Home / Articles / Volume 11 (2014) / A pragmatic investigation of emoticon use in nonnative/native speaker text chat
Document Actions



Using data from college classroom text chats, this study takes a pragmatic, micro-analytic approach to describe how nonnative and native speaker participants use emoticons such as smileys, winkies, and frownies and what emoticons contribute to the verbal message. The pragmatic investigation of emoticons helps clarify their role within a larger conceptual framework of emotive and relational meaning. Moreover, by documenting the multifunctionality of smileys, the descriptive analysis shows that form-meaning pairings (e.g. “:)” means “happy”) cannot be taken for granted. Rather, emoticons are highly context-sensitive and can display affect or serve as contextual cues to signal illocutionary force and/or humor.


In face-to-face communication, nonverbal cues including kinesic and prosodic markers do relational work. For example, a smile, a friendly tone, a nod, or an outstretched hand can signal friendliness, while a frown or an agitated tone can be disaffiliating. In this way, nonverbal markers contribute to maintaining, enhancing, and challenging relationships. Since these nonverbal cues are largely unavailable in text-only computer-mediated communication (CMC), text-based emoticons such as smileys :) =) :-) and their variants :-} :-> :D, winkies ;-) ;) , and frownies :( have long been viewed as compensatory strategies (e.g., Crystal, 2001). Like other nonverbal cues, emoticons are highly context sensitive (e.g., Huffaker & Calvert, 2005; Wolf, 2000) and therefore offer fertile ground for qualitative studies, which are scarce thus far. Such qualitative research could provide for a richer and more nuanced understanding of emoticons and their interpretations in different contexts.

Like nonverbal cues in oral interaction, emoticons are not an add-on feature but are constitutive of CMC, and they occur in both native and nonnative speaker online discourse. As more and more people around the world use web tools to communicate with others in social or institutional settings, many do so in a second or foreign language. The present study investigates the use of these nonverbal cues in text chat. The analysis focuses on the multifunctionality of emoticons, their role in online relational work, and possible connections between emoticon use and language proficiency and thus contributes to a more complete understanding of emotive communication online.



Highly salient and closely associated with online usage, emoticons have been a focus of CMC research since the 1980s – in particular, with respect to their functions and effects in online communication, their relationship to facial expressions, and their conditions of use (for an overview, see Walther & D’Addario, 2001). In what follows, I provide a brief summary of existing research, which comprises mostly quantitative analyses.

Researchers have speculated that emoticons function to express feelings and attitudes as an online substitute for what is expressed in face-to-face (FTF) interaction via facial cues (e.g., Crystal, 2001; Halvorsen, 2012; Provine, Spencer, & Mandell, 2007). Because emoticons (“emot-icons”) are iconic, they are easily recognized as facial expressions (Walther & D'Addario, 2001). By the same token, there is widespread consensus that facial expressions allow for greater subtlety and complexity of emotional expression than emoticons (e.g., Kiesler, Siegel, & McGuire, 1984; Riordan & Kreuz, 2010). Moreover, researchers note that CMC is very much a learned skill, whereas FTF communicative skills are acquired in infancy (e.g., Tidwell & Walther, 2002; cited in Riordan & Kreuz, 2010). Even though emoticons do not carry the same functional load as nonverbal cues in FTF communication, it is not clear if affect is actually lost in CMC. In fact, the social information processing (SIP) model posits that CMC users will achieve relationships much like face-to-face communicators, albeit through different strategies and tools, which may include emoticons (Walther, 1992). Aside from the expression of affect, emoticons have been found to fulfill other functions, such as disambiguating the message, regulating the interaction, and strengthening the message content (e.g., Derks, Bos, & von Grumbkow, 2008; Riordan & Kreuz, 2010).

Several studies have looked at the contextual factors that may impact the use of emoticons. Primarily associated with interpersonal rather than transactional communication, emoticons appear more often in off-task communication (see Huang, Yen, & Zhang, 2008; Walther, 1995). Yus (2011, p. 198), for example, writes that

it is not surprising that using emoticons in task-oriented communication is not considered appropriate and is seen as a source of unnecessary distraction, whereas in socio-emotional conversation their use is not only predictable but expected as happens in most information conversations among adolescents.

What is more, emoticon use has also been linked to the sender's emotional involvement. The more cues a message has, the stronger the recipients judge the sender’s emotions to be (Harris & Paradice, 2007). Along the same lines, studies have found that emoticon users were perceived as more “socially present” (Yamada & Akihori, 2007), dynamic, friendlier, valuable, and talkative than those who did not use emoticons (see Huffaker & Calvert, 2005). The emoticon’s shape also influences the recipients’ perceptions as to its intensity. Yus (2005) found that :-))))) was associated with greater happiness in comparison to the default :-) but not in comparison to other greater-than-default emoticons such as :-))). Other quantitative studies have investigated differences in emoticon use between genders. In general, females use emoticons more often than males (Fox et al., 2007; Herring, 2003; Wolf, 2000; but cf. Huffaker & Calvert, 2005). In sum, researchers have found that gender and medium impact emoticon use, as do contextual factors such as setting and communication purpose.

Whereas existing emoticon research has largely taken the meaning of emoticons for granted, a few studies explore what it is that an emoticon contributes to the message in a given context. These studies contribute to an emerging understanding of the functional variance of emoticons. For example, Dresner and Herring (2010) look at the functions of emoticons and underscore that emoticons are not always iconic of emotion. They identify three core functions of emoticons: "(a) emotion, mapped directly onto facial expression (e.g., happy or sad); (b) nonemotional meaning, mapped conventionally onto facial expression (e.g., a wink as indicating joking intent; an anxious smile); and (c) illocutionary force indicators that do not map conventionally onto facial expression (e.g., a smile as downgrading a complaint to a simple assertion)" (p. 263). The authors stress that in many cases an emoticon does "not convey emotion but rather pragmatic meaning, and thus this function [as an illocutionary force indicating device, or IFID] needs to be understood in linguistic, rather than extralinguistic, terms" (2010, p. 250; see also Derks et al., 2008). Along the same lines, Riordan et al. found that emoticons "are less representative of a present state of emotion than they are of the intention or motivation of a writer. For example, an emoticon used after a negative comment does not indicate one smiling while saying something mean, but that the comment was not intended in a malicious manner" (2011, p. 2415). That emoticons can have pragmatic meaning was also shown in Golato and Taleghani-Nikazm's (2006) empirical analysis. Chat participants employed emoticons "to display their orientation to the dispreferred action [such as a request] […] to their co-participants" (Golato & Taleghani-Nikazm, 2006, p. 309). Generally, smileys and other emoticons are interpreted as affiliative (e.g., Darics, 2010; Golato & Taleghani-Nikazm, 2006; Holmes & Schnurr, 2005), but not always. Derks et al. (2008) report that a smiley placed after a negative verbal message can convey greater sarcasm than a message without an emoticon. Like a smiley, a wink generally has a positive connotation, but it also signals irony and jest (Rezabek & Cochenour, 1998; Walther & D'Addario, 2001). Research thus shows that emoticons complement the verbal message but do not change its valence, i.e., they cannot change the affective direction from negativity to positivity or vice versa (Walther & D'Addario, 2001).

Generally, emoticons are taken to serve a social function and/or promote rapport (Derks et al., 2008; Golato & Taleghani-Nikazm, 2006; Hancock, 2004; Walther & D’Addario, 2001), but their use is highly context sensitive. The substantial body of emoticon research shows that 1) emoticons are not textual equivalents of facial cues; 2) they do not always express emotion; 3) they can enhance or complement the verbal message by expressing the sender's emotions and/or attitudes; 4) they can modify the verbal message but not change its valence; and 5) they are generally affiliative. Even though existing research generally acknowledges the context sensitivity of emoticons, very little research has been done at the discourse level to describe how they are used in context. To date, their role in rapport building and politeness is not yet well understood, in part because qualitative analyses of authentic data remain scarce (but cf. Darics, 2010).

Emotive Communication

Adopting a strictly linguistic perspective on emoticons, this study seeks to situate their analysis within the area of emotive communication. In essence, emotive communication is the display of an affective stance, which Ochs defines as "a mood, attitude, feeling, and disposition, as well as degrees of emotional intensity vis-à-vis some focus of concern" (1996, p. 410). What distinguishes emotive communication is that it refers to "the intentional, strategic signaling of affective information in speech and writing (e.g., evaluative dispositions, evidential commitments, volitional stances, relational orientations, degrees of emphasis, etc.)" (1994, p. 329). From this perspective, emoticons are devices of communication rather than of emotional expressions, i.e., they are "‘signs of affect’, or ... indices of speakers’ feelings, attitudes, or relational orientations toward their topics, their partners, and/or their own acts of communication in different situations" (Caffi & Janney, 1994, p. 329; emphases mine). Viewing emotive communication as linguistic communication, Caffi and Janney persuasively make the case for situating its study firmly within pragmatics.

Emotive communication, like other types of interpersonal communication, allows participants to do relational work. In successful communication, speakers seek to accommodate the often conflicting needs of saying what they want or have to say and building rapport to enhance social cohesion. Politeness researchers have argued for broadening the scope of politeness research beyond its traditional focus on mitigation or avoidance of face-threatening acts (FTAs) (e.g., Locher, 2004, 2008; Locher & Watts, 2005). According to Locher and Watts, politeness requires a discursive approach, which “comprises the entire continuum of verbal behavior from direct, impolite, rude or aggressive interaction through to polite interaction” (2005, p. 11; see also Mills, 2003). From this perspective, linguistic resources for relational communication are as relevant as FTA mitigation to politeness considerations. With the vast majority of existing research on politeness work in computer-mediated communication (Golato & Taleghani-Nikazm, 2006; Herring, 1994, 1996; Morand & Ocker, 2003; Vinagre, 2008; but see Vandergriff, 2013a) grounded in the Brown and Levinson (1987) framework, Locher and Watts' (2005) article opens up new vistas in CMC politeness research.

The Current Study

The current pragmatic investigation of emoticons is designed to document some of the meanings or functions one form can have in different contexts. Its ultimate goal is to help clarify the role of emoticons within a larger conceptual framework of emotive and relational meaning. To this end, this study takes a micro-analytic approach to show how nonnative and native speakers of English use emoticons in classroom text chat.

Research Questions

The following two questions guided the analysis:

  • What meanings do emoticons contribute to the meaning of the verbal message in nonnative/native text chat?

  • How do the previously unacquainted participants use emoticons to manage, enhance, and challenge relationships in the language learning environment?


With examples drawn from a pre-existing corpus of roughly 32,000 words,1 this study documents emoticon use, specifically smileys, frownies, and winkies2 in dyadic text-only chat among unacquainted participants who are native speakers and students at the University of Pennsylvania ("xxx Penn") and intermediate/advanced learners of English at the University of Malmo in Sweden ("xxx Malmo"). The nonnative speaker portion makes up about 63% of the corpus with about 20,000 words, while the native speaker portion contains about 12,000 words. The nonnative speakers had an average of 11 years of formal instruction in English; their English ranged from the intermediate-high to the advanced level. Except for five participants who reported Arabic, Bosnian, or Spanish as their first language, all 23 English learner participants were native speakers of Swedish.3

The corpus contains Live Messenger transcript data that were collected originally for an experimental study on the effectiveness of different kinds of corrective feedback (reported on in Sauro, 2009, 2013). Transatlantic partners worked in dyads on two collaborative writing tasks on the topics of "environmental issues" and "Swedish culture.” For each activity, participants were given word banks and were instructed to use their words to share their opinion on the topic. These word banks were meant to elicit contexts for zero article usage4 to provide opportunities for corrective feedback. Participants had a maximum of 20 minutes to complete the task and used their remaining time to chat freely (Sauro, 2009, p. 114). Data exemplars were chosen to document emoticon use by nonnative and native speakers in on- and off-task contexts in order to illustrate the functional range of their emoticon use.


Emoticon Use in Nonnative/Native Text Chat

Emoticons display the sender's stance, functioning as emotive communication devices, affect signs, or "indices of speakers’ feelings, attitudes, or relational feelings, attitudes, or relational orientations toward their topics, their partners, and/or their own acts of communication in different situations" (Caffi & Janney, 1994, p. 327). In the data, smiley emoticons, including the variants =), :), and :-), constitute the most frequently used emoticon type by far.

Table 1. Number of emoticons by type in nonnative (NNS) and native speaker (NS) on-task and off-task chat

Nonnative speaker participants showed more frequent emoticon use than their native speaker peers, with 2.18 emoticons per 100 native speaker transmissions compared to 4.01 for the nonnative data set. Whereas native speaker participants used smileys and winkies exclusively, the nonnative speaker data showed more emoticon variation. The reason for the differences in frequency and variation are not clear, however. One might speculate that nonnative participants used emoticons more frequently to compensate for linguistic deficiencies, which may be subtle. For example, a learner who is not certain of a word’s connotations may choose to add an emoticon to build rapport. Moreover, the nonnative speaker participants’ interactional role as learners in the learner-expert dyads may also have an effect. A prior analysis of the data (Vandergriff, 2013b) showed that participants orientated to their complementary interactional roles, a finding that suggests that nonnative speaker participants may want to mitigate the imposition on their native speaker peers. Given the relatively small scale of the study, the findings cannot be generalized. Whether or not nonnative speaker participants overall use emoticons more frequently than native speaker participants and, if so, what factors may be responsible for the differential use remain open questions which deserve further study.

Emoticons as Displays of Affect

The qualitative analysis of the data shows that smileys often display positive affect on the part of the sender, such as one might experience with relief from stress or tension, amusement or delight, or a pleasant interaction. In extract 1, Me1 Malmo, an English language learner, states that the task has been completed as the two partners transition to their off-task discourse.

Extract 1: Test is Over

Me1 Malmo sends a series of three transmissions. In transmission 1, s/he is relaying the instructor's ("SS's") permission to move on to less structured talk, and transmissions 2 and 3 show the participant's commitment to the new chat phrase as s/he requests information from Luisa Penn. The analysis focuses on transmission 1 with its transmission-final smiley emoticon. Whereas the verbal part is transactional only, the turn-final smiley icon displays the speaker's stance of relief and marks the turn as an expressive speech act. In enhancing the verbal message the smiley emoticon plays a crucial role.

Returning to the analysis of emoticon functions, extract 2 illustrates how participants used smileys to upgrade the display of friendliness conveyed in the verbal portion of the message. The following extract represents the closing sequence of the chat between Shadow Penn and Me3 Malmo.

Extract 2: I Had Fun

In the process of leave-taking, Shadow Penn affirms that the chat was a pleasure and hopes that Me3 Malmo also enjoyed it. Me3 Malmo chunks his/her message into three transmissions: Transmission 7 is in response to Shadow Penn's "I had fun" in transmission 6 – echoing the evaluation and thereby marking intersubjective alignment. By aligning their stances, the stancetakers themselves also align (Du Bois, 2007). The emoticon enhances this verbal message of alignment and, in this way, upgrades the friendliness of the response. This is followed by a second transmission (transmission 8) that orients to their different participant roles as English language learner and English language native speaker/teacher. Notwithstanding the implied difference in terms of participation roles, Me3 Malmo indexes intersubjective alignment in the structural (I hope + dependent main clause) and semantic parallelism (I hope ...). In the context of interpersonal talk, the emoticon becomes a social cue, much like a smile in face-to-face exchanges, showing pleasure with the encounter. This usage is attested in both the nonnative- and native-speaker data sets. Not surprisingly, social talk seems to be a hospitable context for emoticons, especially smileys, as they occur at higher rates in chat phases of phatic communication, such as openings and closings, than elsewhere.

Compared to smileys, frowny emoticons are relatively rare. In extract 3, Me9 Malmo uses a frowny to express disappointment that the chat has come to an end.

Extract 3: More Time

Here the frowny emoticon, like the smiley, serves to build rapport and enhance the relationship. Extract 3 shows that emoticons can occur in the leave-taking process as participants express pleasure in each other and/or disappointment that the chat is over. In the context of social talk, emoticons display affect in the service of relational work. In sum, both nonnative- and native-speaker participants used emoticons as expressive devices to display feelings or attitudes that have implications for the relationship. This is one way in which emoticons help build rapport.

Participants also used smiley emoticons in compliment responses, as in the following extract.

Extract 4: At Least I Tried

In transmission 4, Steve Penn (the native speaker) compliments the nonnative participant on his/her task performance. The primary interactional goal of giving compliments is to create affiliation, i.e., a feeling of closeness by expressing interest in the hearer. In general, compliments make a response conditionally relevant. Me8 Malmo chunks his compliment response into two transmissions (5 & 6). While Me8 Malmo may agree with the compliment, self-praise is a conversational taboo. S/he downgrades the praise (in transmission 5), then accepts it with a compliment appreciation token in transmission 6 followed by the smiley emoticon.

Emoticons as Contextualization Cues

The data also yield instances of emoticon use that do not display the sender's feelings or attitudes. Instead, the emoticon might best be described as a metamessage that provides the receiver with information on how to interpret the verbal portion of the message. In this section, I analyze extracts that illustrate this use of emoticons as contextualization cues, first in the context of dispreferred action5 (here: requests) and second in the context of conversational humor.

Extract 5 illustrates how the native speaker used emoticons to orient to the dispreferred action and mitigate its face-threatening potential. Recall that the task was designed to elicit potential contexts for the use of zero-article – a tricky usage issue even for advanced learners of English – so that the native speakers could provide feedback. Because of the interactional roles set up by the task, the native speakers took on the role of the teacher vis-à-vis the English-language learner, prompting the EFL participants to produce sentences. In the context of these requests or directives, smiley emoticons occurred. In extract 5, Christie Penn reminds Co4 Malmo of the task instructions.

Extract 5: The Words

With Christie Penn enacting the teacher role, extract 5 resembles formal classroom discourse in a number of ways. The smiley orients to the reminder (“try to use the words”) and, in this way, softens the force of the imperative. As Dresner and Herring (2010) might say, the emoticon mitigates or downgrades the force of the utterance from a reminder to a friendly request. Both the native- and nonnative-speaker data sets yield smiley emoticons in the context of requests.

Emoticons can also play a role in contexualizing conversational humor. Extract 6 illustrates the use of a winky. At the beginning of the text chat, Re2 Malmo indicates his readiness to begin task work.

Extract 6: Born Ready

The verbal portion of transmission 5 is clearly hyperbole but, without the wink, it may be interpreted as self-praise. The winky marks the conversational humor explicitly. Conversational humor relies on a mismatch between what the verbal message seems to convey and what is actually meant. Often, the humor is not formally marked at all; instead, interlocutors must recognize the implicit keying marker, namely contextual inappropriateness (Attardo, 2005; Vandergriff & Fuchs, 2012). If the humor is not recognized, it may result in a face threat to both the humorist and the interlocutor. In oral interaction, to joke and not get a laugh is like "initiating a handshake; only to have one's outstretched hand ignored" (Hay, 2001, p. 58). A joke thus makes a response such as laughter conditionally relevant (Schegloff, 1968). The potential threat to the humorist's face is mirrored by that to the recipient's face, who needs to signal that she "got the joke" in order to fully maintain face (Carrell, 1997). Yet, the potential face threat associated with failed humor may be outweighed by the rapport-building effect of shared laughter. By cueing humor explicitly through an emoticon, the humorist can ensure humor recognition and thus minimize the social risk, even if the recipient does not find the joke funny. Re2 Malmo goes to great lengths to make sure the humor is recognized. In addition to cueing the humor, Re2 Malmo does not wait for a humor receipt token. Instead s/he follows up with an apology and an account.6 It is not clear how successful Re2 Malmo’s attempt at humor was. Shadow Penn’s response of "Great!" (transmission 6) does not clearly signal humor recognition, let alone humor appreciation.

In addition to cueing jest, winkies can display humor appreciation in response to a humor token and, at least in some cases, appear to have overlapping functions with smileys, as can be seen in extract 7 (see also Dresner & Herring, 2010).

Extract 7: Larger City

In echoing both “a bit” and the emoticon, Anna Penn supports Me5 Malmo’s humor and signals alignment (transmission 6, see also extract 2). But instead of repeating Me5’s emoticon, Anna Penn uses a winky to signal humor appreciation and nonserious intent (transmission 7).

Extract 8, viewed alongside extract 7, provides even stronger evidence that smiley and winky emoticons may be interchangeable in cueing and supporting humor.

Extract 8: Staying Home

In transmissions 5 and 6, Re11 Malmo and Luisa Penn joke that they prefer to go to school rather than stay home. In this way, the two co-participants find common ground by co-constructing a type of conversational humor that is (slightly) subversive given the institutional setting of the chat. In transmission 6 the winky cues nonserious intent, followed in transmission 7 by humor support (verbal message and smiley). The smiley shows humor recognition, understanding and appreciation of Re11 Malmo’s humor (see Hay, 2001). The fact that in extract 7 and 8 winkies and smileys occur with apparently similar functions suggests that they are interchangeable in some contexts (Herring, personal communication).

Subversive humor of a different kind can also be found in the following extract. Re9 Malmo appears to attend to task demands but talks about vodka, rather than the assigned topics.

Extract 9: Now You Know That!

Re9 Malmo responds to Alta Penn's invitation to talk about Sweden. Introducing Swedish culture by talking about its most famous vodka brand could be viewed as a breach of appropriateness rules and might therefore be described as slightly subversive in an institutional context that gives priority to weightier cultural topics. At the same time, this kind of humor indexes Re9 Malmo's student identity and seeks to find similarities with his transatlantic peer. S/he takes an assertive volitional stance, exaggerating the directness and aggressive tone in a playful manner ("now you know that,” highlighted by an exclamation mark), then orients to this action with a transmission-final emoticon (transmission 2). The humor thus derives from the exaggerated contextual inappropriateness of the verbal message and is marked formally by the emoticon. The recipient recognizes the humor and responds with a humor appreciation token in transmission 3, including spelled-out laughter and "awesome.”

Conversational humor can build rapport with shared laughter, but it also carries some social risks. The analysis above showed how unrecognized humor may be interpreted as arrogance (extract 6) or nerdiness (extract 9), for example. Whereas contextual inappropriateness eo ipso is a cue of conversational humor, smileys and winkies serve as explicit humor cues to disambiguate the verbal message. As illustrated in extracts 6–9, with the promise of shared laughter and the emoticon to ensure humor recognition, it might appear that humor always pays off socially. But this is not the case, as shown in extract 10.

Extract 10: IKEA Furniture

In transmission 5, Susan Penn attempts to make a joke, marked explicitly as nonserious with smiley. The second-pair part comes in transmission 7 where Co6 Malmo confirms that they are indeed making “IKEA stuff.” The absence of a humor appreciation token is notable in this episode, as well as the serious tone of Co6 Malmo’s sequenced response in transmissions 7 and 8. In light of the emoticon to mark the attempt at humor, Co6 must have recognized the humor but does not find it funny. In fact, it appears that Co6 found the humor offensive, as it suggests that Sweden only produces one product. The expressive “I like IKEA a lot!” may be expressing cultural pride. Susan’s alignment (“me too”) in transmission 9 suggests that she understands that her attempt at humor failed. Whereas emoticons help build and maintain rapport in combination with the verbal message, they cannot always ensure solidarity.


Functions of Emoticons

Extracts 1–10 above illustrate several distinct emoticon functions in the context of nonnative/native speaker chat. Extracts 1–4 show emoticons as markers of affective stance. For example, when a chat participant expresses relief from tension that "the test is over,” the emoticon is utilized to express the sender's emotion or feeling. Such uses, which communicate joy, happiness, or relief, are referred to as markers of "emotional meaning" (Dresner & Herring, 2010). As in oral interaction, where smiles are recruited to serve social functions, smiley emoticons often display affect in order to maintain or enhance the relationship with the co-participant. For example, participants use emoticons to indicate that they enjoyed the chat (extract 2 and 3), that they enjoyed meeting their co-participant, or that they enjoyed receiving the co-participant's compliment on their L2 performance (extract 4).

Aside from affect display, emoticons are also utilized to serve pragmatic functions. The analysis in extracts 5–10 documents how emoticons can function as keying markers or contextualization cues. As contextualization cues (Gumperz, 1982), emoticons constitute an important resource as they allow participants to deliberately manipulate the ‘key’ (Hymes, 1974) or ‘framing’ (Goffman, 1974) of the ongoing interaction. In this way, emoticons communicate the sender's intention or stance with respect to the verbal message and her positioning vis-à-vis the co-participant(s).

What makes this use of emoticons difficult to analyze is that, in their function as contextualization cues, they derive their meaning from context. Always multifunctional, their function can only be determined in concrete contexts. As contextualization cues, emoticons were documented in two functions. First they were shown to orient to a dispreferred action and mitigate the face-threatening potential of requests (extract 5). Extract 5 illustrates what Dresner and Herring (2010) have referred to as an illocutionary force indicating device (IFID). This type of emoticon use has also been described in Golato and Taleghani-Nikazm's (2006) analysis of requests in webchats. The latter authors conclude "that smiley faces and other positive emoticons are utilized by web chat participants as affiliative expressions in order to display their orientation to the dispreferred action (which they have already performed or are about to perform) to their co-participants" (p. 309). Second, smileys and other positive emoticons are often embedded in the transmission following humor and serve to cue nonserious intent on the part of the sender such as in conversational humor or irony (Dresner & Herring, 2010). A number of studies have shown that humor may be cued via emoticons (e.g., Danet, 2001; Crystal, 2005; Derks, Bos & von Grumbkow, 2008; Lo, 2008). In extracts 6–10, for example, the smiley and winky emoticons, respectively, serve to cue contextual inappropriateness (Attardo, 2005), a mismatch of some type, as it were, between what is said and what is meant. Whether as IFID or humor cue, emoticons as contextualization mark this sender's stance with a metamessage that what is said is not what is (strictly) meant.

While a complete account of emoticon usage requires further research that looks at form-meaning mappings in different contexts, this analysis offers a glimpse of how the functional variance of an emoticon may be motivated. Specifically, I argue that the smiley core function of expressing happiness, joy, or relief from tension or stress is extended from the sender-oriented expression of affect to the social sphere, where affect display plays a crucial role in establishing, enhancing, and maintaining relationships. In this way, emoticons are able to convey not only emotion but also stance. An emoticon token such as a smiley or frowny interacts with linguistic features in marking a participant’s orientation to discourse (Du Bois, 2007). A frowny emoticon in the context of a leave-taking sequence, for example, displays how the sender feels about leave-taking (extract 3). In some contexts, the function of winkies may overlap with that of smileys (extract 7 and 8). When used to cue humor or signal humor appreciation, emoticons orient to a message's contextual inappropriateness by indexing the gap between what is said and what is meant. In this way, the emoticon eliminates, or at least minimizes, the risk that a humor token may not be recognized as such, which would constitute a potential face threat to both the humor recipient (for failing to recognize humor) and to the humorist him- or herself. In sum, smiley faces and other positive emoticons are utilized as affiliative expressions in order to display the participant's orientation to what may be perceived as contextual inappropriateness. In this way, the data provide evidence of the emoticon's functional range from "I'm happy" to "I'm friendly" to "I'm kidding," accompanied by a shift from sender orientation to recipient orientation. In the following section, I discuss how these functions do relational work.

Relational Work

This study provides new insights into the ways participants orient to the relationship between co-participants in online learner communication. Adopting Mills’ view of "politeness as practice" (Mills, 2003), it shows that emoticons play an important role in online facework. From this perspective, politeness spans the full range from the unconscious application of scripts to deliberate strategic use, from the unmarked to the explicitly marked (Mills, 2003; Watts, 2003), and from politeness that enhances to (im-)politeness that challenges relationships. Using a smiley emoticon to mitigate a potential FTA has been described before (Golato & Taleghani-Nikazm, 2006) as a politeness strategy, as illustrated in extract 5.

Aside from mitigative strategies, some emoticon uses represent strategies that foreground the recipient’s inferred wants and needs. One such strategy is the display of positive evaluations of the recipient and/or his/her actions. In extract 2, the sender intensifies the verbal message ("Me too") with a smiley to convey appreciation of the chat. In extract 3, a frowny, in spite of its negative value, conveys a positive evaluation of the chat experience in response to the co-participant’s verbal message (“wish we had more time to chat!” – “Me too :(”). This frowny is an example of what Dresner and Herring (2010) classify as “nonemotional meaning,” which maps emoticons conventionally onto a facial expression. In the context of the verbal message, the frowny (extract 3) conveys very similar social intentions as the smiley (extract 2). This interpretation is supported in Walther and D’Addario’s (2001) study of emoticons, which finds that frown emoticons behave differently from smile emoticons, in that the frown emoticon coupled with a negative message does not convey greater negativity than the negative verbal message on its own. In addition, emoticons often occur in the production and reception of humor, which serves to build rapport among participants. In extracts 9 and 10, emoticons serve as cues in the co-construction of humor and thus play a crucial role in avoiding the pitfalls of conversational humor. Regardless of their specific functions, emoticons are affiliative strategies that send a metamessage of rapport alongside the verbal message. Finally, emoticon use itself may index identity and youth (Yus, 2002) in spite of considerable variability across contexts and communities of practice.

Given the social intentions emoticons serve, it would seem that they should occur with greater frequency in interpersonal than transactional communication. In the corpus, emoticons do indeed occur more frequently in off-task than on-task discourse7, in line with previous research (Huang, Yen, & Zhang, 2008; Walther, 1995; Yus, 2011). Moreover, with off-task talk, emoticons appear to occur more frequently in specific activities, such as phatic communication and exchange of personal information, both of which are associated with establishing a relationship between previously unacquainted participants. At the same time, emoticon use was not restricted to this context, attesting to the fact that participants do relational work even in the on-task stretch of chat (e.g., extracts 5–7, 9–10). There is thus little evidence for Yus’ claim that participants consider emoticons in on-task chat inappropriate “and a source of unnecessary distraction” (Yus, 2011, p. 198). Instead it seems that in on-task chat, task demands are given priority, while socio-emotional intentions are backgrounded. Accounting for the relative rarity of emoticons in on-task communication in this way is in line with existing research on language learner discourse that has pointed to a conflict of conversational trajectories in both online and offline settings (e.g., Markee, 2000; Tudini, 2010). It appears that social intentions are often at odds with pedagogical ones and that learners and their co-participants strive to attend to both. The use of emoticons in on-task talk can be viewed as evidence of juggling both.

Emoticons in the Language Learning Context

Situational factors such as the language learning context in general and participant characteristics such as age and language proficiency shape online discourse (Herring, 2007), as do participants' beliefs regarding the norms and tone. The higher frequency of emoticon use in the nonnative speaker participant set suggests that emoticons have a compensatory function for real or perceived second language shortcomings. In other words, nonnative speaker participants may draw on their nonlinguistic semiotic repertoire to compensate for any shortcomings in their use of English, a symbolic system they have not fully mastered. There is evidence that these nonnative speakers themselves are aware of this challenge. Even when communicating successfully, they often oriented to the gap between their own language competence and that of a native speaker (Vandergriff, 2013b). At the same time, transferring communicative skills such as emoticon use into second language practice cannot be taken for granted because we know that learners do not always make use of transferable pragmatic knowledge (see Kasper & Rose, 2001). Instead, they may be highly context-aware in selecting pragmatic strategies in their first language but may under-differentiate context variables such as social distance and social power in the foreign language (Fukushima, 1990; Tanaka, 1988).

Compensatory emoticon use is not always perceived as an index of competence. With respect to online usage, Averianova (2012) notes that students sometimes "overindulge in condensed writing, novel abbreviation, [and] use of emoticons" (p. 16), suggesting that they have not mastered a more appropriate style. From this perspective, emoticons appear more like a pragmatic crutch. It would follow then that in order to signal irony or sarcasm, participants with more language competence would prefer subtler keying markers to the “crude” emoticon. Regardless of whether they are viewed as a tool or a crutch, emoticons may be a preferred nonnative speaker strategy for doing relational work, which native speakers tend to handle through text-only messages. To what extent language competence plays a role in emoticon use thus deserves more attention.

It would be overly simplistic to ascribe the differences in emoticon use between nonnative and native speakers to language competence only. Cultural differences, for example, may play a role in the differential use of emoticons between nonnative and native speaker use. The influence of culture on the form of emoticons is well known. For example, Asian style emoticons are not to be read sideways. A smiley might look like this (*_*), compared to the Western-style :). East Asian emoticons put more emphasis on the eyes, compared to Western emoticons, which emote mainly with the mouth. In general, Asian style emoticons are more elaborate and more varied. Beyond formal characteristics of emoticons, we know that major cultural differences exist with respect to emoticon usage, making transfer errors in emoticon use not unlikely (e.g., Katsuno & Yano, 2007; Nishimura, 2007). Finally, manifestations of relational work differ a great deal across cultures (Mey, 2001). Since participants use emoticons to build and enhance rapport, cultural differences with respect to politeness and facework may also be reflected in emoticon use.

Aside from language competence and culture, the fact that participants enacted different interactional roles in this institutional context is likely to have impacted emoticon use. In spite of their status as peers, much of the interaction between the nonnative and native participants mirrored the formal classroom. Participants themselves were aware of these roles and oriented to them throughout the interactions (Vandergriff, 2013b). Following some brief phatic communication, nonnative and native speakers took on complementary teacher-student roles, with the native speaker acting as an interactional pivot whose primary function was to elicit learner talk, which, in turn, provided opportunities for feedback. It fell on the learner to provide information on culture with very little opportunity for real exchange. Because of these different roles, opportunities for emoticon use differed. For example, native speakers in this study use more direct requests, which they orient to and mitigate by a smiley, while other opportunities for emoticon use are more limited. Among the activities in which both groups of participants engage in comparable ways, co-construction of humor (extract 10) and ludic play stand out. In sum, the differences in the frequency of emoticon use between nonnative and native speakers may be caused by a range of situation factors (Herring, 2007) and their interplay, but language proficiency most likely also played a role. If verbal competence falls short of communicative goals, nonnative speakers may compensate by relying more on nonverbal semiotic resources. Emoticons offer nonnative CMC participants a low-cost strategy to make the most of their developing competence.

Summary and Conclusion

In line with previous research (e.g., Derks, 2008; Derks & von Grumbkow, 2008; Yamada & Akihori, 2007), the findings of this study suggest that emoticons are affiliative strategies that text-chat participants use to build rapport, regardless of whether they are expressing affect, displaying relief, happiness, or friendliness, cueing humor, or mitigating FTAs. This study specifically focused on form-meaning pairings. Most importantly, the analysis documented that emoticons are multifunctional, a finding in line with previous work (Dresner & Herring, 2010). Highly context-sensitive, a smiley emoticon, for example, will interact with linguistic features including syntactic position to serve a range of functions from conveying amusement to mitigating FTAs. At the same time, each emoticon will typically have a number of variants. A smiley, for example, may show up as :) or =) Some researchers (e.g., Dresner & Herring, 2010) who view punctuation as a function of emoticons have speculated that their variation and multifunctionality has to do with the novelty of the medium, and that this seemingly chaotic state of affairs mirrors early writing, which also showed a great deal of variation in orthography and punctuation. Once emoticon conventions become widely accepted, the argument goes, their use will become more regularized. While variant smiley forms :) =) :-) may perhaps fall by the wayside in favor of one dominant smiley form, there is no reason to expect that emoticons will have fewer functions or meanings.

This study also considered the influence of situation factors (Herring, 2007) on emoticon use, such as participant characteristics, including proficiency with language and CMC and intercultural competency; the purpose of the interaction within the educational setting, comprising on-task activities as well as phatic communication; the norms of institutions; and the casual, friendly tone of the exchanges. In order to account for higher rates of emoticon use among nonnative- compared to native-speaker participants and in off-task compared to on-task chat, the discussion suggested ways in which two different contextual parameters, namely the nonnative participants’ competency in English and their interactional roles as language learners vis-à-vis the native speakers, may have affected the use of emoticons. In this way, the discussion cautioned against monocausal explanations and underscored the need for contextual analyses. Given the continuously growing importance of online language use in institutional contexts, emoticons clearly deserve increased research attention, especially qualitative or mixed-methods approaches.

Future research on emoticons should also pay more attention to nonnative speaker use of emoticons. In an increasingly globalized world, more and more people will connect online with speakers of other languages. A more complete understanding of emoticons and their usage patterns in different languages by native and nonnative speakers promises to advance research in emotive communication.


  1. The data were collected by Shannon Sauro for an earlier study of corrective feedback (Sauro, 2009). She provided me access to the data and to a preliminary quantitative analysis of emoticon use. I gratefully acknowledge her generous help.

  2. Nonnative participants also used emoticons such as ^^, which were not included in the analysis.

  3. The gender of the Malmo participants and the age of the Penn participants are not indicated in the analysis unless made relevant by the participants themselves.

  4. With certain nouns English uses so-called “zero article,” i.e., neither the definite article “the” nor the indefinite article “a(n).”

  5. “A dispreferred action is routinely delayed in its turn, routinely prefaced or qualified, is routinely accomplished in a mitigated or indirect form, it’s routinely accounted for” (Liddicoat, 2007, pp. 116-117).

  6. Shadow Penn’s response is sent nearly simultaneously with Re2 Malmo’s; Re2 Malmo follows up on the humor token before reading the response.

  7. The analysis is based on Barbieri and Sauro’s (2012) coding of on-task and off-task interaction. Taking a conservative approach in determining the boundaries of off-task interaction, they coded discourse that preceded the beginning of the task-opening sequences or followed task-closing sequences as off-task.


Attardo, S. (2005) Humor. In Östman, Jan-Ola & Jeff Verschueren (Eds.), Handbook of pragmatics online (pp. 1-18). Amsterdam/Philadelphia: John Benjamins. Retrieved December 3, 2007 from http://www.benjamins.com/online/hop/.

Averianova, I. (2012). The language of electronic communication and its implications for TEFL. Procedia – Social and Behavioral Sciences, 34, 14-19.

Barbieri, F., & Sauro, S. (2012). The linguistic characteristics of on-task and off-task interaction during chat: An exploratory corpus-based study. Paper presented at the annual meeting of the American Association of Applied Linguistics, March 24-27, 2012, Boston, MA.

Brown, P., & Levinson, S. (1987). Politeness. Cambridge: Cambridge University Press.

Caffi, C., & Janney, R. W. (Eds.) (1994). Towards a pragmatic of emotive communication. Special issue of Journal of Pragmatics, 22, 325-373.

Carrell, A. (1997). Joke competence and humor competence. Humor: International Journal of Humor Research, 10(2), 173-185.

Crystal, D. (2001). Language and the Internet. Cambridge: Cambridge University Press.

Crystal, D. (2005). Txting. The gr8 db8. Oxford: Oxford University Press.

Danet, B. (2001) Cyberplay: Communicating Online. Oxford: Berg.

Darics, E. (2010). Politeness in computer-mediated discourse of a virtual team. Journal of Politeness Research, 6(1), 129-150.

Derks, D. (2008). Emoticons and online message interpretation. Social Science Computer Review, 26(3), 379-388.

Derks, D., Bos, A. E. R., & von Grumbkow, J. (2008). Emoticons in computer-mediated communication: Social motives and social context. CyberPsychology & Behavior, 11, 99-101.

Dresner, E. & Herring, S. C. (2010). Functions of the nonverbal in CMC: Emoticons and illocutionary force. Communication Theory, 20(3), 249-268.

Du Bois, J. (2007). The stance triangle. In R. Englebretson (Ed.), Stancetaking in discourse: Subjectivity, evaluation, interaction (pp. 139-182). Amsterdam/Philadelphia: John Benjamins.

Fox, A. B., Bukatko, D., Hallahan, M. & Crawford, M. (2007) The medium makes a difference: Gender similarities and difference in Instant Messaging. Journal of Language and Social Psychology, 26(4), 389-397.

Fukushima, S. (1990). Offers and requests: Performance by Japanese learners of English. World Englishes, 9, 317-325.

Goffman, E. (1974). Frame analysis. An essay on the organization of experience. New York: Harper & Row.

Golato, A., & Taleghani-Nikazm, C. (2006). Negotiation of face in webchats. Multilingua, 25, 293-321.

Gumperz, J. (1982). Discourse strategies. Cambridge, UK: Cambridge University Press.

Halvorsen, A. (2012). Patterns of emoticon usage in ESL students’ discussion forum writing. CALICO Journal 29(4), 694-717.

Hancock, J. T. (2004). Verbal irony use in face-to-face and computer-mediated communication. Journal of Language and Social Psychology, 23(4), 447-463.

Harris, R. B., & Paradice, D. (2007). An investigation of the computer-mediated communication of emotion. Journal of Applied Sciences Research, 3, 2081-2090.

Hay, J. (2001). The pragmatics of humor support. Humor: International Journal of Humor Research 14(1), 55-82.

Herring, S. C. (1994) Politeness in computer culture: Why women thank and men flame. In M. Bucholtz, A. Liang, L. Sutton, & C. Hines (eds.), Cultural Performances: Proceedings of the Third Berkeley Women and Language Conference (pp. 278-294). Berkeley, CA: Berkeley Women and Language Group.

Herring, S. C. (1996). Posting in a different voice: Gender and ethics in computer-mediated communication. In C. Ess (Ed.), Philosophical perspectives on computer-mediated communication (pp. 115-145). Albany: SUNY Press.

Herring, S. C. (2003). Gender and power in online communication. In J. Holmes & M. Meyerhoff (Eds.), The handbook of language and gender (pp. 202-228). Oxford: Blackwell.

Herring, S. C. (2007). A faceted classification scheme for computer-mediated discourse. Language@Internet. 4, article 1. Retrieved May 28, 2013 from http://www.languageatinternet.org/articles/2007/761

Holmes, J. & Schnurr, S. (2005). Politeness, humor and gender in the workplace: Negotiating norms and identifying contestation. Journal of Politeness Research 1(1), 121-149.

Huang, A. H., Yen, D. C., & Zhang, X. (2008). Exploring the potential effects of emoticons. Information and Management, 45, 466-473.

Huffaker, D., & Calvert, S. L. (2005). Gender, identity and language use in teenage blogs. Journal of Computer-Mediated Communication, 10(2), 00.

Hymes, D. (1974). Foundations in sociolinguistics: An ethnographic approach. Philadelphia: University of Pennsylvania Press.

Katsuno, H. & Yano, C. (2007) Kaomoji and expressivity in a Japanese housewives’ chat room. In B. Danet & S. C. Herring (Eds.), The multilingual Internet: Language, culture, and communication online (pp. 278-300). New York: Oxford University Press.

Kasper, G., & Rose, K. M. (2001). Pragmatics in language teaching. Cambridge, UK: Cambridge University Press.

Kiesler, S., Siegel, J., & McGuire, T. W. (1984). Social psychological aspects of computer-mediated communication. American Psychologist, 39, 1123-1134.

Liddicoat, A. J. (2007) Introduction to conversation analysis. Continuum: London.

Lo, S.-K. (2008). The nonverbal communication functions of emoticons in computer-mediated communication. CyberPsychology & Behavior, 11(5), 595-597.

Locher, M. (2004). Power and politeness in action. Berlin/New York: de Gruyter.

Locher, M. (2008). Relational work, politeness and identity construction. In G. Antos & E. Ventola (Eds.), Handbook of applied linguistics. Volume 2: Interpersonal Communication (pp. 509-540). Berlin/New York: de Gruyter.

Locher, M., & Watts, R. J. (2005). Politeness theory and relational work. Journal of Politeness Research, 1, 9-33.

Markee, N. (2000). Conversation analysis. Mahwah, NJ: Erlbaum.

Mey, J. L. (2001) Pragmatics: An introduction. 2nd ed. Oxford/Malden, MA: Blackwell.

Mills, S. (2003). Gender and politeness. Cambridge, UK: Cambridge University Press.

Morand, D., & Ocker, R. (2003). Politeness theory and computer-mediated communication: A sociolinguistic approach to analyzing relational message. Proceedings of the 36th Annual Hawaii International Conference on System Sciences (HICSS'03). Los Alamitos, CA: IEEE Press.

Nishimura, Y. (2007). Linguistic innovations and interactional features in Japanese BBS communication. In B. Danet & S. C. Herring (Eds.), The multilingual Internet: Language, culture, and communication online (pp. 163-183). New York: Oxford University Press.

Paolillo, J. C. (2007). How much multilingualism? Language diversity on the internet. In B. Danet & S. C. Herring (Eds.), The multilingual Internet: Language, culture, and communication online (pp. 408-430). New York: Oxford University Press.

Provine, R. R., Spencer, R., & Mandell, D. (2007). Emotional expression online: Emoticons punctuate website text messages. Journal of Language and Social Psychology, 26(3), 299-307.

Rezabek, L. L., & Cochenour, J. J. (1998). Visual cues in computer-mediated communication: Supplementing text with emoticons. Journal of Visual Literacy, 18, 201-215.

Riordan, M. A., & Kreuz, R. J. (2010). Cues in computer-mediated communication: A corpus analysis. Computers in Human Behavior, 26, 1806-1817.

Riordan, M. A., Dale, R., Kreuz, R. J. & Olney, A. (2011). Evidence for alignment in a computer-mediated text-only environment. In L. Carlson, C. Hoelscher, & T. F. Shipley (Eds.), Proceedings of the 33rd Annual Meeting of the Cognitive Science Society, (pp. 2411-2416). Austin, TX: Cognitive Science Society.

Sauro, S. (2009). Computer-mediated corrective feedback and the development of L2 grammar. Language Learning & Technology, 13(1), 96-120.

Sauro, S. (2013). The cyber language exchange: Cross-national computer-mediated interaction. In K. McDonough & A. Mackey (Eds.), Second language interaction in diverse educational contexts (pp. 129-146). Amsterdam/Philadelphia: John Benjamins.

Schegloff, E. (1968). Sequencing in conversational openings. The American Anthropologist, 70(6), 1075-1095.

Tanaka, N. (1988). Politeness: Some problems for Japanese speakers of English. JALT Journal, 9, 81-102.

Tidwell, L. C., & Walther, J. B. (2002). Computer mediated communication effects on disclosure, impressions, and interpersonal evaluations: Getting to know one another a bit at a time. Human Communication Research, 28, 317-348.

Tudini, V. (2010) Online second language acquisition. London, New York: Continuum.

Vandergriff, I. (2013a). Emotive communication online: A contextual analysis of computer-mediated communication (CMC) cues. Journal of Pragmatics 51, 1-12.

Vandergriff, I. (2013b). "My major is English, belive it or not:)" – Participant orientations in nonnative/native text-chat. CALICO Journal 30(3), 393-409.

Vandergriff, I., & Fuchs, C. (2012). Humor support in synchronous computer-mediated classroom discussions. HUMOR. International Journal of Humor Research, 25(4), 437-458.

Vinagre, M. (2008). Politeness strategies in collaborative e-mail exchanges. Computers & Education, 50, 1022-1036.

Walther, J. B. (1992). Interpersonal effects in computer-mediated communication: A relational perspective. Communication Research, 19, 52-90.

Walther, J. B. (1995). Relational aspects of computer-mediated communication: Experimental observations over time. Organization Science, 6, 186-203.

Walther, J. B., & D'Addario, K. P. (2001). The impact of emoticons on message interpretation in computer-mediated communication. Social Science Computer Review, 19(3), 324-347.

Watts, R. J. (2003). Politeness. Cambridge, UK: Cambridge University Press.

Wolf, A. (2000). Emotional expression online: Gender differences in emoticon use. CyberPsychology & Behavior, 3(5), 827-833.

Yamada, M., & Akihori, K. (2007). Social presence in synchronous CMC-based language learning: How does it affect the productive performance and consciousness of learning objectives? Computer Assisted Language Learning, 20(1), 37-65.

Yus, F. R. (2002). Discourse and identity. In N. J. Smelser & P. B. Baltes (Eds.), International Encyclopedia of the Social and Behavioral Sciences (pp. 3728-3732). Oxford: Pergamon.

Yus, F. R. (2005). Attitudes and emotions through written text: The case of textual deformation in internet chat rooms. Pragmalingüística, 13, 147-173.

Yus, F. R. (2011). Cyberpragmatics. Internet-mediated communication in context. Amsterdam/Philadelphia: John Benjamins.

Biographical Note

Ilona Vandergriff [vdgriff@sfsu.edu] received her Ph.D. from the University of California at Berkeley. She is Professor of German at San Francisco State University. Her research interests include first and second language use, computer-mediated communication, and pragmatics.


Any party may pass on this Work by electronic means and make it available for download under the terms and conditions of the Digital Peer Publishing License. The text of the license may be accessed and retrieved at http://www.dipp.nrw.de/lizenzen/dppl/dppl/DPPL_v2_en_06-2004.html.