The Internet TESL Journal

Online Strategic Interaction:
ESL Role-Playing via Internet Relay Chat

Christopher M. Colburn
dude [at]
Department of Linguistics, University of Delaware


As the age of computer-assisted and adaptive language learning become real concerns of the modern language teacher, it has become necessary to investigate the ways in which the computer can enhance existing teaching methods to improve the overall quality of language instruction. Many researchers have attempted to add their own electronic components to their favorite exercises.

Strategic Interaction (SI) (Di Pietro 1987) was no exception to the electronic bandwagon. In the 1980s, Di Pietro and Arena proposed a number of writing exercises based on his paradigm (cf. Arena and Di Pietro, press pending.) The original concept involved the practice of writing using terminals connected via a PLATO mainframe (Dyer 1983.) With the demise of mainframe technology, Arena sought a modern solution for implementing these exercises.

The emergence of the Internet in the mid 1990s provided a viable way to participate in these exercises in both classroom and distance-learning situations. This paper describes the how, where, and why of implementing Strategic Interaction over local-area and wide-area networks. This description includes not only the goals and merits of the SI writing exercises, but also provides the technical background for those who want to make the exercises work in a local or global electronic classroom environment.


Strategic Interaction: An Overview

Strategic Interaction (SI) refers to a teaching method that was devised by Robert Di Pietro at the University of Delaware (cf. Di Pietro 1987.) SI involves the use of interactive scenarios to foster better communicative competence in the target language. This overview will cover the issues of student roles in the exercise, the types of scenarios that are used, and the three stages that take place in SI exercises.

The SI exercises take place in three stages, or phases. The first stage is known as the Planning stage. During the planning stage, the instructor or moderator splits the class into two or more groups. Each of these groups is given a different role to play within a given scenario. The students are than given some time in which to plan their given role in the scenario. Scenarios usually come from one of four types (Di Pietro 1987: 48-58):

  1. The basic scenario is the Two-Role type, which is suitable for one encounter. In a group situation, there are two groups, and each group has a spokesperson who interacts on behalf of the whole.
  2. Multiple-Role scenarios have more than two direct roles. Di Pietro notes that this type of scenario is only useful for all parties if each role has a distinct reason for being involved. "Background-only" characters are not a good idea, as they do not promote direct interaction among the multiple roles. One idea hinted at by Arena and supported by this author involves using the teacher/moderator as a background character. This can be advantageous for moving the scenario along, especially with less-advanced groups. This character can provide assistance or inject additional tension into the scenario as warranted by the students' performance.
  3. In Group scenarios, all students react in their own ways to an event or situation that involves or affects everyone in the group. This lends itself to use on the Internet, since everyone can make their thoughts known to everyone else (over the IRC network) without anyone's contribution getting lost in the conversation. This is especially important for the more traditionally quiet speakers in the class, who are placed on a more equal footing when presented with the Internet-based classroom.
  4. Open-Ended scenarios promote the development of long-term roles and interrelated episodes that may span a number of class sessions. Di Pietro notes that these scenarios are good for showing the relationships among different events that take place in a target culture.
An example of the kinds of scenarios given to each group can be found at Note that the models on that page were originally conceived to facilitate the Two-Role scenario mentioned above. The exercises, however, can be easily adapted to support any of the above scenario types. For more examples of appropriate scenarios, consult Di Pietro 1987 (which is a must-read for those who wish to make extensive use of SI in the classroom, electronic or otherwise.)

In these exercises, the roles given to the students are different from those in a drama exercise. First, the scope of the role is only limited by the students' abilities. The role is dynamic, and never plays out the same way twice. Also, the roles given to each group are unknown to the members of the other group(s) until the Performance stage of the exercise. This is important to the theme of the exercise, because the roles given to each group are slightly different in focus and/or goals. These differences produce a "twist," or point of conflict, that appears in the Performance stage.

Once the students have prepared their roles, a spokesperson is appointed, and the second, or Performance stage begins. If the scenarios are well-constructed, the "twist" in the scenario appears quickly. As students' expectations of what they were going to say quickly disappear, the students must attempt to maintain communication in the target language while trying to make their case and/or resolve the conflict that has arisen. The goal of the exercises is for the students to work creatively in the target language to resolve the conflict successfully. The end result is that no scenario has the same ending: the outcome depends on the students' abilities to resume and/or maintain communication in the face of a breakdown. By forcing the students to deal with a conflict for which they are not prepared, students are encouraged to think "on their feet" in a manner that they will need to survive in an L2 environment. If the spokesperson needs any help with issues of communication, he or she is free to return to the group, who can then make suggestions that the spokesperson can use in the scenario. The Performance stage continues until the instructor or moderator observes a successful resolution to the scenario, until communication breaks down, or until time constraints force an end to the performance.

The third stage of the exercise is known as the Debriefing stage. At this point in the exercise, the instructor or moderator regroups the class, and discusses the events of the Performance stage. There may be a discussion of discourse, cultural, or grammar problems that appeared. During this stage, the teacher can also provide feedback on how to improve communication and/or performance at key points in the scenario, so the students will be better prepared to perform in future scenarios.

SI's Relationship with Second Language Acquisition and CALL

Part of the original rationale for using CALL in the United States was to provide supplemental exposure to the target language in environments where contact with native speakers and/or instructors was limited, either due to geography or large class sizes. It was hoped that an increase in comprehensible input (cf. Krashen, 1982 and Krashen, 1985) would allow students to progress more quickly.

From the perspective of writing, Ellis (1987) looked at how writing proficiency, like other forms of linguistic competence in a second language, progressed over what he called the interlanguage continuum. When a student first begins to acquire a new language, the student applies what he or she already knows about language to the new one. In effect, his or her writing may use the lexical items of the target language, but uses the structures and idioms of the native (or other second) languages (cf. Colburn 1992.) As the student acquires more of the target language, these L1 structures tend to disappear as the student begins to produce structures that more closely approximate the target language.

These topics relate to Strategic Interaction in that one of the goals of the method is to force the student to use the structures of the target language in a pressure situation. The dialogue provides spontaneous input in manageable chunks. It also allows the student to work with the target language in a group situation that can foster improvement for all of the group members, whether or not they are actors in the scenarios.

Recent Trends in CALL: Transitioning to the Internet

A number of researchers have undertaken work on Computer Assisted Language Learning (CALL) and Computer Adaptive Teaching (CAT) as computers have become more widespread in both homes and schools. Since the early 1980s, researchers have come up with a myriad of exercises to improve communicative competence in foreign language classes. Kenning and Kenning have noted that computer-based interactions "and normal language do not seem to go together very well" (Kenning and Kenning 1990: 63.) Most of the research that they did, however, dealt with the interaction between a computerized "tutor" and a single student. Ahmad and his colleagues mention the importance of CALL in developing writing skills (Ahmad et al. 1985:105,) but make no reference to interactive writing skills between groups of students.

To some extent, we find that literature on CALL in the 1980s makes few references to student interaction over networks, simply because of its impracticality at that time. This was a result of the facts that wide-area networking was far too slow at that time to facilitate interactive written conversations in a reasonable amount of time, and because long-distance communication of this type was severely limited, since the Internet as we now know it did not exist on a widespread scale (and ran at much slower speeds as well.)

One point in the literature, however, shows that CALL, even in the Strategic Interaction context, was being used on local-area networks in the early 1980s. In her dissertation, Pat Dyer recounts the use of the PLATO (Programmed Logic for Automated Teaching Operations) system to implement a number of SI exercises at all levels of instruction at the University of Delaware's English Language Institute (ELI.) The system was used with some success as a component of the Institute's intensive program of English instruction (cf. Dyer 1983: 65-85.)

One of the advantages of the PLATO system was its ability to handle lessons and instructions that were pre-programmed for use on the local network by the students. From a standpoint of current technology, however, this system is not as practical as it once was. The networking and use of the mainframe was confined to a limited area. With the advent of the desktop PC, it became necessary to provide a package that could be used not only while connected to a mainframe, but while connected to a larger network, such as a campus-wide LAN or the Internet. This would allow the students to work in larger groups, and over greater distances, and on a number of operating systems. This need led to the development of the Internet-based SI exercises.

SI on the Intra/Internet: Technical Aspects

When Lou Arena the author with instructions to adapt SI for modern machines, he was presented with limited resources: two Apple PowerBooks and no Internet access. From these beginnings, SI exercises were adapted for LAN use. Based on this original conception, the idea was then expanded for use on the global Internet, as well as for use on multiple platforms. The original conception, as well as the expanded version, are explained below. Note: this section may be more appropriate for the IT managers in the audience, but will also be helpful to the teacher who has to implement this on his or her own.

SI Exercises on a MacOS-based LAN

For the original concept, the lack of Internet access forced the use of AppleTalk networking to connect the two computers. For simplicity's sake, the two computers were linked through the use of an RJ-11 cable attached to two LocalTalk adaptors, which were plugged into the serial port of each machine. Because of the speed limitations experienced on a LocalTalk network, experience has shown that an Ethernet-based AppleTalk network is more effective, and should be used if available. Once AppleTalk networking was enabled on both machines, the only remaining tasks involved supplying the scenarios, as well as the medium of written conversation.

Each computer was supplied with text versions of their respective scenarios. Generic text files were supplied so that any text editor could open the files for viewing. Since all MacOS-compatible machines come with a copy of the SimpleText text editor, it is then easy to open the files when instructed. Note: if the machines in use are running a Language-Kit enabled version of the MacOS (such as the Chinese Language Kit or the Japanese Language Kit,) it is MOST important that the teacher supply an English-language version of SimpleText (or their favorite text editor) for use in the exercises. The versions of SimpleText supplied with these kits are enabled for reading two-byte languages, and they will interpret raw English ASCII text as gibberish.

To provide the conversation environment, the author discovered the ChatNet shareware program, from ELS, Inc. This program can be used freely on an AppleTalk network to link up to three machines in a chat environment, similar to that found on IRC (Internet Relay Chat,) or in a chat room on America Online. A copy of ChatNet is installed on each of the client machines. Each machine is configured as a different user (i.e. one username for each group taking place in the scenario.)

Multi-platform SI Exercises:
Internet-based IRC with WWW Support

The specifics of using these exercises over the Internet are less specific, because there are a number of ways to implement the necessary materials. Some guidelines, however, are presented for teachers who want to use these exercises in a multi-platform or distance-learning environment.

The first concern is that the students' machines are connected to the Internet, either via modem or direct connection. With Internet connections, the physical locations of the students are irrelevant. The only difference involves how the conversational groups collaborate. If the group is in one classroom, they can collaborate verbally. If they are not in the same room, then they can collaborate and prepare their initial roles via a separate chat channel.

Next, the scenarios are presented to each group via a WWW browser. In this way, each student can retrieve the scenario for his/her group from a central location, thus obviating the need for each student to download specialized software for the exercises. Each group is given a different URL from which to retrieve the scenario.

Most importantly, the students need software on which to participate in the chat environment. Internet Relay Chat (IRC) clients are varied in features and performance, depending on the platform. Examples of clients that are usable for this exercise are listed here as guidelines, and not necessarily as requirements or endorsements. Any program with similar functionality can be used. For Windows, programs such as mIRC should suffice. For the MacOS, the author has used IRCle and ChatNet (although the latter needs to be registered to be used for more than ten minutes per session.) For Unix and appropriate clones, ircII, BitchX, cIRCus, and zIRCon have worked well, although ircII and BitchX may require multiple terminal sessions for best results. Any of these software packages are available from the popular Internet software archives for each respective computing platform.

Scenarios in Action

The following provides guidelines for using SI exercises in an online environment. Procedures that have been used on an AppleTalk LAN are provided. In addition, general guidelines for using the exercises on the Internet are also presented.

With the LAN solution, only two computers are required (one for each group.) The teacher then instructs the students to choose one of the scenario files by opening it in the English version of SimpleText. The students view the situation file, and prepare their roles in another file in the text editor. When they are ready, the group can log onto the network to begin the Performance stage of the exercise.

Over the Internet, the students receive their instructions via a central web page. There they are instructed to access the appropriate scenario pages, and to log onto the appropriate chat channels via their preferred IRC client. The group prepares its role over one channel (which could be called the Planning channel.) When they feel sufficiently prepared, they appoint a "spokeswriter," and all students log onto a common channel (the Performance channel) to watch the spokespeople perform their scenarios.

Then, the students act out their roles by typing their lines into the computer. When conflicts arise, the students are forced to deviate from their pre-planned roles in order to come to a resolution. At times, this involves a number of strategies that are only used in a "communication breakdown" situation. In this way, the outcome of a scenario is different for each participating group, due to the lack of preparation inherent in creating the "twist."

If the spokeswriters cannot continue the interaction on their own, they are allowed to collaborate with their respective groups in order to get help. Over a LAN, the group helps the spokeswriter verbally, while over the Internet, the groups can provide written feedback to the spokeswriter over the Planning channel. In this way, the scenario is solved through a group effort, and all of the students get to practice their communication skills.

When the scenario is finished, the Debriefing phase can take place in one of two ways. On a LAN, the students can regroup in a classroom for a traditional debriefing. For a local class, this is advantageous, but for an Internet-based class, this is not feasible. After the Performance stage, the students can migrate to a Debriefing channel for the final stage of the exercise. On this channel, the moderator or instructor can conduct an online chat session that provides feedback for student performance during the scenario.

Advantages of the Exercise

Communicative competence is promoted through the Strategic Interaction exercises. Scenarios promote the kind of "on-your-feet" strategic thinking and communication skills usually associated only with acquisition-based environments. Specifically, these exercises develop a number of skills and perspectives not offered by other types of writing tasks. The major benefits to student writing occur in four main (interrelated) areas:
  1. Linguistic Competence: grammar and writing skills are developed and practiced on a more autonomic level, since it is harder to focus consciously on form when trying to think quickly;
  2. Sociolinguistic Competence: Role playing develops discourse skills and strategies not usually practiced in a written environment;
  3. Strategic Competence: network-based scenarios present a new way of trying to restore communication when it breaks down: strategies for restoring communication, and the duration of their effectiveness, become apparent for the respective scenarios;
  4. Discourse Competence: Scenarios enable students to practice cohesive devices in writing in a fast-paced environment: students learn the limitations and applications of transitions to prolong the communicative exchange.

Suggestions for Further Use

There are a number of directions for this kind of exercise. Research can be done to measure overall effectiveness through a discourse analysis of student writing. Exercises can be expanded to encompass multiple sites and/or multi-faceted scenarios (to add additional conflicts for advanced practice.) This provides for more complicated scenarios (for more advanced students) that encourage participation by more than two groups of students. Other research on the continued use of Strategic Interaction and/or IRC are also encouraged by the author (and email feedback on the topic is both welcome and encouraged!)


Ahmad, K. et al. 1985. Computers, language learning, and language teaching. New York: Cambridge University Press.

Colburn, C.M. 1992. The issues of Interference and Multiple Lf in the Second Language classroom. ms., Williams College, Williamstown, MA.

Di Pietro, R.J. 1987. Strategic interaction. Cambridge: Cambridge University Press.
Note: Excerpts of scenarios in this article and its accompanying web pages are reprinted with the permission of Mrs. Vincenzina (Robert) Di Pietro.

Dyer, P.M. 1983. Instructional procedures for implementing the Strategic Interaction Method in an intensive English as a Second Language program. Ph.D. Dissertation, University of Delaware.

Ellis, R. 1987. Second Language Acquisition in context. Englewood Cliffs, NJ: Prentice Hall International.

Kenning, M.J. and M.M. Kenning 1990. Computers and language learning: current theory and practice. New York: Ellis Horwood.

Krashen, S.D. 1982. Principles and practice in Second Language Acquisition. New York: Pergamon Press.

Krashen, S.D. 1985. The input hypothesis: Issues and implications. New York: Longman.

The Internet TESL Journal, Vol. IV, No. 6, June 1998