leh in Singapore English

For those readers unfamiliar with the language situation in Singapore, it is worth pointing out that in the country there are four official languages and one national language, with English, Malay, Mandarin and Tamil the official languages, and Malay the national language. In Singapore, around 74% of the population are ethnic Chinese, 13% are Malay and 9% are Indian, with some 35% of the population reportedly using Mandarin as a usual home language, and almost 37% use English (Department of Statistics Singapore, 2016). In this current study reported on here, the participants were selected based on the ethnic profile of Singapore in order to ensure that the data is more generally representative of the (‘official’) Singapore ethnic situation. Many studies of English in Singapore (or Singapore English) have argued that there are two varieties of Singapore English, that is, Singapore Standardised English (or SSE), which resembles Standardised British English (Cavallaro, Ng & Seilhamer Reference Cavallaro, Ng and Seilhamer2014), and Singapore Colloquial English (SCE). Features of SSE have been discussed in Wee (Reference Wee and Lim2004), while SCE features have been discussed in terms of lexical productivity (Low & Brown, Reference Low and Brown2005; Leimgruber, Reference Leimgruber2011), substrate-influenced syntax (Bao & Wee, Reference Bao and Wee1999); use of pragmatic and discourse particles (Lim, Reference Lim2004; Wee, Reference Wee and Lim2004; Botha, Reference Botha2018), and phonology (Lim, Reference Lim2004; Deterding, Reference Deterding2005). As such, this article focuses on a specific feature of SCE.

The particle leh in Singapore Colloquial English has received some discussion over the past three decades. Lim (Reference Lim2007: 461) has argued that this particle has a Cantonese origin and notes that a similar particle le ₅₅ (with a ‘high level’ tone marking) exists that is used to form a question, has a comparative function, and that it means ‘what about?’. In Singapore English, Lim (Reference Lim2007: 461) argues that leh serves to ‘mark a question that involves a comparison’. For Lim, the substrate influence(s) on this particle should not be ignored when considering the functional meaning of this (as well as other Singapore English particles). Gupta (Reference Gupta1992) has proposed what she refers to as a ‘scale of assertiveness’, where she situates leh somewhere near the middle of her scale, where the particle can be used to express some degree of assertiveness, and that it indicates a tentative suggestion or request. Wee (Reference Wee and Lim2004) has also suggested that this particle is a marker of tentativeness, as well as used to soften a request or command in Singapore English. Earlier, Platt (Reference Platt, Steele and Threadgold1987) has suggested that this particle indicates a disagreement with a suggestion, conveys information that is assumed to be new, and that it can also mean ‘what about?’; while Kwan–Terry (Reference Kwan–Terry and Kwan–Terry1991) has suggested that leh is used as an emphatic marker (such as in declaratives or imperatives) and used in interrogatives. Table 1 below indicates a summary of the functional uses of leh in the literature on particles in Colloquial Singapore English.

Table 1:

Summary of the functional uses of leh in the literature

^a Note that different spellings of leh are used by some of these authors, with Platt (Reference Platt, Steele and Threadgold1987) using le ₂₂ and le ₅₅, and Kwan–Terry (Reference Kwan–Terry and Kwan–Terry1991) using le ₃₃ and le ₅₅.

In terms of the frequency of use of leh in SCE, some studies have attempted to order the place of Singapore English particles in terms of how often they are used in discourse. Smakman and Wagenaar (Reference Smakman and Wagenaar2013) placed leh at number six in their list of eight investigated particles in terms of frequency of use, similar to Ler (Reference Ler and Fischer2006) which placed this particle at number eight of her list of ten particles in Singapore English. In a previous study of the frequency and distribution of particles in the vernacular speech of students in their social networks, Botha (Reference Botha2018) places leh at number five of the 16 particles investigated. It needs to be noted that the frequency and distribution of these particles often depends on the type of language data that is used for analysis, which reveals that not all SCE particles in such studies are always included in the range of particles that are investigated (for example, those particles of a so-called Malay origin [such as sia and seh] and datasets may often be skewed in terms of ethnicity towards a far greater number of words by a certain ethnic group in the data [as in Smakman & Wagenaar, Reference Smakman and Wagenaar2013]). In the next section I proceed to describe the research methods for investigating the description of leh.

Researching leh in the vernacular speech of students

The social network research approach that was used for this study has already been described in previous publications which investigated particles in the vernacular speech of Singapore students (Botha, Reference Bothain press). This approach has also been developed from my earlier research on language in social networks (Botha, Reference Botha2012, Reference Botha2017; Botha & Barnes, Reference Botha and Barnes2013, Reference Botha and Barnes2015), which focused on the vernacular speech of speakers in South China, and the reader is directed to these references for details regarding the methods that were employed for this current study. However, for the sake of clarity, the approach used for this present study presented is briefly summarised below.

Three ethnic Chinese students, two ethnic Malay students, and one ethnic Indian student participated in the research. Each of these participants acted as the main subject (or ‘ego’) of their individual social networks. The first stage of the study involved establishing the connections (or ‘ties’) that each student had with people that they were in regular contact with. In Figure 1, a visual representation is shown of one of the social networks that were investigated for this study. From this figure it can be seen that there is a distinction made between the first- and second-order network zones in the social networks of the student, where members in the first-order zone all know one another and are in regular contact with one another. In the second-order zone not all the members know one another except the ego of the network, but these members are all in regular contact with the main subject. Interestingly, this study reveals that with the six social networks that were investigated for this study, nearly all the members of each social network are of the same ethnicity as the main subject of the study, especially in the first-order zones, where all the members are of the same ethnicity as that of the main subject.

Figure 1.

Visual representation of one of the Chinese social networks, showing the first- and second-order network (adapted from Botha, Reference Botha2018)

During the next stage of the study, all the participants (that is, social network members of each network) were informed that their conversations with the main subject would at times be recorded, and that they would be made aware of when each recording would be conducted. All the participants were later anonymized in the reporting of results related to this study. The main subject of each network would meet members of their social networks in relaxed social settings and proceeded to record their conversations. These recorded conversations and discussions were on average around 20 minutes in length. The recordings were then transcribed using an adapted version of Du Bois’ (Reference Du Bois1991) transcription method in order to facilitate data analysis. Text messages of and between subjects in their social networks were also collected, consisting of a dataset of some 300,000 texted words, but the spoken discourse was the main focus of the study that is reported in this article. In addition, self-report data on the language use and language backgrounds of the subjects were also collected by means of a short language survey, which was used for obtaining basic demographic data of all the participants. Just over 70,000 words were recorded in the spoken conversations.

A total of 57 subjects took part in the project. The sample characteristics were recorded through the language survey and reveal that 71.9% of the participants were between the ages of 19 and 24, and 28.1% were aged 25 or older. As for the gender of the subjects, 49.1% males and 50.9% females were represented in the study. And in terms of ethnicity, 63.2% of the participants in the sample were Chinese, 22.8% were Malay, 10.5% were Indian, and a further 3.5% were of ‘other’ ethnicities. The sample for this study is to some extent representative of the ethnic makeup of Singapore society when compared to figures for ethnicity in the city-state as a whole (Department of Statistics Singapore, 2016), despite a slight overrepresentation of Malay subjects, and a slight underrepresentation of Chinese subjects.

The use of leh in the social networks

The frequency of leh in the networks

In this part of the article I proceed to present the findings of my results on the social contexts to the use of leh. The results here are discussed with reference to the research aims that were introduced above, where the first aim concerns the frequency patterns of the use of leh in the social networks of students, with reference to various social variables such as modality, gender, and the ethnicity of the speakers. As can be seen from Figure 2, the overall number of these selected particles is 5,355 (or some 1.5% of the total number of words and utterances in the spoken data). From these results, it is clear that certain particles are used much more frequently than others, with ah, lah and eh occurring considerably more frequently than the other particles. These results indicate that leh is the fifth most used particle in the data, occurring 305 times overall, with 176 (6.9%) occurrences in the spoken data.

Figure 2.

Overall frequency distribution of leh, compared with other particles in the social networks (adapted from Botha, Reference Botha2018)

The frequency patterns reveal that there is very little difference in terms of modality (or medium of communication) and gender. The biggest difference was shown to occur with the ethnicity variable, where, of the 176 occurrences in the spoken data, the particle appeared 170 times in the Chinese networks (or 96.6%), five times in the Malay networks (2.8%), and only once in the Indian network (0.6%). These results indicate that ethnicity is an important variable in the use of leh, with Chinese speakers predominantly using this particle. However, the occurrence of this particle in the spoken data in the Malay and Indian networks indicates that this particle is now perhaps being appropriated by some Malay and Indian speakers.

The frequency distribution of the functions

The data was also analyzed to determine the frequency distributions of the identified functions of leh. The results of this analysis are presented in Figures 3 to 6, showing (1) the function patterns for leh in the data according to the previously identified functions identified in the literature; (2) the functions for leh according the additional functions identified in this current study; (3) the distribution of the previously identified functions in the English only exchanges; and (4) the distribution of the additional functions of leh in the English only exchanges. This analysis of the data is motivated to indicate the spread and use of leh in the vernacular language use of speakers in the social networks, and also to compare the variations in the use of leh, according to the main language used in an interchange, with the view to establish the extent to which the functions exist in Singapore Colloquial English.

Figure 3.

The frequency distribution of the functions of leh in the data, based on the functions in previous literature

Figure 4.

The frequency distribution of the additional functions of leh in the data

Figure 5.

The distribution of the previously-identified functions of leh in English utterances

Figure 6.

The distribution of the additionally-identified functions of leh in English utterances

From Figure 3 it can be seen that the overall most frequent use of leh in the previously identified functions in the literature were as an ‘emphatic marker’ (14.2% of overall usage), when used to indicate when ‘information was assumed to be new’ (11.9%), to ‘mark a question’ (11.9%), to mark an ‘interrogative’ (10.2%), and as a ‘declarative/imperative’ (8.5%). The least frequent use of leh was used to ‘soften a request/command’ (2.3%).

The functions for leh according the additional functions identified in this current study are shown in Figure 4. Here it can be seen that the most frequent use of the particle is identified with the ‘softening a disagreement’ (9.1%), to ‘soften an opinion’ (4.5%), and to ‘signal helplessness’ (3.4%). The least frequent use was identified with ‘softening a question’ and ‘softening an agreement’ (0.6%).

In order to capture variation in the use of leh in the expressions where only English was used, or more specifically, where the particle was attached to an English expression, the results in Figure 5 indicate the distribution of the previously identified functions in such expressions. It can be seen in the figure, and similar to the results shown in Figure 3, the most frequent use in this context is as an ‘emphatic marker’ (15.6%), to indicate ‘information assumed to be new’ (11.9%), as an ‘interrogative’ (10.1%), to ‘mark a question’ (10.1%), and to mean ‘what about’ (9.2%).

Finally, Figure 6 shows the distribution of the additional functions of leh in the English only exchanges, where the most frequent uses are identified with ‘softening a disagreement’ (7.3%), to ‘softening an opinion’ (6.4%), and to ‘signal helplessness’ (3.7%). Less frequent use was identified with ‘softening a question’ (0.9%), and no instances of ‘softening an agreement’ were found in the English exchanges.