Defining virtual worlds

As is often the case with emerging and/or evolving technologies, it is sometimes difficult to differentiate between associated programs. This is certainly the case when describing VWs and a closely associated genre: massively multiplayer online role‐play games (MMORPGs). In this chapter, as defined by Sadler (2012a), all VWs will be understood to share the following characteristics:

• Online 3D environment. This may simulate the real world, or it may be wildly creative (examples of these later in the chapter).
• Avatars. Avatars are the in‐world representations of real people who control them (see more on this later in the chapter).
• Real‐time interactivity. VWs include the possibility of interacting with other avatars in the environment in real time (synchronous communication), and usually with a range of objects in that VW.
• 24‐hour accessibility. As opposed to interacting with a friend via a program like Skype™, where the potential for interaction only exists as long as the program is open and both parties are connected, a VW should be open and accessible 24 hours a day.
• Persistence. When a user logs out of a VW, their avatar, and the actions taken by that avatar, are not deleted.
• Social space. Although VWs may vary in look and theme, all VWs are primarily social spaces that exist for the purpose of humans interacting via their avatars.
• Numbers. In most VWs there are many players (sometimes in the hundreds of thousands) online in the world at the same time. Given the social nature of these spaces, underpopulated VWs tend to fade away quickly as users lose interest.
• In many – though not all – VWs, users can also control their own appearance (e.g. height, facial features, eye color), gender, clothing, and even their species. In addition, many VWs allow users some control over their environment. In Second Life™, for example, users have the ability to build their own home, complete with furnishing, landscaping, etc. (Sadler 2012a, pp. 24–25)

Those who have played MMORPGs will almost certainly recognize most of the features mentioned above as also playing a part in popular games such as World of Warcraft™ (worldofwarcraft.com/en‐us; see Thorne 2008 for interesting research on this), Final Fantasy™ (www.finalfantasy.com), or MapleStory™ (maplestory.nexon.net/landing). Indeed, of the characteristics described above, only one is the key in differentiating between these two platforms – VWs as social spaces. While socialization is doubtless an important element in many MMORPGs, they were created as competitive games. In contrast, socializing is the central feature in VWs. As gamers, players in MMORPGs have goals (whether they be gathering gold, slaying an opposing force of orcs, or simply winning) that were created by the game developers to be integral parts of the experience. Socializing with other players, and particularly with teammates (if they exist), may be helpful or even essential in achieving those goals, but it is a secondary mission. There is sometimes also the possibility of game play in VWs, but that, to reiterate, is not the primary purpose – socializing with other avatars is. This focus on socializing with others helps to give VWs such as Second Life, Active Worlds™, There™, etc. (and to a lesser extent MMORPGs) strong potential for education purposes – both formal and informal.

The sim (or island) found in the VW Second Life that is shown in Figure 6.1 is one that its Estate Managers describe as “a helpful place for Japanese‐speaking residents of Second Life” (Linden Research 2019). As seen in the image, this is a 3D environment that meets all eight of the VW characteristics discussed above. The five individuals in this seating area all take the form of human avatars (most VWs, including Second Life, allow for a near infinite variation of avatar forms including human and many other types) and they are engaged in live text‐based communication as seen in the lower‐left section of the screen (verbal chat is also possible in many VWs, including this one). In this case, as shown, their communication is taking place entirely in Japanese. As with most locations in this VW (and others) this island was created to be primarily a social space, with this one including conversation areas, dance spaces, locations where users may practice their building skills, virtual housing, and more. This island, and all others, are available 24 hours per day and are persistent spaces.

Although their names are blurred to protect their identities, each of these avatars had their names appear above their heads in both Japanese characters and in Romaji (romanization of the Japanese written language) and were native speakers of that language. The existence of places such as this in many of the available VWs allows users the possibility of finding native speakers of a wide array of languages. The Second Life destinations guide (which is an admittedly very incomplete list), for example, currently includes international island listings focusing on Arabic, Chinese, French, German, Indonesian, Italian, Japanese, Korean, Norwegian, Polish, Portuguese, Russian, Spanish, and Turkish (Linden Research 2019).

This chapter will continue with an overview of the development of VWs, and an examination of current research on this environment – first that which has been accomplished in more formal (school) settings and then on that which has specifically discussed informal language learning. It concludes with future directions of this technology and suggestions for educators and students who might wish to use a VW to enhance their own informal language learning.

Development and current state of virtual worlds

The idea of VWs, as mentioned in this chapter's Introduction, stretches back far beyond their modern existence. However, their actual history – in more primitive forms – goes back farther than most current users might imagine. Colossal Cave Adventure, created by Will Crowther in 1975–1976, is likely the first step in the evolution of both VW and MMORPG computer‐based environments as it is a text‐based game that sought to recreate the real‐life Mammoth Cave system in the US while adding a fantasy Dungeons & Dragons twist (see Sadler 2012a,b for a more in‐depth discussion). This was soon followed by similar adventures like Zorkin 1979, and the creation of the first Multi‐User Dungeon (MUD) in 1978 by Roy Trubshaw. As implied in its name, MUD included a multiuser element that began a revolution in computer‐based gaming and when it was connected to the ARPANET (a very early version of the internet) in 1980 (as MUD1) it became a true multiplayer game in which one might encounter users from different cities, states, or even countries.

Lucasfilm™, known for its wildly successful Star Wars franchise, created Habitat™ in 1985 and in 1986 it was released on the Quantum Link™ for the Commodore 64 computer. This was their first large‐scale VW environment that was graphically based. Although the interface might seem primitive to users today, it included color, text‐chat between avatars, and more than 20 000 regions at its peak. The creators of this program, Morningstar and Farmer, had as their vision a virtual setting where the “things that are important to the inhabitants … are the capabilities available to them, the characteristics of the other people they encounter there, and the ways these various participants can affect each other” (2001, pp. 173–174). This philosophy is still quite apparent in VWs that come up today.

From the handful of environments available in the 1980s, both VWs and MMORPGs have exploded in popularity. One website, MMOhuts (MMOhuts, 2019), that tracks VWs and MMORPGs listed over 1800 such environments in January 2019, with the number listed eight years earlier set at only 290. While it is possible to ascertain to a relatively accurate degree the number of these worlds, tracking the user statistics for the various programs is much more difficult due to the opacity of the data of the companies controlling them. Nevertheless, there is very strong agreement that users now number well beyond one billion worldwide (KZERO 2015). That same 2015 data indicated that some of the most popular VWs have very large userbases: Club Penguin (225 million; www.clubpenguin.com), Poptropica (313 million; www.poptropica.com), Habbo (295 million; www.habbo.com), Maplestory (140 million), Second Life (41 million). These user numbers, and the popularity of such environments for younger players, make it increasingly likely that classrooms are now filled with students for whom these settings play a major role in their social lives.

General research on virtual worlds

In comparison to education research overall, scholarly examination of the role of VWs in language learning is still relatively young, with very few studies conducted prior to 2000. There have been several notable book‐length contributions to research in this area that are most definitely worth in‐depth examinations for those wishing to thoroughly explore this field. They include Learning and Teaching in the Virtual World of Second Life (Wankel and Kingsley 2009); Learning Online with Games, Simulations, and Virtual Worlds (Aldrich 2009); Education in Virtual Worlds: Teaching and Learning in Second Life (Molka‐Danielsen and Deutschmann 2009); Multi‐User Virtual Environments for the Classroom: Practical Approaches to Teaching in Virtual Worlds (Vincenti and Braman 2011); Virtual Worlds, Telecollaboration, and Language Learning: From Theory to Practice (Sadler 2012a); and a special issue of ReCALL Journal focusing on Interactions for Language Learning in and around Virtual Worlds (Wigham et al. 2018).

One broad category of research findings on this topic addresses the potential for virtual worlds to act either as a substitute for the real world or to serve as a predecessor to activities in real life. Jarmon et al. (2009) provide six advantages of VWs in this area:

(a) the capacity to host virtual social interactions and collaborations, (b) the capacity to allow users to test hypothesis by applying them to an actual project and doing something active (Kalyuga 2007), without the risk and cost of the real world, (c) the possibilities for relevance of their virtual actions to the real world, (d) the capacity to allow for various types of activities to be practiced and demonstrated virtually, (e) the stimulation of imagination, exploration, and creativity, and (f) an increased sense of personal presence and tangible experience in the virtual world.

(p.179)

This idea that a VW, as noted at the end of this quotation, can lead to “tangible experience” is something mentioned by other researchers investigating more formal language learning and their participants as well. The blurring between real life and a virtual life is quite evident in this comment from a student studying Chinese in a virtual setting: “As someone who doesn't really have anyone to practice Chinese I feel that the lab classes allow me to use what I have learnt in a real life setting” (Grant et al. 2014, p. 30). No doubt an added advantage for language instructors is that a virtual environment makes it “relatively easy to adapt and modify the environment to suit a particular lesson design and to create teaching/learning opportunities that might otherwise not be available in the real world classroom (e.g. shops, restaurants, the mazes, etc.)” (Lin et al. 2014, p. 20). On EduNation Island (a sim in Second Life designed by and for language teachers), for example, there are several Star Trek‐style holodecks available that contain more than 100 different scenes ranging from cafés, to hotel registration lobbies, airports, castles, forests, conference centers, libraries, diners, lecture rooms, marketplaces, art galleries, and many others. For some students, practicing language in this type of a virtual settings has definite real‐life implications: “I find these tasks helpful for practicing using the language and they will prepare me to use Chinese in practical situations. Practicing these situations in a 3D environment can make me less nervous if I come across these situations in real life” (Grant et al. 2014, p. 30). This idea has also been applied to other fields of education, particularly for history and art (Bani et al. 2009), architecture (Gu et al. 2009), medicine (Antonacci and Modaress 2005); and business (Belei et al. 2009; Faivre 2009; and Jæger and Helgheim 2009). Given the increasingly global nature of business, science, and education, all of these studies have implications for language learning.

A second major area of findings regarding VWs and formal language learning regards the potential lessening of student anxiety in such settings. As any experienced language instructor can confirm, this is a major issue for many students. Childress and Braswell (2006) found that one explanation for this is that when students communicate in a VW setting utilizing text‐based communication, in comparison to a traditional classroom, they feel reduced anxiety because text chat “allows [students] time to more completely formulate their thoughts as they respond to the class discussion” (p. 166). In addition to the added response time permitted, others found that the nature of an avatar may serve the same function as a mask at a real‐life costume ball, thereby helping the learners to “loosen…up a bit…” as their real life identities are hidden (Love et al. 2009, p. 68) while they are also able “to experiment with new and powerful identities” (Shaffer et al. 2004, p. 6).

Although the research on this topic to data has often relied on qualitative data, Grant, Huang, and Pasfield‐Neofitou performed a statistical analysis to confirm that the college‐age students in their study using the provided virtual environment had significantly lower levels of language‐related anxiety in comparison to students studying solely in a traditional classroom. Their research also examined whether the use of the VW led to greater amounts of technology‐related anxiety for their students (something sometimes mentioned in earlierstudies) but found that “there was not a significant inherent level of technical related anxiety, nor did the technical aspects of interacting in the virtual environment present significant additional levels of technical anxiety” (Grant et al. 2013, p. 7). Similar findings were reported by Abal (2013) who found that students practicing English in a VW had statistically lower levels of anxiety over time in comparison to their counterparts practicing face‐to‐face. The power of the anonymizing nature of VWs was confirmed by Melchor‐Couto, who found that “participants with all anxiety profiles experienced the anonymity afforded by the VW positively” (2018, p. 242) in her study, though she also found that the negative effects of anonymity included the lack of non‐verbal communicative clues due to the limitations of avatars and a preference for some students for traditional face‐to‐face communication.

A final component that has been well‐researched in this area is the potential for VWs to enhance the collaboration for students. Very early research on this topic by Bystrom and Barfield (1999) found that students were better able to accomplish their tasks in VWs with their peers, a finding echoed by Steinkuehler (2004). As might be expected given the widespread applications of Vygotsky's zone of proximal development (1978), this finding has been echoed across a number of VW studies (Churchill and Snowdon 1998; Dillenbourg et al. 2002; Brown and Bell 2004; Price and Rogers 2004; Roussou 2004; Shaffer et al. 2004; Gronstedt 2007; Skiba 2007; Ball and Pearce 2009; Bani et al. 2009; Sadler and Dooly 2012). Some of these researchers, such as Brown and Bell (2004), found that VWs nature as social settings built around virtual objects (their research took place in the VW There) encouraged collaboration. Gronstedt echoes this, maintaining that VWs are a type of “social networking tool” (2007, p. 46).

On neutral ground

VWs are filled with social settings such as that illustrated in Figure 6.1. Just as in Oldenburg's third places, avatars who enter into such settings in a VW are “not required to play host” and “may come and go as they please” (p. 22). In many ways, virtual third places may be considered even more neutral ground than in the real world since users are present in these locales only in the form of their avatars. As discussed in the section “General Research on Virtual Worlds,” avatars give a loose sense of anonymity for their users, resulting in a greater sense of ease and letting them “loosen…up” (Love et al. 2009, p. 68).

A leveler

Third places should be open to the public and without formal membership. All VWs are, in theory, open to the public, but not all areas are 100% open in all such settings. In Second Life, for example, the owner of each island has the option to keep it completely open, open to members only, open to a restricted list, or closed. However, the vast majority of sims in that VW are open and, given the current listing of almost 24 000 sims (for a total virtual land area of over 1557 km²) in that VW alone, there are almost unlimited spaces fully open to any avatar's wanderings (Second Life Grid Survey 2019).

Conversation is main activity

The definition of VWs provided at the beginning of this chapter included as one of the key characteristics (and the one that distinguishes VWs from MMORPGs) that they exist as social spaces and asserted that even though “VWs may vary in look and theme, all VWs are primarily social spaces that exist for the purpose of humans interacting via their avatars” (Sadler 2012a, p. 25). In VWs this interaction may be physical, via text‐chat (as seen between the native speakers of Japanese in Figure 6.1) or via voice‐chat in those VWs that allow it. While Oldenburg lists this conversational component of third places as the third defining characteristic he is quick to note that it is, perhaps, the most critical. In his third places “the talk there is good…it is lively, scintillating, colorful, and engaging” (1999, p. 26).

The importance of communication in a VW such as Second Life may be illustrated via the various ways in which it is allowed. Text‐chat may be typed via the standard interface, resulting in text communication that can be seen by any avatar within 20 virtual meters of the person typing. Text‐chat may also be whispered and be visible by those within 10 m or shouted and seen by any within 100 m. One‐to‐one and group text‐based conferences are also allowed and are visible to anyone in the chat no matter their location in the VW. Voice‐chat by any avatar in that VW can be heard from up to 60 m away and one‐to‐one and group private voice‐chat is also available. Communication can also be accomplished via note cards, signs, images, emojis, facial expressions, gestures, and numerous other ways. A VW without abundant communication and socialization is simply not a VW or an appropriate third place.

Accessibility and accommodation

Although Oldenburg's third places should be available “any time of the day or evening” (p. 32) it is almost certain that he had no expectation that they be open 24 hours per day, 7 days a week, and 365 days a year. On the other hand, VWs (as noted in their definition) meet all three of those criteria. Because VWs are as close as users' computers or in some cases cell phones, their ease of entry is perhaps greater than even the traditional third places whose loss Oldenburg laments.

The regulars

Just as in real‐world third places, it is the Regulars that give a VW location its special flavor. Whether the virtual location is a beach, a ballroom, a dance club, or an education sim, the regular members of the location set the tone, bring in new participants, and help to ensure that they return. In language‐focused regions, these regulars may also set the foundation to provide the sort of Vygotskian zone of proximal development where “the distance between the actual developmental level as determined by independent problem solving and the level of potential development as determined through problem solving under adult guidance or in collaboration with more capable peers” (Vygotsky 1978, p. 86). This is not to say that these individuals are there to act as language tutors (though in some cases they may be willing to do so). Instead, they will almost certainly be willing to converse about things pertinent to the region or to life in general. These “more capable” peers have the ability to help informal language learners with critical “guidance” and “collaboration” due to their familiarity with both the language of interest and, equally important, their knowledge about the third space they occupy.

A low profile

Many locations in VWs are inelegant. However, others are opulent. It is perhaps on this section of the definition that VWs may differ the most from real‐world third places, as defined by Oldenburg. However, one might argue that the very diversity of spaces in VWs – from the informal to the formal – provides a space that suits the taste of every avatar, whether that be a dive bar or a beachside palace.

The mood is playful

Some VW regions might have horseback riding, other might have scuba diving, hang gliding, mazes, dancing, or almost anything else that can be imagined. While VWs are not “playful” in the same way as MMORPGs, they are playful in their own distinct manner.

A home away from home

Most VWs quite literally offer the possibility of a (virtual) home away from home. In Club Penguin each avatar is provided with an igloo that they may customize. Habbo, which is located in a giant virtual hotel, includes a private room for each guest. In Second Life, players may have a home with membership or they may discover that a privately owned sim where they spend time has homes for rent. These “home owners” are often the users who fill the role of third space “regulars.”

Is informal language learning taking place?

While it is clear that VWs have the ability to serve as third places, one key question, of course, is whether informal language learning is actually taking place in VWs and if so, how? There has been surprisingly little quantitative research that directly examines the myriad ways in which VW users engage in informal language practice in those environments that look beyond a single setting such as a virtual classroom or island in Second Life. One study that did examine the larger phenomena (Sadler 2012a) gathered survey responses from 237 avatars in a number of randomly chosen sims (discrete regions in that VW) in order to investigate four primary questions: the backgrounds and nationalities of the users, their language backgrounds, their language use in the VW Second Life, and their second language experiences in that setting.

The respondents included avatars representing users from 19 different countries, with the most frequent listed in Figure 6.2. In addition to those shown, there were two each from Canada, France, and Portugal, and one for China, Egypt, Mexico, New Zealand, Romania, Sweden, and Venezuela. While the countries represented indicate a strong linguistic diversity as only just over 27% of the users came from the US, the more important data is that which indicates their actual language use, both in real life and in the VW. Two additional key questions asked in this research were how often the participants were making use of their primary first language (defined as the language they used the most often in real life) in the VW and how often they used another language in that setting. Figure 6.3 shows that the use of primary and additional languages in this VW was actually quite diverse. While 22% of the participants reported that they never used a second language in the VW, the remaining 78% did so ranging from rarely (24%) to a surprising 12% who usually used a language other than their primary while in the VW (see Sadler 2012a for a more in‐depth discussion of these results).

This finding suggests that VWs are sites of a great deal of multilinguistic language interaction, most of which is taking place in informal settings, as discussed below. Connected to this use of second languages, the survey also asked them how helpful their use of Second Life was in improving their skills in another language. Although 16.5% of the respondents reported that it was not helpful at all, almost half (48%) said it had a positive effect on their proficiency in an L2 and 26.6% said it was very helpful in that process.

The final question asked in this survey is especially pertinent to the topic of this volume as it asked them exactly how they used that VW to improve their second language skills, instructing them to select all pertinent choices (see Figure 6.4). As illustrated, relatively few of the survey takers worked on improving a second language in the more formal settings of a virtual school (8.4%) or via tutoring (6.4%).

Given that VWs are social environments, and that many VWs have groups that users may join – ranging from book clubs, to Lord of the Rings role‐play sims, dance clubs, surf clubs or beaches – it is not perhaps surprising that 73% of the participants mentioned joining a group that used the second language to be one of the most frequent strategies that led to informal language practice. That the most frequent strategy, listening to audio‐chat, was utilized by 83% of the survey takers is not surprising since voice‐chat in that VW carries over a distance, as does text‐chat – used by 71% of them for reading language practice.

Given the findings by Sadler, it seems likely that not only are VWs possible third place settings, but that a number of language students around the world are already using them to engage in informal language learning.