed using the Shapiro–Wilk test, which indicated that the data was approximately normally distributed (p > 0.05). Levene’s test for equality of variances revealed no significant difference in variances between groups (p = 0.45). The L1 Cantonese L2 English L3 Mandarin participants (M = 72.1, SD = 22) performed worse than the native speakers of Mandarin (M = 94, SD = 9), and the difference between the two groups was significant, t(23) = 5.61892, p = < 0.00001.

In the qualitative analysis, in addition to the correct implicature calculation illustrated in (1), (2), and (3) above, the learner responses, although not frequently, were also characterized by some divergence compared to native speaker responses. Based on Scene 9, I illustrate in examples (5)–(9) responses that received a 0. Note that although all these five responses were coded as 0, they show variation in terms of why or where the implicature calculation failed. To illustrate, example (5) indicated that the learner could not comprehend the implicature at all.

Some learners simply repeated the last utterance, and that was coded as simple repetition, thus incorrect, as illustrated in (6).

Some responses showed that the participant understood the literal meaning, but not the implicature, as shown in (7).

A few responses indicated that the participant was aware of the implied meaning that went beyond the literal meaning but they could not calculate the implicature. An example is given in (8).

Finally, some responses demonstrated that the participants did indeed arrive at an implicature, albeit an unintended one given the context, as shown in (9).

Table 2 and Figure 1 illustrate the numbers and percentages of response types based on stages of implicature calculation.

Figure 1:

Percentages of response types based on stages of implicature calculation.

6 Discussion and conclusion

This research investigated whether L1 Cantonese L2 English L3 Mandarin learners were able to comprehend conversational implicatures in L3 Mandarin. The results indicated that the native speakers of Mandarin performed significantly better than the L1 Cantonese L2 English L3 Mandarin learners in implicature comprehension. This is in line with previous research that found that even at high levels of L2 proficiency, L2 learners may not be on a par with native speakers in implicature calculation (Köylü 2018). The results also find support from Wen (2014), who investigated the pragmatic development of request strategies in CFL and found that a higher level of proficiency did not necessarily lead to a target-like pragmatic performance. That provides support for the notion of a threshold proficiency level as a prerequisite for pragmatic competence, postulated by Taguchi (2019). Future SLA researchers could recruit advanced and near-native speakers of L2s or L3s to examine whether such high proficiency levels significantly enhance pragmatic competence in L2s and L3s. Despite not being on a par with native speakers of Mandarin, the L1 Cantonese L2 English L3 Mandarin learners exhibited a wide range of responses in their conversational implicature calculations, ranging from repetitions, literal comprehension, recognition of a gap between the literal and the implied meaning, to unintended implicatures.

One limitation of the current research is lack of a strong connection between L3 acquisition models and the findings of this research. Although I argued at the outset that the L1 Cantonese L2 English L3 Mandarin participants could potentially transfer the processes by which they decode conversational implicatures from their L1 Cantonese into L3 Mandarin, such a prediction is only a speculation since whether language transfer from the previously acquired languages extends to the domain of L3 pragmatics is unclear. Future studies could try to unearth such a connection. Another limitation is the proficiency level of the participants in L3 Mandarin. Due to their low numbers, participants could not be divided into different proficiency levels. Participants at lower (e.g., lower intermediate) and higher proficiency (e.g., near-native) levels could be recruited in future studies to look into the developmental trajectory learners go through in the acquisition of conversational implicatures in L3 Mandarin.

Despite these limitations, the current research constitutes the first attempt to provide experimental evidence for the comprehension of conversational implicatures in L3 Mandarin by participants with a unique L1 Cantonese L2 English background. Another strength of the current study is the utilization of an authentic audiovisual interpretation task with a written production component rather than using a highly controlled, decontextualized listening or reading test with researcher-made dialogues. As Yang (2018: 262) argues, “compared to other areas in Chinese SLA, the investigation of L2 pragmatics (let alone L3 pragmatics) started rather late and has received insufficient attention; only a very small number of studies have been conducted to explore Chinese L2 learners’ pragmatic development”. Thus, the current research contributes to the extremely sparse literature on empirical efforts to investigate the comprehension of conversational implicatures in L3 Mandarin.