Printer Friendly


1. INTRODUCTION. Computer-mediated discourse (CMD) (Herring 2001) is the interaction between at least two individuals through transmitted messages via networked computers, and is a subfield of computer-mediated communication (CMC) (Herring 1999). Although studies focusing on CMD have become increasingly common during the past decade, such as those that have focused on the resources used within CMD that are unavailable in traditional spoken conversation (Dresner & Herring 2010), investigations of chats between native speakers (NSs) and nonnative speakers (NNSs) still lag behind (Belz & Kinginger 2003, Gonzales 2013, Gonzalez-Lloret 2005,2008, 2009, 2011, Osman & Herring 2007, Schonfeldt & Golato 2003), especially with respect to turn organization. Chats are written, synchronous computer-mediated conversations between at least two participants (Herring 1999) and are an increasingly popular communicative medium. In fact, research has called for an increased analysis of second language (L2) behavior exhibited in CMC (e.g. Skehan 2003).

Telecollaboration has been shown to be beneficial as a pedagogical practice in language learning (Belz & Kinginger 2003, Carroll 2005, Gonzalez-Lloret 2011, Kern 1995, Osman & Herring 2007, Sykes 2005, Thorne & Payne 2005). It includes electronic communication between NSs and learners, and offers the opportunity for authentic input and a multiple NS presence in a language classroom, which is especially beneficial in areas where there are few NSs of the target language (Osman & Herring 2007). Within such communicative experiences, text-based chat can be particularly beneficial, as it offers lower-proficiency learners the opportunity to read the informal contributions of the NS interlocutor, such that comprehension difficulties stemming from NS speech rate (Belz & Kinginger 2003) or a lack of dialect familiarity for an NS's more regionally-specific phonetic variants (Schmidt 2009) can be attenuated. Furthermore, undergraduate students report regularly using written, electronic chat with friends and family through programs such as Facebook Messenger[TM] and Google Chat[TM], so these media are familiar and authentic avenues for learners and NSs alike (Tannen & Trester 2013).

Given the lack of insight into what occurs linguistically in NS-NNS chats, the present study helps to fill a gap in the current literature and aims to analyze via conversation analytic methods (Sacks, Schegloff, & Jefferson 1974, Schegloff 2007) the turn structure of the NS and NNS participants within an electronic chat, with a focus on turn organization and sequential organization (i.e. opening, closings, and repairs). Since these details remain unanswered for learners communicating with NSs and NSs communicating with learners alike, both sets of parties will receive attention in the current paper. Furthermore, since written chat can become disjointed (Garcia & Jacobs 1999, Herring 1999) and since learners may have difficulty comprehending authentic NS speech (e.g. Belz & Kinginger 2003, Osman & Herring 2007, Sykes 2005), learner strategies for indicating a lack of comprehension during chat will also be considered.

The current paper is organized as follows: a review of previous literature (based on Conversation Analysis, CMC and CMD, general chat, and NS-NNS chat); the delineation of the goals of the present study; a description of the task, participants, and analysis; the presentation of the results; the discussion; and limitations of the present study and avenues for future research.

2. BACKGROUND. The review of previous literature begins with a summary of key concepts of Conversation Analysis (CA), followed by a review of studies divided among three different emphases: studies that have had a general focus on CMC or CMD, those that have analyzed synchronous chat, and, lastly, those that have specifically analyzed conversations between NSs and NNSs via synchronous chat.

2.1. CONVERSATION ANALYSIS. The current study utilizes a CA framework (Sacks et al. 1974; Schegloff 2007) in order to analyze in detail what occurs in each chat, including the surrounding context. The principal focus of CA is to demonstrate how the technical aspects of interactive talk are the structured and socially organized resources by which participants realize activities in interaction. Accordingly, a primary requisite of CA is to use natural data and to implement detailed transcriptions, including gestures, laughter, prosody, and overlaps (Schegloff 2007). Such designs also enable the study of in-depth interaction within its original context. CA lends itself well to analyses of CMC because it enables the researcher to discover patterns of how interaction takes place and how participants orient to sequences (e.g. Gonzales 2013; Gonzalez-Lloret 2011) and also because it is effective in revealing trouble sources and repairs (e.g. Gonzalez-Lloret 2005). Thus, an implication for L2 work that places learners and NSs together to communicate is to have the participants engage in naturalistic, spontaneous conversation, with minimal or no intervention from the researcher.

ADJACENCY PAIRS are the basic units of sequential constructions and indicate how the discourse is organized. They consist of the first pair part (FPP) and second pair part (SPP), although pre-expansions, insertion expansions, and post-expansions may occur (Schegloff 2007).

SEQUENCES are groups of actions that realize something via ordered turns. Some classic examples consisting of adjacency pairs include question-answer and offer-acceptance sequences (Schegloff 2007). These examples illustrate how the first part has great influence on the second, resulting in conditional relevance, such that the FPP limits what can relevantly follow it.

TURN CONSTRUCTIONAL UNITS (TCUS) are the basic units that form turns and project what information is to come (Schegloff 2007). In the utterance Hola, buenos dias, como estas? 'Hello, good morning, how are you?', three TCUs occur within this single turn: 1) hola 'hello' 2) buenos dias 'good morning,' and 3) como estas? 'how are you?' Note that each TCU is syntactically, pragmatically, and phonologically complete.

In CA, REPAIRS can be initiated by the original speaker (e.g. reformulations) or by the addressee (e.g. by asking for clarification) and normally consist of the trouble source (i.e. where the misunderstanding originates), the repair itself, and the recognition of the repair (Schegloff 2007). Repair work usually involves the use of an insertion sequence.

Now that some of the key concepts of CA have been presented, a review of previous studies that have analyzed CMC, CMD, and chat will be presented. As many of these studies have used the CA framework, these concepts will reappear both in the subsequently reviewed literature and in the current analysis.

2.2. CMC AND CMD. Computer-mediated communication (CMC) has received increasing attention during the past two decades. Maintaining coherence within the medium can be difficult, as there is a high degree of disrupted adjacency (with responses frequently separated from the turns to which the speaker responds), overlapping exchanges, and topic decay, such as when a speaker deserts a particular topic or introduces a new one while a prior topic remains unresolved (e.g. Herring 1999). Disrupted adjacency occurs because CMC modes are less-tightly held together than face-to-face conversation, detracting from overall coherence, but offering the compensatory benefit of a persistent on-screen textual record of the conversation (Herring 1999).

Computer-mediated discourse (CMD) is the interaction between at least two individuals through transmitted messages via networked computers (Herring 2001). CMD is a subfield of CMC and has a focus on language use via electronic communication, utilizing discourse analysis as its analytic framework. The majority of CMC is text-based, as messages are typed on a computer and later read by the interlocutor, and thus, language is presented visually. Computer-mediated language is sensitive to multiple technical and situational factors and is therefore quite complex and variable according to the exact medium used (e.g. email, chat, etc.). CMD is notably faster than traditional written exchanges, yet still slower than spoken interaction. It also allows multiple participants to communicate simultaneously. In CMD, information is only available visually (via typed text), so users are required to compensate textually for missing auditory and gestural cues. Whereas some authors emphasize the commonality between CMD and oral speech (Chun 1994), others differentiate CMD more strongly as a separate genre (Johanyak 1997, Yates 1996). Regardless of features that it might share with written and oral communication, CMC is its own, distinct form of communication (for more on the uniqueness of CMC, see Thorne & Payne 2005).

The systems of turn-taking in CMC and oral conversation were compared in Garcia and Jacobs (1999). The investigators recorded four quasi-synchronous CMC conversations between students in a college classroom and found that the turn-taking system differs from that of oral conversation in that there are multiple, concurrent speaker selection options (instead of the standard sequential turn-taking rules for oral conversation). Since more than one participant may be constructing a message at a time, participants cannot control the placement of their message relative to others and messages may appear disjointed and unclear for the interlocutors. Despite this shortcoming, CMC offers the advantage of creating an ongoing document to which participants can refer during interaction.

The CMC research of Dresner and Herring (2010) demonstrated that the traditional characterization of emoticons (i.e. emojis) (1) as iconic indicators of emotion conveyed via a written channel fails to account for some of their most important uses. Via speech act theory, the investigators provided an additional account for emoticons, which includes their function as indicators of illocutionary force. The authors also noted commonalities between emoticons and utterance-final punctuation marks, as both can be used to indicate pragmatic illocutionary force and are parts of text. They recommend that future research investigate whether emoticons are becoming increasingly conventionalized as textual markers (for more on the effects of new media on language use, see Tannen & Trester 2013).

2.3. SYNCHRONOUS CHAT. Electronic chat has been analyzed in numerous recent studies. Noblia (2004) focused on the fundamental role of participants' face (Goffman 1981), since it is the element that participants jeopardize or negotiate in this context. Within chats, language is assigned an exclusive role in the definition and negotiation of face, which is recoverable only via mutual interaction, as each speaker expresses herself and conducts discursive interaction with her interlocutor. As the interaction is very personal, the risk of threatening the interlocutor's face is quite high. Specifically, Noblia highlighted the use of irony both to preserve and threaten the interlocutor's face, to negotiate interactional guidelines and resolve conflicts, and to contextualize and demonstrate group membership.

In a follow-up study of synchronous chat, Noblia (2009) utilized CA for its interactional perspective. She noted that the plane of interpersonal relationships changes, which, in turn, enables and restricts particular ways for participants to mark and negotiate their positions, resolve misunderstandings, and define interactive roles. Thus, while chatting, participants orient their contributions, defining and negotiating their roles and identities.

The role of repair in chats was the focus of Schonfeldt and Golato (2003), who used CA to analyze the interactional organization of a German chat in comparison to traditional oral conversation, with a focus on the positions from which repairs can be initiated and the types of trouble sources that the repairs address. The authors also investigated when, why, and how interlocutors adjust repair practices based on traditional oral conversation to fit the chat medium. Since chat is more restricted than oral conversation, participants rely solely on written messages and sequential ordering to make repairs, as they lack access to prosody and gestures.

Golato and Taleghani-Nikazm (2006) used CA methods to analyze the notions of face and politeness in chats (e.g. Goffman 1981, Haverkate 1994), tying face and social solidarity to the CA concept of preference organization. They also analyzed the technological constraints of chat that affect the way in which participants provide PREFERRED and DISPREFERRED RESPONSES. Specifically, interlocutors followed preferred/affiliative conversational moves in a manner similar to traditional conversation, creating solidarity while doing so. Participants did make modifications which are not possible in spoken interaction, however, such as the use of emoticons and written pauses (via ellipses) when providing dispreferred responses, which roughly corresponded to facial expressions and interactional delays, respectively, in spoken interaction, and performed the same function of supporting solidarity and providing orientation to the interlocutor's face (for more on politeness in CMC, see Locher 2010).

Finally, the structure of OPENINGS in chats was analyzed in Rintel et al. (2001). The authors reported that turn coordination during openings is quite ambiguous, can potentially be disruptive of the developing relationship, and can lead to a substantial emphasis on interactional strategies for the ordering of opening phases. Results also revealed the repetition (or slight modification) of the interlocutor's opening by speakers within chat.

2.4. NS-NNS CHAT. Belz and Kinginger (2003) used chats as a medium to study L2 acquisition in their influential early work. Specifically, the authors investigated use of formal and informal pronouns of address by English-speaking learners of German. Results demonstrated that telecollaboration between learners and NSs in Germany facilitated the development of appropriate use of forms of pronominal address. In short, the social meaning contained in the pronominal forms was made more apparent following the use of authentic CMC with native speakers of German.

The beginnings of TCUs within turns in conversations between NSs and NNSs were the emphasis of Carroll (2005), who reinterpreted NNS 'false starts' as skilled interactional achievements that are the result of monitoring, which are similar to what NSs do, as opposed to breakdowns or failures. Furthermore, such restarts contribute to the achievement of speech without gap or overlap, a key component of speech between NSs. Additionally, NNS intra-turn pauses were shown to function in an interactionally relevant way, as opposed to simply being evidence of psycholinguistic overload (for more on NNS conversations, see Gardner & Wagner 2005).

The potential for synchronous chat to be used for pedagogical purposes was considered by Osman and Herring (2007) within the context of a distance learning program between two different university groups (four learners in Azerbaijan and two facilitators in the United States). The study focused on the use of chat for increased interaction between speaker groups and the facilitation of relationships between the groups' members. Findings indicated that the quality of the interaction was limited by the task, linguistic shortcomings, and different cultural expectations regarding instruction. Nevertheless, conceptual negotiation did increase over time. Overall, the researchers provided further evidence that synchronous chat can facilitate learning, but warned that the costs (e.g. large amount of instructor effort and shared knowledge of subject matter and culture) may outweigh the benefits when linguistic and cultural barriers are high.

Learning across modalities was the focus of Sykes (2005), who found that a group of learners who had performed written chats outperformed oral chat and face-to-face groups in terms of utilizing less-direct refusals and providing more supporting moves. The author also noted the use of laughter in the written chat group. Kern (1995) showed other benefits of chat for learners, including fostering a greater quantity of linguistic production and more sophisticated morphosyntax. Ortega (1997) has noted, however, that early studies of NS-NNS chat were often highly variable in terms of group size and communicative task, meaning that detailed analyses of what occurs in NS-NNS chat are still needed. In fact, Smith (2003) posited that there are additional stages in the negotiated interaction that occurs in chat (i.e. confirmation and reconfirmation), which are not present in face-to-face negotiation interaction (e.g. Varonis and Gass's original model of negotiated interaction, 1985).

Moreover, the current paper follows Gonzalez-Lloret's (2011) suggested technique of not imposing full CA turn-taking norms on participants and, instead, simply noting what occurs in chat discourse and analyzing it accordingly. In this manner, as has been reported, CA can be an appropriate tool for analyzing electronic chats and for discovering how interaction is carried out and how participants orient to sequences (Gonzalez-Lloret 2011). (2)

3. PRESENT STUDY. Whereas studies focusing on CMD have become increasingly common, including investigations of the resources used that are unavailable in traditional spoken conversation (e.g. Dresner & Herring 2010), studies of chats between native speakers and nonnative speakers are still infrequent (e.g. Osman & Herring 2007, Schonfeldt & Golato 2003). Given the lack of insight into what occurs in NS-NNS chats and the increasingly popular nature of the medium, the present study fills a key gap in the literature.

The current study aims to analyze via conversation analytic methods (Sacks et al. 1974, Schegloff 2007) the turn structure of the NS and NNS participants within an electronic chat, with a focus on turn organization and sequential organization (e.g. openings, closings, and repairs). Additionally, given the lack of oral prosody available in electronic chat, the study aims to document the features that emerge in the discourse of the participants and the corresponding functions fulfilled, and how learners resolve any potential lack of comprehension. (3)


4.1. PARTICIPANTS. The data come from four chats (from Google Chat[TM]) between NSs and NNSs. The four NNS participants (mean age: 20.75 years) were intermediate learners, all in the process of completing their second or third Spanish language course at a large Midwestern university, and each chatted with a different NS (mean age: 30.75 years). None of the learners reported having spent two or more weeks abroad or even having conducted extended conversation with NSs outside of the classroom setting prior to participation in the study. The four NSs were originally from Mexico, Argentina, and Puerto Rico (n=2). Each NS was a graduate student with experience teaching Spanish to speakers of English. Consequently, this group of speakers was representative of the make-up of the institution's Spanish Department, in which learners receive instruction from speakers from a variety of national origins. All participants completed two tasks: a language background questionnaire and an electronic chat. The language background questionnaire is the source of the demographic information included in the current section; it enabled the researcher to verify that the learners had not spent time abroad. Each chat lasted approximately thirty minutes (mean length: 33.75 minutes) and participants were told that they could discuss whatever topics they found most interesting, although they were given a list of additional topics in order to stimulate conversational ideas (see Appendix), which included their studies/careers, hobbies, activities of the past two weeks, and additional themes of interest. Interactions took place in the researcher's office for the learner participants (while the investigator was present in another section of the office, in case technical difficulties arose), while the NSs participated in their location of choice, most commonly from their homes.

4.2. DATA ANALYSIS. A CA framework is used to analyze the four NS-NNS chats in order to study in a detailed manner what occurs in each chat and to utilize surrounding context (Sacks et al. 1974, Schegloff 2007). The behavior of both the NSs and NNSs is of interest in the current data. Comparisons are made between the NS and NNS groups in terms of the turn structures utilized, repair sequences (including trouble sources), and manipulation of turns. In addition, the elements used in the absence of prosody and gestures are discussed. Furthermore, since written chat can become disjointed (Garcia & Jacobs 1999, Herring 2009) and learner lack of proficiency may impede comprehension (e.g. Belz & Kinginger 2003, Osman & Herring 2007, Sykes 2005), strategies used by learners to indicate a lack of comprehension are also considered.


5.1. TURN ORGANIZATION AND ITS EFFECT ON SEQUENCING. Our analysis of the NS-NNS conversations begins with turn organization and its effect on sequencing. Notably, there were two general patterns of turn construction used by participants. Firstly, and most typically, participants created one 'entry' turns (Noblia 2004). In such cases, A-B adjacency pairs were frequently the sequential outcome, as seen in the conversation between Isabel (NS) and Danny (NNS) in (1), with the English translation directly below. In order to provide authentic data, all excerpts are unedited, and thus the grammar, vocabulary, spelling, accentuation, and capitalization (or lack thereof) reflect the original content. All names used in the current paper are pseudonyms, with Spanish pseudonyms assigned to the NS participants and non-Spanish names to the learners.


1 Isabel: mucho gusto :)

2 Danny: encantante

3 Isabel: tuviste clase hoy?

4 Danny: si tuve un clase de 'documentary production'

5 Isabel: si? es una clase de cine?

6 Danny: el profesor me ensena sobre peliculas

7 Isabel: que interesante!

8 Danny: cual es tu carrera

9 Isabel: yo estudio espanol

1 Isabel: nice to meet you :)

2 Danny: nice to meet you

3 Isabel: did you have class today?

4 Danny: yes I had 'documentary production' class

5 Isabel: yeah? Is it a film class?

6 Danny: the professor teaches me about movies

7 Isabel: how interesting!

8 Danny: what's your major

9 Isabel: I study Spanish

Repeated use of A-B adjacency pairs in the four sequences in (1) is rather straightforward. For example, the conversation begins with an A-B greeting sequence (greeting-greeting) (lines 1-2). This is followed by an A-B question-answer sequence (lines 3-4). Next, there is a slight deviation, as we have a question-answer-comment sequence, with a first pair part (question), a second pair part (answer), and a third pair part (comment) in lines 5-7, respectively. Another A-B question-answer sequence can be seen in the fourth and final sequence in lines 8-9.

The other common pattern of turn construction was the use of multiple entries within each turn, which presents a slightly more complex take on the A-B structure, as can be seen in the dialogue between Susana (NS) and Stephanie (NNS) in (2).


1 Susana: mi hermano siempre tenia que ir al dentista cuando era chico, yo no... iba, pero no

2 lo necesitaba tanto como el

3 el ahora los odia, claro...

4 Stephanie: los odia?

5 Susana: bueno... odiar es muy fuerte... jaja... podemos decir que les tiene miedo

6 o que tiene malos recuerdos

7 visitaba mucho a su dentista...

8 Stephanie: donde esta tu hermano?

9 Susana: mi hermano vive en Miami, FL

10 trabaja ahi, en una agencia de publicidad

11 Stephanie: bien bien

12 tu hermano trabaja con actors?

13 Susana: jaja no... el hace las publicidades, pero no actua en ellas

14 en realidad, se que su trabajo es mas coplicado

15 que eso, pero nunca entendi exactamente que es lo que hace

16 Stephanie: tu hermano te gusta agencia de publicidad?

17 se*

18 Susana: si, por suerte, le gusta mucho su trabajo

19 Stephanie: te gusta trabajar por Springfield University? (4)

20 te gusta tus estudiantes?

21 Susana: me gusta trabajar aca, si. Y mis estudiantes... me caen bien la mayor parte del

22 tiempo... jajajaja

23 no, en serio, me caen bien

1 Susana: my brother always had to go to the dentist when he was a boy. I didn't... but I

2 didn't need it as much as he did

3 now he hates them, obviously...

4 Stephanie: he hates them?

5 Susana: well... hate is very strong... haha... we might say that he's afraid of them

6 or that he has bad memories

7 he visited his dentist a lot...

8 Stephanie: where is your brother?

9 Susana: my brother lives in Miami, FL

10 he works there, in an advertising agency

11 Stephanie: good good

12 your brother works with actors?

13 Susana: haha no... he does the ads, but he doesn't act in them

14 actually, I know that his work is more complicated,

15 but I never understood exactly what it is that he does

16 Stephanie: your brother likes your ad agency?

17 se* (means to correct to 'his')

18 Susana: yes, luckily, he likes his job a lot

19 Stephanie: do you like working for Springfield University?

20 do you like your students?

21 Susana: 1 like working here, yes. And my

22 students... I like them most of the time...hahaha

23 no, seriously, I like them

The use of multiple entries per turn is prevalent in (2), as can be seen from the very beginning. Susana's first turn is a comment that consists of two entries (one in lines 1-2 and the second in line 3), which is followed by Stephanie's clarification question in a one-entry turn (line 4). Susana responds with a three-entry turn (one entry each in lines 5-7), in which she clarifies her original comment, in response to the clarification question. We then have an A-B question-answer sequence in which the question (A) occurs in a one-entry turn (line 8), followed by the answer (B) in a two-entry turn (lines 9-10). The next sequence is an A-B question-answer sequence, started by a two-entry turn that contains a response token entry (line 11) and then a question entry (A) (line 12), which is followed by the answer (B), developed over two entries (one in line 13 and the other in lines 14-15). This is followed by a question-answer sequence, which begins with a question (A) that occurs in a two-entry turn, as the question is asked in line 16 and then repaired in an entry in line 17. Repair will be discussed in greater depth in its own section, but here we can note a self-initiated repair that attempts to correct a perceived grammatical error. In line 18, we see the answer (B) in a one-entry turn. The final sequence in this excerpt is another question-answer, although it also contains an insertion. Lines 19-20 present a two-entry turn that contains two consecutive questions, each realized via a separate entry (the second question can be viewed as an insertion, following the first question). This is followed by an answer that is also developed via a two-entry turn (lines 21-23, with the first entry in lines 21-22 and the second in line 23). Here, the second entry (line 23) in some ways is a comment on and response to the first turn (lines 21-22), as the speaker reiterates that she was joking in the first turn and self-corrects the original response. There is also laughter during this turn, which will be further discussed in the section devoted to that topic.

Another notable aspect of sequential organization that occurred in the data was the use of latching (i.e. the entire copying) of NS structures by NNSs, in which the NNS organized her turn by completely copying that of the NS, as illustrated in (3).


1 Susana: claro, y ya sabes que tipo de doctor quieres ser?

2 pedriatra? cardiologo? o...?

3 Stephanie: dentist?

4 Susana: si, dentista

5 Stephanie: :)

6 Susana: y por que te gustaria ser dentista?

7 Stephanie: me gustaria ser dentista porque un dentista trabajaria con chicos y chicas

1 Susana: of course, and do you already know what type of doctor you want to be?

2 pedriatrician? cardiologist? or...?

3 Stephanie: dentist?

4 Susana: yes, dentist

5 Stephanie: :)

6 Susana: and why would you like to be a dentist?

7 Stephanie: I would like to be a dentist because I would work with boys and girls

Here, in line 7 (me gustaria ser dentista porque i would like to be a dentist because'), the NNS (Stephanie) completely copies the structure used by the NS (Susana) in line 6 (por que te gustaria ser dentista? 'why would you like to be a dentist?'). Moreover, this latching is notable in that it is the NNS's only use of a conditional form, providing further evidence that she likely replicated what the NS speaker had said, instead of generating the complex word-form independently. (5)

Having reported the two major patterns of turn organization (i.e. single-entry turns leading to A-B adjacency pairs and multiple-entry turns with more complex sequential organization), we turn our attention to the organization of openings, closings, and repairs.

5.2. OPENINGS. Openings were quite comparable in each of the four chat conversations. General A-B greeting-greeting sequences held. In fact, in all four conversations, repetition of a greeting/comment from the interlocutor occurred, usually with each participant creating a one-entry turn, as seen in (4-7).


1 Guillermo: hola

2 Jill: hola come te llamas?

1 Guillermo: hello

2 Jill: hello what's your name?


1 Isabel: Hola

2 Danny: hola

1 Isabel: Hello

2 Danny: hello


1 Stephanie: hola

2 Susana: hola, que tal?

3 Stephanie: bien

4 y tu?

5 Susana: bien

1 Stephanie: hello

2 Susana: hello, how are you?

3 Stephanie: well

4 and you?

5 Susana: well


1 Emilia: Hola!!! como estas!

2 ??

3 Andrew: Hola. Estoy muy bien, y tu?

4 Emilia: muy bien, gracias... tengo muchas cosas que hacer pero estoy bien

1 Emilia: Hello!!! how are you!

2 ??

3 Andrew: Hello. I'm very well, and you?

4 Emilia: very well, thanks... I have many things to do but I'm good

Thus, in (4-6) we see greeting-greeting sequences with the repetition of hola 'hello' by the interlocutor (B) (line 2, in each case) in a one-entry turn, directly following its prior use (A) (line 1, in each case) in a one-entry turn by the other speaker. In 6, we also see the repetition of bien 'well' (line 5) after its prior use (line 3). In excerpt 7, hola is again repeated by the learner interlocutor (line 3), but in this case it follows a two-entry turn of the NS, who entered question marks in an additional entry, likely to correct the original use of an exclamation point used with a question. We then see the repetition of muy bien 'very well' (line 4), following its original use by the interlocutor (line 3). Interestingly, after repeating muy bien (line 4), the speaker elaborates, providing more information and then only using bien in her description, providing possible further evidence that she may have been simply repeating what her interlocutor had said in her original use of muy bien.

Having considered the general pattern of greeting sequences of one entry repetitions of greetings, we move to closings in the NS-NNS conversations.

5.3. CLOSINGS. Similar to openings, repetition occurred in closings, in the form of providing thanks or farewell wishes from the interlocutor, as seen in (8-10).


1 Stephanie: pero bien habla con tu

2 muchos gracias

3 :)

4 Susana: lo mismo digo

5 fue un placer

6 mucha suerte con tus cursos en la uni!

7 adios

8 Stephanie: adios

1 Stephanie: but good speaking with you

2 thanks a lot

3 :)

4 Susana: I say the same

5 it was a pleasure

6 lots of luck with your courses at the university !

7 bye

8 Stephanie: bye


1 Andrew: Yo neceisto salir. Gracias para me hablas.

2 Emilia: Gracias a ti Adios!!! :)

3 Andrew: Adios!

1 Andrew: I need to leave. Thanks for talking to me.

2 Emilia: Thank you to you Bye!!! :)

3 Andrew: Bye!


1 Isabel: si, me dio gusto hablar contigo

2 espero que todo te vaya super bien en tu carrera

3 Danny: gracias para tu tiempo

4 adios

5 Isabel: adios :)

1 Isabel: yes, it was a pleasure speaking with you

2 I hope that everything goes really well for you in your major

3 Danny: thanks for your time

4 bye

5 Isabel: bye :)

Firstly, in these A-B closing sequences, we see learner repetition of adios 'good-bye' (B) in a one-entry turn directly after its use by the NS interlocutor in the previous entry (A), a common theme of each of the three excerpts, occurring in lines 7-8 of (8), lines 2-3 of (9), and lines 4-5 of (10). Further, thanking of the interlocutor occurred during the closing sequence in each excerpt, as well. Interestingly, in excerpts 8 and 10 we see use of the word gracias 'thank you' by the learners (line 2 in 8 and line 3 in 10), accompanied by more syntactically complex forms of thanking by the NSs, including fue un placer 'it was a pleasure' and me dio gusto hablar contigo 'it was a pleasure speaking with you', respectively.

Now that the common practice of one-entry repetitions of closings has been considered, the organization of repairs will be analyzed in the following section.

5.4. REPAIRS AND UNKNOWN LEXICAL ITEMS. Repairs were also quite common in the data. In (11), we see an example of other-initiated repair, with the trouble source emerging in line 1, as the learner interlocutor does not understand the meaning of the question and makes this explicit in line 2. In line 3, the NS attempts repair by reformulating the original question in an insertion sequence. However, we see that the learner stops attempting to understand the question and changes the topic in line 4, and the NS accepts this change by asking a follow-up question related to the new topic in line 5. Thus, the repair made in line 3 was not particularly successful, as the learner was unable to grasp the meaning of the question, even following the reformulation. As the repair occurs, one can also note the phenomenon of topic decay, or the shifting of topics prior to resolution of a prior topic (Herring 1999).


1 Susana: hace mucho vives en springfield?

2 Stephanie: no intendiendo?

3 Susana: desde cuando vives en springfield?

4 Stephanie: yo come en mucho restaurante en springfield

5 Susana: tienes un favorito?

1 Susana: have you lived in Springfield for a long time?

2 Stephanie: I don't understand?

3 Susana: since when do you live in springfield?

4 Stephanie: I eat in many restaurants in springfield

5 Susana: do you have a favorite?

Another instance of repair occurs in (12), as the learner inaccurately phrases her question in line 3, creating a trouble source in that line and thereby asking an unintended question ('Where are you?' rather than 'Where are you from?'). The NS responds to the question as written in line 4, leading the learner to realize that she had improperly formed the question, and thus she reformulates in line 5. This can be viewed as self-initiated repair to the extent that the NS had not realized that the learner had asked an unintended question (e.g. 'Where are you?') in line 3. There is also self-repair in the original two-entry turn by the NS in lines 1-2, as he makes a typographical error in line 1 and then repairs it in line 2, highlighting the repair with an asterisk.


1 Guillermo: mi nombre es guillermo, como tu te Hams?

2 llamas*

3 Jill: soy Jill, donde esta?

4 Guillermo: estoy en la universidad, todavia :S

5 Jill: o que es tu origen?

1 Guillermo: my name is guillermo, what's your name?

2 name*

3 Jill: I'm Jill, where are you?

4 Guillermo: I'm in the university, still :S (emoji)

5 Jill: or what is your origin?

An additional example of repair can be found in (13). Here, the trouble source occurs in line 4, as the learner asks a question using singular instead of plural verbal morphology ('How old are you?' instead of 'How old are they?'). The NS once again interprets the question as written, and thus answers with his age (line 5). The learner then attempts repair in line 7, making explicit the grammatical subject of her original question. This is self-initiated repair, once again, because the speaker repairs her own question. The repair is successful, as the NS responds to the intended question in line 9.


1 Jill: tu hermanos tiene nins?

2 ninos*

3 Guillermo: si, mi hermano de nueva york tiene dos nenas

4 Jill: cuanto anos tiene?

5 Guillermo: yo? yo tengo 28

6 estoy viejo jajajajaja

7 Jill: no, los ninos..

8 jajajaja

9 Guillermo: oh jajajajaj tienen 14 y 12

1 Jill: your brothers has kids?

2 kids*

3 Guillermo: yes, my brother from new york has two girls

4 Jill: how old are you?

5 Guillermo: me? I'm 28

6 I'm old hahahahaha

7 Jill: no, the kids

8 hahahaha

9 Guillermo: oh hahahahah they're 14 and 12

Now that some general examples of repair have been analyzed, we move to a specific type of repair involving an unknown lexical item in (14). Here, in line 5 the learner specifically asks for the meaning of a lexical item, referring back to the trouble source, the use of the lexical item in line 4. The repair is successful, as the NS explains the word's meaning in a two-entry turn in lines 6-7. The learner notes that he now understands the meaning in his subsequent contribution (line 8).


1 Danny: estudio la red y production de videos

2 Isabel: te gusta?

3 Danny: me gusta mi carrera mucho y me encanta mi computadora tambien

4 Isabel: jaja, que bien! y tienes que hacer tus propios videos?

5 Danny: que significa propios?

6 Isabel: mmm, quiero decir videos que realizas tu mismo

7 o que haces tu mismo

8 Danny: si, entiendo, los videos son interestante para me porque son el primero modo de

9 communicacion en la red

1 Danny: I study networking and production of videos

2 Isabel: do you like it?

3 Danny: I like my major a lot and I love my computer too

4 Isabel: haha, how good! and do you have to make your own videos?

5 Danny: what does 'propios' mean?

6 Isabel: mmm, I mean videos that you yourself make

7 or that you make yourself

8 Danny: yes, I understand, the videos are interesting because they are the first mode of

9 communication on the internet

The fact that misunderstandings in the data were often lexical is similar to the findings in the NS-NNS service-encounter data of Gonzalez-Lloret (2005). Following this overview of lexical and grammatical repairs, we now turn to strategies used by the participants to compensate for the lack of oral prosody. The strategies included laughter, emojis, all-capitalization, and multiple punctuation marks.

5.5. LAUGHTER. In the absence of (oral) prosody, a common strategy was the use of laughter, in order to mitigate a negative response or to show solidarity (as also seen in Glenn 1995). Laughter was used in three of the four chats in the present study. In (15), we see laughter used by both participants. Firstly, we see the NS using laughter in line 4 to mitigate her somewhat negative response about only liking her students most of the time, which could be offensive to the NNS, a student. Thus, the laughter serves to show that the NS is not being completely serious and that the learner should not be offended. She adds further mitigation in line 5. We can see that the learner is not offended, as she offers mutual laughter in line 6.


1 Stephanie: te gusta trabajar por Springfield University?

2 te gusta tus estudiantes?

3 Susana: me gusta trabajar aca, si. Y mis estudiantes... me caen bien la mayor parte del

4 tiempo... jajajaja

5 no, en serio, me caen bien

6 Stephanie: jajaja bien

1 Stephanie: do you like working for Springfield University?

2 do you like your students?

3 Susana: I like working here, yes. And my students... I like them most of the time...

4 hahahaha

5 no, seriously, I like them

6 Stephanie: hahaha good

Another use of laughter can be seen in (16), as the NS uses laughter to mitigate her negative, and possibly face-threatening, response (line 5). Rather than simply stating that the answer to the NNS's question is 'no', she softens the negative answer, and face threat, by first providing laughter.


1 Susana: mi hermano vive en Miami, FL

2 trabaja ahi, en una agencia de publicidad

3 Stephanie: bien bien

4 tu hermano trabaja con actors?

5 Susana: jaja no... el hace las publicidades, pero no actua en ellas

1 Susana: my brother lives in Miami, FL

2 he works there, in a public relations agency

3 Stephanie: good good

4 your brother works with actors?

5 Susana: haha no... he makes the ads, but he doesn't act in them

Another instance of laughter by both participants is found in (17), which is interesting because laughter fulfills distinct functions in different contexts. Firstly, the NS says something rather negative about himself in line 6 and then somewhat mitigates it with laughter, which also functions as an attempt to create solidarity with the interlocutor. Interestingly, in the following turn (lines 7-8), the learner indicates in the first entry that the NS had answered the wrong question, but in the second entry provides laughter, showing solidarity with the NS and also mitigating her negative, and potentially face-threatening, response in the prior entry. The NS then returns the laughter in line 9, which likely softens the misunderstanding from the previous question.


1 Jill: tu hermanos tiene nins?

2 ninos*

3 Guillermo: si, mi hermano de nueva york tiene dos nenas

4 Jill: cuanto anos tiene?

5 Guillermo: yo? yo tengo 28

6 estoy viejo jajajajaja

7 Jill: no, los ninos..

8 jajajaja

9 Guillermo: oh jajajajaj tienen 14 y 12

1 Jill: your brothers has kids?

2 kids*

3 Guillermo: yes, my brother from new york has two girls

4 Jill: how old are you?

5 Guillermo: me? I'm 28

6 I'm old hahahahaha

7 Jill: no, the kids

8 hahahaha

9 Guillermo: oh hahahahah they're 14 and 12

Now that the use of laughter for mitigation and the expression of solidarity has been analyzed, another strategy used by participants in the absence of prosody will be discussed--the utilization of emojis.

5.6. EMOJIS. Similar to laughter, emojis (i.e. emoticons) were used to mitigate negative responses or to show solidarity. In fact, emojis were used in all four chats. In 18, the NS realizes that the conversation has begun to shift rather drastically from information about the participants' town of residence to discussion about a drug problem in another city, so she redirects the conversation in line 4, by mentioning that it would be better to talk about their town instead. Since this is a somewhat face-threatening statement, she mitigates it with an emoji, in this case a smiley face.


1 Andrew: Ellos dicen que las personas son locas y la ciudad les danan una nariz y la ciudad

2 tiene una problema druga

3 Emilia: si hay muchas metanfetaminas... y la ciudad tiene un olor feo... tambien el sonido

4 de los trenes es horrible! todo el dia y la noche... pero mejor hay que hablar de Springfield,

5 :) te gusta?

1 Andrew: They say that the people are crazy and the city hurts their noses and the city has

2 a drug problem

3 Emilia: yes there are a lot of methanphetamines... and the city has a bad odor... also the sound of

4 the trains is horrible! all day and night... but better that we talk about Springfield, :) do you

5 like it?

Another example of emoji use can be seen in (19). In this example, the NS interestingly uses emojis in three consecutive entries (lines 5-7) in response to the learner's grammatical question. Since the NS is correcting the learner's grammar, she uses an emoji in line 5 to soften this correction. She provides further information about what the form that he had used meant in line 6 and includes an emoji to again prevent herself from seeming too rigid in her correction and to minimize any face-threat. Then, we see a slightly different emoji function in line 7, as she explains to the learner that she will be able to understand him and that he does not need to worry about his grammar. She finishes that entry with a direct, linguistically unmitigated command (relajate 'relax'), so the emoji serves to soften the force of the command (and perhaps the prior command no te preocupes 'don't worry').


1 Andrew: no le conozco a Smith. Yo se que antropologia tiene una clases pocas en

2 linguistica. Pero, Semestre futuro, yo tome la clase de lingiistica antropologia.

3 Emilia: ok :)... y te gusta la uni?

4 Andrew: un momento, es futuro tomare o tome?

5 Emilia: tomare :)

6 tome es preterito ;)

7 pero no te preocupes yo entiendo todo, relajate :)

8 Andrew: okay, gracias.

1 Andrew: I don't know Smith. 1 know that anthropology has a few classes in linguistics.

2 But, future semester, I took the linguistic anthropology class.

3 Emilia: ok :)... and do you like the university?

4 Andrew: one moment. Is the future 'tomare' or 'tome'?

5 Emilia: 'tomare' :)

6 'tome' is preterit ;)

7 but don't worry I understand everything, relax :)

8 Andrew: okay, thanks.

This section has illustrated examples of emoji use for the purposes of solidarity or mitigation; we now move to instances of all-capitalization in the data.

5.7. ALL-CAPITALIZATION. In addition to laughter and emoji use, the utilization of all-capitalization fulfilled some of the functions associated with prosody. Nevertheless, rather than to mitigate, as was typically seen with laughter and emojis, all-capitalization served to intensify the propositional content, while also fulfilling solidarity functions, similar to the prior two features. All-capitalization was used in one of the four chats. In (20), the NS reiterates the learner's line 1 comment, repeating in line 2 what the former had said, but adding an adverb to intensify, and using all-capitalization in her realization of the adverb to further the intensification. This also fulfills a solidarity function, as she shows agreement with her interlocutor.


1 Danny: bien pero estoy en Adams y es un poco frio

2 Isabel: es MUY frio!

1 Danny: well but I'm in Adams and it's a little cold

2 Isabel: it's VERY cold!

In (21), the same speakers who had discussed the cold in the previous excerpt are discussing New York. The NS brings up the cold again in line 6 and uses all-capitalization to indicate the degree of difference (in terms of temperature) between the participants' town and New York. We see that the strength of her turn (via all-capitalization) is perfectly acceptable, as the learner agrees with her in the following turn (line 7) and the NS, in turn, provides a third-pair part further agreeing with him (line 8).


1 Danny: si llegue a NY en sept, y tengo un trabajo (internship) jan-jun ' 11

2 Isabel: que bien! entonces, no vas a estar aqui el proximo semestre

3 yo voy a ir a NY en diciembre de este ano :)

4 Danny: si voy a estudiar el proximo ano

5 ah:)

6 Isabel: que bien! pero en NY es MUCHO mas frio

7 Danny: si el oceano tiene cosas mal y bueno

8 Isabel: exacto!

1 Danny: yes I arrived (6) at NY in sept, and I have a job (internship) jan-jun ' 11

2 Isabel: how great! so, you're not going to be here next semester

3 I'm going to go to NY in december of this year :)

4 Danny: yes I'm going to study next year

5 ah:)

6 Isabel: how good! but in NY it's MUCH colder

7 Danny: yes the ocean has good and bad things

8 Isabel: exactly!

Now that examples of the intensifying function of all-capitalization have been considered, the final prosodie feature will be discussed--multiple punctuation marks.

5.8. MULTIPLE PUNCTUATION MARKS. The final prosodie feature, the use of multiple punctuation marks, served to intensify the content of a contribution, similar to all-capitalization. This phenomenon occurred in two of the four chats. In (22), once again, the temperature is discussed and the NS has already stated his dislike for the cold (line 3), with the learner showing agreement (line 4). In line 5, the NS repeats his previous sentiment in an even stronger form, using a verb with strong semantic content (odiar 'to hate') and further intensifying his comment with the use of multiple exclamation points. Since he already knows that the learner shares his assessment, given her previous agreement in the preceding line, he can comfortably make this strong claim without the fear of offending her or threatening her face. Furthermore, since he knows that she agrees with his idea, his intensification via multiple punctuation marks serves to create further solidarity between them. In fact, following this comment, she continues to agree with him, responding with further solidarity (line 6).


1 Jill: muy habladora y extrovertido y apasianado

2 te gusta springfield?

3 Guillermo: si me gusta mucho springfield menos el frio

4 Jill: yo tambien

5 Guillermo: odio el frio!!!!!

6 Jill: no me gusta el verano

1 Jill: very talkative (FEM) and extroverted (MASC) and passionate (MASC)

2 do you like springfield?

3 Guillermo: yes I like springfield a lot minus the cold

4 Jill: me too

5 Guillermo: I hate the cold! ! ! ! !

6 Jill: I don't like it..I like summer

As has been reported in the preceding sections, laughter and emojis were frequently used to mitigate statements, whereas all-capitalization and multiple punctuation marks were used to intensify. However, regardless of mitigation or intensification, all four of these elements served to create solidarity between the participants, as when the former two mitigated negative comments and the latter two intensified positive statements.

6. DISCUSSION. The current discussion section highlights the present study's most important findings, interpreting them in line with previous literature on CMC and CMD. Turn structure will first be discussed, followed by openings, closings, and repairs, and then by an analysis of the discourse features used in the chats. The use of the small corpus created for the current study offers many of the benefits posited by Vaughan and Clancy (2013), including familiarity of the investigator-compiler with the data, knowledge of the backgrounds of the participants, and awareness of the setting and purpose of conversation. The small data set also enabled the inclusion of representative excerpts from multiple conversations, such that contextual information that surrounded the selected discourse could be included.

6.1. TURN STRUCTURE. AS noted in the results section, two major patterns of turn structure emerged in the data: 1) the use of A-B adjacency pairs (usually question-answer) that consisted of single-entry turns and 2) the use of multiple-entry turns. The frequent utilization of the latter upholds previous findings of the commonality of multiple-entry turns (Noblia 2009). Whereas at times entries and turns were one and the same (in those cases of single-entry turns, as described in the first general pattern of turn structure), multiple-entry turns were prevalent, which supports the re-definition of the basic contribution of each chat participant as an entry, and not a turn. Thus, an important difference between electronic chat and face-to-face conversation is the basic unit of contribution that constitutes each.

The fact that both multiple-entry turns and single-entry turns were prominent in the data indicates that chat users, at least in the present study, implement both types of turn structure, as all eight participants in the present study exhibited at least some use of both patterns. Given the synchronous nature of the chats and the time pressure that accompanies the medium, it is not surprising that participants at times typed short entries and then expounded on the information via multiple entries to complete the turn. Besides enabling faster responses, providing short entries also enables the participant to have a more likely opportunity to place his contribution directly after the entry to which it corresponds, rather than after a new entry that may have been crafted by the interlocutor in the meantime.

6.2. OPENINGS AND CLOSINGS. NSS tended to have greater control and were more likely to initiate sequences in the openings and closings of the chats. The use of repetition, usually by the learner, was also quite prevalent during such sequences. Repetition of the opening was highly common, which supports Rintel et al.'s (2001) finding that exact or slightly varied repetitions were frequent in internet relay chats. In fact, in the present study, the use and repetition of hola occurred in all four chats, although in one of the four, it was actually the learner who initiated with hola and the NS who repeated it. The fact that openings were so similar across conversations and by both NSs and learners likely indicates the similarity of openings across cultures and Lis, similar to what has been reported for telephone conversations (Hopper & Doany 1989; Hopper et al. 1990), although it is also possible that the learners in the current study simply followed the lead of their NS partners. Nevertheless, the fact that hola is a greeting used by all of the current participants and that an NS repeats it to a learner in one of the conversations likely indicates that such repetition is not problematic for the NS interlocutors. The current data also demonstrated instances of the participants reporting their present condition via repetitions of their interlocutor in two of the four chats, as either bien or muy bien, through exact repetitions. The use of repetition may be used to create solidarity, as the learner indicates that he is feeling similarly to the NS, or it may simply be a linguistic tactic used to compensate for a lack of lexical alternatives. Furthermore, the use of relatively simple and pithy openings between the participants during this initial interaction is consistent with findings that telephone-conversation greetings between strangers were more reduced than those between intimates (Hopper et al. 1990).

Closings were quite similar in the present data in that repetitions were prevalent in closing sequences as well. The use and repetition of adios occurred in three of the four chats (and was initiated by the NS in two of the three cases). Repetition in chat has been reported in prior work (Rintel et al. 2001), although the authors focused on openings in their study, whereas the present data demonstrate that this finding also applies to closings. Consequently, the strategy of repeating the NS's contribution may continue throughout the conversation and is not simply a feature of early conversational entries. Although this may also be an example of building solidarity with the interlocutor, it is likely further evidence of a limited learner repertoire of lexical items that fulfill greeting and closing functions. Furthermore, learner closings were quite brief, as evidenced by the frequent use of adios without additional lexical material, which is consistent with Gonzales' (2013) case-study finding of foreshortened closings, in which her learner provided a closing and immediately ended the chat within the same turn, especially during early chat sessions. It is hypothesized that the learners in the current study would develop a longer closing sequence over repeated chat sessions, as was the case with Gonzales' longitudinal participant.

6.3. REPAIRS. Repairs were quite common in the present study. The fact that misunderstandings were frequently lexical in the data is similar to findings for NS-NNS service-encounters (Gonzalez-Lloret 2005). Instances of both self-initiated and other-initiated repair abounded, which were also similar to those found by Gonzalez-Lloret (2005), although self-initiated repair was especially common in the present data set. Moreover, initiation of repair by both the NSs and NNSs is consistent with Gonzalez-Lloret's findings, as are the numerous clarification requests and confirmation checks.

Self-initiated repair was most commonly used by the learner and usually took the form of his originally attempting to ask a question via non-targetlike grammar that actually resulted in inquiring about something else entirely (i.e. creating a trouble source). Then, the NS would respond to the question as written, at which time the learner realized that he had asked something unintended and self-initiated repair by reformulating the question. Thus, repairs in the data were most commonly self-initiated and typically performed by the learners following some slight grammatical difficulty. It appears that the learners assumed that the NSs were always accurate in their interpretations, and so if the answers that they provided seemed to answer a different question, the learners interpreted this as a shortcoming of the original question, and thus they self-repaired and reformulated. This high use of self-repair is consistent with previous research (Markman 2005; Schonfeldt & Golato 2003). Given that the learners were only in their second or third semester of Spanish, their faith in the NS responses, or lack of confidence in their own questions, is not surprising. We might hypothesize that more proficient learners would not abandon original questions as abruptly.

Additionally, another form of self-repair which had not been previously attested in the CMC literature appeared in the data--the use of asterisks to repair typographical errors or misspeaking (i.e. stating something that participants did not intend to say). This self-repair strategy was quite common and was used by both NSs and NNSs. Given this finding, it will be interesting to note whether future studies report the use of similar strategies.

6.4. 'LAUGHTER' AND EMOJIS. The use of typed 'laughter' was found to be common in all four chats. The major functions of laughter in the chats were 1) to mitigate negative responses and 2) to show solidarity, often in the form of one speaker repeating laughter after another speaker had laughed (i.e. 'laughing with,' according to Glenn 1995). These two principal functions of laughter have both been previously reported (Glenn 1995), although in the present chats, laughter for the sake of solidarity was more common, as there were relatively few instances of negative responses or potential conflict, and thus, mitigation via laughter was less necessary. The repetition of laughter specifically in chat has also been reported (Noblia 2009), and this was supported in the present study. The frequent use of laughter may also result from the fact that the NS and NNS interlocutors did not know each other prior to chatting, causing a greater need to create initial solidarity and perhaps a greater fear of offending the other party.

Emojis fulfilled many of the same functions as laughter in the current chats. As with laughter, emojis were used both to mitigate negative responses and to show solidarity with the interlocutor (often being replicated by one speaker after use by the previous speaker), although the repetition of emojis was less common than the repetition of laughter in the present data. The present study supports findings by Dresner and Herring (2010) that emoticons can be used as indicators of illocutionary force, either mitigating or aggravating the utterance in which they are used. The authors also hypothesized that emoticons were becoming increasingly conventionalized, and their use by nearly all speakers in the current data indeed aligns with this hypothesis. In fact, only one speaker did not use emojis and he was the one who 'spoke' the least, in general.

6.5. ALL-CAPITALIZATION AND MULTIPLE PUNCTUATION MARKS. All-capitalization, although less frequent than the use of emojis or laughter, was also used to perform the function of intensification in the chats. Although this phenomenon has received less attention in the literature, Noblia (2009) does note its use in her chat data, specifically when it accompanies laughter, as when laughter was occasionally typed via all-capitalization, which served to indicate louder volume. Whereas Noblia does not otherwise discuss the general use of all-capitalization to intensify, it does seem to perform this function in her data as well, similar to the current study, although it was not used by all participants in the present chats.

Multiple punctuation marks were also occasionally used in the current data (although, similar to all-capitalization, were less common than emojis or laughter). They, like all-capitalization, served to intensify the propositional content of the sentence in which they occurred and also to demonstrate greater interest in the information being discussed (e.g. the use of multiple question marks to indicate great interest in a particular question). In fact, parallels have been drawn between the functions of emoticons and multiple punctuation marks (Dresner & Herring 2010), as both affect pragmatic illocutionary force, and multiple punctuation marks in the present study (especially multiple exclamation points) clearly served to intensify utterances.


7.1. PEDAGOGICAL IMPLICATIONS. The current study offers a number of pedagogical implications. Given the strong interest expressed by the NNSs in participating in the electronic chats, it appears that for many learners, having the opportunity to engage in natural communication with an NS offers a very desirable learning experience.

Thus, as early as possible, instructors should consider granting learners the opportunity to interact directly with NSs, rather than delaying such activities until a time of greater proficiency. Furthermore, at lower levels, where learners may still be apprehensive or lack the ability to easily understand or respond to NS speech, the use of written media, such as chats, enables learners to have natural dialogue with NSs, while still being able to read the NS's contribution and have the time to process the message and craft their own response.

Another pedagogical implication of the current study is that more classroom emphasis should be placed on greetings and closings in Spanish, as the learners utilized a very limited set of forms. Examples of authentic speech should be presented in class, and given the frequency and importance of greetings and closings, this material can be introduced from the very beginning of language instruction.

7.2. LIMITATIONS AND FUTURE DIRECTIONS. The present study found that NSs generally had more control over topic selection and the initial asking of questions within each topic. Additionally, two general turn structures emerged: 1) an overall occurrence of A-B question-answer adjacency pairs and 2) multiple entries forming a turn. The NS tended to be in greater control of the opening and closing of the chats, offering the initial greeting in each case and the start of the closing (or pre-closing). Self-repairs were also quite common, especially for the learners, and learners were quite successful at obtaining meaning when unknown lexical items emerged, either via asking related follow-up questions or specifically stating that they did not understand.

Despite its contributions, the study has limitations which bear further consideration. Firstly, given the qualitative nature of the data, there was a small set of participants and conversations were limited to approximately 30 minutes. Although interesting interactions occurred during this time frame, extending the chats to longer sessions and to a longitudinal design of multiple sessions would yield richer data in which development could be charted. Additionally, although the inclusion of NSs from multiple nations of origin reflected the make-up of the Spanish Department of the institution in the study, future work that limits NS participants to one particular region may lead to more uniform NS linguistic behavior. (7) The fact that the NSs were also instructors of Spanish also likely affected the conversations, such that future work analyzing NS-NNS chats might uncover greater need for repair based on including NSs who are less experienced in communicating with NNSs or speakers of LI English. The use of post-chat stimulated recall could also be beneficial (especially for the learners) in ascertaining what participants were thinking at certain times and how they combatted comprehension difficulties.

There are many related avenues for future research. A pragmatic instruction treatment (e.g. of greetings/closings) could be implemented in order to follow how learners perform prior to and after instruction and in comparison to a control group that did not receive instruction. Also, analyzing NS-NNS discourse in other electronic media (e.g. Facebook, Skype) could be a fruitful endeavor in documenting what discourse features emerge via those media and how they might differ from those of chat. Lastly, the use of more controversial themes could allow for richer data in terms of mitigation and intensification, as participants would have opinions that have greater potential to offend, and would thus need to be more careful in forming their contributions. Further research will certainly build upon the findings of the current study as new technology that is popular with students is used not only as a medium of study but also as a conduit for interaction in the target language.

In the current study, intermediate learners were able to sustain conversations of over 30 minutes on familiar topics with NSs, even in the absence of previous experience maintaining discourse with NSs in 'real life' settings. When faced with difficulties, learners were able to obtain meaning following a lack of comprehension via various strategies, including asking content-related follow-up questions or directly questioning the meaning of a lexical item. Furthermore, learners followed the NS's lead in each case in terms of topic, lexical items (in greetings, closings, etc.), and syntactic structures. Given the instructional value of electronic chat and the uniqueness of chat as a medium of analysis, future research will do well to expand on the findings of the current study across different Lis and L2s, discourse topics, theoretical approaches, and longitudinal interactions.


Temas de conversacion--Aqui se encuentran algunos temas si los quiere consultar. Deberia ser una conversacion, no una entrevista, asi que espero que no tenga que hablar mas que su interlocutor.

- Nombre/origen

- Especializacion / profesion futura

- Pasatiempos

- Estudios en el extranjero?

- Familia/hermanos

- Fin de semana tipico

- Clases este semestre / favorita?

- Musica/grupos/generos preferidos

- Peliculas/actores preferidos

- Programas de television preferido

- Deportes/ equipos preferidos

- Restaurante favorito aqui

- Otros sitios preferidos aqui?

- Diferencias entre EE.UU. / tu pais / (cosas que extranas)

- Comida preferida

- Donde quieres vivir en el futuro

Translation: Topics of conversation--Here, a few topics can be found, if you would like to consult them. It should be a conversation, not an interview, so I hope that you do not have to speak more than your partner.

- Name / hometown

- Major / future profession

- Hobbies

- Studies abroad?

- Family / siblings

- Typical weekend

- Classes this semester/ favorite class?

- Favorite music / groups / genres

- Favorite movies / actors

- Favorite TV shows

- Favorite sports / teams

- Favorite restaurant here

- Other favorite sites here?

- Differences between the U.S. and your country / things you miss

- Favorite food

- Where you want to live in the future


BELZ, JULIE A., and CELESTE KINGINGER. 2003. Discourse options and the development of pragmatic competence by classroom learners of German: the case of address forms. Language Learning 53.591-647.

CARROLL, DONALD. 2005. Restarts in novice turn beginnings: disfluencies or interactional achievements? Second language conversations, ed. by Rod Gardner and Johannes Wagner, 201-20. London: Continuum.

CHUN, DOROTHY M. 1994. Using computer networking to facilitate the acquisition of interactive competence. System 22.17-31.

DRESNER, ELI, and SUSAN C. HERRING. 2010. Functions of the non-verbal in CMC: emoticons and illocutionary force. Communication Theory 20.249-68.

GARCIA, ANGELA CORA, and JENNIFER BAKER JACOBS. 1999. The eyes of the beholder: understanding the turn-taking system in quasi-synchronous computer-mediated communication. Research on Language & Social Interaction 32.337-67.

GARDNER, ROD, and JOHANNES WAGNER (eds.) 2005. Second language conversations. London: Continuum.

GLENN, PHILLIP J. 1995. Laughing at and laughing with: negotiation of participant alignments through conversational laughter. Situated order: studies in the social organization of talk and embodied activities, ed. by Paul ten Have and George Psathas, 43-56. Lenham, MD: University Press of America.

GOFFMAN, ERVING. 1981. Forms of talk. Philadelphia: University of Pennsylvania Press.

GOLATO, ANDREA, and CARMEN TALEGHANI-NIKAZM. 2006. Negotiation of face in web chats. Multilingua 25.293-322.

GONZALES, ADRIENNE. 2013. Development of politeness strategies in participatory online environments: a case study. Technology in interlanguage pragmatics research and teaching, ed. by Naoko Taguchi and Julie M. Sykes, 101-20. Amsterdam: John Benjamins.

GONZALEZ-LLORET, MARTA. 2005. Reconstructing NS/NNS communication. The consequences of mobility: linguistic and sociocultural contact zones, ed. by Bent Preisler, Anne H. Fabricius, Hartmut Haberland, Susanne Kjasrbeck, and Karen Risager, 1-27. Roskilde, Denmark: Roskilde University, Department of Language and Culture.

GONZALEZ-LLORET, MARTA. 2008. Computer-mediated learning of L2 pragmatics. Investigating pragmatics in foreign language learning, teaching and testing, ed. by Eva Alcon Soler and Alicia Martinez-Flor, 114-32. Bristol, UK: Multilingual Matters.

GONZALEZ-LLORET, MARTA. 2009. CA for computer-mediated interaction in the Spanish L2 classroom. Talk-in-interaction: multilingual perspectives, ed. by Gabriele Kasper and Hann thi Nguyen, 281-316. Honolulu, HI: NFLRC and University of Hawaii Press.

GONZALEZ-LLORET, MARTA. 2011. Conversation analysis of computer-mediated communication. CAUCO Journal 28.308-25.

HAVERKATE, HENK. 1994. La cortesia verbal: estudio pragmalinguistico. Madrid: Gredos.

HERRING, SUSAN C. 1999. Interactional coherence in CMC. Journal of Computer Mediated Communication 4.0.

HERRING, SUSAN C. 2001. Computer-mediated discourse. The Handbook of Discourse Analysis, ed. by Deborah Schiffrin, Deborah Tannen, and Heidi E. Hamilton, 612-34. Oxford: Blackwell.

HOPPER, ROBERT, and NADA KOLEILAT DOANY. 1989. Telephone openings and conversational universais: a study in three languages. Language, Communication and Culture, ed. by Stella Ting-Toomey and Felipe Korzenny, 157-79. London: Sage.

HOPPER, ROBERT; NADA KOLEILAT DOANY; MICHAEL JOHNSON; and KENT DRUMMOND. 1990. Universais and particulars in telephone openings. Research on Language and Social Interaction 24.369-87.

JOHANYAK, MICHAEL F. 1997. Analyzing the amalgamated electronic text: Bringing cognitive, social, and contextual factors of individual language users into CMC research. Computer and Composition 14.91-110.

KECSKES, ISTVAN, and JESUS ROMERO-TRILLO (eds.) 2013. Research trends in intercultural pragmatics. Berlin: De Gruyter.

KERN, RICHARD G. 1995. Restructuring classroom interaction with networked computers: effects on quantity and characteristics of language production. The Modern Language Journal 79.457-76.

LOCHER, MIRIAM A. 2010. Introduction: politeness and impoliteness in computer-mediated communication. Journal of Politeness Research 6.1-5.

MARKMAN, KRIS M. 2005. To send or not to send: turn construction in computer-mediated chat. Proceedings of the twelfth annual symposium about language and society-Austin Vol. 48, ed. by Chiho Sunakawa, Tomoko Ikeda, Shannon Finch, and Malavika Shetty, 115-24. Austin, TX: Texas Linguistic Forum.

NOBLIA, MARIA VALENTINA. 2004. La ironia en los chats: una forma de preservar las imagenes y administrar los conflictos en las conversaciones mediadas por computadoras. Pragmatica sociocultural: estudios sobre el discurso de cortesia en espanol, ed. by Diana Bravo and Antonio Briz Gomez, 371-97. Barcelona: Ariel.

NOBLIA, MARIA VALENTINA. 2009. Modalidad, evaluacion e identidad en el chat. Discurso & Sociedad 3.738-68.

ORTEGA, LOURDES. 1997. Processes and outcomes in networked classroom interaction: defining the research agenda for L2 computer-assisted classroom discussion. Journal of Language Learning and Technology 1.82-93.

OSMAN, GIHAN, and SUSAN C. HERRING. 2007. Interaction, facilitation, and deep learning in cross-cultural chat: a case study. The Internet and Higher Education 10.125-41.

RINTEL, E. SEAN; JOAN MULHOLLAND; and JEFFERY PITTAM. 2001. First things first: internet relay chat openings. Journal of Computer-Mediated Communication 6.0.

SACKS, HARVEY; EMANUEL A. SCHEGLOFF; and GAIL JEFFERSON. 1974. A simplest systematics for the organization of turn-taking for conversation. Language 50.696-735.

SCHEGLOFF, EMANUEL A. 2007. Sequential organization in interaction. Cambridge: Cambridge University Press.

SCHMIDT, LAUREN BETH. 2009. The effect of dialect familiarity via a study abroad experience on L2 comprehension of Spanish. Selected proceedings of the 11th Hispanic Linguistics Symposium, ed. by Joseph Collentine et al., 143-54. Somerville, MA: Cascadilla Proceedings Project.

SCHONFELDT, JULIANE, and ANDREA GOLATO. 2003. Repair in chats: a conversation analytic approach. Research on Language & Social Interaction 36.241-84.

SKEHAN, PETER. 2003. Task-based instruction. Language Teaching 36.1-14.

SMITH, BRYAN. 2003. Computer-mediated negotiated interaction: an expanded model. The Modern Language Journal 87.38-57.

SYKES, JULIE M. 2005. Synchronous CMC and pragmatic development: effects of oral and written chat. CALICO Journal 22.399-431.

TANNEN, DEBORAH, AND ANNA MARIE TRESTER (eds.) 2013. Discourse 2.0: language and new media. Washington, DC: Georgetown University Press.

THORNE, STEVEN L., and J. SCOTT PAYNE. 2005. Evolutionary trajectories, internet-mediated expression, and language education. CALICO Journal 22.371-97.

VARONIS, EVANGELINE MARLOS, and SUSAN GASS. 1985. Non-native/non-native conversations: a model for negotiating meaning. Applied Linguistics 6.71-90.

VAUGHAN, ELAINE, and BRIAN CLANCY. 2013. Small corpora and pragmatics. Yearbook of corpus linguistics and pragmatics: new domains and methodologies, ed. by Jesus Romero-Trillo, 53-76. Dordrecht, Netherlands: Springer Science and Business Media.

YATES, SIMEON J. 1996. Oral and written aspects of computer conferencing. Computer-mediated communication: linguistic, social and cross-cultural perspectives, ed. by Susan C. Herring, 29-46. Philadelphia: John Benjamins.

Department of Linguistics

University of Pittsburgh

Pittsburgh, PA 15260



University of Pittsburgh

(*) I am grateful to Cesar Felix-Brasdefer, editors Jill Brody and Jeremy King, and two anonymous reviewers for their feedback and suggestions. I also thank the native-speaker and learner participants for their time and spirited participation. All errors are mine alone.

(1) Since terminology has changed somewhat since the publication of the cited research, in the present paper the term EMOTICONS will be used to refer to claims and findings of prior work, whereas EMOJIS will be employed when discussing the current findings. This also reflects a change in the way that such icons are conveyed (i.e. now with their own buttons).

(2) For more on the variety of research methods that researchers have implemented in the study of intercultural pragmatics, see Kecskes & Romero-Trillo (2013).

(3) Note that unlike other frameworks of analysis, CA does not typically utilize pre-formulated research questions.

(4) University and location names have been changed to preserve confidentiality.

(5) Nevertheless, as an anonymous reviewer notes, it is possible that the learner simply did not encounter other contexts for using the form during the chat session.

(6) One reviewer notes that the form could be interpreted as subjunctive (prescriptively llegue 'I might arrive') or preterit (prescriptively llegue 'I arrived'). Although the use of a preterit form would not be target-like here, the learner produced numerous preterit tokens in his chat, did not produce any subjunctive tokens, and generally did not use accent marks. Therefore, this is more likely a context of non-target-like preterit production, although it remains possible that the learner intended to use the present subjunctive here.

(7) Nevertheless, as one anonymous reviewer notes, the NSs formed part of a speech community of educated NSs in a US university setting, which may have reduced even greater potential variability.
COPYRIGHT 2014 Linguistic Association of the Southwest
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 2014 Gale, Cengage Learning. All rights reserved.

Article Details
Printer friendly Cite/link Email Feedback
Author:Kanwit, Matthew
Publication:International journal of the Linguistic Association of the Southwest
Geographic Code:4EUSP
Date:Dec 1, 2014

Terms of use | Privacy policy | Copyright © 2020 Farlex, Inc. | Feedback | For webmasters