Telling People to Change Their Behaviour Through Implications: An Implicature Analysis on Covid-19 Public Service Announcements in Indonesia

Public service announcements (PSAs) are the official way for governments to inform, educate and change public behaviour in order to reduce public health issues, such as Covid19. This study aims to analyse the types of implicatures in the Covid-19 PSAs published by the Ministry of Health Republic of Indonesia in their Instagram account, @kemenkes_ri, in order to fill the research gap on implicature studies in a public health and political contexts. This qualitative descriptive research analysed a data of eleven Covid-19 PSAs which were published from March 2020 to January 2021 according to Grice’s implicature theory. The researchers discovered that PSAs employed conversational implicatures more frequently than conventional implicatures (18%) to deliver their messages, with generalised conversational implicature being more prevalent (64%) than particularised conversational implicature (18%). Conventional implicature was used only in earlier posts when Covid-19 has not been a common knowledge among the public, and once people are already used to the “new normal”, alter PSAs used conversational implicature because the public already have the context of Covid-19. Results of this study illuminated the differences between each type of implicatures and also contributed to the lack of studies of PSAs’ implied meanings, the dearth of implicature studies in a non-classroom context.


Introduction
Currently, the coronavirus disease 2019  is the most urgent threat to global public health, becoming a hot issue in most, if not all, information sources around the world due to its thorough impact in every aspect of people's lives worldwide. As a global threat, the present population are forced on a daily basis to know information related to the Covid-19 pandemic (Barnes, 2020). People of all ages and all background around the globe are expected to not only actively obtain pandemic-related information, but are also given such information by authorities of all levels as a necessity (Naeem & Bhatti, 2020). The information, ranging from health advice to the level of Covid-related emergency, are advertised through various means, such as newspapers, magazines, radio, television and social media. The researchers focused on the official means of information that governments use to spread information, namely Public Service Announcements (PSA).
A PSA is known as an advertisement that is supported by the government, although it is not the same as paid advertisement. Government organizations act as the ones who launch PSAs to the public. The primary purpose of a PSA is to change public behaviour by announcing certain types of information (Pratiwi & Hidayat, 2020). In other words, PSA information must be understood by the public. However, the information conveyed by government's PSAs can fail to change the public's behaviour, leading to unfavourable circumstances. For instance, despite government's announcements for people to wear masks during the current pandemic, the public does not obey the message, leading to a rising count of Covid-19 infected individuals (Djaafara et al., 2020;Nasir, Baeguni, & Nurmansyah, 2020). This study intends to address the problem by investigating the implicature in Covid-19 PSAs, specifically ones launched by the government of the Republic of Indonesia. Results of this study is important considering that Covid-19 is currently the leading barrier to the country's well-being and prosperity.
As one of the hottest issues in research, studies on Covid-19 has proliferated exponentially in in the past year. Most were undeniably from researchers with medical backgrounds because of the demand to learn, identify and prevent the spread of the virus (Bentlage et al., 2020;Drissi et al., 2020;Rahmani & Mirmahaleh, 2020;Zheng et al., 2020). Management of environment, economy and education (Butler-Henderson et al., 2020;Castro & Zermeño, 2020;Golan, Jernegan, & Linkov, 2020;Talib, Bettayeb, & Omer, 2021) have also been productive in producing and publishing papers regarding how people perceive Covid-19 and how the pandemic affected various parts of the system and daily life. Studying information distribution of Covid-19 has also been carried out by many researchers. To name a few, Bento et al. (2020) took an information technology angle in studying Covid-19 by investigating people's information-seeking responses to news of local COVID-19 cases, Kim et al. (2020) explored the misinformation on people's information seeking, Wang et al. (2020) observed how Facebook users cope with Covid-19 in relation to information sources. In the political side, majority of studies explored government responses to Covid-19 in a national and international scale (Greer et al., 2020;Hale et al., 2020;Haug et al., 2020). The studies cover a very wide range of responses in order to understand, inform and rate the effectiveness of government's responses. This study takes on a more specific route by examining a particular type of government response, namely the government's response to Covid-19 pandemic by educating their people using public service announcements. shows an example of a Covid-19 PSA published on May 22, 2020 by the Ministry of Health Republic of Indonesia or Kementerian Kesehatan Republik Indonesia (Kemenkes-RI). The message displayed, "Maskerku melindungimu. Maskermu melindungiku" which means "My mask protects you. Your mask protects me," accompanies the picture of a diversity of people wearing masks. Taken literally, the information published only delivers what someone's mask can do, which is 'protecting'. However, this PSA carries the implied message of reminding the public to wear masks. The context of this announcement is the fact that a significant amount of Indonesian society still neglects to wear masks, thus contributing to the increase of positive Covid-19 patients on a daily basis. The phrases "your mask" and "my mask" implies that everyone needs to use masks, and that the government expects the community to do so in order to protect themselves, their families and other people. Through this example, it can be seen that what is said literally is different from what is implied.
The pragmatics study of implicature, for the most part, were explored in the context of classrooms. While the premise of these studies of implicature is often concerned with the fact that what speakers say means much more than what they were actually saying, majority of previous studies were concerned with evaluating students' comprehension of implicatures. Recent studies have applied implicature theory to analyse literature works to offer insight on the implied meanings of authors' writings beyond the lens of critical analysis or psychoanalysis (Igwedibia, 2017;Tiara, 2018). However, as far as the researchers are aware, studies on implicature very rarely explored other contexts, such as implicatures in politics or media. This current study will use Grice's theory on implicature (Grice, 1975;Syafitri et al., 2019) as the foundation for the investigation on Covid-19 PSAs. This study might be one of the firsts to explore implicatures in the matter of government service and Covid-19, and its contribution will be significant in opening the way for more studies on implicatures in similar languagedependent communications.

Public Service Announcement
Countries around the world have many problems which stemmed from human behaviour, including but not limited to smoking, substance abuse and suicide among others. The prevention of these problems must be preceded or complimented with raising people's awareness of the problems in question, and this role is fulfilled by public service announcements (PSAs). Crozier, Berry, & Faulkner (2018), Gordon, Bell, & Ranta (2019), Wakefield et al. (2013), and many others have showed that PSAs can be effective in changing people's attitudes and behaviours regarding some health problems.
From what can be found in the literature on PSAs, the researchers note that PSAs themselves seem to be seldom studied. Most studies use PSA as a catalyst for an effect and were more concerned in investigating those effects rather than the PSAs, usually by distributing surveys to a great number of people. For example, Crozier, Berry, & Faulkner (2018) explored parents' attitudes towards children's physical activities after viewing a PSA called Mr. Lonely, which highlights the negative effect of screen time on children's outdoor playtime. The study's examination found that the PSA's message variables (i.e., personal relevance, novelty of information, and feelings toward sponsoring organization) showed an overall positive relationship with parents' attitudes. Reidenberg & Berman (2020) highlighted the scarcity of studies evaluating the impact of suicide prevention PSA campaigns as most only demonstrated how it can increase awareness or knowledge. Results of the study offered initial evidence on how suicide prevention PSA campaigns influenced audience to engage in help-seeking behaviours in the short-term. Similarly, Lim, Lee, & Lim (2020) surveyed more than 600 people to examine the firstperson effect of anti-panhandling PSA messages on promotional behaviours and donation intentions. The firstperson effect had a bigger impact on promotional activities than it did on contribution behaviours and largely mediated the influence of message desirability on promotional behaviours. Liem, Marta, & Panggabean (2019) critically analysed the sanitation behaviour of Indonesians through conversations regarding the effectiveness of a PSA meant to deal with the prevalence of stunting in Indonesia by informing people about poor sanitation behaviour and stunting. Rooted in the fact that PSAs can influence behavioural change, a study proposed a way to improve the automatic smoking detection system by automatically displaying a PSA about the hazard of smoking in public when the sensors triggered an alarm after detecting a certain level of smoke (Arief et al., 2020).
When it comes to Covid-19 as a public health concern, this study found only found one study which explored people's perceptions of Covid-19 public service advertisements, not public service announcements (Putri, 2021). Though the distinction is small, a PSA is not a paid advertisement as the organisation that produces the PSA does not pay to have it aired. Putri (2021) distributed questionnaires via Google Form to a variety of age groups, from 15 to 60-year-old Indonesian citizens, and the results culminated to a confirmation that most respondents often access, receive and share public service advertisements regarding Covid-19 on social media.

Implicature
In the study of pragmatics, other authors focused on principles of politeness and communicative efficiency, but Herbert Paul Grice was the first scholar to systematically study cases where what a speaker utters is different with what they mean, and he was the one who coined the term "implicature" (Wang, 2011). Implicature described the implicit meaning or the meaning implied. It is a perceived meaning that extends beyond the literal meaning. According to Grice's theory, there are two types of implicature, namely conventional and conversational implicatures. The difference lies in the need for context; the former is independent of context, whereas the latter requires people to have the context to understand the implied meaning. Moreover, conversational types can be divided into two more subtypes, namely generalised conversational implicature which does not require a particular background knowledge and particularised conversational implicature which does require a particular background knowledge.
A significant amount of implicature literature were focused on seeing and evaluating students' comprehension of implicature, usually students from English, English Teaching or English Literature programs. This is in conjunction with the recent attention on the ability to teach pragmatic theory in classroom settings (Murray, 2011). According to Samaje & Ariyanmanesh (2018), pragmatic competence is one of the most important aspects in language learning, yet it is understudied in comparison to other aspects of communicative language competence. In the existing, most focused on speech acts framework, while implicatures and other features of pragmatic knowledge are still currently being evaluated.
Derakhshan & Eslami (2020) also tried to investigate Iranian English major students' comprehension of literature by using fifty-six videos from the TV series Friends and Desperate Housewives. They tested the students with a multiple-choice implicature listening test and found that students' comprehension can be effectively promoted if they are provided with contextually appropriate input through video using methods of pragmatic instruction (e.g., metapragmatic consciousness-raising, translation, and discussion). Their results were a continuation and proof of their previous study on the effectiveness of video-input for teaching pragmatics (Derakhshan, Mohsenzadeh, & Mohammadzadeh, 2014). Alsmari (2020) tries to see if flipped classroom instruction affected Saudi EFL learners' comprehension of conversational implicatures and found that effective out-of-class preparation and appropriate manipulation of in-class time did increase pragmatic competence.
Other than trying to see if some form of technology can be effective in teaching pragmatics, other studies tried to see if understanding of certain knowledge or possession of certain level of skills can affect comprehension on implicatures. Abdelhafez (2016) had English Teaching program students interpret conversational implicature and investigated the effect of their training, specifically on whether their understanding of conversational implicature developed their pragmatic competence and language proficiency. Results revealed a significant correlation with both. A similar study administered multiple-choice test on conversational implicature to see if students' English proficiency correlate with their comprehension of implicature and found that the case is indeed so (Samaje & Ariyanmanesh, 2018). Instead of relying on one variable such as language proficiency, Zand-Moghadam & Samani (2021) ambitiously pursued the effect of multiple variables, namely information-gap, reasoning-gap, and opinion-gap tasks, on EFL learners' pragmatic production, metapragmatic awareness, and comprehension of implicature.
A considerable number of studies on implicature comes from the literature discipline. Budiaty (2014) carried out a content analysis aimed to identify the types of implicatures used in The Princess Diaries novel. Results found that there were more conversational implicatures than conventional implicatures. Another study tried to analyse the types of implicatures that could be found in the classic, international movie owned by Disney called Beauty and the Beast (Tiara, 2018). Results of the study noticed that there is more evidence of generalised conversational implicature than particularised conversational implicature, positing that because in a certain social group, there are some factors that allow the existence of an implicature in the communication process. Jane Austen's Pride and Prejudice is also another novel that has been extensively analysed, and Wibawa (2017) contributed in analysing its implicatures, citing that their presence contributes in the process of character development and influenced by the setting and plot of the novel.
Though poetries have been observed to usually be studied through the lens of literary and critical analysis, they have recently begun to be analysed pragmatically. Igwedibia (2017) highlights how many have been written about an American writer, feminist, womanist, librarian, and civil rights activist named Audre Lorde's poetry, but there is absolutely no evidence of a pragmatics study of her work and so the study tried to fill this gap. The study concluded that Audre Lorde's poems sometimes adheres and sometimes violates Grice's theory. Similarly, script for plays or theatres are frequently analysed with psychoanalysis, literature and even theology, but seldom with pragmatics. Megawati (2019) addressed that gap by analysing the implicature within the script of Oedipus Rex, revealing passages that contained conventional or conversational implicatures.
Few, if any, studies ever explored implicatures in contexts beyond the classroom or learning pragmatics. This study only found Mayora (2010), who examined the use of conversational implicatures in Kenyan print media interviews. The study's data were from two Kenyan newspapers, and the results showed the interviewees in one media gave more indirect "yes" or "no" to yes/no questions than direct answers, while the case is the opposite with the other newspaper media.

Method
This study is a qualitative descriptive research as the data were collected in the form of words and pictures rather than statistics (Thao & Herman, 2020;Sinaga et al., 2020), specifically the utterances in the Kemenkes-RI's Instagram account's public service announcements regarding Covid-19. The researchers explored @kemenkes_ri's posts from March 2020, which is when Covid-19 became a national public health concern in Indonesia, to January 2021 (eleven months). All instances of Covid-19 PSAs were collected with screen-capture or in screenshots. Then, content analysis is used to identify the data's implicature referring to Grice's (1975) classification of implicature.

Results
This study found a total of 11 Covid-19 public service announcements that had been published by the Kemenkes-RI since March 2020. Results showed that one type of implicature is more present than others, as illustrated in Figure 2.

Figure 2. Number of Implicature Types from 11 Covid-19
PSAs on @kemenkes_ri Compared to conventional implicature, this study found that conversational implicature is more commonly used by the government to convey information about Covid-19 to the public. According to Grice (1975), this type of implicature has two sub-types, namely generalised conversational implicature and particularised conversational implicature. The former depends on the context and does not require audience or viewers to have any particular background knowledge to understand the implied meaning. This type is found to be the most dominant-seven out of eleven PSAs (64 per cent). The researchers postulated that it is due to the fact that the knowledge of Covid-19 is now a very common information among the public as Indonesia has been living with Covid for more than ten months to date. This means that most people can understand what is implied in the PSAs generally.  The description of the data illustrated in Figure 3 is "Hidup Normal Baru, Hidup Lebih Sehat dan Produktif", which means "The New Normal Life is a Healthier and More Productive Life". The message is accompanied with an image of a person working on their laptop in front of a clock. This public service announcement was posted on July 3 rd , 2020, four months after the country has urged the public to work from home.
The intended message of this PSA is apparent, which is to tell people to continue all of their activities and work-life by following the health protocols. The goal of Kemenkes-RI in publishing this post is to make sure that economic activity will not be negatively impacted further by Covid-19, but in fact, increase as people can be more consciously productive and healthier. Another example of generalised conversational implicature is depicted in Figure 4. The message "Jaga jarak untuk melindungi sesama" tells people to "Keep your distance to protect others". From the accompanied caption in the middle-bottom of the picture, "Jarak 1-2 meter" which means "Distance 1-2 metres", it can be inferred that the Government and the Health Ministry Indonesia suggested people to remain keeping one till two metres from another person.
This PSA was posted on September 23 rd , 2020. At this point, people have been living with Covid-19 for at least half a year. Indonesian citizens have had to adjust to living with social and physical distancing, so the public is presumed to have already understood the importance of keeping distance. This PSA's implied meaning is to remind people to keep doing so, as they do not know of the health status of people around them. By keeping their physical distance, the action can protect everyone, including siblings, parents, friends and everyone in their community from being infected by the virus.
The other sub-type of conversational implicature is particularised conversational implicature. Opposite to the first sub-type, this type does require the reader or interpreter to have a particular background knowledge to understand the implied meaning. Only two out of eleven PSAs published from March 2020 to January 2021 were found to contain this type of implicature (18 per cent). Figure 5 contained a long message announced to the public regarding a specific type of tradition that is unique to Indonesian people. The image of a man and a woman, presumably in love or in a romantic relationship, were accompanied with the hashtag of "#TidakMudik". This hashtag has the equivalent meaning of "not going home". This PSA is categorised as a particularised conversational implicature because mudik is a specific culture and tradition for the dominantly Muslim people in Indonesia to return to their hometown during Lebaran Idul Fitri or Eid al-Fitr. At this time, every Moslem visits their family, both as an obligation and a voluntary and beloved tradition to stay in touch with relatives, known as silaturahmi.

Figure 5. An example of Particularised Conversational Implicature
The message of PSA tells people to not go home, but not because the government forbids silaturahmi. This is implied in the message, "Bukan tak ingin bertemu, tapi demi masa depanmu", which means "Not that I don't want to meet, but for your future". The ban on mudik acknowledges mudik's importance, but highlights the importance of safety above it.
"Kamu kangen sama dia? Jangan mudik dulu yaa, karena siapa tau dirimu malah dibawa "oleh-oleh" corona dan buat dia dan keluarganya. Kangenmu simpan dulu di kulkas biar awet sambil tetap #Dirumah Aja." This long description means "Do you miss her? Don't go home first, because who knows, you might even give the "gift" of corona for her and her family. Keep your longing in the refrigerator so that it lasts while still #StayingAtHome". This part of the message contained another hashtag to popularise the movement of staying at home, as well as a couple of jokes that can be understood by the locals who are familiar with the situation. The message posits its position as if they are advising the male in the picture to "freeze" his longing for the female in a refrigerator, implying that people can control their feelings much like they can control food from spoiling so they do not have to feel forced to travel to their hometown. Another joke is also about giving gifts, which is a normal part of mudik, but in this case, the gift could be the coronavirus. This part of the message warns people of the consequences of visiting their families and loved ones, which is that they might be infecting them in the end and risking their lives.
This PSA was posted on April 18 th , 2021, a little under a month before Eid al-Fitr 2021 began in the evening of May 12 th , 2021 and ended in the evening of May 13 th , 2021. The spread of Covid-19 in Indonesia, which began in March 2020, continued until Eid 1441 H took place. Kemenkes-RI posted this message to ask the public, particularly Muslims, to not take part in mudik, not because staying in touch with their loved ones is no longer alright, but because keeping their loved ones healthy and safe from Covid-19 is more important than meeting them.
Last but not least, this study identified two out of eleven Covid-19 PSAs published by the Kemenkes-RI contained a different type of implicature, known as conventional implicature. Where conversational implicatures require people to know the context, conventional implicature is independent of context and non-truth conditional inferences. This is illustrated in Figure 6. This public service announcement presents the question "Siapa yang perlu memakai masker?", which means "Who needs to wear a mask?" Two answers are offered for this question in the bullet points below it, "Jika sedang sakit flu atau batuk" and "Jika sedang pemulihan dari sakit", each respectively means "If you have a cold or cough" and "If you are recovering from illness".
Covid-19 is never explicitly worded in the message of this PSA. This study noted that it was posted on March 2 nd , 2021, which marks the beginning of the spread of Covid-19 in the country and when the government acknowledged it as a threat. At this point, the public is not familiar with Covid-19. Social and physical distancing has not been implemented yet. However, this PSA is informing people about the virus by telling them to wear masks, an object meant for people with certain sicknesses that are symptoms of the virus. This announcement is context-independent as the use of mask is not yet known to be the provision for Covid-19 but also the provision for other illnesses, so it can be said that the context is independent. Thus, it can be inferred that the implied meaning is that people who have a cold, cough or recovering from illness need to wear masks as they may be infected by Covid-19, although the public lacks the context of Covid-19 at this point, thus this kind of announcement is a conventional implicature.

Discussion
Public service announcements were launched by government bodies to change public behaviour by publishing information that contains different types of implicature. This study is contributed in the exploration of pragmatics in the political and public health context. Over the course of eleven months, from March 2020 when Covid-19 is acknowledged as a national health emergency in Indonesia to January 2021, this study found only eleven Covid-19 PSAs were launched in the Instagram account of the Ministry of Health Republic of Indonesia (Kemenkes-RI). Nine out of eleven contained conversational implicature, with generalised conversational implicature being more dominant (64 per cent) than particularised conversational implicature (18 per cent). The conventional implicature is only found in two of the eleven Covid-19 PSAs (18 per cent), both of which were posted on March 2021. The rest of the PSAs published on April 2021 and onwards are conversational implicature, because at that point the public has acquired the background knowledge of  Reflected in the literature review and the results of this study, conversational implicatures are more present than conventional implicatures. This is because "conversational implicatures are ever present and commonly used in our dayto-day conversations. They are of importance in our conversations and it is almost impossible not to use them when it comes to communication," (Mayora, 2010, p. 1). Studies on conventional implicature are scarce, even in studies that investigates students' comprehension of implicatures. At most, conventional implicature appears on literature when researchers tried to identify types of literature, such as in Budiaty (2014) and Megawati (2019). The findings of this study illustrated the distinction clearly through the timeline of the public service announcements' publications. Earlier posts of Covid-19 PSAs were the conventional type of implicature because at the time, the public did not have the general context of Covid-19 as the pandemic was still new and physical distancing has yet to become the norm. Once physical distancing type of life became the "new normal" for many people, later Covid-19 PSAs contained conversational implicatures as people have become more familiar. The results of this study may contribute in illuminating a little bit more on the nature of conventional implicature for future studies to explore more in-depth.
Regarding conversational implicatures, the findings of this study echoed previous studies that tried to identify types of implicatures. Similar to Tiara (2018), generalised conversational implicature is considerably more evident than particularised conversational implicature. The data indicates that this is most likely attributed to the nature of public service announcements. PSAs are essentially created to inform and educate the public at large, so the message should necessarily be framed in a way that can reach the largest amount of audience. This explains why there are more generalised conversational implicature than particularised conversational implicature. When the government needs to convey an instruction by implying it in the PSA to a specific community of the public, such as Indonesian Muslims who are getting ready for mudik (Figure 5), then they switch to particularised conversational implicature.
This study also sheds light on the importance of ensuring that the messages implied in public service announcements can be truly realized. Figure 5 is a PSA that asks Indonesian citizens to forgo mudik in order to not risk the health of their loved ones. However, the meaning of this message fails to reach many of the public, as many Indonesians still travelled to their hometowns on May 2021 which leads to an exponential spike on the Covid-19 positive cases in Indonesia by June 2021 (Andi, 2021). Taking a preventive measure beyond announcing information is certainly expected to ensure that the implied message in government's PSAs is followed by the public, but Nadine & Imtiyaz (2020) showed that government-issued regulations still have problems during their implementation.
Overall, there is little the researchers can do in terms of critically comparing the results of this study with previous studies since it explores a rather new territory. Studies on implicatures have mostly focused in the language learning classroom context, either in evaluating students' understanding and comprehension of implicatures (Alsmari, 2020;Derakhshan & Eslami, 2020) or investigating the effect of certain teaching models or learning technologies in the teaching of this pragmatics feature (Abdelhafez, 2016;Samaje & Ariyanmanesh, 2018). Many previous studies that investigated types of implicatures did not look at public service announcements, but works of literature such as The Pride and Prejudice novel (Wibawa, 2017), Princess Diaries novel (Budiaty, 2014), Beauty and the Beast movie (Tiara, 2018), Oedipus Rex play script (Megawati, 2019) and poems written by Audre Lorde (Igwedibia, 2017). Even then, the first three studies were dissertations, with the fourth being a conference proceeding and the fifth being the only peerreviewed journal article.
There is also a dearth of public service announcement studies that are similar to this study since most of them were concerned about the effect of the PSAs rather than the PSAs themselves. As far as the researchers are aware, they also have yet to focus on Covid-19 PSAs, but on other public issues such as excessive screen time (Crozier, Berry, & Faulkner, 2018), suicide (Reidenberg & Berman, 2020), panhandling (Lim, Lee, & Lim, 2020), stunting (Liem, Marta, & Panggabean, 2019) and smoking (Arief et al., 2020). The most similar previous study is Putri (2021), who was also concerned about Covid-19 PSA. Although this study is entirely different since Putri's (2021) aim was to collect data on Indonesian citizen's perceptions of Covid-19 PSAs through online questionnaires. This previous study was not at all concerned about types of implicatures or other pragmatic features. Meanwhile, the only study on implicature that used data from the media was Mayora (2010), who wasn't concerned about Covid-19 PSAs but rather the tendency of Kenyan interviewees' answers to yes or no questions. Therefore, this study believes that it contributes significantly in showing new paths for future researchers who are interested to explore implicatures or pragmatics using public service announcements as their base of data. Furthermore, the interest of studying implicatures often use English language data, so this study also contributed in exploring implicatures using data derived from non-English language.

Conclusion
This study collected evidence of public service announcements on Covid-19 that were distributed by the Ministry of Health Republic of Indonesia (Kemenkes-RI) on their Instagram account, @kemenkes_ri. After analysing a total of eleven Covid-19 PSAs which were posted from March 2020 to January 2021, the researchers identified that the PSAs rarely used conventional implicatures to convey their messages (18 per cent) and dominantly used the conversational type of implicature, with the subtype generalised conversational implicature being more present (64 per cent) than particularised conversational implicature (18 per cent). While previous studies have attempted to identify types of implicatures, this study offers a different perspective by its unique data and context, revealing new insights that may explain why one type of implicature is more frequently used than the other. Moreover, this study offered findings to fill the research gap of implicature studies in a non-classroom context.