First version presented at ReLive 08 at the Open University, UK, November 2008.

This is the longer version prepared for publication as follows

Carr, D. Oliver, M., Burn, A. (in press 2010) ‘Learning, Teaching and Ambiguity in Virtual Worlds’, in Researching Learning in Virtual Worlds. Peachey, A, Gillen, J, Livingstone, D, Smith-Robbins, S. (eds) UK : Springer  

Learning, Teaching and Ambiguity in Virtual Worlds

Diane Carr, Martin Oliver and Andrew Burn

London Knowledge Lab, Institute of Education, University of London





What might online communities and informal learning practices teach us about virtual world pedagogy? In this chapter we describe a research project in which learning practices in online worlds such as World of Warcraft and Second Life were investigated. Working within an action research framework, we employed a range of methods to investigate how members of online communities define the worlds they encounter, negotiate the terms of participation, and manage the incremental complexity of game worlds. The implications of such practices for online pedagogy were then explored through teaching in Second Life.

Second Life eludes simple definitions. Users, or ‘residents’, of Second Life partake of a range of pleasures and activities – socialising, building, creating and exhibiting art, playing games, exploring, shopping, or running a business, for instance. We argue that the variable nature of Second Life gives rise to degrees of ambiguity. This ambiguity impacts on in-world social practices, and has significant implications for online teaching and learning.     

Background: the ‘learning from online worlds’ project

This chapter focuses on research undertaken during a small project called ‘Learning from Online Worlds; Teaching in Second Life’. This project involved the investigation of learning practices in online worlds and the extension of this work through the design and delivery of taught classes in Second Life. The project’s aims included theorising the learning that happens in social worlds, exploring how any such learning might inform teaching in virtual worlds, and investigating the various cultural factors (such as subjectivity and identity) that might impact on pedagogy in these contexts.

Given that there was a relative shortage of available research that might provide a theoretical grounding for the project, an action research approach was adopted (Zuber-Skerrit, 1992). This provided a framework in which experience could be gained, theorised and used to plan further investigation in a structured yet emergent manner. The first iteration of the action research process was primarily of value in establishing the issues that the remainder of the project should pursue.

The first phase of the research was thus experiential, and began with exploring Second Life as users. We socialised, attended seminars and meetings, experimented with in-world creative practices (from making clothes, to creating machinima), and kept ‘game diaries’ of our experiences, which allowed us to move into a reflective phase of work. The following is a sample of this early material:  

“I’ve been spending time on the] customising of my avatar. It’s easy, quick, and the improvements are obvious. That leads to learning (the acquisition of a basic familiarity with the edit appearance settings, the inventory). And from there to despair (‘oh, I look like a duck’) and from despair to shopping, and from shopping to freebies […] In retrospect I think of this as my pre ‘pain barrier’ phase […] Because I battered away and persevered […] I eventually scraped some skills together and stumbled across this pain barrier, at which point SL became funny, enjoyable and potentially interesting.” (DC’s ‘game diary’ March-June 07)

The team compared diaries, and as a theorising and planning step, identified a set of issues relevant to the project’s central questions. These issues included:

  • Expertise (how is it demonstrated, measured or performance?)
  • Conventions (socially produced) relating to expertise, identity, etiquette, trust, etc.
  • Learning curves and the ‘Second Life pain barrier’
  • Credibility, ‘noobs’ (new users) and hostility or ‘gatekeeping’
  • Self-presentation, representation.
  • Drama and performance
  • Public spaces, social constructions and ritual spaces (the‘magic circle’)
  • Voice and access issues
  • Creative and collaborative practices (such as machinima).

These issues were then explored in greater depth through particular research activities, each of which built on the previous iterations of the research process.

We referred to the emerging research in this area, including educators’ blogs, wikis, reports (such as Kirriemuir, 2008) and conference proceedings (such as the annual Second Life Community Conference Education Track). At this stage much of the reportage is exploratory, and remains difficult to coherently review due to the range of educational contexts (from classroom based building work, to distance learners’ media production, for example) and the variety of learners (from children to adults), as well as the diversity of disciplinary affiliations and perspectives of those engaged in research (from computer sciences to media practice for instance). These resources are, however, evidence of the amount of interest in this area, and the range of work being undertaken. In addition to reviewing this material we drew on literature from digital game studies, media and cultural studies, Internet studies, drama education theory; ICT and education research, and communities of practice theory (Wenger 1998).

Defining Second Life; Doing Second Life research

During our individual experiences we had identified an initial period in which using Second Life was frustrating, even annoying; this ‘pain barrier’ had to be overcome before we felt comfortable as users of the environment. We provisionally defined this pain barrier as the moment when sufficient learning or competence has been accrued to tip the new user from bafflement and annoyance, to pleasure, or even ‘flow’ (Csikszentmihalyi, 1990).

Given that most of our students would be entering Second Life for the first time, and would be doing so at our instigation, we had an obligation to understand the difficulties posed by this virtual world for beginners. Such difficulties included issues about the interface, as well as the public and potentially intimidating nature of this virtual world. Additionally, we expected that an understanding of the pleasures and the potential frustrations of Second Life should inform the planning of our classes.

First we needed to ascertain if pain barrier experiences were at all commonplace. To find out we proposed to gather short accounts of the journey from newbie to resident, from self-described residents. We wished to be non-intrusive, and thus we posted a message identifying ourselves and making our request at a Second Life forum, rather than approaching subjects in-world. Requests from students and researchers are common at the various Second Life forums. Here is a sample response in which a pain barrier moment is described, while pleasure and frustration are attributed to particular aspects of Second Life. SV wrote that 

“[at first] I hated it! […] All I saw was walls! I had no idea what was where, it was totally disorientating! […] I just couldn’t get used to it. It was only when one of the guys I came here to be with from the old chat [room] asked me to come in for another’s birthday that I did and it just clicked, it was then, in March, I felt ‘right’, it all came together”. (SV by email)

This response suggests that there is value in the notion of a pain barrier. Meanwhile, however, a dozen contributors at the forum had responded to the eliciting message with scepticism or outright disbelief. Posters suspected, for instance, that our message was from impostors engaged in some form of ‘scam’. Furthermore, it was suggested that we were not conducting proper research which (as some of the posters explained) involves clinics, large amounts of data and/or observing people without their knowledge in order to ‘get at the truth’. While these forum responses were of limited value in relation to our interest in the idea of a Second Life pain barrier, they were intriguing in terms of our interest in hostility, control and credibility in virtual communities. We were also interested in the respondents’ comments about research.

Prompted by the forum respondents’ suggestions as to what constitutes ‘proper research’ we found ourselves reflecting on the manner in which we ourselves had implicitly constructed Second Life in this instance. We had focused on the experiences of individuals, and had assumed a ‘human subject model’ (see Bassett and O’Riordan, 2002). What alternative approaches might there be to the analysis of hostility, territorialism, community or credibility in Second Life?

One alternative would be to analyse these issues while conceptualising Second Life as a ‘collective text’. Within this text, particular discourses emerge and circulate. Within this discourse, certain constructs or agencies might emerge. For example, an agency – invisible, vague yet omniscient – could manifest in discourse as ‘we’ or a ‘them’, as ‘the community’ or an ‘everyone’. Such agencies would function as figures or reference points, and would be used by participants attempting to, for instance, establish the terms of legitimate (and hence illegitimate) participation. The point of these speculations is not just to suggest the value of alternative methodological approaches to virtual worlds. What this is intended to highlight is the potentially circular relationship between a preferred yet perhaps implicit definition of Second Life, and research-as-practice.

Second Life can be used in various ways and thus there is scope to define it in various ways (as collective text, programme, social networking platform, tool, public space, etc.).  This suggests that definitions will be provisional, and reflect the perspective of the user (or the disciplinary perspective of the researcher). The preferred definition of Second Life will inform the research questions that are devised, and the methodologies that are adopted. All of which will impact on analysis and findings.  None of this is ‘bad’ of itself, but failing to recognise this circularity could be detrimental.

Furthermore, we found that the difficulties associated with defining Second Life can be associated with tensions within what might be described as the Second Life community. These issues of fluidic definition, ongoing negotiation, varying expectations, different practices, dissent, and emergent conventions together contribute to what we have described as the ambiguity of Second Life.   

Contested definitions and ‘community’

Our attempted explorations of the pain barrier had raised the ambiguity of Second Life as an issue. This suggested the importance of exploring how residents were constituted in (and in relation to) Second Life. Our next phase of research was designed to focus on this process of community constitution and maintenance. Conflict over the definition or ‘real meaning’ of Second Life are present in Second Life residents’ discussions. We looked for an example of this that would allow us to investigate how users negotiated their community involvement, and what kinds of things counted as legitimate participation for them. Tensions relating to these issues were heightened during 2007 as a result of the proposed introduction of an integrated voice feature by the developers, Linden Labs. In discussions at Second Life forums during this period there was anxiety that fundamental aspects of Second Life would be altered as a result of the new feature.

We appreciate that forums have their own conventions and that Second Life forums should not be conflated with Second Life itself. However, what the forum offered was a ready-made display of contested definitions of Second Life and references to legitimacy and practices of exclusion, authored by Second Life users.  We explored this in more detail while drawing on notions of communities of practice and legitimate participation drawn from Wenger (1998).

To do this, we identified a particular 21 page thread on a popular Second Life forum. The originating post in this case was from a self-described deaf Second Life resident expressing dismay at the imminent arrival of the integrated voice feature. In an iterative process, we analysed 13 posts on this thread to generate a set of identifiable rhetorical and discursive strategies. We then reviewed 100 posts to expand and clarify this set.  We looked to the identification of objects and agents, claims about these, the classification of claims, counter-claims, and the construction of self, other and difference through these manoeuvres.  This material was then organised according to its including or excluding function. By these means it was possible to map the work undertaken by forum participants as they performed particular identities, set up and attempted to enforce the terms of community and legitimacy, and defined Second Life itself. 

Across the posts it was possible to identify inclusive rhetoric, where affiliation was claimed. Membership was performed, for instance, through declaration or reference to Second Life ‘belonging’ – in assertions about how long a person has been a resident, through mutual recognition, or by displays of technical vocabulary and in implicit or explicit references to ‘us’. Affiliation and inclusion was expressed in terms of affect (expressions of pleasure and contentment relating to Second Life use), and in relation to popularity and friends. 

Similarly, it was possible to identify attempts to exclude, as when posters made declarations of apartness or distinction, for example, through reference to self ‘I’m not…’ and through phrases such as ‘your kind’ or ‘them’; in expressions of discomfort, or even in apparently supportive efforts which in actuality emphasized the distance between those identities constructed as normal, whole, or neutral, and those not (‘I can’t begin to imagine what it is like for people like you’).  Posters would situate themselves ‘within’ the Second Life community, positioning themselves and others in certain ways, while harnessing particular strategies. The original post became the ground from which claims and counter claims were made about what is realistic, fair or democratic; what is controversial, acceptable or admirable.

The introduction of the integrated voice feature stirred these debates because it threatened to undermine what were held to be definitive aspects of Second Life: its separateness from participants’ ‘first’ lives, and the option of anonymity. If voice became the default mode of communication in social settings, those not using voice (it was proposed) would be compelled to explain their choice. The use of voice was linked to issues of disclosure and trust, and such concerns linked to discussions about what Second Life is.

One definition in the thread that surprised us was the description of Second Life as a game. We had been certain that Second Life was not a game, because we share a background in game studies and we are accustomed to defining games according to criteria drawn from game studies literature (see, for example, Salen and Zimmerman, 2004). Games are played, and incorporate various modes of play. Games have goals, chance, rules, and discernable outcomes – which are absent in Second Life. So we were struck when posters to the forum referred to Second Life as a game or even a ‘videogame’ (and these claims remained uncontested). For these posters, it was the separateness of Second Life from real life, the value of this separation, and the play in fantasy and identity that the distinction affords, that meant that Second Life could be categorised in this way. As an aside we would note that this raises interesting questions about our reliance on structure-orientated definitions within game studies, and the ‘grey area’ where games and play combine.

These issues emerged in the context of a discussion about the introduction of the integrated voice feature, because posters considered that voice was being introduced to service those sectors, such as business, that were being courted by the developers at the expense of ‘real residents’. Across this thread, then, it was possible to identity tensions relating to identity and ambiguity, the defining of Second Life, and references to its ‘proper use’ being aired by participants in the context of a discussion where legitimacy and inclusion were at stake.

During the project we looked at inclusive and exclusive patterns at the forum, rather than specifically at the issues referred to in the original post such as deafness and access. It was clear, however, that online communities and online worlds offer researchers a location to examine socio-cultural aspects of disability. At the conclusion of this project we returned to these concerns and conducted follow-on research into the impact of the voice feature on deaf residents (Carr, submitted 2009), and issues pertaining to voice, offline identity and ‘immersion’ in Second Life (Carr and Oliver, in press).

Investigating tensions on the Second Life forums was useful in terms of our understanding of patterns of participation and community within this virtual world. The variability in modes of participation, however, can complicate attempts to document specific instances of learning (outside of educational settings). To investigate learning practices within virtual worlds, yet outside of educational contexts, we turned to the Massively Multiplayer Online Role-Playing Game, World of Warcraft

Learning in online game-worlds

World of Warcraft offers multiple modes of participation, yet play involves interacting consciously with particular structures and constraints. While the game incorporates variability and different modes of play, play styles and preferences, particular constraints or structures will – in part at least – define the experience and the ‘text’. It has a generically definitive rule set, for example, that marks it as a role-playing game. This means that characters ‘level up’ in experience points by performing particular tasks, and while specialising in certain traits (such as physical power or particular forms of magic, for example).

Play within World of Warcraft can be dramatic, variable, collaborative, social and expressive, but aspects of the game-play (rules, progression, goals, for example) are more defined than that which is on offer in general within Second Life. The game world has a geographic cohesion, with roads and transport hubs, major cities and public spaces as well as wild habitats and borderlands. Avatars or characters in the game do reflect player preference, in that customization involves the ‘look’ of an avatar (its hair colouring, for example), as well as choice of species or profession, and the ongoing collection of particular equipment to further support specialist skills (such as better healing powers, or greater resilience). While there is variation, the avatars conform to particular, customizable templates. The game is ‘owned’ by its developers, but it only ‘lives’ because of the input of its subscription-paying users. (For MMORPG analysis, see Taylor 2006, or Ducheneaut et al, 2006). 

While learning obviously happens in this game world (otherwise players would not progress), studying learning practices is not straightforward. We did not want to divorce players from the real-world contexts of play, yet we wanted to focus on learning that emerges through players’ participation with a game world and other players. Our solution was to focus on couples who play the game together, while sharing a real space.

We recruited through online guilds and real-world social networks, and interviewed ten people (four heterosexual couples and one mother-son pairing). In doing this, our previous experience in research and playing online games – we had been casual players of World of Warcraft for around 18 months when this project began, and we still play – proved extremely valuable in making initial connections to willing participants.

We interviewed our couples in game at a location of their choosing, and chat-logged the text-based semi-structured interviews, each of which lasted between 60 and 90 minutes.  The resulting transcripts were split between the authors, and the contents categorised. These separate categories were then jointly reviewed and reconciled to produce our initial categories, which included references to:

  • Who got who started and how
  • Assessments of increasing competence (and incompetence)
  • Help – from mentoring to ‘backseat driving’
  • Guilds
  • Affect
  • Domestic space and assets (‘best’ chair or computer, for example)
  • Alts (second or alternative characters)
  • Gender
  • Time constraints
  • Relationships (in game, in guilds, being a couple, etc.)

After organising the data into these categories, they were reviewed to identify common elements. This led to the identification of ‘management of resources’ as the central or organising concept, to which the others were all related. As we explained:

“Through an engagement with and analysis of this material we developed a framework through which to consider learning practices. This involved looking at increasing competence in relation to the management of resources, where ‘management’ involved the recognising, negotiating, accessing and applying of ‘resources’, which might be categorised as material, ludic or social. In actuality these three classes of resource constantly intermingle, yet these provisional distinctions allow us to gesture to specific examples of learning practice, and to the complexity of competence in this context.” (Carr and Oliver, 2009)

In other words, as players’ competence developed, they were able to identify and leverage an increasingly complex array of in-game resources, while negotiating real-world resources and demands. In fact we identified three inter-relating categories of resource: ludic, social and material.

Ludic resources involved the mechanics and economics of the game itself; learning how to play, how to ‘level up’ and negotiate with the various options (forms of specialization and strategy, for example). Social resources include peer assessment and mentoring, collaboration, player relations and networks (such as guilds) that can be drawn upon to support action. Actual or material resources, meanwhile, included the setting of play and the various factors (time, technology, game accounts and childcare) that may impact on play. Through this process, the interconnections between the real world contexts of play, and activities within a game-world were explored. 

In a second phase of this research, our understanding of player competence in relation to these resources was developed further using concepts drawn from Wenger’s work on Communities of Practice (1998). Specifically, players’ accounts were explored in terms of trajectories of participation and multi-membership, looking for evidence that practices had been performed in a way that met the expectations of the communities which the players were part of (see Oliver & Carr, 2009). Typically, that would involve balancing the demands of the partner (e.g. to ensure housework is done, time spent together, and so on) against those of other players (e.g. to take up an important role in a guild event or raid), for example, in terms of how much time can be spent on each. Other examples included creating several characters, so that one could be played with a partner and another with other groups; or developing complementary areas of expertise, so that one partner could advise the other on useful User Interface ‘Mods’ which would customise the game in a helpful way. 

The general pattern reported in these players’ accounts was of successful learning – understood as taking an increasingly responsible role in a community’s practices, and requiring the acceptance of that community both of the person and their actions. However, there were exceptions. For example, one couple actively withdrew from a guild where other members had different priorities as regards spending time in the game; another participant had chosen to give up playing temporarily while their partner continued. This had involved ending some in-game relationships, and shifts in the time spent on other activities including house work. In both these cases, from a community of practice perspective, the participants were still learning as a result of the game. In the first example, what the couple learnt was about who or what they were not (e.g. that they were casual players not end-game raiders). In the second, the participants learnt about different priorities, and had to negotiate a way of balancing these that they both found to be acceptable.

While some of the ludic elements that featured in this study are clearly specific to online multiplayer games such as World of Warcraft, many of the social and actual resources involved would be the same for the use of any virtual world. Second Life users might manage different resource sets, or in different proportions, but we would nevertheless expect that learning practices in relation to the management of resources could be documented in Second Life. For the sake of coherence it may be necessary to narrow the inquiry to identifiable communities of practice (see, for example, Burn, forthcoming). In such a case, the terms of legitimate participation might function as the implicit curriculum, for instance. To return to the example of World of Warcraft, it is interesting to consider this game as a combination of structures that support social and experiential learning and to ask – then – what games might teach us about curricula and pedagogic design. Questions of design are returned to in the discussion that follows, where learning and teaching are considered in relation to Second Life’s amorphous tendencies.   

Teaching in Second Life 

In this section we give an account of our teaching in Second Life, with an emphasis on the issues of definition and ambiguity discussed thus far. This ambiguity should not be assumed to be a problem in a teaching context. On the contrary, as we have argued elsewhere (Carr, 2009), ambiguity has the potential to unsettle or de-naturalise aspects of our roles (as teacher, learner or researcher) – which could be considered one of the most interesting aspects of a virtual world for educators.

We taught four sessions in Second Life over two terms (see Table 1). These taught sessions were designed to fit into existing MA (postgraduate) programmes, rather than being designed to fully exploit or exhaustively test the various potentials of Second Life. We documented this work on a report posted to the project blog, ‘Learning to Teach in Second Life’ (Carr, 2008). The following discussion incorporates material drawn from this report.

Table 1: The four classes in Second Life


Class no. Course module Class format
1 Computer Games, Gaming Culture and Education(from the MA in Media, Culture and Communications)

Topic: Machinima 

Guest presenter and studio tour

4 students plus a guest, 2 facilitators, 1 guest presenter, 1 host (during tour)

2 Computer Games, Gaming Culture and Education(from the MA in Media, Culture and Communications)

Topic: Fan practices and role-play

Guest presenter

3 students plus 1 guest/informant, 2 facilitators, 1 guest presenter.

3 Computer Mediated Communications(from the MA in ICT and Education)

Topic: Virtual world research i. Ethics

3 facilitators, approximately 15 students

4 Computer Mediated Communications(from the MA in ICT and Education)

Topic: Virtual world research ii. Discussion

3 facilitators, approximately 15 students


Methodologically, this phase of work echoed the reflective, experiential model of the first phase; however, we also drew on students’ evaluations and general reflections about the Second Life sessions. These were posted by students to forums at Blackboard, the more conventional Virtual Learning Environment that was used for the course as a whole. Additionally, we chat-logged our sessions in Second Life (i.e. recorded the live text-chat that participants typed), and conducted post-session interviews with a course tutor, as well as follow-up interviews with students. The students were informed of our research at the start of their module, and the sessions in Second Life were elective. Students who did not wish to be involved with this research could opt out and instead follow the sessions as chat-logs that were posted to Blackboard.

The majority of participating students (especially the distance learners) found the Second Life sessions enjoyable and productive. The students compared Second Life to Blackboard, commenting that while Blackboard offered structure, it was socially ‘dry’. Second Life, on the other hand, was highly motivating but occasionally anarchic or chaotic. The Second Life sessions offered very welcome real-time and virtual-space contact with peers and tutors. Most of the students, however, did not value Second Life over Blackboard. They appreciated the offers of both and described the identified differences as complementary. See, for example, this student’s comments:

“So I thought the second [Second Life] session was great with the small groups. In comparison to Blackboard, there is that ‘live’ element – instant reactions in SL, compared to […] Blackboard. Usually when making a posting on Blackboard, I do put a lot of thought into it – I often type them up in Word first, read them, correct mistakes […] Blackboard gives me time to think.” (Student interview, AM)

While most of the feedback was very positive, the students were not uncritical. The few students who had real-space access to peers and tutors were unimpressed by the Second Life sessions. Some of the students struggled with following text discussions, and it became clear that participating in discussions in Second Life with confidence is an acquired skill. While we had calculated on students having to familiarize themselves with the interface and the basics of avatar movement, we had not fully appreciated the problems associated with text-chat for beginners.

In terms of learning and content, the Second Life sessions worked because the topics (ethics in social world research, for example) were no longer abstract: the students were considering notions of privacy, consent, avatars, self-presentation and identity, while they were themselves in-world, as avatars. As this suggests, working and learning in Second Life can involve shifts in perspective (and in perceptions of research). Such shifts, while very productive, may prove disorientating for some students. Any such disorientation might be very productive if, for instance, it leads to discussions about the nature of research, or the relationship between a researcher and his or her object of study. 

For tutors, Second Life sessions require significant amounts of preparation (emailing reminders, posting material to the VLE, writing and distributing notecards), and they are quite labour intensive to run. When possible we divided roles across two or three tutors. This left one tutor free to ‘direct’ the session, greet stragglers or manage technical problems using instant messenger; one to lead the discussion or present and, if possible, another tutor free to be immersed in the discussion itself, responding to the multiple input from the students with counterpoints or questions.

Both students and tutors found that the Second Life sessions to be quite intense and draining. While four students might constitute a very small group in a real classroom, it can feel like a sizable group in Second Life – and it is probable that the larger the group, the more necessary some kind of formalized structure becomes (especially in those situations where the participants are not experienced users of Second Life). As this suggests, classroom conventions do not translate in a predictable fashion. Over our four classes we moved from less structured, small groups, to more structured larger groups, and from a more exploratory format (tours and guests) to a more conventional format (group discussions held in-the-round, for example). These are not evaluative descriptions – we do not propose that one of these approaches is ‘better’ than the other on the basis that it is more innovative, or less creative. What we find important, rather, is the manner in which a set of factors (see Table 2) combined during a session to engender particular affect.

 To identify these factors, we reviewed students’ comments and identified instances where affect (confusion, anxiety) could be linked to specific aspects of a session, and considered such elements across all four of the sessions. Session 1 and 2 were less formally and less familiarly structured than session 3 and 4. Additionally, the relationship between Second Life as phenomenon and the content of the meeting itself were less obvious in sessions 1 and 2 than they were in sessions 3 and 4. The various factors we found to contribute to greater and lesser degrees of ambiguity are summarised in Table 2.

Table 2: The factors in the sessions that contributed to or countered ambiguity


Classes 1 and 2

Classes 3 and 4

Elective activity

Core activity (but still voluntary)

Smaller group

Larger group

Guest presenters (with tutors present)

2 of the 3 tutors were known to the class

Course taught in mixed mode of delivery: via conventional VLE (First Class), 2 days face-to-face residential teaching, 2 elective sessions in Second Life.

All distance learning (on Blackboard) with 2 Second Life sessions.

Mix of full time/on campus students, with part-time distance learners

All distance learners

Less straightforward relationship between course content (computer games, gaming and education), and Second Life as a whole, or as a phenomena 

Obvious links between course content (computer mediated communications and education) and Second Life as phenomena

Some of the course literature (on simulations and role-playing for example) does apply to Second Life.

Clear relevance of much of the course literature and set reading.

Session format: Tours, guest presenters, projected images, various locations.

Familiar discussion and presentation format, in one location.

                    (towards greater ambiguity)

                      (towards reduced ambiguity)


Depending on the degree to which a session is structured, and depending on the management of elements such as those mentioned in the above table (and this is not an exhaustive list) we would propose that ambiguity (of place or role, for instance) might be either amplified or suppressed according to the teacher’s needs. A degree of disorientation or ambiguity might be productive in one learning context yet completely counter-productive in another. When planning teaching in virtual worlds, then, pedagogic structures could be imagined as constituting an aperture through which ambiguity can be incorporated and managed. ‘Managed confusion’ can be pedagogically useful if, for instance, it confounds expectations, exposes assumptions and promotes reflection. One difficulty, however, is that effectively managing these affective aspects of learning can be a problem in a virtual world where the teacher might have surprisingly little access to real-time feedback.

The fact that so many things are possible in Second Life means that every decision is open to question. Why take the appearance of a humanoid? Why ‘sit’ in a chair? Why reproduce a lecture theatre? The unfamiliar format rendered pedagogic design ‘visible’ to our students. This is potentially significant in relation to education studies. The students questioned the various pedagogic decisions that had been made, wondered about the role of student (or ‘good’ student), and the expectations of their tutors:

“The ‘real class’ occurred when we were asked to sit around [the tutors…But what] was the tutor’s perception towards the students’ participation? Did they give credit to students who answered most of the questions [?…What if] Second Life crashed and it took a long time sometimes to restart? The tutor might think that this student participates less” (Ae’s report)

We propose that it would be productive to further explore the uses and management of ambiguity in virtual world learning contexts. Furthermore, we wonder if it would be illuminating to consider these issues while drawing on theories of game design. Second Life is not a game per se, but nonetheless it is arguable that game design theory (see Salen and Zimmerman 2004, for example) could usefully inform and help to develop our thinking on virtual world pedagogy.  The relevance is not due merely to interactivity or ‘immersion’. What is significant, we suggest, is that successful, engrossing games blend the conditional with the experiential. ‘Conditionality’, in this instance, involves rules, etiquette, software or curriculum, whereas the ‘experiential’ involves play, agency, learning, improvisation and collaboration.  Game designers have to think about real-time experience, location, context and constant feedback. Games such as World of Warcraft involve the clear communication of goals, yet also allow scope for the player to exercise prerogatives and preferences, experiment, and make mistakes. Such games require effort and persistence. Learning during play involves (as we have argued) the discovery and application of various resources.  Successful games mix specific yet potentially adaptive structures (such as rules), with the timely delivery of information. All of which is relevant to teaching in virtual worlds. 

Finally we would note that the Second Life sessions contributed to the social aspects of the distance course, supporting student confidence and engagement (and thus possibly retention) in the process. One student commented, for example, that the sessions were ‘interesting in that I felt as though I was actually meeting the rest of the class for the very first time’ (from Af’s report), while another remarked that

“I could feel the ‘real class’ when I saw bunch of you gathering at the outside of the ground floor. I felt that finally I would meet all my classmates (even though it was not real). Can you imagine in real life when you meet your classmates for the first time and you will automatically introduce and ask around about people? It was fun.” (Ae’s report)

Thus the impact of the Second Life sessions went beyond the learning that took place in terms of content delivery, to alter social dynamics and participation in the course more generally. By supporting informal, social peer-to-peer contact, these sessions enhanced the students’ experience of the course as a whole. Second Life’s ability to support peer-to-peer informal contact between distance learners should not be underestimated.


The pedagogy that is emerging from this work may best be understood in terms of managing the ambiguity that virtual worlds bring, rather than necessarily removing it. As the research described in this paper makes clear, the object of study (a virtual world) is constructed and enacted in different ways according to the setting in which it is encountered, and the interests of the user. This cannot be avoided. It can, however, be worked with pedagogically. For instance, this variability or ambiguity renders roles, teaching designs and practices unfamiliar or visible, and thus it may be used to draw attention to issues of pedagogic importance.

This was most evident when the educational process itself was the focus for teaching. In such a case, the various design decisions and the perceived potentials of virtual worlds were questioned and challenged by participants, thus heightening their awareness of the issues involved in ‘education and computer mediated communications’ debates, for example. However the same complexities and ambiguity can be seen in relation to other topics, and in other contexts. The forum responses to the issue of deafness provide a case in point. In the discussions that we reviewed, deafness was constructed as a disability, and a tangled relationship between actual hearing loss, online identity (both ‘normal’ and ‘other’), technology, social inclusion and virtual community was evident. Contested definitions, assertions regarding more or less legitimate participation, and the carrying over of real-world expectations and identities into online identities all contribute to the ambiguity of Second Life

While there is much here that might be regarded as critical, we would repeat that the overall response from students was very positive, to the extent that Second Life sessions in one form or another will be integrated into at least two of the Institute of Education’s MA programmes during the coming academic year. This presents us with an opportunity and we hope to extend this action research into a next iteration, continuing in our role as educators while further refining and extending our research into virtual worlds, learning, ambiguity and affect. 



The project ‘Learning from Online Worlds, Teaching in Second Life’ was supported by the Eduserv Foundation. Commentary and papers from the project can be found at:


Bassett E H and O’Riordan K (2002) ‘Ethics of Internet Research: Contesting the Human Subjects Research Model’ Ethics and Information Technology 4 (3) 233-249. Accessed February 2009. Online at  

Burn A (2009 – forthcoming) ‘Machinima, Second Life, and the pedagogy of animation’ in Burn, A. (forthcoming) Making New Media: Semiotics, Culture and Literacies. New York. Peter Lang.

Carr D and Oliver M (2009) ‘Tanks, Chauffeurs and Backseat Drivers: Competence in MMORPGs’. Eludamos. Journal for Computer Game Culture. Vol 3, No 1. Accessed May 2009. Online at  .

Carr D (2009) ‘Learning in Virtual Worlds’ for inclusion in Education 2.0? A commentary by the Technology Enhanced Learning phase of the Teaching and Learning Research Programme, N.Selwyn (ed) pp 17-22

Carr D (submitted 2009) Technology, Inclusion and Social Practices in Virtual Worlds. London Review of Education Special Issue ‘Being Online: A Critical View of Identity and Subjectivity in New Virtual Learning Spaces’  2010

Carr D and Oliver M (in press) ‘Second Life, Immersion and Learning’ in Social Computing and Virtual Communities. P. Zaphiris and C.S. Ang (Eds). Taylor and Francis. 2009/10

Csíkszentmihályi M (1990) Flow: The Psychology of Optimal Experience. New York: Harper and Row

Ducheneaut N Yee N Nickell E and R J Moore (2006) ‘Building an MMO With Mass Appeal: A Look at Gameplay in World of Warcraft’ in Games and Culture 2006 1: pp 281-317

Kirriemuir J (2008 ) A Spring 2008 ‘snapshot’ of UK Higher and Further Education developments in Second Life. Accessed May 2009. Online at

Oliver M & Carr D (2009) Learning in virtual worlds: using communities of practice to explain how people learn from play. British Journal of Educational Tecnology, 40 (3), 444-457.

Salen K and Zimmerman E (2004) Rules of Play. Cambrige, Mass. The MIT Press

Second Life Community Conference Education Track (2006, 2007) are online at and Accessed Sept 2008

 Taylor T L (2006) Play Between Worlds. Cambridge, Mass. The MIT Press

Wenger E (1998) Communities of Practice: Learning, Meaning and Identity. Cambridge: Cambridge University Press.

Zuber-Skerrit O (1992), Action Research in Higher Education, LondonKogan Page

3 Responses to “3. The ReLive paper (Ambiguity in SL)”

  1. […] 3. The ReLive paper […]

  2. […] Learning from social worlds […]

  3. Thanks for making this great resource available online. Interesting reading :-).

Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google photo

You are commenting using your Google account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s