Abstract
This article focuses on response comments related to the use of offensive language in Ghanaian political discourse on GhanaWeb (www.ghanaweb.com). Data were obtained from the Corpus of GhanaWeb Comments on Ghana's 2016 Election (CGCGE16). The article examines the various ways some commenters react to inappropriate use of language on this online platform. It demonstrates that although interaction on GhanaWeb is characterised by pseudonymity and asynchronicity, there are some commenters who play the role of regulatory figures and traditional gatekeepers of appropriate communicative behaviour as they disapprove of the use of abusive language. Practically, these commenters do not contribute to the topic under discussion in a comment thread. Rather, they focus on the negative communicative practice of others and attempt to reinforce the social expectation that Ghanaians must exercise the right to freedom of speech within the limits of their socio-cultural norms and values. This research can serve as a springboard for further research on this communicative behaviour in other socio-cultural contexts.
Introduction
For many Ghanaians, the current state of political discourse is an issue of concern, with major stakeholders such as traditional rulers, clergy, politicians, academics and media critics frequently condemning the use of offensive language in the media (Ofori, 2015). The reason is that in Ghana, it is socio-culturally unacceptable to criticise a person of a higher status in a direct manner or in public. It is even more so when the criticism is couched in potentially offensive language (Thompson, 2021a). Ofori (2017) states that the use of unsavoury comments against politicians and other powerful people is considered disrespectful in Ghana because of the conservative cultural concerns of the society. However, this behaviour appears to be a “normal” phenomenon during political discussions in recent times, especially on online news platforms like GhanaWeb (www.ghanaweb.com) (Thompson, 2019).
GhanaWeb is a news website which offers news items about current events in the fields of politics, sports, entertainment etc. in addition to operating as an internet radio station for Ghanaians living abroad (Otu, 2021). The website features a comments section where users share their thoughts on the news stories. As the topics are primarily contentious, the comments generated are usually polarised and sometimes devolve into verbal attacks. This can be attributed to the website's relaxed moderation and pseudonymity (Groshek & Cutino, 2016; Hess, 2015; Otu, 2021; Sambaraju & McVittie, 2020; Thompson, 2019, 2020, 2021a, 2021b). It is, therefore, reasonable to assume that the fear of negative consequences that inhibits people from engaging in such behaviour is non-operational on this website (Otu, 2021; Thompson, 2019, 2020, 2021a, 2021b). However, participants who rely on abusive language do not always get away with it, as some users perform the role of “regulatory figures” who evaluate this communicative behaviour by countering it.
The regulatory role played by these commenters essentially involves the use of language, which is the focus of the present study. Language use on GhanaWeb, like other web-based platforms, falls within the broad domain of computer-mediated communication (CMC), defined as “an inclusive umbrella term for multimodal human-to-human social interaction mediated by information and communication technologies (ICTs)” (Meier & Reinecke, 2021, pp. 2–3). CMC has been a central topic in recent research in socio-linguistics, pragmatics, and language and communication, with the use of abusive language on these platforms gaining prominence in these studies (e.g., Badarneh, 2020; Faria, 2023; Granato & Parini, 2015; Pacheco-Baldó, 2023; Rabab’ah & Alali, 2020; Thompson, 2019, 2020, 2021a, 2021b). Such research has generally revealed that commenters use linguistic strategies such as insults to attack and portray people in authority as incapable, unserious and involved in socially unacceptable conduct. While this line of inquiry has enhanced our understanding of this online behaviour, how the targets of such behaviour and their associates respond to it has not received much attention. This is unfortunate, because as several studies (e.g., Culpeper et al., 2003; Culpeper & Hardaker, 2017) have clearly shown, since such verbal interactions are two-dimensional or dyadic, until we consider how the addressees or their associates respond to impolite language, we can only have a partial understanding of such interactions.
Of the few studies on response strategies by addresses of impoliteness (e.g., Harb, 2020; Haugh & Sinkeviciute, 2018; Talebzadeh & Khazraie, 2023), research by Haugh and Sinkeviciute (2018) remains very significant to the present study. The authors have reported that responses are framed as accusations in the forms of denials, counter-accusations and challenges which function as public condemnation of another party. Similarly, work by Harb (2020) has pointed to the situatedness of impoliteness and its responses in specific socio-cultural settings. For instance, his study showed that the use of supplications as a response strategy is a reflection of the Islamic social domain where the data were sourced. The present study contributes to this emerging research by examining the responses to insult-filled comments directed at national leaders on GhanaWeb so as to enhance our understanding of how this linguistic behaviour works in the unique socio-political environment of Ghana. The article specifically draws attention to how online participants respond to the inappropriate use of language by their co-participants. It also highlights the motivation behind such responses. It is hoped that this study can provide a fair understanding of some speech practices of Ghanaians and contribute to research on the dynamics of online discourse, impoliteness, and media and culture.
Theoretical framework
This investigation is grounded on the theory of impoliteness, which offers an excellent way of understanding face-threatening language behaviour and its response options (Culpeper, 2010, 2011, 2016; Culpeper et al., 2017; Culpeper & Hardaker, 2017; Culpeper & Tantucci, 2021). According to Culpeper (2011, p. 23): Impoliteness is a negative attitude towards specific behaviours occurring in specific contexts. It is sustained by expectations, desires and/or beliefs about social organisation, including, in particular, how one person's or a group's identities are mediated by others in interaction. Situated behaviours are viewed negatively–considered “impolite”–when they conflict with how one expects them to be, how one wants them to be and/or how one thinks they ought to be. Such behaviours always have or are presumed to have emotional consequences for at least one participant…
Culpeper (1996) taxonomised impoliteness into five supra-strategies (i.e., bald on-record, positive, negative, off-record and withhold impoliteness) and one meta-strategy (i.e., sarcasm/mock impoliteness). In response to several criticisms of this taxonomy, Culpeper (2011) later refined his strategies into formulae, as shown in Table 1.
Conventionalised impoliteness formulae.
Source: Culpeper (2011, pp. 135–136) and Culpeper and Hardaker (2017, p. 212).
As hinted earlier, since the response to an impolite verbal behaviour is as important as the impolite behaviour itself, Culpeper et al. (2003) proposed a typology of response options to impoliteness. According to Culpeper et al., when someone's face is attacked by impoliteness, the person may decide to respond or not respond, and a person who decides to respond may either accept or counter the impoliteness. A person who accepts may take responsibility for the impolite act being issued in the first place by, for instance, apologising to the attacker and, by so doing, increasing the damage to his or her own face. On the other hand, a person can counter the attack by being either offensive or defensive. Offensive strategies involve responding to the impoliteness using the same or similar impoliteness strategies (see Table 1). Thus, someone may offensively respond to a personalised negative vocative (e.g., you fucking moron) using a dismissal (e.g., fuck off). Defensive strategies, on the other hand, aim to deflect, block or manage the face attack. For example, a person attacked by You are a moron may defend by saying No, I’ve got a PhD (Culpeper & Hardaker, 2017). Figure 1 shows Culpeper's model for responses to impoliteness:
Impoliteness response options. Source: Culpeper and Hardaker (2017, p. 213) and Culpeper et al. (2003, p. 1563).
Culpeper et al.'s (2003) approach to impoliteness has witnessed some refinements over the years. Bousfield (2007), for instance, has added the concept of an offending event, which entails any action or event that triggers an initial impolite response. Of most importance to the present research, however, is the work by Dobs and Blitvich (2013) and Langlotz and Locher (2012) on the interactional order of polylogal disagreements (i.e., disagreements in interactions involving multiple participants; see Kerbrat-Orecchioni, 2004) and response options for witnesses of impoliteness, respectively. Importantly, Langlotz and Locher have added that responses to impolite acts are not merely targeted at the interactional partner but to a wide range of social agents. These include characters/persons in the news story, the author of the article, co-participants and other related matters generated by the news story. In the present study, we take a cue from this to analyse the types of response comments (RCs) based on their attentional focus and the targets of the comments.
On their part, by analysing data from small-group interactions, Dobs and Blitvich (2013) concluded that in polylogal encounters, responses to impolite acts do not always come from the attacked but also from witnesses of the event. These witnesses may choose to either respond or not, and if they choose to respond, they have the option of reacting, corroborating opposition or denying opposition by proposing a compromise or countering the impolite act. Like Culpeper et al. (2003), Dobs and Blitvich explained that counters may be done either offensively or defensively. In the present study, given that news articles reported on GhanaWeb are usually on socio-political occurrences, actors of the offending event are usually not on the news website, so they themselves do not respond to comments that are triggered by their actions. Thus, it is usually the witnesses (i.e., other users of the platform) that respond. In the present study, we are particularly interested in the strategies used by the witnesses in countering impolite language on the platform.
Methodology
Data for this article were obtained from the Corpus of GhanaWeb Comments on Ghana's 2016 Election (CGCGE16). The corpus building was carried out from September 1, 2016 to December 6, 2016, a period that witnessed intense campaigning and citizen participation in Ghana. The corpus consists of 1,882 insult-filled reader comments as well as reader responses to the insults. Though quite old, this meticulously built corpus is still relevant to offer fresh and insightful perspectives on a rarely researched issue of this nature (Cowton, 2019; Ketchen Jr et al., 2023).
The key figures who are the targets of the insult-filled comments analysed in this study are John Dramani Mahama (Mahama or President–the president at the time who was seeking a second term. He was the presidential candidate of the National Democratic Congress [NDC]); Nana Addo Danquah Akufo-Addo (Nana or Nana Addo–the main opposition leader and the presidential candidate of the New Patriotic Party [NPP]); Nana Konadu Agyeman Rawlings (Nana Konadu–a former first lady of Ghana and the presidential candidate of the New Democratic Party [NDP]); Ivor Greenstreet (Greenstreet–the presidential candidate of the Convention People's Party [CPP]) and Charlotte Osei (EC Chair–the chairperson of Ghana's Electoral Commission). Apparently, the presidential candidates were targeted because they belonged to a different political party and had different views from the insulters, and for the EC, she was targeted because she was suspected to be supporting the sitting president who appointed her to that position.
In this study, we employ discourse analysis (Fairclough, 2003). Discourse analysis is a strategy for language studies that focuses on how language is used in specific contexts to attain specific objectives (Gee & Handford, 2023). In the present study, discourse analysis is appropriate to examine how the insult-filled posts aimed at key figures involved in Ghana's 2016 general elections are deemed as unacceptable and condemned by co-participants on GhanaWeb.
In particular, the data were analysed at the micro level, considering the various responses to impolite expressions. The analysis was guided by the extant literature on the interactional order of online interactions as well as forms of impoliteness and available response options and strategies (e.g., Bou-Franch & Blitvich, 2014; Bousfield, 2007; Culpeper, 2010; Dobs & Blitvich, 2013; Langlotz & Locher, 2012). The first and second authors conducted the coding, compared the results and resolved disagreements through discussions. Though we gained insights from previous studies, given that the socio-political context was different, our approach to coding was inductive and flexible, allowing codes to emerge from the data without necessarily relying on a codebook. This analytical procedure turns out to be employed in most recent studies of this kind (e.g., Gao & Liu, 2023; Khazraie & Talebzadeh, 2020; Perelmutter, 2018).
In order to enhance understanding, the RCs that are direct replies are listed together with the initial impolite expressions or preceding comments (PCs) while the RCs that are directed at a group of people or the general speech behaviour on the platform are listed alone. In some cases, the headline of the news report (NH) that prompted the insult-filled comment is also provided. All the comments are presented with the pseudonyms of the commenter, the date and the time that the comment was published on GhanaWeb. All the comments are presented without any form of correction. However, where necessary, the correct forms of some words or phrases are put in square brackets and inserted into the comments; where understanding is not hampered, [sic] is placed right after the incorrect expression.
Findings
This section focuses on the findings of the study. We first identify the types of RCs as well as their targets. Then, we examine the discursive strategies used by the responders, followed by their motivation for the responses. Thus, the first two sub-sections rely on insights from Langlotz and Locher's (2012) interactional order and targets of polylogal responses while the third is based on the literature on impoliteness and response options (e.g., Bou-Franch & Blitvich, 2014; Bousfield, 2007; Culpeper, 2010; Dobs & Blitvich, 2013).
Types of RCs
The RCs in CGCGE16 are of two types: either part of an on-going discussion (i.e., on-topic response) or parallel with it (i.e., off-topic response). In this section, we explain these with illustrations from comments on an article titled “It's Impossible for gov’t to Influence Election–Charlotte Osei.” Consider excerpts (1)–(4) below:
Excerpt 1
Excerpt 2
Excerpt 3
Excerpt 4
Excerpts (1) and (2) above can be categorised as on-topic responses as both align explicitly with the verbal abuse against the Chairperson of the Electoral Commission. Sir Levels in excerpt (1) implies that Tish's behaviour is a sin and draws attention to the fact that the target of the verbal attack could be Tish's mother. In excerpt (2), Stoway retorts with similar abusive language which even extends to the mother of Tacos. On the other hand, excerpts (3) and (4) are off-topic responses as the commenters do not address the attack on the Chairperson of the Electoral Commission. They rather signal that abusive comments against people of higher status (in terms of rank or age) are bad and unacceptable.
Targets of RCs
This section concentrates on the targets (referents/addressees) of the RCs in CGCGE16. The RCs can be broadly categorised as you-directed or we-directed in terms of their attentional focus. The you-directed comments are marked by the pronouns, you and your while the we-directed ones are typically characterised by the pronouns, we, us and our.
You-directed RCs
In CGCGE16, the focus of the you-directed RCs is on the co-participant(s) and some other people who may not even be on the interactive platform. These comments connote the idea of “pointing a finger” at another (Pennebaker, 2011, p. 75). They suggest accusation, blaming, confrontation and criticism (van Swol et al., 2017).
The use of you shows that a speaker's utterance is directed at an individual or a group of people. It can help to individualise the person or the group and elicit their attention, as it addresses them directly (Teterina, 2012). The you-directed comments can be directed at a specific participant on the platform. In the instances where you is not explicitly used, the you-directed comments involve the calling out of specific names. Consider excerpts (5)–(7).
Excerpt 5
Excerpt 6
Excerpt 7
The excerpts above are all direct replies to specific comments. KK and Agumu in excerpts (6) and (7) respectively direct their comment to a specific individual by mentioning their name (i.e., Kofi and Otanii ba). While Kofi is an actual name, Otanii ba is not a recognised given name; it may be a nickname or a nonce pseudonym. Agumu in extract (7) adds or whatever you call yourself to highlight that the target's actual name is not presented on the platform.
The you-directed comments can also target a specific group of participants, as in excerpts (8) and (9), or to all the abusive commenters on the platform, as in excerpt (10):
Excerpt 8
Excerpt 9
Excerpt 10
The excerpts above are not in reference to a particular person, but to a specific group of people such as Ashantis 1 in excerpt (8) and members of the NPP in excerpt (9) or to all the abusive commenters on the platform in excerpt (10).
In some cases, the you-directed comments extend their focus beyond the commenters on the platform to three other categories of people. First is the family members of the commenter, as exemplified below:
Excerpt 11
Excerpt 12
Excerpt 13
Notice from excerpts (11) to (13) that the same lexicon of insults, that is, words like useless, bitch and kwasiafoɔ (i.e., fools) are employed by the responders against the preceding commenters’ father, mother and family, respectively.
Second, some responses extend their focus to members of political parties that their comments suggest they are affiliated to, as in excerpts (14) and (15) below:
Excerpt 14
Excerpt 15
In responding to Oku who calls Nana Konadu, ofui “idiot,” Vicent in excerpt (14) labels the top members of the NDC National Executive Committee including Kofi Adams (National Organiser), Koku Anyidoho (Deputy General Secretary) and Asiedu Nketia (General Secretary) as useless people. In the same excerpt, Ruby's reply to Oku implies that John Mahama, the presidential candidate of the NDC, is useless. In excerpt (15), the supporters of NPP are characterised as so disrespectful.
Finally, some responses in CGCGE16, as demonstrated in (16) and (17), are directed at the staff of GhanaWeb. The commenters suggest that the negative linguistic behaviour on GhanaWeb is massive because the site is not moderated:
Excerpt 16
Excerpt 17
We-directed RCs
The focus of the we-directed RCs is on the entire online community. These comments are more collaborative and signal the commenters’ affiliation with all other participants (Kane & Rink, 2015), as shown in excerpts (18)–(20):
Excerpt 18
Excerpt 19
Excerpt 20
By using we and us in excerpts (18)–(20), the commenters create an equal social relationship between themselves and the other participants. These commenters are not just invoking shared identity, but they are drawing attention to the fact that the use of abusive language in Ghanaian politics is an issue that has to be dealt with collectively. Apart from using the inclusive pronoun, we, the use of please in these excerpts implies that the commenters would want their contributions to be appreciated as an appeal rather than an imposition on the other participants.
Countering impoliteness as a response strategy
This section examines the strategies used by some commenters to counter the use of abusive language against leading political figures on GhanaWeb.
Offensive responses
Some of the offensive responses involved the use of impolite formulae to question the proper upbringing of co-participants. This is evident in excerpts (21)–(23) below:
Excerpt 21
Excerpt 22
Excerpts 23
Addo, in excerpt (21), describes the abusive commenters on the platform as people who were reared. In excerpt (22), Kwame's description of Vida well explains the concept of being reared among Ghanaians (i.e., one who has not gone through any cultured, disciplined and ethical training). In many Ghanaian languages, there is a distinction between being reared and being properly nurtured or trained, as shown in Table 2.
The distinction between being reared and being cultured in Ghanaian languages.
A person who was reared is one who violates socio-cultural norms at every given opportunity. Such a person is considered as one who was only fed to grow, like an animal, in their formative ages and, thus, did not undergo the needed enculturation and socialisation. Yevugah's rhetorical question, in excerpt (23), so those who insult [the] president do they also have that primary socialisation?, which exemplifies what Culpeper calls unpalatable questions (Culpeper, 2011; Culpeper & Hardaker, 2017), attests to this. A reared person is often disapproved of and seen as not having the relevant knowledge, norms and values to participate in social activities. By contrast, a person who is considered as well-nurtured or properly trained is highly regarded in society (Opuni-Frimpong, 2012).
There were some cases where the offensive responses contained insults that signal disapproval and attempt to compel the target to stop subverting social norms and values. Excerpts (24)–(26) show offensive responses employed to draw attention to the targets’ specific wrongdoing:
Excerpt 24
Excerpt 25
Excerpt 26
It can be seen in the excerpts above that the responders do not just abuse their targets without supplying any justification. Here, the purpose of the abusive language is to signal disapproval and attempt to compel the target to stop subverting social norms and values. In excerpts (25) and (26), for instance, the offense is specified as insulting the president and Nana Addo, respectively.
In some cases, the offensive responses involve calling a misfortune on the target, that is, what Culpeper (Culpeper, 2011; Culpeper & Hardaker, 2017) refers to as curses and ill-wishes, as shown in excerpts (27)–(29) below:
Excerpt 27
Excerpt 28
Excerpt 29
Excerpt (27) involves the invocation of a supernatural agent (i.e., God). This often happens when the speaker acknowledges that they are not in the position to punish the offender but believes that there is an agent (specified or unspecified) who is able to do it for them (Agyekum, 1999). Where no supernatural agent is literally invoked, as in excerpts (28) and (30), the cultural assumption is that expressions like die and bad luck are inherently potent. Thus, in excerpt (29), the commenter redirects any ill-wish on the electoral commissioner (this woman) back to those declaring it, as well as their families.
Defensive responses
The Ghanaian culture encourages people to show respect for seniority in age (Forson et al., 2017). It is required that both advanced age and relative age are respected. In the data analysed, there were instances where the responders were defensive by expressing the need to show respect for the elderly by not insulting them. This is illustrated in excerpts (30)–(32) below:
Excerpt 30
Excerpt 31
Excerpt 32
The excerpts above draw attention to the fact that the manner in which opinions and views are shared on the platform is not in line with the Ghanaian culture of civility and respect for the elderly.
Some defensive RCs involve the religious formulaic expression may God forgive you. Excerpts (33)–(35) present what looks on the surface like a wish that God forgives the abusive commenters involved:
Excerpt 33
Excerpt 34
Excerpts 35
Many religious people believe that God can forgive any transgression or wrongdoing, so the expression may God forgive you is an invocation which also points to the fact that the referent has misbehaved or done something very offensive (Samarah, 2015). As used in the excerpts above, it is a device which implies that the target(s) did something extremely bad. It is a way of condemning a person's behaviour without using any negative language.
Motivation for responses on GhanaWeb
It can be observed from the preceding sections that the responders on GhanaWeb usually evaluate the linguistic behaviour of other participants negatively and/or appeal to these participants to express their opinions in a more acceptable manner. Some go to the extent of rebuking the participants and/or their associates with harsher or equally unacceptable linguistic forms. On the platform, it is uncommon to find a response to a RC that seeks to justify an initial abusive comment or defend its author. This may be because the authors of the negative comments would rather not engage in a rebuttal. It may also be that due to the asynchronous nature of the interaction on GhanaWeb, as some of them do not come back to view the reactions of other participants to their comments.
In general, other commenters who respond to the RCs make similar responses, thereby creating a form of a sequence. For instance, in the excerpt (36) below, Amenyo describes Nana Akufo-Addo as a short man devil and irresponsible. The first RC comes from Chris who describes Amenyo as a disgrace to his family. Togbega, who responds to Chris, does not engage in favour of Amenyo but rather, he also condemns the comment of Amenyo. Chris and Togbega's comments form a sequence of response acts beneath Amenyo's comment. It is clear that although Chris and Togbega are both participating on the platform as regulatory figures and their comments are meant to serve the same purpose, Togbega's choice of language is civil and relatively more appropriate than that of Chris, who employs counter-abusive language.
Excerpt 36
It has already been established that GhanaWeb is characterised by effective pseudonymity. Moreover, interaction on the platform is asynchronous. To the advantage of the responders, the asynchronous nature of the interaction allows them to reflect on and make possible interpretations of the contributions of other participants before posting their own comments (Kleinke & Bös, 2015; Tanskanen, 2007). On a negative note, their responses may not have any influence on the subsequent post of their targets, because these targets may never notice the reproof or criticism directed at them. The question, therefore, is: what motivates the commenters to respond to the posters of insult-filled comments?
Research suggests that sometimes, people expect others to follow the rules of acceptable communication of face-to-face interactions even when they are communicating online (see Chen & Abedin, 2014; Darics, 2010; Graham, 2007). As part of the socio-cultural mechanisms of Ghana, older people are obliged to criticise any form of behaviour that is considered as inappropriate around them. If they do not do that, they have failed in their duty to uphold the principles of the society. This is reflected in the Akan maxim, panyin a ɔtena fie ma mmɔfra we nanka no, yebu nankawefoɔ a ɔka ho “if an elderly person stays in the house unconcerned for the children in the household to eat a python, he will also be viewed as a python eater.” Bearing this maxim in mind, it is practical to say that the responders react to the abusive commenters as “elders” (i.e., traditional gatekeepers) who have to discharge their social responsibility. Their reactions create the awareness that although one can exercise one's right to freedom of speech on the online platform, it is expected of them to do so within the limits of the Ghanaian's socio-cultural norms and values.
Discussion
In this section, we discuss our findings in light of the existing literature. The study found that the responses can be on-topic or off-topic, which echoes the finding of Langlotz and Locher (2012) that responses can be related to the topic of the article or to the real-world affairs triggered by the news article. The study also found that the responses can be targeted at specific individual(s) (i.e., you-directed) or to all users of the platform, including the responder (i.e., we-directed). In their study on disagreements on online platforms, Langlotz and Locher also noted that users have a range of targets for their comments, including key participants in the news, the author of the article, issues triggered by the article's content or the contributions of other users of the platform. This is in line with the findings of the present study as the responses were targeted at various categories of people including the writer of the article (i.e., staff of the platform) and other users of the platform.
The you-directed responses, in particular, involve what is known among Ghanaians as the name-and-shame approach (Ofori, 2015). With this approach, the names of political activists who engage in insulting others during public discourse are mentioned or published in the media for everyone to identify them. This is to serve as a form of public punishment for transgressors and a system of reinforcing moral boundaries. The belief is that people would want to avoid public humiliation. This finding agrees with that of Haugh and Sinkeviciute (2018), who found that responses are framed as accusations that function as public condemnation of the abuser. On the other hand, the use of we in we-directed responses can depict a kind of collectivity that may be national, religious, socio-cultural or ideological (Adam-Troian et al., 2021). It suggests that the speaker and the referents/addressees share a communal experience. In the present study, the kind of collectivity that is being projected can be regarded as national or socio-cultural. By the use of we, the responders claim that they belong to the same community (i.e., Ghana) and share the same beliefs, norms and values as their co-participants (van Swol & Carlson, 2017).
As for the offensive strategies used, insults were remarkable. Even though the use of insults during a speech event is considered as unacceptable behaviour, it is one of the conventional means of protesting against a person (a target) who engages in untoward acts that challenge the values of Ghanaian society (Sekyi-Baidoo, 2009; Thompson, 2020; Yankah, 2002). Yankah (1998, p. 23) explains that the act of pointing out people's flaws by means of insults is “expected to lead to the reformation of wayward behaviour for the general good.” In agreement with the work by Thompson (2019, 2020, 2021a, 2021b), we found that some insults compare the targets to animals (e.g., you were reared). In the Ghanaian context, animals are considered to differ from humans in many respects, including human's ability to reason and distinguish right from wrong, which, among the Akans, is attributed to having tiboa (i.e., conscience) which the individual acquires through the socialisation process. Thus, referring to the abuser as an animal is a way of labeling the individual as lacking the ability to reason because he or she lacks proper socialisation and, therefore, behaves in a way that does not conform with human expectations (Agyekum, 2010; Gyekye, 1995; Ofori, 2017).
In Ghana, people are usually well-enculturated with communicative norms and values in the early stages of their lives and are motivated to conform consistently (Awedoba, 1996; Salifu, 2010; Yankah, 1991). During naming ceremonies, babies are first initiated into “the essence of truth and discreet speech, the need for care, truth, firmness and social sensitivity in the exercise of the spoken word” (Yankah, 1991, p. 48). The parents, grandparents, uncles, aunts and other adults in the community then share the responsibility of their upbringing. They seize different opportunities daily to transmit socio-cultural norms and values to the children (Salifu, 2014). In line with the collectivism of Ghanaian society, an individual's poor public speech reflects badly on their parents, family or community (Yankah, 1998). This explains why the offensive responses are sometimes extended to the families of the targets. This finding, therefore, agrees with Agyekum's (2004a) view that apart from not maintaining the dignity and good reputation of their relatives, those who often speak indecorously cause their relatives embarrassment.
Another important response strategy is the curse or ill-wish. According to Agyekum (1999), utterances that specify a misfortune or an undesirable circumstance on an individual or a group of people can be used to express one's reaction or feelings about the (in)actions of others. Often, saying that something bad should happen to a person is a way of pronouncing judgment on the person for having done something offensive or objectionable (Tweneboah, 2017). The social effect is to restore moral order (Teilanyo, 2015). Many African societies, including Ghanaians, believe in the “power of words.” Thus, whether as an invocation of misfortune or an expression of anger, the import of these negative utterances is understood by interlocutors on the basis of their shared background knowledge (Teilanyo, 2015).
We also found that the defensive strategies are grounded in the socio-cultural norms that strongly promote respect for the elderly. Regarded as “the symbol of wisdom” and “society's memory databank,” the elderly are relied upon for important decisions that can serve the best interest of the people around them (Agyekum, 2004a, p. 137). For this reason, younger persons are obliged to concede precedence and accord them some form of deference when addressing them. It is against the communicative values to insult the elderly during a speech event. In other words, a younger speaker must avoid being offensive and must be able to properly manipulate their linguistic actions and expressions in accordance with acceptable norms (Agyekum, 2004b; Salifu, 2014; Yankah, 1995). Interestingly, some defensive strategies involve religious expressions. These expressions, which resemble what Harb (2020) refers to as supplications, reflect the religious beliefs of Ghanaians and also point to the situatedness of impoliteness and its responses in specific socio-cultural settings.
Conclusion
This article has explored how some participants on GhanaWeb attempt to ensure civil and smooth communication flow among interactants in political discussions through the use of RCs. It examined the evaluative reactions to the use of abusive language against some presidential candidates and the electoral commissioner involved in Ghana's 2016 general election by commenters who perform the role of traditional gatekeepers of appropriate communicative behaviour. Generally, their reactions are prompted by all the abusive comments in a discussion thread or by a particular one.
The RCs on GhanaWeb are, on the one hand, on-topic or off-topic and, on the other, you-directed or we-directed. The you-directed comments are with direct reference to a specific commenter (i.e., you-singular), a group of people or all of the abusive commenters (i.e., you-plural) on the platform. Sometimes, the content of the comments extends to non-participants (i.e., people who may not be on the platform), such as the relatives of commenters, affiliates of the political party to which an abusive commenter may belong or the staff of GhanaWeb. The we-directed comments implicate not only the users of abusive language on the platform but also the readers and the responders themselves. Even though these online participants may not know one another, with the we-directed comments, some responders attempt to establish some form of solidarity with others.
These commenters who play regulatory roles on GhanaWeb, often, are not concerned with or do not contribute to the main issue under discussion. Their focus is on the communicative practice on the platform as they attempt to reinforce Ghanaian cultural values. They construct their message in ways that imply their co-participants’ choice of language is inappropriate and thus, they should desist from such choices. Some of these online regulatory figures call for comments that show respect for the elderly or show that their co-participants have had a proper upbringing in terms of the socio-cultural norms of Ghana. The others employ negative language to show their disapproval. All in all, the paper has established that among Ghanaians, the act of using abusive language, especially against the elderly, is not applauded in any way. It has further demonstrated the idea that although the interaction on GhanaWeb is pseudonymous and asynchronous, these regulatory figures react to abusive commenters on the platform so as to fulfill a social obligation.
This study is the first to extend the impoliteness theory to RCs in an online context in Ghana. The study demonstrates that the speech practice on online platforms patronised by people in Ghana is unique in its adherence to Ghanaian communication norms. This study is, therefore, inclined towards the notion that “understanding local discourse and ideologies of media technology is crucial since speakers incorporate new technologies of communication from existing communicative repertoires, which influence new and emerging cultural practices” (Wilson & Peterson, 2002, p. 461). In addition, the study contributes to the theory by categorising the response types by their focus: on-topic and off-topic. This study also contributes to work that draws attention to news sites in Ghana as another platform where the use of unmitigated invective and insults against authority figures and people of higher social status is increasingly becoming the norm. The response expressions in the corpus are analysed to elucidate the ways in which they reflect cultural assumptions in Ghana.
Despite the considerable contributions of this work, it is crucial to highlight that the findings need to be interpreted with caution. In particular, the present study relied on the qualitative method to examine data generated from one news website (i.e., GhanaWeb) and from a limited period of time (i.e., about 4 months). It is, therefore, necessary to not over-generalise the findings to other online news platforms. In light of this limitation, we argue that the study can serve as a springboard for future research. In particular, future studies can draw attention to the dynamics of political discourse on online interactive platforms in other countries with similar cultures (e.g., Nigeria) or countries with different cultures (e.g., Australia) by exploring the reader comments on their news sites, as well as social media platforms.
Footnotes
Acknowledgments
We would like to thank the anonymous reviewers and the Editor for their constructive comments and suggestions.
Declaration of conflicting interests
The author(s) declared no potential conflicts of interest with respect to the research, authorship and/or publication of this article.
Ethical statement
The study was approved by the Griffith University Human Research Ethics Committee (30-08-2016). Informed consent was obtained from the administrators of GhanaWeb prior to the commencement of the study. Nearly all commenters on GhanaWeb use pseudonyms and very few, if any, use personal names (i.e., full names) that could be linked to certain individuals. Such personal names were not disclosed in the paper.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship and/or publication of this article: This work was supported by the Griffith Graduate Research School with support from the School of Humanities, Languages and Social Science, Griffith University.
