BUY MY EMOJI: AN ANALYSIS OF ADS IN TWITTER/X

Compre o meu emoji: uma análise de anúncios no Twitter/X

Compra mis emojis: un análisis de anuncios en Twitter/X

Abstract

Emojis, a visual representation of feelings, ideas, entities, statuses, or events, first appeared in 1995 and became available in 2011 for mobile devices. They are frequently found in messages on social networks and also in posts with marketing purposes. The present study aims to investigate the linguistic uses of emojis in brand posts in Brazilian X (former Twitter). We adopted the Cyberpragmatics approach (Yus, 2011) to data analysis in addition to previous studies on emoji use. Brands and/or industries from 3 different segments that have a Twitter profile in Brazilian Portuguese were included in the analysis. Data comprised 150 posts from 15 Brazilian brands. Emojis are primarily used to convey attitude and emotions, acting as pragmatic markers and occasionally as word replacements or reinforcement.

Keywords:
Online communication; Emoji; Cyberpragmatics; Language function

Resumo

Emojis são uma representação visual de sentimentos, ideias, entidades, status ou eventos, que apareceram pela primeira vez em 1995 e foram disponibilizados para dispositivos móveis em 2011. São frequentemente encontrados em mensagens nas redes sociais e em postagens com fins de marketing. O objetivo do presente estudo é investigar os usos linguísticos dos emojis em postagens de empresas no X (antigo Twitter) brasileiro. Adota-se a abordagem da Ciberpragmática (Yus, 2011) para análise de dados, além de estudos anteriores sobre o uso de emojis. Foram incluídas na análise marcas e/ou indústrias de três segmentos diferentes que possuem perfil no Twitter em português brasileiro. Os dados incluem 150 postagens de 15 marcas brasileiras. Os resultados mostram que emojis são usados ​​principalmente para transmitir atitudes e emoções, atuando como marcadores pragmáticos e, ocasionalmente, como substitutos ou reforço de palavras.

Palavras-chave:
Comunicação online; Emoji; Ciberpragmática; Função da linguagem

Resumen

Los emojis son una representación visual de sentimientos, ideas, entidades, estados o eventos, que aparecieron por primera vez en 1995 y estuvieron disponibles para dispositivos móviles en 2011. A menudo se encuentran en mensajes y publicaciones de redes sociales con fines de marketing. El objetivo del presente estudio es investigar los usos lingüísticos de los emojis en las publicaciones de las empresas en el sitio brasileño X (antes Twitter). Para el análisis de datos se adoptó el enfoque Cyberpragmatics (Yus, 2011), además de estudios previos sobre el uso de emojis. Se incluyeron en el análisis marcas y/o industrias de tres segmentos diferentes que tienen perfil de Twitter en portugués brasileño. Los datos incluyen 150 publicaciones de 15 marcas brasileñas. Los resultados muestran que los emojis se utilizan principalmente para transmitir actitudes y emociones, actuando como marcadores pragmáticos y, ocasionalmente, como sustitutos o refuerzo de palabras.

Palabras clave:
Comunicación en línea; Emojis; Ciberpragmática; Función del lenguaje

1 INTRODUCTION

Evans (2017, p. 19) defines emoji, which is an anglicized version of two Japanese words - e for picture and -moji for character - as “a visual representation of a feeling, idea, entity, status or event.” They first appeared in 1995, becoming available in 2011 for digital keyboards in our mobile devices, and have been used for a variety of purposes since then. They have become ubiquitous in social networks. Around 5 billion emojis are used on Facebook and Facebook Messenger daily around the world (Statista, 2021), and most internet users (92%) send emojis regularly (Evans, 2018). The ‘laughing face with tears of joy’ emoji was chosen as the word of the year in 2015 by the Oxford Dictionary. It is the most popular emoji on Facebook and Twitter, while the heart is the most frequently used emoji on Instagram (Statista, 2021).

Emojis are derived from emoticons, which are “graphic signs, such as the smiley face, that often accompany computer-mediated textual communication” (Dresner; Herring, 2010, p. 1). Emoticons were first used in 1982 when the smiley face - :), made from rotated printed text came along with the frowny face :-( ; the latter meant to indicate that a message sent by a professor should be taken as a joke (Dresner; Herring, 2010; Tang; Hew, 2018). Emoticons are literally an ‘emotional icon’ representing a facial expression with a corresponding emotion. Dresner and Herring (2010) argued that emoticons were more than expression of emotions; they could also function as illocutionary force markers. Over a decade later, emojis appeared and became widely used by Internet users. Figure 1 shows the timeline of emojis:

Figure 1
History of emojis

After the first face-based emojis were introduced, several additions were made by Unicode, and now one can find animals, food, flowers and plants, and a wide array of symbols and icons that can be used to enhance communicative exchanges. For Alshenqeeti (2016, p. 56), emojis “provide greater nuance and clarity to text messages and are an expression of creativity in language [...].” It is unquestionable that emojis are used in CMC to aid comprehension, by conveying emotions and attitudes (Escouflaire, 2021).

While emojis may not be regarded as a new ‘language’ - or even worse, a setback in literacy - they help interactants cope with the limited affordances of computer-mediated communication (CMC) when we compare to face-to-face communication (Evans, 2018; Crystal, 2011). The lack of eye contact, body language cues, and prosody may hinder online communication; nevertheless, users may resort to strategies to cope with the specificities of the online medium to express emotions and make their meanings across, to mention a few possibilities. One such strategy is by incorporating emojis in the flow of interaction to provide non-verbal cues when communicating online. In this respect, emojis may be considered additions to the language (Evans, 2017, 2018).

There have been investigations conducted so as to describe the communicative properties of emojis and their linguistic functions (Tang; Hew, 2019), which include their pragmatic function (Dresner; Herring, 2010; Oliveira; Cunha; Avelar, 2018; Li; Yang, 2018; Thompson; Filik, 2016; Weissman; Tanner, 2018), marketing purposes (Ge; Gretzel, 2018), socio-semiotic relations (Alshenqeeti, 2016), and syntactic function (Cohn; Engelen; Schilperoord, 2019).

A recent systematic review of 51 studies on the contribution of emoticons, emojis, and stickers in online communication revealed that they aid interactants in expressing emotions and managing relations as well as contribute to message comprehension when used as words (Tang; Hew, 2019). It is precisely these aspects that the present research seeks to explore; in other words, how emojis are used pragmatically and how they contribute to message construction in tweets posted by Brazilian corporate brands.

The present study aims to investigate the use of emojis in brand posts with a social media presence; that is, to describe and discuss the use of emojis for marketing purposes. So, we will adopt Yus’ (2014) taxonomy of pragmatic functions (later adapted by Li and Yang, 2018) that emojis may exhibit to first identify and describe their usage in tweets posted as a marketing tool for corporate brands in Brazilian X¹ (former Twitter). In addition, we incorporated Escoulfair’s (2021) analysis to include the referential function. We will also resort to recent literature on social media marketing, trying to describe its discourse features and the roles emojis are likely to play.

2 CYBERPRAGMATICS

Pragmatics is concerned with language in context; that is, it analyses language in terms of the context it was produced. It is known that humans are rarely literal when they use language to communicate - part of their intended meaning is implicit when they perform a speech act. Speakers consider their interlocutors’ ability to interpret what they say adequately, when they are composing their utterances (Yus, 2011). In this context, people can do things with words, or, to put it in Austinian terms, when they perform speech acts. Among the things humans can do with sentences, we can cite apologizing, giving orders, making requests, and many other functions (Austin, 1962).

The advent of the Internet in the 1990s and its rapid development has had an impact on the way people communicate. The internet gave rise to types of communication that may be considered ‘new’ when compared to prevailing ones (Crystal, 2011; Barton; Lee, 2013). This impact on language and communication is also “an inherent object of analysis for pragmatics,” according to Yus (2011, p. 13). Pragmatic models of communication do not easily explain linguistic behavior found online (Crystal, 2011). Such need gave rise to Cyberpragmatics, a term coined in 2001 by Yus to address the cognitive aspect of pragmatics studying Internet-mediated communication. Cyberpragmatics aims to provide an account of how information is produced and interpreted in the online environment as well as how context aids in filling the gap between what is written and the underlying communicative intention.

Yus (2011) lists four hypotheses which makes up the foundation of Cyberpragmatics. The first one is related to the message and how it is composed to lead to the intended interpretation; the second one regards the inferential strategies adopted by users when they interpret a message; the third refers to the contextual information users expect to be available to interlocutors so as the former will be able to interpret the utterance correctly; and finally, the fourth related to the varied possibilities available to users, be it webpages, social media, email and others, and how these possibilities affect user’s access to contextual information, amount of information and the cognitive aspects of such use.

One of the central objectives of Cyberpragmatics is related to the fourth hypothesis: what the role of media is and how media features affect the effort made to interpret the message in addition to how media alters the perception of relevance. A second objective of Cyberpragmatics is concerned with the strategies used by interlocutors to compensate for the lack of contextual cues that are missing in virtual communication compared to face-to-face communication. Since the text typed from the keyboard may lack the user’s feelings and emotions, an extra burden on cognitive and language processing may be present (Yus, 2011). As previously mentioned, some strategies are employed by Internet users to compensate for the lack of immediate context (Crystal, 2011; Ge; Herring, 2018), including the use of emojis, which is the focus of this paper.

Previous research has shown that emojis can be used to show how the speech act should be understood; in other words, including an emoji in a message can influence its interpretation. In this regard, emojis may be considered pragmatic markers (Escoulfaire, 2021; Evans, 2018). An example can be seen in Figure 2 (personal data). The ‘grinning face with sweat’ emoji tells the reader that the tweeter is just joking and should not be taken seriously.

Figure 2
‘grinning face with sweat’ emoji

Another example is seen in (1) from Escouflaire (2021, p. 208). The laughing with tears of joy emoji at the end of the written message tells the reader that the writer is being sarcastic.

(1) I hate you so much

Escouflaire (2021) categorized this emojis’ function as ‘interpretative,’ since, as previously mentioned, CMC lacks non-verbal cues present in face-to-face interactions. Emojis would not only be used to express emotions - as their predecessors, the emoticons, do - but also to change the tone of the message, telling the reader how it should be interpreted. In such uses, emojis are placed at the end of the written message, as shown in Escouflaire (2021) and Ge and Herring (2018).

3 EMOJIS AND ONLINE COMMUNICATION

Several studies have explored the motivations underlying emoji use in online communication. These studies have looked at the pragmatic/semantic as well as the syntactic domain. Dresner and Herring (2010) discussed the contribution of emoticons as illocutionary force indicators. Oliveira, Cunha and Avelar (2018) also argued that emojis may be used as illocutionary force indicators in apology requests following refusals. In other words, emojis can be used as repair strategies in face-work. Irony and sarcasm can also be achieved when emojis are present in the interaction. Weissman and Tanner (2018) investigated brain responses to the processing of emojis conveying irony, which yielded similar responses to verbal-conveying irony while Thompson and Filik (2016) conducted an experiment in which they revealed that emoticons may be used to clarify message intention; that is, identify sarcasm. Both the tongue face and the wink face were sarcasm indicators.

The systematic review conducted by Tang and Hew (2019) showed that emoticons, emojis, and stickers supplement the lack of cues in online communication by expressing emotions and affecting human relations as well as serving as ‘words,’ that is, vocabulary items to aid comprehension. In addition, the expression of positive emotions through emojis can improve the favorable impression the recipient of the message has on the sender. The perception of intimacy and friendship formation is also noted when emojis are present. Emojis are also used to avoid misunderstandings, that is, they establish the illocutionary force of the utterance, mitigate impoliteness as in flaming, and may act as punctuation. What is interesting to notice is that when the studies on users’ motives were reviewed, interactants reported that emojis would avoid misunderstandings regarding feelings or meanings, help them socialize by creating more interaction in conversation, create enjoyment in the communication and offer cues to compensate for the lack of nonverbal elements.

Li and Yang (2018) investigated the use of emojis among Chinese users of a chat app. The authors concluded that emojis were frequently used with a variety of pragmatic functions, including showing and/or enhancing emotion/stance; as an interaction device for turn-taking and as a backchannel device; and modifying illocutionary force as the most common uses. Over 50% of emojis were used as emotion signifiers. The authors also concluded that the positive emojis were more frequent than the negative ones in socio-emotional settings, such as the chat app they investigated, which comprised interactions between family members and friends. Despite being frequent, emojis do not mirror real-world interactions. They compensate for the lack of contact to some extent.

Escouflaire (2021) analyzed a corpus of 1200 Facebook and Twitter posts containing emojis written in English and French. He aimed to propose a typology of their linguistic uses. Seven linguistic functions were described: expressive, interpretative, relational, politeness, emphatic, structural, and referential. Among these functions, only the referential would be used as word replacement and/or reinforcement. The first five, which were more prevalent, would add or modify the content of the message and were mainly represented by smiley and people emojis, while the structural and referential were mostly objects. The 10 most used emojis were smileys and were used in the first five functions. The most frequent one was ‘laughing with tears of joy’, which also happens to be the most frequently used emoji according to the emojitracker website and Statista (2023).

Emojis have also been seen under the lenses of their syntactic constraints in language use. Two patterns of use have been described: emoji used in combination with text, which prevails, and emoji used in sequence, in which there is either repetition/ reduplication of the same emoji or clusters of semantically related emojis. Cohn, Engelen, and Schilperoord (2019) investigated the grammatical system that may underlie emoji usage. Two experiments were conducted to describe their grammar and how they interact with the grammar of a sentence; that is, the authors wanted to determine whether emojis have a syntax of their own or whether they follow the grammar of the participants’ language or any other order. The authors found that participants used simple constructions, the most common being those that express responsive emotions at the end of a sequence to convey emotions and attitudes. Another finding was the presence of reduplications, aiming at emphasizing or increasing the magnitude of what was previously mentioned.

In essence, research conducted so far seems to point to the expression of emotions and the pragmatic/ relational function as the most used when we look at the linguistic function of emojis in online communication. Smileys are usually found performing such functions. The referential function, that is, replacing words or reinforcing information already given is less common, and usually objects and symbols are used. We now turn to look at how emojis have been used as a tool to convince consumers.

4 EMOJI AS MARKETING TOOL

Emojis can also be a tool to persuade consumers. Big companies, such as Sony, Chevron, Coke, Burger King, and Taco Bell use emojis to communicate with customers on social media. Since 2015, the number of tweets with emojis in brand names has increased by 49% (McShane et al, 2021). Social media platforms have been used by brands to build and strengthen the relationship between themselves and consumers.

Ge and Gretzel (2018) conducted a study in which they identified and classified emojis in terms of their persuasive role, that is, emojis as rhetorical devices. They investigated digital influencers’ posts that were aimed at fostering engagement and/or initiating conversations with their followers on the Chinese social network Sina Weibo. Their goal was to establish how emojis, embedded in the written texts and images from a post, were employed to establish influencers’ credibility (ethos), to touch readers’ emotions (pathos), and to present logical reasons (logos). Emojis were used to support the expression of attitudes and emotions 31.7% of the time, followed by delivering factual information in around 18% and making questions and requests in 16.2%.

Ass for the more descriptive and structural nature of emojis, data from Ge and Gretzel (2018) also revealed two patterns of usage, namely, emojis used alone or in combination with text. For the combination pattern, they may repeat words to emphasize the speech act and are usually placed at the end of the post. In addition, they may be used to replace words and those are placed in the middle of the written part. When emotions are conveyed, emojis are placed frequently at the end of the post as sentiment enhancers. If emojis are used in isolation, a mixture of them is found. In this case, they are arranged so that they follow the standard order of a sentence. Their data also showed that emojis “add, help strengthen, or modify persuasive appeals aimed at triggering engagement” (Ge; Gretzel, 2018, p. 16). They also suggest that “firms can use emojis to deliver product- or brand-related information via introductions and/or compliments, and to maintain their social media presence by creating affiliation with consumers through phatic posts” (Ge; Gretzel, 2018, p. 16). This is to say that emojis are potentially useful marketing tools aimed at creating engagement with future consumers.

McShane et al. (2021) studied the influence of emojis in brand communication on consumer engagement (likes and shares) with brands on Twitter. They found that consumer engagement is higher when an emoji is present in a tweet, and, interestingly, the increase in the number of emojis results in more engagement in existing tweets of both celebrity brands and corporate brands. In a controlled experiment, they also found that emojis that are related to the textual element of the tweet (that is, they somehow substitute or reinforce the written content) enhance engagement. In addition, emoji position also seem to play a role. When they precede related text, engagement is enhanced.

The language brands use may also impact consumer engagement, and emotion is to be considered. Informality and respect to social media norms, including a style that resembles oral conversation is expected by consumers when brands communicate with them. As a result, according to Deng et al. (2020), the adoption of a casual style in brand posts would meet consumers expectations and ultimately enhance consumer engagement. Emojis, contractions, and personal pronouns are considered informal features of discourse. As for emotion, the authors predict, based on previous marketing research, that posts containing emotional appeal, especially those conveying positive emotions, may enhance consumer engagement. Their results revealed that consumers on Facebook tend to ‘like’ more posts that convey positive emotions. Since emojis may convey or reinforce emotions or attitudes, we may expect emojis to contribute to consumer engagement.

5 METHODOLOGY

To investigate how emojis were used by brand names in Brazilian Twitter, we created and analyzed a small corpus of tweets. Brands and/or industries that have a Twitter profile in Brazilian Portuguese were included in the analysis. To be included, the profile should have at least 20,000 followers and be active on Twitter; that is, they should have tweeted at least once a week. They also should be able to sell and/or deliver their products anywhere in Brazil. The profiles corresponded to those involved in producing/selling food and cosmetics as well as department stores. Five profiles from each category were selected. We collected the last 10 tweets containing emojis that were published within the last 6 months from each profile, generating a corpus of 150 tweets. Replies, retweets, and survey results were not included in the analysis. Data was collected in the first semester of 2021.

Tweets were collected as screenshots to preserve their original display of information, which includes emojis and images, links, hashtags, and other elements that may be part of the tweet. In addition, screenshots avoid data loss in the event of it being deleted from the user’s account. All tweets were publicly available.

To analyze emojis’ pragmatic contribution to tweets, we adopted and adapted Li & Yang’s taxonomy in their analysis of emojis on WeChat to analyze our data set. The seven categories were as follows: Attitude/emotion signal, Attitude/emotion intensity enhancer, Illocutionary force modifier, Humor, Irony, Turn-taking/giving, and Backchannel device.

Since we are not analyzing the interactions that might have developed as a response to the initial tweets, we did not include the last two categories - turn-taking/giving and backchannel device in our categorization. Thus, the following categories were used in our analysis: i) Attitude/emotional signal, ii) Attitude/emotion intensity enhancer, iii) Illocutionary force modifier, iv) Humor, and v) Irony. We also adopted Escouflaire’s (2021) typology of emojis, considering his referential function, making this the 6th category. In essence, we analyzed data trying to identify when emojis were used as pragmatic markers and when they were used as words, either to replace or reinforce what was previously stated.

We included the following profiles, grouped according to the segment they belong:

• i) Food and drink: Burger King Brasil @BurgerKingBR; Ambev @Ambev; McDonald’s Brasil @McDonalds_BR; Garoto @garoto; Cacau show @cacaushow;

• ii) Department stores: Magazine Luiza @magazineluiza; Shoptime @canalshoptime; Casas Bahia @casasBahia; Submarino @submarino; Americanas @americanascom; and

• iii) Cosmetics: O Boticário @oBoticario; Natura @naturabroficial; Vult @vultcosmetica; Avon @AvonBR; Salon Line @SalonLine Brasil.

After data collection, the 150 tweets containing emojis were analyzed to determine their main function. The analysis was conducted manually, and the frequency of occurrence of emoji usage per category was recorded. We also recorded the frequency of each category by brand segment, that is, cosmetics, food and drink, and department stores. In the following section, we present the results, including the frequencies and a qualitative analysis of data.

6 DATA ANALYSIS

Our analyses revealed that emojis were mostly used as attitude/emotional signals and attitude/emotional enhancers. An example of an emotion signal can be seen in the following tweet from a beauty brand:

Figure 3
Vult post

The red heart following ‘your favorite foundation with you every day’ indicates a positive emotion related to the product. Hearts and faces with hearts were used to reinforce positive emotions, as the second tweet shows:

Figure 4
Boti post

The smiling face with hearts after the word ‘admire’ and the green heart after the word ‘love’ in this beauty brand tweet enhances the positive emotion and attitude that the tweet requests from their followers. Asking followers to ‘tag someone you admire’ and requesting them ‘to start the week by spreading love’ are in themselves requests for positive attitudes and emotions, which are reinforced by the emojis placed at the end of the textual elements.

A similar use is found in the following tweet by a beauty brand:

Figure 5
Boti post #2

The green heart at the end of the text in Figure 5 indicates that the beauty brand is heartfelt to have the celebrity advertising their products. The crown emoji refers to the fact that they consider her a ‘queen,’ as stated in the textual element - ‘the best reality commentator is our glamour queen,’ followed by her profile name and ‘what an honor to have you with us.’ The heart expresses the brand’s feelings toward Marilia and the crown reinforces the idea that she is considered a queen.

The illocutionary force of an utterance was also established by emojis. In the following tweet (fig. 6), offering someone a chocolate candy should be understood as ‘I am offering you, but I hope you do not accept it’ when we look at the ‘pleading face’ emoji as if the tweeter was pleading or begging to have his offer refused. In this way, the ‘pleading face’ determined how the offer should be understood.

Figure 6
Garoto post

The same use was found in the next tweet in Figure 7. The ‘rolling on the floor laughing’ emoji should tell readers that the request for saying ‘the name of your favorite character from a Mexican soap opera’ should be understood as a joke. It is only when readers see the emoji at the end of the sentence that the illocutionary force of the utterance is determined. We should keep in mind that the emoji might have been used to amuse tweeters, that is, with a humorous intent, but it is, primarily, an illocutionary force modifier. Humor may be a side effect of tweet comprehension. Such use of emojis is in line with what Yus (2011) described as one of the objectives of Cyberpragmatics, that is, to investigate the strategies used to compensate for the lack of contextual cues in CMC. By doing so, users do not place an extra burden on their interlocutors language processing by making it clear how the message should be understood.

Figure 7
Casas Bahia post

A similar case can be seen in Figure 8. The expectation x reality contrast was presented not through images, as it usually is in social networks, but with a sequence of emojis in this post by a beverage company.

Figure 8
Ambev post

The expectation was to have joined a carnival ‘block,’ followed by smiley faces, hearts, and party face emojis while the reality was to have received a funding request, followed by sad and angry faces. Not only do emojis act as illocutionary force determinants but also depict feelings/attitudes.

In the next two tweets (Figures 9 and 10), examples of word substitution and reinforcement are shown. The cosmetics brand Avon was claiming peace, referring to ‘fire’ as a heated discussion. “What about leaving ‘the fire in the park’ there?” The fire emoji replaced the word fire (fig. 9), while the wave emoji reinforced and is the referent used to the ocean, the name of the perfume being advertised by Natura (Figure 10).

Figure 9
Avon post

Figure 10
Natura post

The pictures represent the character as she is and how she feels when she is wearing the ‘Kaiak Ocean’ perfume.

Similarly, the next tweet contains much of its content relying on emojis substituting words. The tweet reads “What would keep the peace between Brazilians and Italians?” and the handshake emoji represents peace, followed by a sequence of 4 pasta emoji (Figure 11).

Figure 11
Americanas post

The repeated emojis seem to convey the idea of large quantity, as suggested by Cohn, Engelen, and Schilperoord (2019). We could also hypothesize that the sequence of similar emojis in Figure 8 expresses similar ideas.

The following graph revealed that the tweet corpus we collected contained emojis being used mostly as attitude/emotional signals (43%), followed by word substitution/repetition (37%). When we consider the emotional function of tweets (grouping the attitude and emotional signal and the attitude/emotion intensity enhancer categories), almost 60% of emojis were used to convey or reinforce emotions and/or attitudes, which may confirm previous studies such as the one conducted by McShane et al. (2021), and Deng et al. (2020). The former showed enhanced consumer engagement when positive emotions were conveyed by emojis. The latter argued for the effect that the number of emojis in a tweet as well as they are being used as a substitute or reinforcement have on consumer engagement. Therefore, it is not unexpected to find such figures in our data.

Graph 1
General emoji function distribution

The functions of irony and humor were not found in the data. We speculate that this may be because the irony and illocutionary force modifier overlap, given the fact that emojis depicting irony would change how the message should be read, or, in other words, would change the illocutionary force of the utterance, as in figure 6. We also believe that humor and attitude/emotion functions overlap, making it hard to distinguish between them. We suggest that humor may be included in some instances of attitude/emotion markers, as in Figure 7. In other words, we did not find, in our data, an emoji being used with the sole purpose of creating humor. The same was found in Escouflaire’s (2021) analysis, in which he categorized the primary and secondary functions of emojis, showing that such functions may overlap. We looked for and categorized the primary function of emojis.

When we look at the structure of the verbal elements, we found that similar to what was argued by Cohn, Engelen, and Schilperoord (2019) simple constructions were common, with emotional emojis being placed at the end of the sentence. The same can be said of emojis used to reinforce what was previously stated. In a few instances, the substitute/reinforcement emoji was placed before and after the verbal element, as in Figure 12 (“Are you looking for beautiful and comfortable chairs to your home?”), preceded and followed by a chair emoji.

Figure 12
Shoptime post

When looking at each segment, we could notice that the results reflected the general frequency of emoji functions.

Graph 2
Emoji function distribution across cosmetic brands

Graph 3
Emoji function distribution across food and drink brands

Graph 4
Emoji function distribution across department stores

In all 3 categories, both attitude/emotional signal and attitude/emotion intensity enhancer comprised over 50% of emoji functions. Illocutionary force modifier was present in less than 12% of our data in the food and drink segment and accounted for only 4% of data in the department store and cosmetics segments. Such results are similar to what was found in previous research, evidencing the emotional aspect that pervades emoji use in CMC.

7 DISCUSSION

The present paper aimed to investigate how Brazilian brands use emojis in their tweets, that is, what were the linguistic functions performed by emojis. Results revealed that emojis were mainly used to convey attitude/emotions, or, in other words, acted as pragmatic markers (Yus, 2011). In addition, emojis were, to a lesser extent, seen as word replacements or reinforcement, mirroring what was found in previous studies (Escouflaire, 2021; Dresner; Herring, 2010; Li; Yang, 2018).

As found in Ge and Gretzel (2018), in which 31% of the emojis were used to touch readers’ emotions, our analysis also showed that emojis were used to convey emotion in 43% of the occurrences found in the corpus. In addition, 13% of emojis were found in posts to convey attitude and as emotion intensity enhancers. Taken together, these two categories account for over 50% of data, revealing that emojis were mostly used to convey emotions, a finding also reported by Escouflaire (2021) and Li and Young (2018). We also found emojis either replacing and/or repeating words 37% of the time, which represents the least prevalent use of emojis, also shown in Escouflaire (2021). According to Cohn, Engelen, and Schilperoord (2019), emojis lack complex grammar; the most common form is placing an emoji at the end of a sentence to express emotions and reduplications, aiming at emphasizing the previous content of the message.

Even though we did not calculate the frequency of single emojis that appeared in the corpus, we may notice that most posts contain faces and/or hearts. These categories are usually used when the expression of emotions is the goal (Escoulfaire, 2021). Future studies could investigate which types of emojis, together with their function, are the most frequently used in the corpus, similar to what was described in Escoulfaire (2021), in which the ten most frequent emojis belong to either the face or heart type. Such a finding echoes what can be seen in the Unicode webpage (2021), being the ‘face with tears of joy’ the most common emoji.

Engagement with consumers is a goal that may be achieved by brands when emojis, especially those conveying emotions, are present in posts on social media. Both Ge and Gretzel (2018) and McShane et al. (2021) have shown that consumer engagement is higher when emojis are incorporated into the posts. Also, readers are more likely to ‘like’ posts conveying positive emotions, and thus enhance engagement through Facebook posts containing emojis, as Deng et al (2020) revealed. Such findings may account for the fact that most Brazilian brand posts in our corpus conveyed emotions and/or attitudes. Future research could investigate how the expression of emotion and positive attitude relate to consumer engagement through social media.

Deng et al. (2020) also showed that the informal and casual tone of the posts such as contractions and personal pronouns - in addition to emojis - help enhance consumer engagement. We could also see in our data the same features described by Deng et al. (2020), especially in the beauty brands posts, even though that was not our aim. A more informal style seems to be present in most posts in the corpus.

Our study has contributed to the comprehension of emojis in online communication by investigating how emojis added pragmatic meaning to Brazilian brand posts. It has shown that emojis express emotion and a positive attitude, in addition to replacing and reinforcing words/ideas as well as determining illocutionary force. Further research could investigate the patterns of interaction and emoji functions when potential consumers interact or react to such brand media posts.

REFERENCES

• ALSHENQEETI, H. Are Emojis Creating a New or Old Visual Language for New Generations? A Socio-Semiotic Study. Advances in Language and Literary Studies, v. 7, n. 6, p. 56-69, 2016.
• AUSTIN, J. L. How to Do Things with Words. Oxford: Oxford U.P., 1962.
• BARTON, D.; LEE, C. Language Online: Investigating Digital Texts and Practices. London: Routledge. 2013.
• COHN, N.; ENGELEN, J.; SCHILPEROORD, J. The Grammar of Emoji? Constraints on Communicative Pictorial Sequencing. Cognitive Research: Principles and Implications, v. 4, n. 33, p.1-18. 2019. DOI. https://doi.org/10.1186/s41235-019-0177-0.
  » https://doi.org/10.1186/s41235-019-0177-0.
• CRYSTAL, D. Internet Linguistics: A Student Guide. New York: Routledge, 2011.
• DENG, Q.; HINE, M.; JI, S. WANG, Y. Investigating the Impacts of Brand Social Media Posts’ Linguistic Styles on Consumer Engagement. Hawaii International Conference on System Sciences, 53^rd Proceedings of the…, p. 2741-2750, 2020.
• DRESNER, E.; HERRING, S. C. Functions of the Nonverbal in CMC: Emoticons and Illocutionary Force. Communication Theory, n. 20, p. 249-268. 2010.
• ESCOUFLAIRE, L. Signaling Irony, Displaying Politeness, Replacing Words: The Eight Linguistic Functions of Emoji in Computer-Mediated Discourse. Lingvisticæ Investigationes, v. 44, n. 2, p. 204-234, 2021.
• EVANS, V. Breaking the Emoji Code. 2018. Available in: https://www.psychologytoday.com/us/blog/language-in-the-mind/201804/breaking-the-emoji-code Accessed on: Nov 22nd, 2023.
  » https://www.psychologytoday.com/us/blog/language-in-the-mind/201804/breaking-the-emoji-code
• EVANS, V. The Emoji Code: The Linguistics behind Smiley Faces and Scaredy Cats. New York: Picador, 2017.
• GE, J.; GRETZEL, U. Emoji Rhetoric: A Social Media Influencer Perspective. Journal of Marketing Management, v. 34, n. 2, p. 1-24. 2018. DOI: https://doi.org/10.1080/0267257X.2018.1483960.
  » https://doi.org/10.1080/0267257X.2018.1483960.
• GE, J.; HERRING, S. C. Communicative Functions of Emoji Sequences on Sina Weibo. First Monday, v. 23, n. 11, p. 1-62, 2018. DOI: https://doi.org/10.5210/fm.v23i11.9413
  » https://doi.org/10.5210/fm.v23i11.9413
• LI, L.; YANG, Y. Pragmatic Functions of Emoji in Internet-Based Communication: A Corpus-Based Study. Asian-Pacific Journal of Second and Foreign Language Education, v. 3, n. 16, p. 1-12. 2018. DOI: https://doi.org/10.1186/s40862-018-0057-z
  » https://doi.org/10.1186/s40862-018-0057-z
• McSHANE, L.; PANCER, E.; POOLE, M.; DENG, Q. Emoji, Playfulness, and Brand Engagement on Twitter. Journal of Interactive Marketing, v. 53, p. 69-110, 2021. https://doi.org/10.1016/j.intmar.2020.06.002
  » https://doi.org/10.1016/j.intmar.2020.06.002
• OLIVEIRA, A. L. A. M. O.; CUNHA, G. X.; AVELAR, F.T. Emojis como estratégias de reparo em pedidos de desculpas: um estudo sobre conversas em ambiente digital Trab. Ling. Aplic., Campinas, v. 3, n. 57, p. 1615-1635, set./dez. 2018.
• STATISTA, 2021. Available at: https://www.statista.com/topics/11194/emoji-usage/#topicOverview Accessed on: July 17th 2022.
  » https://www.statista.com/topics/11194/emoji-usage/#topicOverview
• TANG, Y.; HEW, K. F. Emoticon, Emoji, and Sticker Use in Computer-Mediated Communications: Understanding Its Communicative Function, Impact, User Behavior, and Motive. International Journal of Communication, v. 13, p. 2457-2483, 2019.
• THOMPSON, D.; FILIK, R. Sarcasm in Written Communication: Emoticons are Efficient Markers of Intention. Journal of Computer-Mediated Communication, v. 21, n. 2, p. 105-120, 2016.
• UNICODE, 2021. Available at: https://home.unicode.org/emoji/emoji-frequency/. Accessed on: Nov 21st, 2023.
  » https://home.unicode.org/emoji/emoji-frequency
• WEISSMAN, B.; TANNER, D. A Strong Wink between Verbal and Emoji-Based Irony: How the Brain Processes Ironic Emojis During Language Comprehension. PLoS ONE, v. 13, n. 8, 2018. DOI: https://doi.org/10.1371/journal.pone.0201727
  » https://doi.org/10.1371/journal.pone.0201727
• YUS, F. Cyberpragmatics. Internet-Mediated Communication in Context. Amsterdam: John Benjamins, 2011.
• YUS, F. Not All Emoticons Are Created Equal. Linguagem em (Dis)curso, v. 14, n. 3, p. 511-529, set./dez. 2014.

• 1
  Twitter became X in July 2023. Since we collected the data before this date, we decided to keep Twitter and tweets while acknowledging that just the name, and not the affordances, has changed in this social network.

Editor de Seção:

Fábio José Rauen

Publication Dates

• Publication in this collection
  27 May 2024
• Date of issue
  2024

History

• Received
  11 Dec 2023
• Accepted
  08 Mar 2024