Singapore Colloquial English or more fondly known as Singlish - contains a number of pragmatic particles that have come under much scrutiny and discussion over the years. These particles have received much attention, and become a distinctive representation of Singlish.
Many studies have been done over the years on the use and functions of these particles (Platt 1987; Platt and Ho 1989; Wong, 2004; Gupta, 2006; Lim 2007, 2011). Much work and research have been done on the classification, history, origins and reasons for the emergence of these particles in the use of Singlish (Lim 2007, Gupta 1992). Gupta (1992) proposed a scale of assertiveness of which eleven SCE particles could be placed on in terms of different degrees of assertiveness. Wee (2002) however, took a different stance, and sought to highlight and substantiate the development and possible evolution of Singlish particles in his analysis of the particle lor using the grammaticalisation framework developed by Traugott (1982, 1988, 1989). Platt (1987: 392) gave a more simplified explanation on the function of these pragmatic particles, and claimed that in a sentence, these pragmatic particles 'convey additional meaning over and above that expressed by the rest of the utterance'. This meant that a particle holds by itself independent meaning, and when different particles are attached to an utterance, it would produce different meanings.
Lee (2007: 2) illustrated this point by using the following example (1):
a. I want to drink mah.
b. I want to drink lah.
c. I want to drink leh.
d. I want to drink lor.
e. I want to drink hor.
f. I want to drink a.
In this example, the head clause of 'I want to drink' does not change. According to Lee (2007: 2), 'the clause-final particles are syntactically and semantically optional, as their omission affects neither the grammaticality nor the basic meaning of the matrix clause'. Therefore in SCE, each clause would possess a different meaning and communicative function by virtue of the clause-final particle attached to it.
For example in 1(a), mah performs what Gupta (1992: 43) calls a 'contradictory function'. By adding mah, the clause would suggest that it is a contradictory response to a prior utterance, by presenting what is being said as an 'absolute and obvious fact'. In contrast, while lor in 1(d) can perform a similar function to mah of presenting what is being said as an obvious fact, it generates and encompasses other different meanings. According to Gupta's (1992: 43) definition again, 'the central function of the more mildly assertive lo is to mark a directive or to create a suggestion'. In my opinion, the point that Gupta makes about a particle having a 'central function' is not necessarily accurate, as a particle's usage is based on users' preferences and choices, and not based entirely on the effectiveness of a particle in an utterance - as suggested by Gupta in her allocating of a primary serving function to the particles. Going back to the main point, 1(d) can now mean that the speaker is suggesting an activity (to drink), probably in response to a question such as 'What do you want to do there?'. Thus, it is clearly evident that these particles can change the entire meaning of an utterance.
1.2 Tone in SCE Pragmatic Particles
However, the interesting thing about these pragmatic particles is that they, individually, possess many tonal variants. Gupta's suggestion that the particles have a 'central function' also reflects the evidence that these particles have multiple meanings and functions. Taking the much studied particle lah for example, Loke and Low (1988) identified a total of nine tonal variants of lah, classifying them into three major groups termed 'high', 'mid' or 'low'. However there is a general lack of agreement in terms of recognising the tonal quality of particles. There have been studies that suggest the alternative view is true (Detarding, 1994), and that a particle, 'especially when sentence final, functions as the carrier of the intonation contour of the utterance' (Gupta, 1992). This disagreement meant the 'number of different particles lah remain a mystery' (Wong, 2004: 760). However, what is clear from this is that it is very possible for a particle to have at least two different meanings and functions - which some have argued, is differentiated by tone.
Focus has tended to be placed more on Singlish particles as a group, rather than dissecting their various meanings and grammaticality individually. While there is now growing attention and research done in analysing individual Singlish particles (Wee, 2002, 2003; Wong, 2005; Lee, 2007), data and analysis are still lacking in terms of quantity. In this paper I will focus on the particle leh, which is sometimes also written as lei.
However, while I am of the view that pragmatic particles in SCE do possess tonal variants, I would like to challenge the notion that the different meanings and functions leh can be categorised into just tonal variants (Lee, 2007), and I argue that there are actually two different particles in leh, which possess tonal variants of their own. This is a significant distinction because these two particles of leh, can be produced with the same tone, yet their meanings differ. I also seek to add on and complement previous research in this area of study.
2. Neglected variant of leh
2.1 Unpopular leh?
Leh is seemingly a less frequently used particle, and Wong (2004) suggests that less popular particles such as leh - as compared to particles such as lah - could be less popular due to it 'lacking certain components that allow users to express their cultural norms of interaction in everyday speech' (2004: 764). In fact, an analysis of the ICE-SIN corpus (containing about 600,000 words) showed leh to be the third least commonly used particle out of 10 particles selected (Ler, 2005). The particle's seemingly low occurrence in Singlish utterances and conversations could also be one of the reasons for a lack of attention and research in this area.
2.2 /lÉ›/ and /le/
It is notable, and also the central point of focus in this paper, that leh exists in more than one form. By form, it is meant that leh possesses a different version of itself which differs in not just tone, but pronunciation, and also in meaning. When the particle leh is mentioned, the typical representation of it is that of /le/. In this paper, I seek to bring added attention to its heteronym, /lÉ›/. It must also be added that within the two separate versions of leh, different meanings and forms of usage can be derived as well, depending on context.
Studies in the past have typically mentioned or focused simply on the /le/ version of leh, failing to distinguish the difference between the two versions (Gupta 1992, 1994; Lim, 2004, 2007). Even when the difference in function and meaning is noticed and investigated, a distinction between them fails to be made (Lee, 2007), thus hampering further analysis of /lÉ›/ as a separate particle. I contest that previous assumptions are incorrect, and will seek to show the marked distinction in the usage of /le/ and /lÉ›/ in this paper.
I believe one of the main contributing factors for the lack of distinction of these two different particles, is due to them having an entirely similar Romanised form of leh. I have mentioned earlier that these two particles function as heteronyms. Due to the fact that they have identical Romanised forms, and tone is often not encapsulated in a Romanised script, it breeds a sense of the two particles being the same. While they do share certain traits in terms of meaning, I would like to prove in this study that the range of their meanings would point to them being separate, independent particles.
It is also worth noting that the fact that little attention has been paid to the /lÉ›/ form of leh could possibly be attributed to it being a somewhat recent development. As mentioned by Lim (2011: 13), 'Singlish is capable of displaying a continuum of possibilities in prosodic phonology, of more mesolectal and more basilectal features', and 'Singlish may in fact be viewed as changing, in the more Sinitic-dominant ecology of the recent era, to display more Sinitic features'. It is thus worth hypothesising that change had possibly occurred to the particle itself, in the process changing its pronunciation and range of attitudes conveyed. This however, will not be the main focus of the paper.
3. Previous Studies
3.1 Gupta (1992), and Lee's (2007) arguments against Gupta (1992)
Gupta (1992) analysed a total of 11 different particles in Singlish, and categorised them on a scale of assertiveness. Gupta's claim is that these particles can be marked as belonging to three main categories of 'contradictory', 'assertive' and 'tentative'. According to her, contradictory particles are used in cases where there is an explicit contradiction to something that had been previously said. In her terms, these particles are maximally assertive. Next down the scale is the assertive group, this group consists of particles which 'express speakers' positive commitment' (Gupta, 1992: 37) to a particular utterance. Finally, there is the minimally assertive tentative group. Particles in this group present a less positive commitment as compared to the assertive group.
Gupta's attempt to categorise these particles on a scale of assertiveness have met with some contention and disagreement (Wee, 2002; Wong, 2004). In particular her choice of labelling Singlish particles as being assertive in nature has been challenged, and Wong (2004) suggests that this classification displays an ethnocentric view in that Gupta is taking an 'Anglo cultural perspective' (Wong, 2004: 752). Moreover, I will not be dealing with the contention surrounding Gupta's scale of assertiveness in this paper.
Gupta (1992) also included a non-pragmatic classification of leh (appears as lei in Gupta's paper), which she lists as being 'used in x-interrogatives which do not have a wh-word, and recaptures a contextually understood antecedent x' (Gupta, 1992: 36). This version of leh in question is classified as a high-level tone leh (Platt 1987; Platt and Ho, 1989), and Gupta claims in this way, leh can be translated as 'What about?' when used as a question particle. This is a view first mentioned by Platt (1987), but is not an accurate analysis of the particle. As mentioned by Lee (2007), this translation 'cannot adequately capture the meaning of every occurrence of leh'. As with many other pragmatic particles in SCE, contextual factors play a huge role in the interpretation of the meaning of the particle. While leh can carry the meaning of 'What about?', it is by no means restricted to such a definition. Lee (2007) takes issue with this particular definition of the question particle leh, and tried to explain how 'What about?' cannot be used in certain contexts with question particle leh. I list out two examples he used to illustrate his point:
A: Everyone's here. Let's go.
B: Wait. Siew Lian leh?
A: Alan will wipe the tables, and John will sweep the floor.
B: Then Siew Lian leh?
Lee (2007) claims that in (2), the utterance with leh should be 'Wait. Where is Siew Lian?', and in (3) it would then take the suggested translation of 'What about?'. He goes on to say that this is proof of how high-level tone leh does not necessarily carry the meaning of What about?'. What he is trying to explain is that how an utterance is shaped with the use of leh will be shaped by the context of which it is used in. I agree with this view, however, I disagree with how he illustrates his point.
This is because in trying to argue for the flexibility of high-level tone leh, he has neglected to acknowledge a similar flexibility in terms of meaning for 'What about?' as well. For example in (2), 'Wait. What about Siew Lian?' can still be construed and understood as a question asking for the whereabouts of Siew Lian. It does not necessarily have to be framed as an explicit question demanding for the whereabouts or location of Siew Lian in 'Where is Siew Lian'. The converse is true for (3), and we can flip Lee's (2007) claim against himself by showing that 'what about Siew Lian' could also be translated explicitly as 'Then what would Siew Lian's duties be?'. Thus, similar to high-level tone leh as a question particle, 'What about?' is also dependent on context. This can be further illustrated in the following:
A: We will all be going to the party at 10.
B: What about Tim?
B's utterance can be understood as asking whether Tim would be going to the party as well. However, supposing Tim has fallen very sick and needs someone to be by his bedside. Both A and B know it would be near impossible for Tim to go to the party, and the most immediate concern would be having someone look after Tim. In this case, B's utterance would be taken to be asking 'Then who's going to take care of Tim?'. Thus, we can see that 'What about?' functions similarly with high-level tone leh in that they depend upon contextual factors for a correct interpretation. As such, Lee's (2007) usage of a narrow definition of 'What about?' as an argument is not conclusive evidence for the incorrectness of this definition for high-level tone leh.
However, this is where the similarities between them end. Using the following example:
A: If Sam comes tonight, we can finish.
B: Then he don't come leh?
In (5), 'What about?' cannot be used anymore. In such a scenario, 'What if' would be more appropriate, as in 'What if he doesn't come?'. 'What if' and 'what about' have markedly different meanings, and while we will not delve deep into their range of meanings and functions, it can be agreed that they do not carry the same meaning. As we can see then, high-level tone leh can be used more than as 'What about?'. As such, it is incorrect - as suggested by Platt (1987) and Platt and Ho (1989) - to translate high-level tone leh as simply 'What about?'. While leh in this form can possess such a meaning, it does not necessarily always translate to such a definition.
Gupta (1992) also included the 'maximally assertive' leh, and claims that 'both declaratives and imperatives with leh frequently, but not always, function as directives' (1992: 42). She also manages to recognise and identify the /lÉ›/ form of leh when she used the following data:
[YG finds passing-out parade picture]
YG: Soldier is like that one leh? [high rise]
She stated that in (6), it meant the statement with leh was made with an expression of surprise. In her words in such a case, the speaker 'makes an observation, of which there is no doubt, but which is unexpected' (Gupta, 1992: 42). This is different from leh the question particle, and leh the assertive particle which she identifies in her paper. Despite identifying this, she fails to distinguish the difference between /lÉ›/ and /le/, and treats it as a deviation of the meaning of leh as a result of context, rather than see it as a separate particle. Lee (2007: 6) brings up another point of contention when he suggests this example is a case of 'misguided induction, caused by the inadvertent use of an isolated example'. He goes on to say this is most likely a 'performance error, whereby the particle is being misused', and claims this is 'typically unacceptable to a native SCE speaker'.
First of all, it is not clear as to which pronunciation of leh Gupta is referring to in her excerpt. If the version of leh used is /le/, then indeed the particle does seem out of place in the utterance. In this case, I would agree with Lee that there is a performance error. However, supposing the /lÉ›/ version is the one used by the speaker, there would be nothing wrong at all, and it is definitely acceptable. As I will show further in the study, there are examples of native SCE speakers reproducing this leh in other instances. Lee (2007: 6) also continues and suggests the use of the particle meh instead - in place of leh - as being more appropriate. This is highly incorrect, as this would give the utterance an entirely different meaning. Since this study does not concern the study of meh, I will not dwell on this issue.
Moreover, this confusion highlights one of the problems of not distinguishing clearly the two different particles of leh - /lÉ›/ and /le/. The fact that one of the variants of leh would be out of place in a utterance that would be appropriate with the other, further supports the argument that these two should be treated as two different, separate particles.
3.2 Wee (2004)
Wee (2004) in his paper included a different function of leh. He suggests that leh marks an assertion or request as being tentative, and therefore working as a pragmatic softener (2004: 122). In his analysis, Wee (2004) used an excerpt from the GSSEC to display such a softening function:
A: Actually… come to think about it actually, er, this movie speaks very badly about men leh.
According to Wee (2004), leh softens the opinion that A makes, and signifies that it is a weak opinion, therefore explaining the speaker's hesitance and sheepishness in making the statement. However, it is not clear what version of leh the speaker used. In fact, both /lÉ›/ and /le/ can be used in such in instance without affecting Wee's interpretation. Despite this, I argue that /lÉ›/ and /le/ exhibit different states of minds and attitudes of the speaker, and as such cannot be used interchangeably.
Both versions exhibit a 'softening' function, however to different degrees. When /lÉ›/ is used, it suggests more assertion rather than softening, and the opinion is more forceful than when /le/ is used. For this reason, it is assumed that the /le/ version is the one Wee (2004) is referring to. Wee (2004:122) also used the following:
(A and B are talking about a movie)
A: But so few people lah, maybe because it has been running for quite some time lah.
B: Actually two weeks only leh.
Wee again uses this as an example of leh acting as a softener. Once again, it is not clear which leh is used. In this example, if /lÉ›/ is used, the assertion of it being 'two weeks only' is much more forceful than that of /le/. It further goes to show how these two particles should be defined properly. I will address the issue of leh being used as a 'softener' later in this paper.
3.3 Lee (2007)
Lee (2007) identifies three tonal variants of leh in his paper, and suggests that each particle 'has its own pragmatic function realised in specific speech contexts'. He claims they can be realised in tone 1, tone3 and tone 4 of Mandarin Chinese respectively.
In addition to identifying the more commonly known /le/ as leh1, Lee (2007) also correctly identifies /lÉ›/ in his paper. He found two tonal variants of it - leh3 and leh4. However he too, makes no mention of the differentiation in pronunciation. For the sake of discussion, we will assume that he has made the correct distinction between /lÉ›/ and /le/. In marking the variants of /lÉ›/, Lee (2007) categorises leh3 as being a 'marker of intent' and leh4 as a 'marker of assertion'. One of the functions Lee (2007) has accorded the use of leh3 is that of reporting a new state of affairs which is 'assumed to be beyond the addressee's knowledge'. However, as I will show later in the presentation of data, that this is not necessarily true. Even with speakers knowing that each other has knowledge of a particular piece of information, the particle can be used.
Lee (2007) also does not fully expand the section on the usage of the leh4 variant that he calls marker of assertion. He suggests only a 'subtle distinction between the third of fourth tones of the particle' (2007: 15). This is not true, and as data will show, leh4 can capture and reflect different meanings and attitudes of the speaker, including the suggestion of exaggerated emphasis or sarcasm. This as a result would mean more than just a subtle distinction between the two particles.
4. The different forms of leh
The data used and presented in this paper have been taken from conversational interactions between Singaporeans using SCE. Because of the naturally occurring circumstances of which these data have been obtained, they are authentic instances of the use of SCE (and thus the use of the leh particles). These data would be analysed and they form the basis of presenting the different forms of leh in this paper.
In Lee's (2007) study, he separated leh into three different tonal variants, where within one tonal variant the particle could take on different meanings. I would take a similar approach, however I make a marked distinction between the /le/ and /lÉ›/ forms of leh.
The analysis would be separated into two major parts, with the first part being the more commonly known /le/ version of leh, and the second detailing the /lÉ›/ of leh which I will represent with lea.
4.1 /le/ leh
The /le/ leh consists of a number of variants, and I will attempt to separate them into their particular functions.
4.1.1 Leh as a Softener - realised as tone 1 in Mandarin Chinese
As has been categorised by Lee (2007: 7), this leh 'occurs in the second part of an adjacency pair, whereby the speaker is unable or unwilling to provide a preferred response with respect to a proposition introduced by the addressee in the first part of discourse'. Lee calls this the 'dispreferred second' (2007: 7). According to Yule (1996: 79), 'the preferred is the structurally expected next act and the dispreferred is the structurally unexpected next act'. Therefore, disagreements and refusals are next acts which are unexpected. Yule (1996) has found that in English, hesitation and prefaces are used to make a response (as a dispreferred second) less challenging to the first, thereby softening an unexpected next act. In SCE, leh performs a similar function.
Thus as mentioned, this leh is used by a speaker in response to something said prior that the speaker does not agree with. This does not necessarily have to be an opinion of which one can blatantly state an agreement or disagreement with, but also requests or false statements. The leh is tagged to the dispreferred second of an utterance, and thus performs a function of softening the blow of an unexpected next act.
A: Eh, later after this we go get something to eat.
B: Later? I got something on leh.
In (9), A is suggesting to B that they go for a meal after their current activity. However, B is unable to make it. In this instance, B is offering a refusal to A's invitation. He does not refuse the invitation outright. Rather, he answers the question indirectly by stating that he has 'something on', which naturally would be taken to mean that he wouldn't be free for A's meal invitation. This indirect refusal could be taken to be a way of making the refusal less challenging. Even so, leh is still used to further soften the refusal.
Leh is able to perform this softening function because the usage of leh suggests an attitude of a willingness to compromise, or to negotiate a position - whether it is an intention the speaker is trying to convey or not. For instance in (9), because the use of leh softens the unexpected next act of refusal, we see the following exchange:
A: Eh, later after this we go get something to eat.
B: Later? I got something on leh.
A: Huh? What thing? Eat lah!
B: Cannot lah.
The use of leh by B suggests a position of the possibility of negotiation rather than closing the door on any possibility of accepting the invitation with an outright rejection, which explains why A continued by attempting to change B's mind and getting him to accept the invitation. B clearly did not have the intention to negotiate because he promptly put an end to A's attempts to persuade by refusing the invitation outright in his second utterance. Thus, leh performs a softening function in suggesting a position of compromise and willingness to negotiate, therefore making a refusal or rejection less challenging.
(A and B are discussing the price of a new phone. A thinks the price is expensive, while B thinks it is a reasonable price)
A: Cheap? Then you buy me one lor.
B: Please leh, cheap also need money. Buy for you I might as well buy more for myself.