CURRENTS. A Journal of Young English Philology Thought and Review, vol.1, no.1/2015 Ed. by E. Bodal, A. Jaskólska, N. Strehlau & M. Włudzik, www.currents.umk.pl ISSN 2449-8769 || All texts licensed under CC BY-NC-ND 3.0
Katarzyna Rogalska
Nicolaus Copernicus University
CONFLICT TALK AND AGGRESSIVE VERBAL INTERACTION IN COMPUTERMEDIATED COMMUNICATION: AN ANALYSIS OF AGGRESSIVE ONLINE INTERACTIONS IN A MITT ROMNEY SUPPORTERS’ FACEBOOK GROUP Keywords: computer-mediated communication, conflict talk, virtual community, discourse analysis
Introduction Computer-mediated communication studies (CMCS) have recently become one of the leading approaches in the research of human behaviour. Millions of people interact daily via social media services, and, although the context of their interaction changed, they still engage in socially meaningful activities, such as conflict talk. Even though conflict has an extensive history of investigation in social sciences, the study of the actual discourse within conflict episodes and its features is relatively new. In contrast to naturally-occurring spoken interactions, those computer-mediated leave a textual trace and allow behaviour and discourse researchers to employ more detailed methods of analysis. The aim of this paper is to analyse and interpret a sample set of virtual aggressive conversations between members of a Facebook group—one of the few Facebook utilities accessible to all members of a given community, and, consequently, to the scrutiny and reflection of scholars. The analysis concerns occurrences of verbal aggression among English-speaking Mitt Romney supporters and opponents in the 2012 presidential election in the USA group. The results of the preliminary study indicate that there are numerous factors influencing the manner of conducting virtual disputes, such as gender, ethnicity, and seniority in the group, which are not that easily detectible in a typical study of naturally-occurring verbal conflict. The paper provides an 89
CURRENTS, vol.1, no.1 / 2015
explanation why Computer-Mediated Communication (CMC) constitutes a valid sample material for discourse analyses, reviews the history of conflict talk studies and offers a thorough analysis of a three verbal disputes among Facebook users. Computer-Mediated Communication and Facebook As one of the biggest social networking services on the Internet, Facebook has become an invaluable source of analysable data for CMC researchers. It allows its users both to generate their own content communicated to other users (which in turn can take various forms) and to interact socially following the structures provided by Facebook itself. No surprise then that having a toolkit perfectly applicable to social networking service communication, CMCS, Computer-Mediated Discourse Analysis (CMDA, cf. Herring 2001, 2004, 2011, 2013) among them, finds Facebook one of the richest sources of research material. Following Herring, over time it [Facebook] has added and combined so many features (including embedded graphics, games, polls, and various modes of CMC) that […] not just because of the number of features, but because the whole is qualitatively different from the sum of its parts, […] it has no single offline or online precedent. (2013: 8)
Computer-mediated discourse (CMD) has great potential as an object of any discourse analysis studies. Nowadays, communication and, consequently, discourse, is successively moving from the “real” world to the virtual one. It all started in the late 1960s in the USA as a national defence utility facilitating the transfer of computer programs and data between remote computers (Levy 1984; Rheingold 1993 in Herring 2001). The new possibility of interpersonal communication was almost immediately noticed by scientists, academics and businessmen, and, in the 1990s, it gained wide recognition and came into popular use. CMD itself became a research subject as early as in 1984, when its potential in language change was first noticed by Naomi Baron. Her article initiated the first wave of CMD descriptions which can be characterised as limiting CMC to 90
CURRENTS. A Journal of Young English Philology Thought and Review
the written medium (e.g. Ferrara et al. 1991). It was not until the era of Susan Herring’s studies, who invented an approach called Computer-Mediated Discourse Analysis, that the discourse of computer-mediated interaction was given thorough scrutiny and examination. According to Herring, CMDA is any kind of analysis of online behaviour based on textual observations. It can be either qualitative (observations of discourse phenomena in a sample text may be made, illustrated and discussed) or quantitative (phenomena may be coded and counted, and summaries of their relative frequencies produced) languagefocused content analysis which applies methods adapted from languagefocused disciplines, such as linguistics, communication studies, and rhetorics. Furthermore, it is grounded in methodological paradigms related to the study of spoken and written language, such as Conversation Analysis, Critical Discourse Analysis, Interactional Sociolinguistics, Pragmatics, Text Analysis, etc. (Herring 2001, 2004). Bearing in mind that a CMDA researcher has to consolidate knowledge from quite a few language-related paradigms, there are also some basic tasks that must be fulfilled in order to perform a valid analysis. First of all, one poses an answerable research question and selects methods that address the research question. Then, the researcher applies them to a sufficient and appropriate corpus of data. Finally, findings are interpreted responsibly and in relation to the original research question. On the other hand, the question posed must also meet some requirements: it must be based on prior observation and motivated by a hypothesis, ask about empirically-observable phenomena, be non-trivial, answerable and open-ended (Herring 2004: 345–352). Although CMDA draws on numerous methods of discourse analysis, for many years the major problem of the approach has been whether computermediated discourse can be classified and analysed in terms of naturallyoccurring discourse. On the one hand, there is no oral production or auditory reception; turn-taking patterns are disrupted and there is no simultaneous feedback. That is probably why previous approaches treated the notion of 91
CURRENTS, vol.1, no.1 / 2015
”conversation” as a metaphor for CMC. However, it must be noted that Internet users refer to textual exchanges as conversations and employ oral discourse strategies. What is more, they experience CMC in similar ways to spoken conversation, regardless of synchronicity, and, in fact, online exchanges are textual representations of auditory information2. For these reasons, Herring proposes a new definition of conversation, in which conversation is seen as “any exchange of messages between two or more participants, where the messages that follow bear at least minimal relevance to those that precede or are otherwise intended as responses” (2011: 4). As a result, the traditional understanding of the term can be questioned. Computer-mediated discourse that can be found on Facebook is a bottomless well of analysable data; therefore, in order not to lose context by random sampling, a researcher must choose a convenient method of data selection. Following Herring, there are at least four data sampling techniques, motivated, respectively, by theme (e.g. all messages in a particular thread), by time (e.g. all messages in a particular month), by phenomenon (e.g. instances of conflict talk only—the technique to be used in the present study) or by individual or group (e.g. all messages posted by an individual or by members of a given group) (Herring 2004: 349–350). A researcher must also pay attention to whether the mode of digital communication he/she decides to choose is appropriate to the analysis he/she wants to conduct. The basic distinction made by CMDA scholars is between synchronous and asynchronous communication. The former refers to messages that cannot be modified after being sent, such as turns in spoken conversation (e.g.: chat, instant messaging), while the latter allows users to edit their messages for a longer time (e.g.: mailing lists, news groups, discussion boards). Consequently, a distinction between one-way message transmission (single unit) and two-way message transmission (chunks) can be made (Herring 2011:2).
92
CURRENTS. A Journal of Young English Philology Thought and Review
Conflict Talk In order to provide a trustworthy analysis of conflict talk, one needs to find an appropriate methodology, usually related to aggression. It seems obvious that such a condition can be fulfilled only by naturally-occurring interpersonal conflict. Nevertheless, numerous researchers conducted studies based exclusively on artificial data. An interesting example of that phenomenon could be the study by Wathlawick, Beavin, and Jackson (1967 in Bavelas, Rogers & Millar 15) who analysed fictional arguments present in Who’s Afraid of Virginia Woolf? by Edward Albee (1962). They came to a conclusion that interpersonal conflict is a non-chaotic speech event with pattern and structure, which was later supported by studies based on naturally-occurring data. One of the first studies of interpersonal conflict, which served as a prototype for a wide variety of initial investigations, was proposed by Lein and Brenneis (1978). They arranged a conflict by asking children from grades one to eight from different speech communities to role-play an argument about, for example, who is the strongest or whose ball it is. They claim to have come with spontaneous data, which was tape-recorded. Lein and Brenneis found three sequential patterns: repetition, escalation, and inversion, the most commonly used one being escalation (both in content or by volume). Therefore, they concluded that verbal conflicts are sequentially organised, “rule-governed, socially organised and frequently quite complex events;” competitive, but “also cooperative performances” (Lein & Brenneis 308). A study containing analyses of both verbal and nonverbal behaviours was offered by Gottman (1979), who analysed sequences of couples’ discussions on conflict-provoking topics, applying to “thought units” eight content codes for verbal behaviours, and three nonverbal codes (positive, neutral, or negative). Moreover, he used a statistical analysis, called lag sequential analysis, which provided information about the probability of given behaviours following a selected criterion behaviour at different time-ordered behavioural steps (lags) in an ongoing interaction (Sackett 1979 in Bavelas, Rogers & Millar 16). Such 93
CURRENTS, vol.1, no.1 / 2015
lag profiles can be later used to test a research hypothesis concerning the dynamics of discourse and specified outcome variables (e.g.: resolution or nonresolution of conflict). When it comes to computer-mediated conflict talk, the only data available for analysis relates to verbal aggression. Nevertheless, the lack of body language is to a certain degree compensated with the use of emoticons and other graphic elements. They convey information similar to that of face expressions and gestures, and often change the classification of a given utterance. Similarly, such classification can be modified with accordance to the conflict phase in which the utterance in question occurred. According to Eisenberg and Garvey (150), “an adversative episode is a sequence which begins with an opposition” in reply to an action, a request for action, or an assertion. However, this approach is challenged by Maynard, who states that in such a case any previous claim could be interpreted as an initiator of a dispute (1985 in Leung 5). Consequently, it is the interpretation of a concrete participant that is of importance, not all possible interpretations. Moreover, conflicts may also be a result of cultural differences: lack of skills in understanding interpretive conventions or a mismatch in communicative styles (Gumperz 1982 in Leung 10). Scholars agree that the characteristic distinguishing an argument from a mere disagreement or claim of opposition is a necessity of at least three moves (Antaki 163; Muntigl & Turnbull 227). It is in the third move, produced by the first speaker as a response to an oppositional claim of the second speaker, where it is clearly visible that the participants consider the interaction “argumentative.” Such exchanges can be either weak or strong disagreement claims, the latter being usually provoked by “irrelevancy claims” (IR), in which the speaker denies the relevancy of the other party’s claim. It is worth mentioning that some sequence types lead to a quarrel more often than others. Among them, there are undoubtedly complaints and accusations, as a denial constitutes a common answer to them. 94
CURRENTS. A Journal of Young English Philology Thought and Review
Conflict maintenance and escalation constitute one of the most challenging aspects of aggressive behaviours for discourse analysts. Conflict talk may advance in any number of directions, escalating and de-escalating in numerous unpredictable ways. Following Coleman, they may expand in focus (the original focus may become neglected or forgotten), change in focus (they may become more personal than initially), and spread along existing class of social boundaries (1957 in Grimshaw 296). Once an argument is started, the principle of the machinery that drives its structure enables the participants to fall into the argumentative interaction trap—an absence of counterattack could be interpreted as lack of ability to do so, and, in consequence, to the loss in the dispute. What follows is the fact that a conflict episode does not necessarily terminate when there is a shift in content, just the opposite. Nevertheless, a clear positioning of a loser and a winner is rarely the case. According to Vuchinich, a single dispute usually becomes the resource for constructing another one (119–138). The sequential continuity of arguments, being an example of the notion of CA adjacency pairs, cannot be “naively” ignored by participants. However, the concept of adjacency is used here in a modified form from the usual question-answer type, namely, the action-opposition sequence. The three-turn structure, as mentioned above, is helpful in recognising an argument between two parties. There is an antecedent of an arguable event, an oppositional utterance, and a reaction phase, in which the opposition itself is responded to by the first speaker with one of the various ways to reject the second speaker’s opposition. However, the model is not followed in a multiparty context. A multiparty argument requires collaboration, which can be offered from the third party, or be solicited or invited by the already “insider” of the emerging dispute. When it comes to alignment strategies applied by the “outsider,” there can be distinguished alignments against a position (taken by the first speaker), and alignments with a counter position (taken by the insider partner to the first speaker) (Maynard 1986: 265). 95
CURRENTS, vol.1, no.1 / 2015
It is also worth mentioning how externally-shaped public arguments, such as political debates and interviews are maintained. The unique contextual setup for this particular social speech genre usually requires the participants to act accordingly with pre-given roles. What is characteristic of debates is the presence of the audience and a moderator who casts the topic and delays the immediate production of the second adjacency pair. The result is often an indirect interpretation of the audience and the media. This artificial structure makes a televised debate different from disputes embedded in everyday conversation. The question whether Facebook disputes should be treated the same way remains open. Traditionally, it was thought that the primary form of conflict termination is a movement of the participants towards the same position. However, studies of naturally-occurring conflict show that agreement among participants and resolution of the central issue are rarely accomplished. Nevertheless, in order to end the conflict, opponents have to collaborate, which can take various forms. Vuchinich, after observations of American family disputes at dinner table, postulates five main termination formats: – submission, when one party accepts another party’s position, usually by means of a weak argument, such as: “well, maybe,” or “there may be some truth in that,” – dominant party intervention, when the opponents submit to a third party, – compromise, when a concession is offered by one party and it is accepted, – stand-off, when no submission or compromise is reached and both parties continue to hold contrary views, – withdrawal, when one of the parties withdraws from the verbal conflict or from the environment. According to Vuchinich the most common form of conflict termination is stand-off, being the one that allows bringing the conflict to an end without losing one’s face. However, it should be noted that family disputes constitute a 96
CURRENTS. A Journal of Young English Philology Thought and Review
very rare context in which face is not as important as power or affect (118– 138). What should be elaborated on is the role of a third party in conflict termination. A third party can help cease an argument also in a more institutionalised manner, for instance, during a family therapy or divorce mediation sessions. Greatbatch and Dingwall report that mediators use various formulas in order to de-escalate arguments, such as seeking clarification, shifting the topics, or negatively sanctioning the opponents, at the same time eliminating specific conflictual processes from the interaction and preventing formation of argumentative adjacent turns (151–170). It is also worth noticing that resolving an underlying conflict does not always constitute the main goal of an argument. The display of aggravated disagreement is an important sociable feature of peer culture, treated as nonserious and cooperative by the participants, but, at the same time, allowing for experiencing power and domination. For instance, children often do not take an opportunity to terminate conflict talk, but instead continue in efforts to maintain the dispute (Goodwin 1990 in Leung 8). Additionally, arguments serve to gain power and control in a given group creating social organisational structures. As conflict behaviours may be characterised either as functional or as dysfunctional, they give various outcomes. On the one hand, they may cause damage to relationships and produce other negative consequences, but on the other, they can provide positive outcome and even strengthen interpersonal relationships. Typically, conflicts end in impasse, resolution, or face-saving stand-offs, allowing their participants a return to their activities (Vuchinich 118–138). This illustrates the complex, and, at the same time, paradoxical nature of conflict.
97
CURRENTS, vol.1, no.1 / 2015
An Analysis of Selected Examples of Conflict Talk in CMC Environment The theoretical considerations about CMDA and conflict talk offered in the previous sections serve to introduce the analytical intention of the present paper. While making an attempt at analysing a set of Facebook data, a researcher can come across numerous problems. First of all, data collection may seem trivial at first sight, but in fact, it is not. The main problems are related to the appropriate size and representativeness of data samples, data processing techniques, kind and amount of contextual information necessary, and, as usual when data samples are derived from the Internet, ethical issues (anonymity, privacy protection). Above all, however, it must be noted that there are no standard guidelines for CMD corpus design and of publicly-available CMD corpora (Androutsopoulos & Beisswenger 1), which means that the kind of data chosen for analysis depends on the researcher exclusively. The analysis in question is a qualitative analysis of logs of verbal interaction in Mitt Romney for President 2012 Facebook group. Such a corpus has been chosen as the present paper’s author found the group members particularly expressive and emotional, which may provoke aggression. The discussion in the group is organised around posts and links of its members. The sampling technique chosen is, following Herring’s classification, by phenomenon. As it is only a preliminary study, the sample size is not very representative—the present analysis takes into account three instantiations of conflict talk among group members, while the total number of aggressive verbal interactions found is 23. The research questions posed are: what factors influence the manner of conducting the three phases of virtual disputes? and in what ways does the virtual context of conflict talk differentiate it from naturally-occurring verbal disputes? The first conflict talk to be analysed occurred between two senior members of the group, Henrique and Daniel. Although both of them participate actively in all group discussions, this one seems to be their first conflict. It initiates as presented in Figure 1. 98
CURRENTS. A Journal of Young English Philology Thought and Review
Figure 1. Example 1. Conflict initiation
As mentioned previously, an argument should be initiated with least three moves. Figure 1 illustrates this fact perfectly. Here, Daniel denies the relevance of Henrique’s claim that Obama does not fulfil basic requirements for becoming a US president. What is more, Daniel’s claim is a dispreferred course of action for the first part of an adjacency pair (Henrique expected group members to support him, not oppose). Aware of the fact that they are no longer engaged in a neutral conversation, Henrique, in a short utterance, emphasises his nationality and accuses Daniel of being an immigrant without providing any supporting arguments. This face-threatening act leads to conflict escalation (see Figure 2).
99
CURRENTS, vol.1, no.1 / 2015
Figure 2. Example 1. Conflict escalation and maintenance
Following Coleman’s classification of conflict escalation directions, the episode in question has both expanded in focus (the original focus on Obama’s nationality became neglected) and, later, changed in focus (it became more personal than initially). Daniel responds to the face-threatening act using typical techniques for conflict maintenance, such as name-calling (underlined in Figure 2). His turn is very long (which emphasises the asynchronous mode of the dispute) and has nothing to do with the original oppositional claim. It is followed by another turn by Daniel which has an offline component, namely the reference to Henrique’s employer, found on his profile. On the other hand, Henrique’s answer resembles synchronous communication; his turn is short and contains capital letters associated with shouting. As far as conflict termination is concerned, a clear positioning of a loser and a winner of a verbal dispute is rarely the case. Figure 3 presents the final turn taken by Henrique. 100
CURRENTS. A Journal of Young English Philology Thought and Review
Figure 3. Example 1. Conflict termination
Henrique’s technique of conflict termination is consistent in almost all of his disputes with group members. He posts an utterance similar to that from Figure 3 and gives an ignored person a number. It is hard to classify such behaviour in terms of typical conflict termination strategies listed by Vuchinich. It can be treated as either a stand-off, because no submission or compromise is reached and both parties continue to hold contrary views or as withdrawal, because he withdraws from the verbal conflict, even though Daniel continues the discussion (he even tagged Henrique in his post, so that he notices it). The second example illustrates a conflict which occurred between James (a senior member of the group) and Kate (a rather passive contributor) in response to the link posted by Joe, who does not take part in the subsequent course of action. James suggested that the information being spread by Joe is fake. Kate must hold an opposite claim, as she accuses James of being “a brainwashed idiot.” The initiation of their conflict is presented in Figure 4. Figure 4. Example 2. Conflict initiation
101
CURRENTS, vol.1, no.1 / 2015
When it comes to conflict escalation and maintenance, this adversative episode is a rather extended one (this phase’s initial fragment is presented in Figure 5). Kate keeps questioning James’s intelligence, while he defends himself with (at least seemingly) reliable arguments. The participants replies are segmented which suggests a more synchronous exchange than in the previous example. One of Kate’s weapons against James is irony (underlined in Figure 5). Figure 5. Example 2. Conflict escalation
James and Kate’s conflict talk and the sequential continuity of their arguments is a perfect example of the modified notion of adjacency pairs, where the usual question-answer sequence is replaced with the actionopposition one. The conflict is terminated with a typical stand-off—neither James nor Kate accepts another party’s position. Kate withdraws, saying that she has better things to do (see Figure 6). James does not comment on this.
102
CURRENTS. A Journal of Young English Philology Thought and Review
Figure 6. Example 2. Conflict termination
The third example to be analysed is the most complex one. It contains two separate conflict episodes with alternating turns. James from the previous example reacts aggressively to Joseph’s post. He is supported by Philip, which in turn provokes conflict between Philip and Joseph. In the meantime, Kate attempts to continue her episode with James analysed in previously. (see Figure 7 for this conflict’s initiation). Figure 7. Example 3. Conflict initiation
As the exchange between James and Kate is shorter it shall be analysed first. It is initiated with Kate correcting James’s spelling mistake and trying to prove his lack of proper education, which was the main subject of their conflict in the previous example. Consequently, Vuchinich’s observation that a single dispute usually becomes the resource for constructing another one is confirmed (1990: 119–138). Nevertheless, James replies to Kate’s attack only once to engage in another conflict between Philip and Joseph. Although Kate wants to continue, he ignores her, which can be treated as withdrawal from the verbal conflict (but
103
CURRENTS, vol.1, no.1 / 2015
not from the environment). Philip’s “hmmm” can be read as a significant turn, he seems to be suggesting that Kate’s contribution is out of place (see Figure 8). Figure 8. Example 3. James and Kate’s conflict termination
At the same time, Joseph and Philip’s conflict escalates. It is worth noticing that the original focus of the conflict is maintained and that, contrary to the previous examples, the argument does not become personal (Joseph and Philip continue to talk about politics, although, as underlined in Figure 9, Philip suggests that Joseph’s views and arguments are outdated). Figure 9. Example 3. Joseph and Philip’s conflict maintenance
With the course of time, however, Joseph and Philip’s conflict deescalates (see Figure 10). The initial focus is lost and their arguments become less coherent. Finally, James terminates the episode, which can be read as dominant 104
CURRENTS. A Journal of Young English Philology Thought and Review
party intervention. As he recognises the argument as pointless, it is not continued. Figure 10. Example 3. Conflict de-escalation and termination
Conclusion Although the preliminary study presented here is small-scaled, it makes some general conclusions regarding conflict initiation, maintenance, and termination in the CMC context possible to be drawn. First of all, when it comes to conflict initiation, an important observation is that the three-turns rule is preserved, as in all of the examples analysed it was the third turn that revealed the exchange’s adversative character. Interestingly, although women tend to use literal aggression more often than men in real-life context (Tannen 1998), Kate is among the very rare examples of female conflict initiators among group members (the group has over 20,000 members; however, gender proportions cannot be established). Moreover, from the first analysed example, it is clear that ethnicity issues provoke conflict talk (related both to the subject of discussion and to the participants themselves). Finally, senior members of the group seem to be more willing to take part in verbal duels than newcomers. As far as conflict escalation is concerned, all of Coleman’s possible real-life conflict maintenance directions easily found in virtual disputes. However, turns tend to 105
CURRENTS, vol.1, no.1 / 2015
be much longer than in naturally-occurring situations, which makes simultaneous and unrelated conflicts possible. On the other hand, adjacency pairs can be elicited without problems. As in real-life conflicts, those virtual ones can advance in any number of directions, escalating and de-escalating in numerous unpredictable ways involving various conflict maintenance techniques. Conflict termination analysis revealed that virtual conflict does not conclude in any preferred way; however, intuition suggests that an analysis of a larger sample of date would provide similar results to real-life conflict analyses, with stand-off and withdrawal as the most popular conflict termination techniques and dispute winners difficult to detect. However, further research is needed to confirm this intuition. An upcoming election of 2016 and the newlyemerged Facebook groups devoted to it could constitute an optimal corpus for such a study. Endnotes 1. An early version of the present article has been presented at a conference Meaning Context and Cognition (MCC) organised by the University of Łódź in 2013. 2. E.g. by means of referring to one’s written utterance “he said that […].”
References Androutsopoulos, J. & M. Beisswenger. 2008. “Introduction: Data and Methods in Computer-Mediated Discourse Analysis.” Language@Internet, 5, 1–7. Antaki, C. 1994. Explaining and Arguing: The Social Organisation of Accounts. London: Sage. Barab, S.A., R. Kling, & J.H. Gray (Eds.) 2004. Designing for Virtual Communities in the Service of Learning. New York: Cambridge University Press. Baron, N.S. 1984. “Computer Mediated Communication as a Force in Language Change.” Visible language, XVIII 2, 118–141. Bavelas, J.B., R.L. Edna, & F.E. Millar. 1985. “Interpersonal Conflict,” in: T.A. van Dijk (Ed.), 9–26. Brown, P., and S.C. Levinson. 1987. Politeness: Some Universals in Language Use. Cambridge: Cambridge University Press. van Dijk, T.A. (Ed.) 1985. Handbook of Discourse Analysis. Vol. 4: Discourse Analysis in Society. London: Academic Press. Eisenberg, A.R., & C. Garvey. 1981. “Children’s Use of Verbal Strategies in Resolving Conflicts.” Discourse Processes, 4, 149–170. 106
CURRENTS. A Journal of Young English Philology Thought and Review Ferrara, K., H. Brunner & G. Whittemore. 1991. “Interactive Written Discourse as an Emergent Register.” Written communication, 8(1), 8–34. Gottman, J.M. 1979. Marital interaction: experimental investigations. New York: Academic Press. Greatbatch, D., & R. Dingwall. 1997. “Argumentative Talk in Divorce Mediation Sessions.” American Sociological Review, 62, 151–170. Grimshaw, A. (Ed.) 1990. Conflict Talk. Sociolinguistic Investigations in Conversations. Cambridge: Cambridge University Press. Herring, S. 2001. “Computer-mediated discourse,” in: D. Schiffrin, D. Tannen, and H. Hamilton (Eds.), 612–634. Herring, S. 2004. “Computer-Mediated Discourse Analysis: An Approach to Researching Online Behaviour,” in: S.A. Barab, R. Kling, & J. H. Gray (Eds.), 338–376. Herring, S. (Ed.). 2011. “Computer-Mediated Conversation, Part II.” Special Issue of Language@Internet, 8. Herring, S. 2013. “Discourse in Web 2.0: Familiar, Reconfigured, and Emergent,” in: D. Tannen and A.M. Tester (Eds.), 1–25. Lein, L., and D. Brenneis. 1978. “Children’s Disputes in Three Speech Communities.” Language in society, 7, 299–323. Leung, S. 2002. “Conflict Talk: A Discourse Analytical Perspective,” http://journal.tclibrary.org/ojs/index.php/tesol/article/view/20/27, DOA 10.07.2011. Levy, S. 1984. Hackers: Heroes of the Computer Revolution. New York: Dell. Maynard, D. 1985. “How Children Start Arguments.” Language in Society, 14, 1–21. Maynard, D. 1986. “Offering and Soliciting Collaboration in Multiparty Disputes among Children (and Other Humans).” Human Studies, 9, 261–285. Muntigl, P. & W. Turnbull. 1998. “Conversational Structure and Facework in Arguing.” Journal of Pragmatics 29: 225–256. Rheingold, H. 1993. The Virtual Community: Homesteading on the Electronic Frontier. Reading, MA: Wesley. Schiffrin, D., D. Tannen, & H. Hamilton (Eds.) 2001. The Handbook of Discourse Analysis. Oxford: Blackwell Publishers. Tannen, D. 1998. The Argument Culture. London: Virago Press. Tannen D. and A.M. Tester (Eds.) 2013. Georgetown University Round Table on Languages and Linguistics 2011. Discourse 2.0: Language and New Media. Washington, DC: Georgetown University Press. Vuchinich, S. 1990. “The sequential organization of closing in verbal family conflict,” in: A. Grimshaw (Ed.), 118–138. Abstract Computer-mediated communication studies (CMCS) have recently become one of the leading approaches to the research of human behaviour. Millions of people interact daily via social media services, and, although the context of their interaction changed, they still engage in socially meaningful activities, such as conflict talk. In contrast to naturally-occurring spoken interactions, those computer-mediated leave a textual trace 107
CURRENTS, vol.1, no.1 / 2015 and allow behaviour and discourse researchers to employ more detailed methods of analysis. The aim of this paper is to analyse a sample set of virtual aggressive conversations among members of a Facebook group of Mitt Romney supporters in the 2012 presidential election in the USA. The results of the preliminary study indicate that there are numerous factors influencing the manner of conducting virtual disputes, such as gender, ethnicity, and seniority in the group, which are not that easily detectible in a typical study of naturally-occurring verbal conflict.
108
