Home / Articles / Volume 5 (2008) / Conversation analysis and community of practice as approaches to studying online community
Document Actions



During the last decades, online community has been approached in various ways, depending on the research discipline or question (e.g., Baym, 1998; Cherny, 1999; Döring, 2003; Herring, 2004; Höflich, 1996). Herring (2004) offers an overview of paradigms and methods useful for studying online community from a discourse perspective on computer-mediated communication (CMC). One of the proposed paradigms is Conversation Analysis or CA (Psathas, 1995). In this article, CA is applied to explore what it can reveal about online community. I focus on interaction in a web-based setting; more precisely, I attempt to answer the question whether and how interactive practices constitute community. First, I relate Herring's (2004) approach to analyzing online community to interaction analysis, followed by a brief consideration of the transferability of CA concepts to CMC data. One of these concepts is discursive identity, which is addressed through analyzing data from an online forum on eating disorders. A link between discursive identity and (certain criteria of) community is found, confirming that interaction analysis is useful for studying online community.

The analysis then goes further to show that the discursive invocation of online community in the web-based forum is often accompanied by reference to forum rules. These rules appear to serve as a resource for the moderator to set the limits of the community. The forum rules are drawn upon in the interaction, but they are also discourse products external to the interaction in the forum. Finally, the concepts of participation and reification from Community of Practice (CoP) theory (Wenger, 1998) are considered for a theoretical understanding of the close connection between members’ interaction and listed forum rules.


Studying Online Community

Studying online community by using interaction analysis implies that we are situated in the realm of discourse analysis. The model of Computer-Mediated Discourse Analysis (CMDA) (Herring, 2004) was developed to facilitate the study of online interactive behavior, both on the micro-level of linguistic phenomena and the macro-level of concepts such as community and identity. Herring proposes CA as an appropriate paradigm in the discursive domain of interaction. This article sets out to explore what an interaction analysis, drawing on CA, can actually reveal about online community.

In an attempt to synthesize previous theories and concepts of virtual or online community, Herring (2004, pp. 351-352) operationalizes the concept of virtual community as six sets of observable criteria. These are:

  1. active participation and a core of regular participants

  2. shared history, purpose, culture, norms, and values

  3. solidarity, support, reciprocity

  4. criticism, conflict, means of conflict resolution

  5. self-awareness of group as entity distinct from other groups

  6. emergence of roles, hierarchy, governance, rituals

Despite the apparent clarity of these criteria, Herring notes that the question of when a web-based environment can justifiably be labeled a virtual community poses "interpretive challenges" (Herring, 2004, p. 359). In fact, the establishment of objective criteria is necessarily arbitrary, and therefore she suggests that the perceptions of the participants themselves with regard to their felt senses of community membership should also be taken into account.

According to Herring, each criterion requires its own analytical method. However, in this article I argue that an interaction analysis can 1) reveal various dimensions or aspects of community simultaneously and 2) facilitate addressing the interpretative challenges mentioned by Herring. Such an analysis would thus transcend distinguishable criteria, instead illuminating online community by means of an analysis of the discursive practices in web-based environments. It is not the aim of this article to test whether CA can handle all facets of the criteria for online community; rather, it explores how the CA paradigm and method can shed a certain kind of light on each of these criteria. The methodology of CA (Drew, 2005; Sacks, 1992) constitutes the methodical paradigm for an interaction analysis. It focuses not on the researcher’s categories or criteria for community, but on members' orientations. With its strong focus on spoken interaction, CA is not transferable to web-based interaction at first sight, but it can nonetheless serve the study of online community.

CA for CMC

CA investigates social action with a focus on participants' understanding of one another’s conduct. This premise (language use as social action), as well as certain aspects of sequence organization, which refers to relations between turns such as adjacency pairs, can be adopted for an analysis of CMC. Through their online messages, participants accomplish actions and display their understanding of one another’s actions, and these actions can stand in a particular relation to one another. Moreover, for CA the research data are typically naturally occurring interactions, which are easily collectable in the case of CMC.

Another concept that has been invoked in CA studies is identity (Antaki & Widdicombe, 1998; Schegloff, 1997). Various types of discursive identities can be distinguished; for the scope of the article, it suffices to mention categorical identities and situational identities. Categorical identity refers to how actors attend to and are attended to by categories in their naturally-occurring interactions (Benwell & Stokoe, 2006; Schegloff, 2007). Some examples of categories are ‘girl,’ ‘hairdresser,’ and ‘anorectic.’ Categories as they are used in interaction display various regularities, one of which is that they generate inferences that can be drawn upon during talk.

The concept of situational identity was proposed by Zimmerman (1998) in his analysis of emergency calls. Situational identities come into play in a particular situation; in the case of emergency calls, the situation can be described as one in which the identities ‘citizen’ and ‘complainant’ are relevant. Certain expectations are tied to these identities. For instance, the citizen is expected to describe some kind of emergency (more or less) adequately. If s/he does not realize this category-bound feature of being a citizen caller, it may have consequences for the conversation. In specific settings, such as mental hospitals and web-based forums, the situation can be a resource for the invocation of the identity of ‘new patient’ or ‘help seeker.’ These can be connected to categorical identities such as ‘mentally ill’ or ‘sufferer of breast cancer.’

How to deal with social (hierarchically-organized) identities within CA has occasioned heavy debate (see Billig & Schegloff, 1999; Schegloff, 1997; Wetherell, 1998). In my data, I choose to consider demonstrable identities, where demonstrability is different than it is in spoken data as a consequence of the computer medium of interaction. In CMC, transcription is not needed (hence, conversation is not dependent on audibility), pitch and voice (which frequently indicate gender) are irrelevant, and the design of the eating disorders forum renders names and membership positions visible in each contribution to members and researchers alike, which largely solves the problem of how to name study participants. It follows that the relevance of categorical and situational identities should be observable both by members and by researchers in what is readable/visible on the computer monitor.

Previous research has demonstrated that a CA framework can be fruitful for understanding CMC (e.g., Antaki, Ardévol, Núñez, & Vayreda, 2005; Benwell & Stokoe, 2006). Several CA-related studies have paid attention to the work of identities in CMC. Lamerichs and Te Molder (2003), for instance, have shown that identities in web-based interactions on depression are used to attend to contradictory normative requirements. A similar study by Sneijder and Te Molder (2006) showed how discursive psychology (Edwards, 1997), which is methodically related to CA, can be used to analyze identity work in online discussions on veganism. The authors refer to discursive identities or speakers’ identities in order to emphasize that identity is an interactional tool. In this sense, interactants may construct themselves as ignorant and others as experts on the issue of veganism. Hence, identity is linked to a socially "known" category (veganism), including its negative inferences, but also to local orientations to being either uninformed about the category or a specialist.

The research on identities in CMC suggests that community may also be a relevant concept for analysis. The identities ‘ignorant’ and ‘expert’ with regard to veganism can be argued to play a role in community management; experts inform novices and thereby perform a group role, enact solidarity, explain norms and values, etc. While the occurrence of expert and ignorant identities alone does not explain online community, the discursive use of identity can illuminate various aspects of community. As orientations in interaction, these identities can contribute to community construction. The actions conducted may draw on and simultaneously bring forth community in the turn-by-turn process of online interaction. In the two analyses presented below, I demonstrate, starting from a CA-inspired paradigm, how discursive identities and community are used as interactional resources, and how this reveals specific characteristics of the community included in the sets of criteria proposed by Herring (2004).

Analysis 1: How Discursive Identities Construct Community in Interaction

The extracts are taken from the Hungrig-Online forum, a German forum on eating disorders. Hungrig-Online (HO) is an Internet portal that offers chat sessions and information sites, as well as a forum for persons struggling with eating disorders. HO aims at reaching not only eating-disordered persons who engage in therapy, but also those that have not sought medical assistance yet. Approximately 97% of the participants are female, almost 50% are under 23 years old, and the participants suffer predominantly from anorexia and bulimia (Leiberich et al., 2004). HO members are not identifiable except through their nicknames, which often display telling features (Stommel, 2007). I conducted a micro-analysis of 30 threads from the Hungrig-Online sub-forum for novice members called Neu Hier.

My first goal is to demonstrate how members enact a discursive identity in HO. Frequently, members who post an initial contribution in this particular sub-forum display insecurity with regard to the actual step to join the forum. The situational identity of being new is acted out by verbalizing insecurity or hesitancy in the subject line of the posting or directly after (or even in) the greeting. In extracts 1 and 2, the posters Doreen and Katrina1 construct themselves as novice members by formulating a grave uncertainty or insecurity with regard to joining the forum. (Note that all extracts below are in English translation; the original German-language texts are provided in the appendix.)

Extract 1 (thread no. 22)


quite unsure
Hi everyone!
I’m not really sure if I’m in the right place, [...]

Extract 2 (thread no. 7)


am I in the right place?
So, hello.
so I thought for quite a long time about whether I should register here or not. [...]

It appears that the situational identity of being a new and inexperienced member in this specific setting is relevant for joining HO. The interactional question is how other members deal with this specific discursive identity. Extract 3 shows one way of acting in response to an insecure novice member: offering a warm welcome, invoking the community to make the novice member feel comfortable. In this extract, the abbreviation e.d. stands for eating disorder, SIV for self-inflicted violence, and BW for best wishes.

Extract 3 (thread no. 33)


#Antonia: Hello I’m new…
Hello you guys,
I’m still a bit unsure what I should write. I am [18-22] years old and have
had an e.d. for 8 years. It’s a bit like bulimia, but sometimes I also have
phases of starvation. But recently it’s been ok with the e.d.. Borderline is
also not unknown to me, because I’m in therapy for that, amongst other things.
Before Christmas I had a SIV relapse, but that’s also ok again now. Oh well, I
don’t even want to talk about my other symptoms, because then you’ll
definitely think I consist only of symptoms.
Love, Antonia
#Joe: Hello Antonia,
firstly I also wish you a warm welcome here in the forum.
I don’t even want to talk about my other symptoms, because then you’ll
definitely think I consist only of symptoms.
I also think that about myself – that I only consist of symptoms. But I have a
great deal of respect that you are still strong despite everything and have
made the decision to fight – I read that you are also in therapy?! Because of
the SIV?
I’m also [18-22] and have also suffered from an e.d. for some time also with
phases of starvation.
I believe nobody here thinks about whether we have this or that many symptoms.
After all, we’re actually here to be able to talk about everything, aren’t we?
Would also be happy to learn something more about you.
till then
BW Joe
#Antonia: […]@ Joe:
You know, it’s just so embarrassing that I drag so much around with me. I just
always think that others could then tell me whether there is anything else to
come, and that I already have the e.d. and whether that’s not enough already.
But I really don’t choose the symptoms. But then I have such strange thoughts.
No idea why. […]
#Joe: yoohoo,
I can understand that well, know that from myself I’m also ashamed sometimes
about talking to my friends again about my little problems or when I get low
again and could just bury myself when I’m in a depressed mood I don’t let it
show so people don’t tell me again I’m being bleak again, and what then I know
that but sometimes it’s important to let everything out that can work wonders.
I can only do that here in the forum, because nobody I know has the problems I
have with e.d. and everything.
Hope I can help you a bit or give you some ideas till then and
Love Joe

Let us briefly consider the most important aspects of the exchange between Antonia and Joe, focusing on identity and community. Antonia opens her posting with a verbalization of insecurity about what to write (situational identity: novice). She continues with a self-presentation in terms of identity categories: age and (the duration of) her eating disorder. The subsequent lines nuance the extent to which she fits into single categories, such as bulimia, borderline, and auto-mutilation. Joe speaks on behalf of the whole forum in claiming that HO members do not judge others members in terms of (numbers of) symptoms. Joe's speaking in the name of the forum both works to embrace Antonia as a novice member and enables him to act as an experienced member, since adopting the role of speaking for all HO members is a means of constructing his own identity (which is particularly notable, considering that he is a novice member himself).

The enactment of these identities can be interpreted as evidence of roles in the forum (cf. criterion 6). By making a general statement about the character of the forum (line 25), Joe also constructs the forum as a community with a certain purpose (cf. criterion 2): ‘After all, we’re actually here to be able to talk about everything, aren’t we?’ The possibility to talk about everything effectively means that according to Joe, one will not be judged or laughed at in this forum, which creates security. By stressing that ‘here’ one can talk about everything, a boundary is set with the world outside the forum. The community is constructed in contradistinction to the out-group (cf. criterion 5). In other words, Joe treats Antonia's orientation to being accountable for having many illnesses as not necessary or relevant in the forum. In this way, community is used as an interactional resource for denying the relevance of many diagnoses and thus for accepting a novice member.

In her subsequent response, Antonia addresses Joe (@Joe, line 30) in a colloquial, personal style. The start with ‘You know’ (line 31) signals that Antonia is about to confide something to Joe. This stands in contrast to ‘you’ in her first posting (line 2 and 8), in which Antonia expected that ‘they’ would reduce her to her ‘symptoms.’ She now displays feeling safer and more secure in the forum. The fact that she is embarrassed about dragging around so much (line 31) also exhibits Antonia's preoccupation with how others perceive her and her problems. The ‘you’ from the first posting is now replaced by ‘others’ (line 32). It is unclear who these others are: forum members other than Joe, or others outside the community? Both possibilities are left open by the ambiguity in the reference to ‘you.’

After the warm responses that she has received, this ambiguity enables Antonia on the one hand not to offend other forum members (by implying that they are critical about her many symptoms). On the other hand, it contributes to demarcating “outsiders” to the forum and thereby identifying the community as such. Hence, both Antonia and Joe invoke community to enable them to perform certain actions in the exchange. Various criteria for community are thereby implied or manifested: roles and hierarchy (the novice versus the forum representative), distinction from outsiders, purpose of the forum, solidarity, and support. In this way, an insecure novice member is interactionally welcomed to HO with the help of a community construction.

As shown in the next section, a CA perspective can also shed light on the criterion of norms and values, condensed as forum rules.

Analysis 2: Forum Rules in Interaction

Web-based forums frequently have a list of forum rules (or netiquette), according to which certain styles of communication are either encouraged or inhibited. Moderators may check whether participants take these rules into account and intervene when necessary. Rules are likely to affect the interaction in terms of style or content, although it may be difficult to determine exactly how. Herring's (2004) second criterion for online community (norms and values) is revealed by an examination of these forum rules, as well as by the analysis of verbal reactions to violations of appropriate conduct. Such reactions (e.g., the display of being offended by a certain behavior) are frequently uttered in the forum itself. An interaction analysis can thus also shed light on the practice of dealing with forum rules, and thereby on the aspect of norms and values. In the next sections, I explore the interrelation of forum rules and the invocation of community. It appears that discussion of forum rules draws heavily on ideas or statements about community and thereby co-constitutes the community. Again, discursive identities (including roles and hierarchy) play a role. This underscores that separately observable criteria may in fact be rather intricately interconnected in web-based forum interaction.

Consider the first posting by novice member Sylvia. The following analysis focuses on the opening, the closing, and the first response, in which community, discursive identities, and the moderator's intervention on the basis of forum rules are articulated interrelatedly.

Extract 4 (thread no. 31)


#Sylvia: am I in danger?
Hello you forumers…
so I don’t actually know what I should write here exactly
actually I want to know whether others find my thoughts abnormal
so I’ve never really revealed my thoughts to anyone on this subject
don’t really know where to begin. I weigh myself [*] times a day.
[21 lines omitted]
now is that ill? I mean somehow it is actually ill…but is it REALLY
is it ill to think to hope that one somehow will eventually get
thanks in advance
[Dear Sylvia. Please don’t give any figures for your weight or
anything to do with it and no descriptions of how you harm yourself.
Read through the rules again. Love Marcia]
@Marcia: I’m sorry, I had read the rules, but somehow I didn’t really
pay attention… well, I did try but this one thing I must have

Marcia's intervention (lines 19-21) literally breaks into Sylvia's posting (i.e., Marcia has replaced a number by an asterisk in line 8 and adds a few lines at the bottom of Sylvia's posting). She kindly reminds Sylvia that numbers are not permitted in the forum and indirectly justifies her correction of Sylvia's posting with reference to the forum rules. Although the intervention is formulated politely (dear, please, love), it does not include a welcome to the forum, which is offered in other cases.2 Marcia's advice to read the rules again implies that she is redirecting Sylvia to the registration process in which the forum rules have to be read and virtually signed. By doing so, Marcia treats Sylvia as a (still ignorant) applicant and not as a member; this would imply that Sylvia is not (yet) accepted as a new member and that her position in the community is uncertain.

The fact that the intervention is signed with the nickname of the moderator is salient, since this is not always the case. It renders the intervention an observable activity between identifiable members of distinct statuses. From a CA point of view, it is important to understand why Marcia thought Sylvia's posting in extract 4 was a violation of norms and values and thus felt she should intervene. Sylvia's concern seems to be to what extent she is ill (categorical identity). In the course of her posting, she builds a relation between herself and the forum members by referring to the community and situational identities. The greeting with forumers is ironic or funny, but also distanced. More precisely, it stresses the institutional location of the interaction: the forum. Line 2, and probably also the dots after the greeting, construct Sylvia as an insecure novice member (situational identity).

Line 6 constructs the forum participants as a special audience. The forum members are addressed as entitled, or experienced enough, to have a grounded opinion on this highly personal issue that Sylvia has never disclosed before. By asking to what extent she is ill, she treats the forum members as in position to judge on her degree of illness. She thus positions herself outside the community. Her concern is reformulated in lines 12 to 15. Sylvia acknowledges that something about her thoughts is ill (line 12). The fact that she has never disclosed these thoughts before, along with the question marks in line 15, indicates that this is a serious problem. It seems that Sylvia's questions are extremely urgent and important, even though she seems to know the answers already. Thus, she constructs a certain distance towards the community through her interactional behavior, while through her insecurity and emphasis on urgency, she seems to pursue acceptance by or inclusion in the community.

In responding to Marcia's intervention, Sylvia displays awareness of Marcia's “control” in lines 24-26. She apologizes and counters the suggestion that she has not gone through the registration process correctly by asserting that she has read the rules and was aware of them while typing her first contribution. The interaction between Sylvia and Marcia illustrates how reference to forum rules is done in this forum, but also that novice members can thereby be put in a not-yet-accepted position. This means that not only is it relevant which rules are operative in the forum, but how they are used and what actions they are used for. Marcia attends to the rules as an execution of a certain power in directing a novice member and possibly thereby as a corrective to inappropriate behavior in the community (cf. Herring’s fourth criterion).

I propose to extend this argument and show that not only reference to rules but also “defining” what forum participation means (cf. criterion 2: purpose and culture) is used to manage inappropriate behavior by a novice member (cf. criterion 4) and to draw limits of community membership (cf. criterion 5). In the thread initiated by Sylvia, the moderator decides not only to intervene when necessary (which is her job), but to contribute a posting herself after a discussion between Sylvia and other members that circled around the question of whether Sylvia was ill or not (including numerous further contradictory descriptions of her behavior and thoughts). If Marcia had not intervened earlier, her posting could be read like any other contribution, since moderators are also HO members who may participate in discussions when not “on duty.”3 But since she already intervened in this thread twice, Marcia's contribution has a particular weight and authority. It can be described as interactionally setting the limits of, and thereby constructing, the HO community.

Extract 5 (thread no. 31)


#Marcia: Hello Sylvia,
so what do you want to hear from the users here in the forum?
I think the fact you have registered here in the forum and actively take part
shows you know yourself that your behavior is not actually normal.
[7 lines omitted]
All the things you describe: constant weighing, wish for an illness to lose
weight, self-hate, all point to the fact that you have problems.
Admit them to yourself and don’t play them down.
I think you can also answer your question whether you are ill or not yourself.
Love Marcia

In lines 4 to 5, Marcia claims that registration for and active participation in the HO forum implies that a person knows that his/her behavior is not normal (although this is not stated on the HO website or in the forum rules). This act of defining membership in the community serves an interactional goal. The statement that participation in the HO forum implies that a person already knows he/she is ill tells Sylvia that she asked the wrong question. Marcia displays irritation about Sylvia’s numerous repetitions of the question of whether she is ill during the thread (line 3). In fact, she seems to reprimand Sylvia for her refusal to hear the responses she received (not shown in extract 3) as answers to her questions. However, the irritation does not seem to be caused by the repetition of the question per se, but by the kind of question. Hence, Sylvia is rebuked for asking an inappropriate question. Marcia indicates the correct way to deal with Sylvia's question/problem, which is admitting that she is ill. Line 14 refers to the question again and rejects it because the author knows the answer already.

Obviously, Marcia sees herself in a position to judge the appropriateness of questions on the HO forum, and, indirectly, Sylvia is considered unsuitable for participation in the forum. It is perhaps unsurprising that Sylvia does not post any response after this and does not contribute to the HO forum again, which means that a person seeking help has been put off. Breaking the rules and repeatedly asking the wrong question has resulted in a failed entry into the forum. This thread thus demonstrates how an interaction analysis based on CA can reveal various aspects or dimensions of online communities, such as norms, values, and purpose.

Participation and Reification from Community of Practice

Not all aspects (of criteria) of online community can be illuminated by an interaction analysis. The criterion of norms and values tied to forum rules refers not just to netiquette or rules as an interactional resource; the list of rules itself is also a piece of discourse. The CA paradigm can be supplemented by an understanding of the status of forum rules adopted from the theory of Community of Practice (Lave & Wenger, 1991; Wenger, 1998). The advantage of this approach is that the interconnection of criteria for community is not dependent on distinct approaches, but the interconnection can be theoretically explained. Referring to the rules and defining what registration to the forum means with regard to identity (e.g., illness) are discursive practices, but they are tied to the fact that there are forum rules and that there is an association (HO) that has installed moderators to check forum contributions. Hence, forum rules are used in the interaction, but they also exist external to interaction. Practice and experience with forum interactions may lead to modifications or changes of the rules, while other practices in forum interactions are sanctioned on the basis of these rules. This interrelation between forum rules and (interactive) practices needs to be taken into account in studying online community. The conceptual framework of Community of Practice offers an understanding of how reified rules and statements co-occur with participation in communities and thus can be used for studying online community.

Community of Practice theory claims that people form communities as they pursue shared enterprises over time (Wenger, 1998). CoPs develop in socio-historical, cultural, and institutional contexts. In this theory, the intersection of issues of community, social practice, meaning, and identity are explored. Using CoP as a theoretical concept renders community and identity non-essentialist, situated, and practice-oriented, as well as intricately connected and mutually constitutive.

The conceptual pair ‘participation-reification’ within the theory of CoP can explain the observations concerning forum rules made in the analyses above. Participation (together with reification) describes the social experience of living in the world in terms of membership in social communities and active involvement in social enterprises. It is both personal and social. Reification refers to the marks produced by human activity, and refers both to the product and to the process of this marking. Reification and participation, as fundamental to the concept of Community of Practice, are not mutually exclusive, but complementary and always in interplay. In fact, participation and reification together construct who we are through the negotiation of meaning. The question whether interaction is a form of participation or a form of reification cannot be answered unambiguously. Wenger (1998) states that words as projections of human meaning are a form of reification, but that at the same time words in speech "affect the negotiation of meaning through a process that seems like pure participation" (p. 62). In other words, interaction is the negotiation of meaning and therefore encompasses processes of both reification and participation. Hence, not only documents such as forum rules are reifications, but also verbal interaction itself has reificative dimensions.

Generally, in online interaction the reificative dimension is stronger than in ephemeral face-to-face conversation. In many web-based forums, interactions are automatically stored in an archive, and are therefore retraceable. As a result, the history of the forum is publicly available. Nonetheless, these reified interactions result from the negotiation of meaning in time and space by various participants. In HO, threads stored in the archive can be modified and deleted by participants themselves. Thus, the reified forum interactions remain subjected to participatory activities. Furthermore, the forum rules have been shaped by the experience of participation of the moderators. Reified in a web-based document, these rules are markers of both the forum’s history and culture (Herring’s criterion 2).

In HO, the most important rules are related to information in postings that may harm or trigger other participants to increase eating disordered behavior. This includes information on amounts of food, detailed descriptions of eating and purging behavior, weight figures, BMI (body mass index), as well as bra cup size and school grades. Low weight figures may cause emulation, panicking, and/or further food refusal; low calorie numbers may lead to fear of edibles. Past experience with forum participation has shown that numbers may trigger instigation, comparison, or competition between members.

Reification and participation are not only closely tied with regard to forum rules. They are also intricately related in the pervasive influence of the institution that employs the forum. The HO forum closes during weekends, temporarily blocking participation. Furthermore, forum moderators have the task of screening postings for rule violations. Since postings can be “broken into” by moderators, participation is also constrained in this sense. However, despite these conditions, members shape their own practice. In other words, despite forces of institutional power, a community can enable member actions and agency. Members develop their own shared norms that are not stated in the forum rules, for instance with regard to which questions are appropriate to ask and which are not.

Analysis 2 above can thus be reanalyzed as follows. Marcia, by referring Sylvia to the rules, invokes the authoritative status of rules, which are reifications of the community's do's and don'ts. Hence, reifications can be drawn upon in participation as authoritative instances. Simultaneously, Marcia's advice to re-read the forum rules makes clear that participation (reading) is required for the reified rules to be operative. In her account (extract 4, lines 24-26), Sylvia confirms this by stating that she has read the rules and that moreover, she had the intention of obeying them. Marcia's posting (extract 5), the typing of which can be seen as a participatory act, includes a definitional statement concerning the nature of participating in the forum. Such a statement can be seen as a reification of the forum's character, since it is done by a moderator and prescribes what members should do (i.e., admit that they are ill).

Note that these postings are publicly available and stored in an archive, which means that they can be read by many other readers in the future. Therefore, a definitional statement such as Marcia's is persistent, which is another argument for calling it a reificative act. At the same time, this act of reification is used for a participatory (interactional) goal, namely sanctioning Sylvia. This exploration implies that if we adopt CoP's understanding of the intricate connection between reification and participation, analysis of members’ interaction in web-based environments and analysis of the influence of extra-interactional rules can be fruitfully combined. This seems worthwhile for the study of online community.


This article has explored the use of the CA paradigm for the analysis of online community. The analyses of a German eating disorders forum illustrated that through the enactment of discursive identities, community may be attended to and invoked. Novice and experienced member identities indexed members' positions in the forum, and the forum was ascribed community characteristics that made a novice feel welcome and safe.

Such discursive identities and their deployment to accept (or not accept) a novice were also accompanied by reference to forum rules (extracts 4 and 5). Although interaction analysis revealed how forum rules were articulated to negotiate membership identities in the community (both the moderator's authoritative position and the novice's as not-yet-accepted), the workings of forum rules cannot be understood adequately from a CA perspective alone. I proposed that CoP's dual concept of participation-reification can help to understand and theorize the phenomenon of forum rules, as well as other reificative aspects of online community, such as archives. Hence, CoP seems helpful for the study of online community, especially when forum rules are included as relevant.

The extent to which CoP and interaction analysis can be combined further (for instance regarding membership, membership boundaries, and the role of learning) requires further investigation. In addition, more systematic research on web-based platforms is necessary to illuminate the exact relation between interactional activities and the criteria for online community, so as to understand more precisely what CA and CoP can reveal about online interactive behavior.


  1. For ethical reasons, all names used in extracts are pseudonyms. For examples of “real” HO nicknames, see Stommel (2007).

  2. For instance: ‘Welcome to the forum. I took the figures and food proportions from your contribution, see our rules, in order to avoid that other users can compare.’ [German original: Herzlich willkommen im forum. ich habe die zahlen- und lebensmittelangaben aus deinem beitrag genommen, siehe unsere regeln, damit andere usern nicht zu vergleichen verleitet werden. (thread no. 19)]

  3. Nonetheless, even when moderators post “as a member” they are recognizable as moderators, since their nicknames, appearing with the contribution, are marked with the label ‘moderator.’


Antaki, C., Ardévol, E., Núñez, F., & Vayreda, A. (2005). "For she who knows who she is:" Managing accountability in online forum messages. Journal of Computer-Mediated Communication, 11(1), article 6. Retrieved July 29, 2008 from http://jcmc.indiana.edu/vol11/issue1/antaki.html

Antaki, C., & Widdicombe, S. (1998). Identities in talk. London: Sage.

Baym, N. (1998). The emergence of online community. In S. Jones (Ed.), Cybersociety 2.0: Revisiting computer-mediated communication and community (pp. 35-68). Thousand Oaks, CA: Sage.

Benwell, B., & Stokoe, E. (2006). Discourse and identity. Edinburgh: Edinburgh University Press.

Billig, M., & Schegloff, E. (1999). Critical Discourse Analysis and Conversation Analysis: An exchange between Michael Billig and Emanuel A. Schegloff. Discourse and Society, 10(4), 543-582.

Cherny, L. (1999). Conversation and community: Chat in a virtual world. Stanford, CA: CSLI.

Drew, P. (2005). Conversation analysis. In K. L. Fitch & R. E. Sanders (Eds.), Handbook of language and social interaction (pp. 71-102). Mahwah, NJ: Lawrence Erlbaum.

Döring, N. (2003). Sozialpsychologie der Internet; die Bedeutung des Internet für Kommunikationsprozesse, Identitäten, soziale Beziehungen und Gruppen (2nd ed.). Göttingen: Hogrefe.

Edwards, D. (1997). Discourse and cognition. London: Sage.

Herring, S. C. (2004). Computer-mediated discourse analysis: An approach to researching online behavior. In S. Barab, R. Kling, & J. Gray (Eds.), Designing for virtual communities in the service of learning (pp. 338-376). New York: Cambridge University Press.

Höflich, J. (1996). Technisch vermittelte interpersonale Kommunikation. Opladen: Westdeutscher Verlag.

Lamerichs, J., & Te Molder, H. (2003). Computer-mediated communication: From a cognitive to a discursive model. New Media & Society, 5(4), 451-473.

Lave, J., & Wenger, E. (1991). Situated learning: Legitimate peripheral participation. Cambridge: Cambridge University Press.

Leiberich, P., Nedoschill, J., Nickel, M., Loew, T., & Tritt, K. (2004). Selbsthilfe und Beratung im Internet; Mündige Benutzer können die Arzt-Patienten-Beziehung neu gestalten. Medizinische Klinik, 99(5), 263-268.

Psathas, G. (1995). Conversation Analysis: The study of talk in interaction. Thousand Oaks, CA: Sage.

Sacks, H. (1992). Lectures on conversation. Oxford: Blackwell.

Schegloff, E. (1997). Whose text? Whose context? Discourse and Society, 8(2), 165-187.

Schegloff, E. (2007). A tutorial on membership categorization. Journal of Pragmatics, 39, 462-482.

Sneijder, P., & Te Molder, H. (2006). Disputing taste: Food pleasure as an achievement in interaction. Appetite, 46, 107-116.

Stommel, W. (2007). “Mein Nick bin ich!” Nicknames in a German forum on eating disorders. Journal of Computer-Mediated Communication, 13 (1), article 8. Retrieved September 3, 2008 from http://jcmc.indiana.edu/vol13/issue1/stommel.html

Wenger, E. (1998). Communities of practice: Learning, meaning and identity. Cambridge: Cambridge University Press.

Wetherell, M. (1998). Positioning and interpretive repertoires: Conversation Analysis and Post-Structuralism in dialogue. Discourse and Society, 9(3), 387-412.

Appendix: Original German Extracts

Extract 1 (thread no. 22)


ziemlich unsicher
Hallo mal an alle!
Ich bin mir nicht wirklich sicher ob ich hier richtig bin, [...]

Extract 2 (thread no. 7)


bin ich hier wohl richtig?
Hallo erstmal.
also ich habe mir ziemlich lange überlegt, ob ich mich hier anmelden soll oder nicht. […]

Extract 3 (thread no. 33)


Hallo bin neu…
Hallo ihr,
Ich bin noch ein bisschen unsicher was ich schreiben soll. Ich bin [18-22] Jahre alt und habe seit 8 Jahren eine ES. Das geht in Richtung Bulimie, manchmal habe ich aber auch Hungerphasen. Aber mittlerweile geht es mit der ES. Borderline ist mir auch nicht unbekannt, weil ich deswegen unter anderem eine ambulante Therapie mache. Vor Weihnachten hatte ich einen SVV-Rückfall, aber das ist jetzt auch wieder ok. Naja, von meinen anderen Symptomen will ich gar nicht erzählen, weil ihr dann sicherlich denkt, dass ich nur aus Symptomen bestehe.
Liebe Grüße, Antonia





Hallo Antonia,
erstmal auch von mir Herzlich willkommen hier im Forum.
von meinen anderen Symptomen will ich gar nicht erzählen, weil ihr dann sicherlich denkt, dass ich nur aus Symptomen bestehe.
Das denke ich auch manchmal von mir das ich nur noch aus Symptomen bestehen. Aber ich hab einen großen Respeckt das du trotz allem immernoch stark bist und den entschluß gefasst hast zu kämpfen, wie ich gelesen habe bist du auch in Thera?!Wegen der SVV?
Ich bin auch [18-22] und leide auch an ES seit einiger zeit auch mit Hungerphasen.
Glaube keiner denkt hier über einem man hat der oder die viele Symptome. Schließlich sind wir doch hier um über alles reden zu können oder?
Würd mich freuen auch von dir noch was zu erfahren. bis denn
LG Joe



[…]@ Joe:
Weisst Du, es ist mir nur so peinlich, dass ich so viel mit mir rumschleppe. Ich denke halt immer, dass andere dann sagen könnten, ob da noch mehr kommt und dass ich ja schon die ES habe und ob das nicht schon reichen würde. Aber ich suche mir die Symptome ja auch nicht aus. Aber mir kommen dann so komische Gedanken. Keine Ahnung wieso.[…]



ich kann das gut verstehen, kenn das von mir ich schäme mich auch manchmal wieder über meine Problemchen mit freunden zu sprechen oder wenn ich mal wieder ein Tief habe und mich nur vergraben könnte wenn ich mal Depressive stimmung habe lasse ich mir das auch nicht anmerken um ja nicht wieder gesagt zu bekommen man ist der öde wieder was, und was denn jetzt ich kenn das aber manchmal ist das wichtig alles raus zu plaudern das kann wunder wirken. Ich kann das auch nur hier im Forum, denn Privat hat niemand die probleme die ich habe, was jetzt ES und so angeht.
Hoffe ich kann dir was helfen oder anregungen geben bis dann und
Liebe grüße Joe

Extract 4 (thread no. 31)


bin ich gefährdet?
Hallo ihr forumaner...
also eigentlich weiß ich gar nicht genau was ich hier schreiben soll eigentlich will ich wissen ob andere meine gedanken krankhaft finden also ich hab meine gedanken noch nie wirklich jemanden so offengelegt was dieses thema betrifft
weiß nicht so genau wo ich anfangen soll. ich wiege mich [*] mal täglich
[21 lines omitted]
ist das jetzt krank? ich mein irgendwie ist es schon krank...aber ist es SO krank?
ist es krank zu denken zu hoffen das man irgendwie doch noch dünner wird????
ich dank euch schonmal
[Liebe Sylvia. Bitte keine Zahlenangaben zu deinem Gewicht oder was damit zu tun hat und keine Beschreibungen wie du dich Verletzt. Lies dir doch nochmal die Regeln durch. Liebe Grüße Marcia]


@Marcia: entschuldigung, ich hatte zwar die regeln gelesen aber ich hab dann irgendwie nich so genau drauf geachtet..also ich habs schon versucht aber das muss ich dann wohl überlesen haben

Extract 5 (thread no. 31)


Hallo Sylvia,
also was willst du denn von den Usern hier im Forum hören? Ich denke, dass du dich hier in diesem Forum registriert hast und dich aktiv beteiligst zeigt doch, dass du selbst weißt, das dein Verhalten eben nicht normal ist.
[7 lines omitted]
Diese ganzen Dinge die du beschreibst: andauerndes Wiegen, Wunsch nach einer Krankheit zum Abnehmen, Selbsthass, deutet alles darauf hin dass du Probleme hast.
Gesteh sie dir ein und spiel sie nicht herunter.
Ich denke deine Frage ob du krank bist oder nicht kannst du dir auch selbst beantworten.
Liebe Grüße Marcia

Biographical Note

Wyke Stommel [wykestommel@yahoo.com] recently completed her Ph.D. in Sociology at the University of Frankfurt, Germany. Her research interests include interaction, discourse and computer-mediated communication, and illness, identity, and community.


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