Member Login

Reset Password



Vol 19, No. 2 (January 2002)

[article | discuss (0) | print article]

Negotiation of Meaning in Nonnative Speaker-Nonnative Speaker Synchronous Discussions

Marisol Fernandez-Garcia
Northeastern University
Asuncion Martinez-Arbelaiz
Michigan State University

Research on negotiation has thus far focused on oral conversations/interactions. This study expands on this line of research by investigating whether learners engage in negotiation when exchanging ideas in synchronous computer-mediated interaction. Four groups of learners of Spanish discussed a number of content questions about a reading assignment using an Open Transport (OT) Chat. The analyses of the transcripts of the interactions showed that instances of negotiation as operationalized in Varonis and Gass (1985b) do occur in the electronic medium. A limited repertoire of types of primes reoccurred, due in part to the nature of the medium and the academic context of foreign language learning in which the interactions took place. Of special concern was the tendency to use the native language in the response of the majority of the routines since this tendency does not result in target language modified output, which is claimed to be fundamental for second language acquisition (SLA) (Swain, 1985).



Input, Output, Negotiation of Meaning, Learner-Centered Instruction, Computer-Assisted Classroom Discussion.


In the last few decades there has been a shift from a traditional teacher-centered instructional setting to one that is mainly learner-centered (Nunan, 1988). Given the emphasis that recent approaches to language teaching


place on small group work, it seems essential to expand on the line of research that investigates the characteristics of the discourse generated during learner-learner interaction. As several researchers have noted (Ellis, 1994; Long, Adams, McLean, & Castaños, 1976; Long & Porter, 1985; Porter, 1986), the literature on small-group work and interaction suggests that interaction between learners is more effective than teacher-led interaction1 in providing the conditions which have been hypothesized to facilitate second language acquisition (SLA).

According to research on interaction, the conditions for SLA are enhanced by the presence of discourse moves that allow interlocutors to ensure message comprehensibility. Specifically, several studies have focused on the interactional modifications that take place when a communication problem arises in a conversation. These studies propose that interactional modifications can help to make input more comprehensible which, in turn, assists in language learning (Gass & Varonis, 1985b, 1986; Pica, 1994; Pica & Doughty, 1985; Pica, Kanagy, & Falodun, 1993; Scarcella & Higa, 1981; Varonis & Gass, 1985a; 1985b).

Certain applications of computer technology appear to go hand in hand with learner-centered instruction. In particular, local networks, which link computers in a laboratory to each other, have made possible one-to-one and many-to-many synchronous interchanges, thus affording language learners new opportunities for communicative practice. A few studies that have examined computer-assisted-classroom discussions (CACDs) suggest that the electronic environment provides optimal opportunities for language development (Beauvois, 1992; Chun, 1994; Kelm, 1992; Kern, 1995; Warschauer, 1996; Sullivan & Pratt, 1996). Nevertheless, as Warschauer (1997) notes, the "growing interest in computer-mediated collaborative language learning has not been matched by sufficient research and theory." Thus, if CACDs are to become a common practice in the foreign/second language classroom, it is crucial to describe the nature of the discourse generated through the electronic medium. In particular, research needs to ascertain whether the features relevant to the processes involved in SLA are present in CACD's discourse. In this study we examined whether interactive discourse features of oral discussions that are said to foster inter-

language development are also present in electronic interactions.

The Study of Learner-Learner Interactions

Several studies have called to our attention the advantages of learner-

centered instruction. These studies have compared the interaction in teacher-centered lessons with that found in group work. Long, Adams, McLean, and Castaños (1976) found that students working in small groups performed better than students in a teacher-fronted classroom in terms of


quantity and quality of language used. More precisely, small group work not only provided more opportunities for language production,2 but also a greater variety of discourse moves in initiating discussion, asking for clarification, interrupting, competing for the floor, and joking. Similarly, Rulon and McCreary's study (1986) revealed that a small group discussion generated significantly more negotiation of content than a teacher-led discussion.

Other studies (Varonis & Gass, 1985b; Porter, 1986) have pointed to the value of learner-learner conversations, specifically the interlanguage talk generated in such discussions, as a source of opportunities for meaning negotiation. For example, participants in a conversation may experience problems in understanding or breakdowns in communication. To remedy this situation, they often engage in negotiation of meaning, interactional modifications that aim at ensuring shared understanding.

Interactional modifications can be analyzed in terms of discourse functions (e.g., requests for clarification or confirmation) and in terms of a specific discourse structure. A number of studies have provided taxonomies of discrete discourse functions (Long, 1983) as well as models of foreign talk discourse structure (Varonis & Gass, 1985b; Ehrlich, Avery, & Yorio, 1989). Ellis (1994) has pointed out that the use of models that account for discourse structure "constitutes a definite advance, as it enables researchers to examine the 'pouring back and forth' … consider[ed] essential for investigating how learners acquire language."

Varonis and Gass (1985b) proposed a model that shows how the discourse structure unfolds during the negotiation of meaning. According to this model, the discourse of conversation advances in a linear fashion, represented by a horizontal line in their model. When an instance of nonunderstanding occurs, speakers may engage in a series of exchanges with the purpose of resolving that particular breakdown in the conversation. These instances are viewed as vertical sequences along the horizontal line.

In Varonis and Gass's model, a negotiation routine consists of two parts: a trigger and a resolution. The trigger (T) is "an utterance or portion of an utterance on the part of the speaker which results in some indication of non-understanding on the part of the hearer." The second part of the routine, the resolution, consists of two primes: an indicator (I), by which one of the conversational partners lets the other know that something was not clear, and a response (R), which acknowledges the request for information. An optional prime, the reaction to the response (RR), may tie up the routine. In addition, Varonis and Gass offered a useful analysis of the types of primes used within a negotiation routine. This analysis reveals how interlocutors employ their linguistic resources in order: (a) to let each other know that something has not been successfully understood and (b) to solve the communication problem.


Varonis and Gass's model proves to be a useful tool to characterize and understand a particular type of interaction that is said to promote SLA. Although the synchronous electronic medium has been suggested to provide an environment that affords ample opportunities for learners to interact with each other, the actual advantages of this environment in terms of the opportunities it affords to negotiate meaning has not been addressed to date.3 Within this context, the study presented here explores whether this environment affords opportunities for meaning negotiation. The following section summarizes the findings of studies that suggest that computer-assisted classroom interaction facilitates interlanguage development.

Computer-Assisted Classroom Interaction

Recent studies suggest that the use of computer-assisted interaction may be beneficial in the language acquisition process. In particular, computer-assisted communication seems to allow for a more equal pattern of participation to the point that the instructor may become a mere member of the group (Kelm, 1992). In other words, the electronic medium has the potential to subvert the traditional roles enacted by teachers and students. In addition, it seems to afford more opportunities for learner output than oral discussions and to support a greater range of discoursal moves. In Kern's (1995) study, students produced more turns, words, and sentences in CACD than in face-to-face whole-class discussion, which suggests that CACD affords more opportunities for learner output than oral discussions. Two studies (Sullivan & Pratt, 1996; Warschauer, 1996) compared small group interactions in the oral and the electronic modes and found greater equality of participation in the electronic discussion. Chun (1994) found that electronic discussions prompted discoursal moves such as topic initiation and expansion, interactional moves (e.g., clarification requests, comprehension, and confirmation checks), and repairs in case of misunderstanding.

The present study adds to this line of research focusing on the discourse generated through computer-assisted interaction. While previous research on CACD offers a very encouraging picture of the synchronous written interaction, from an interactionist perspective, it should be noted that it is not only the amount of participation and/or production that matters but also the specific structure that the interaction displays. According to a growing body of research (Gass & Varonis, 1985b, 1986; Pica & Doughty, 1985; Pica, 1994; Pica, Kanagy, & Falodun, 1993; Scarcella & Higa, 1981; Varonis & Gass, 1985a; 1985b), the type of interaction that has been identified as "negotiation of meaning" is the one that provides optimal conditions for language acquisition since it offers opportunities to generate both comprehensible input and modified output. Given the relevance


of this interaction, the main goal that the present study pursues is to investigate whether negotiation of meaning occurs in the electronic synchronous written medium. A second aim of the study is to characterize the linguistic means conversational partners used to achieve message comprehensibility.


We follow the model for the negotiation of meaning proposed by Varonis and Gass (1985b). Table 1 shows an example of one of the routines used by two nonnative speakers to negotiate a nonunderstanding.

Table 1

Discourse Model of the Negotiation of Meaning With Example

0x01 graphic

Note: adapted from Varonis and Gass, 1985b.

In Table 1, the first utterance "My father now is retire" serves as a trigger in the routine. The NNS2 repeats the word "retire" with rising intonation, "retire?" Thus, this utterance functions as an indicator in the sequence. The NNS1's utterance, "yes," confirms that "retire" was the word she had used and constitutes the response. Finally, the example includes the optional unit of the routine, the reaction to the response, exemplified by NNS2's last turn, "oh yeah," which ties up the negotiating exchange.

The current study examines whether these negotiation routines emerge in the synchronous electronic medium. If found, of special interest would be to analyze the type of primes used in this medium as they can offer insights that explain why misunderstandings occur as well as which kinds of means interlocutors use to resolve them.


The participants of the study were foreign language university learners of Spanish. All of them were native speakers of English and were enrolled in a third-year course on grammar and composition at the time the study took place.



The task consisted of discussing several content questions about a reading assignment.4 Both the reading and content questions had been assigned as homework in the previous class. After the chat group discussion, students wrote a paper and pencil summary about the content of the reading. The study focuses on the first part of the task, the Open Transport (OT) Chat group discussion.5 The goal of using a chat session was to provide an opportunity for learners to work in collaboration so that they could clarify or develop ideas that they had not been able to work out on their own.


All subjects participated in the chat group discussion in two different sessions approximately 20 days apart from each other. The 28 students enrolled in the class were randomly assigned to one of four groups. Since there were some absences, the class was divided so as to include at least six learners in each group.

The instructions were presented in English and in written form on the computer screen followed by a set of content questions about the reading assignment. Students were informed that they would be able to switch back and forth from their chat room to the screen with the instructions by pressing a specific button in an adjacent control panel.

The written instructions explained to students they would have 20 minutes for the discussion part and 15 minutes to work on the paper and pencil summary of the reading assignment. Students were also told when to start and stop working on each part of the task.

Results and Discussion

We found instances of negotiation in all groups in both sessions except for one group in the second session. In what follows, we present and comment on excerpts that illustrate how nonunderstandings were resolved in the context of group interaction through the electronic medium. Example 1 presents one of the negotiation routines in the discourse generated in computer-mediated communication.

Example 1

0x01 graphic


NNS1 gives his/her opinion about one of the characters in the novel. This turn activates a nonunderstanding and, therefore, acts as the trigger. This utterance's function as a trigger is established in retrospect by considering the turn of NNS2 that explicitly asks for the meaning of the word mentiroso. Thus, NNS2's question functions as the indicator and points to the specific source of the nonunderstanding. In the last turn, NNS1 responds by giving the English translation of the Spanish word.

In the majority of the negotiating routines of this study there was a single indicator coming from one conversational participant, as in example 1 above. However, a few routines included several indicators which came from either one or from more than one member of the chat group. Example 2 contains three indicators (the turns preceded by an arrow) that clearly show that the meaning of the word tallarines is unknown by two of the participants.

Example 2

0x01 graphic

In order to understand how this negotiating routine unfolds, let us consider how the turn(s) of each of the five participants contributes to this exchange. The first turn, Que es el significado de los tallarines?7 is not a metalinguistic question; NNS1 is not asking for the meaning of an unknown word. The question refers to the role that the noodles play in the chapter of the novel students had been assigned as homework. Thus, que es el tallarines is the first indicator in this routine. The appearance of Sson tallarines "cards"?, the second indicator, may seem surprising given that an English equivalent of tallarines has been provided in the previous turn. This particular sequence of turns may be explained by the fact that the OT Chat does not register simultaneous turns as such. While two chatters may write simultaneously, the transcript shows the turns in the order they have been sent, not in the order they have been written. A second possibility is that NNS3 had actually read the response of NNS1. Her question would simply reflect she had in mind another alternative that was viable for her in this context.


The utterance, El tallarines cards o pasta, is the third indicator in this routine and suggests that NNS2 needs confirmation after having read two alternatives. What is interesting is that while the previous turns have worked to resolve the nonunderstanding for NNS3 (Ah, veo) they did not have the same effect on NNS2. The frequent absence of question marks in some of the turns may have led her to interpret El tallarines es "pasta" as a question rather than as an assertion. It is the intervention of NNS5, an advancement in the horizontal line,8 which seems to finally clarify the meaning of tallarines for NNS2. To sum up, this routine shows that the OT Chat allows each participant to engage in the negotiation of meaning, that is, to indicate a breakdown in communication and to work towards its resolution at different stages in the unfolding of the group discussion.

Some of the types of primes found by Varonis and Gass in oral interactions are also present in the electronic medium. In our data, some types of primes occurred more frequently than others. With respect to the indicator, Varonis and Gass identified several types: echo, explicit statement of non-understanding, no verbal response, and inappropriate response. Some of these types also appear in synchronous written group interaction. In example 3, the indication of nonunderstanding is expressed by an echo, the repetition of the unknown lexical item, followed by a turn with a question mark.

Example 3

0x01 graphic

The overwhelming majority of the indicators found in our data (19 out of 21) are what Varonis and Gass call "explicit statement of non-understanding." Most of them were expressed in the form of a direct appeal for assistance, for example: ¿qué significa X? 'what is the meaning of X?' as in example 4, ¿es X Y? 'is X Y?' and ¿es X Y o Z? 'is X Y or Z?' as in example 5, and ¿Qué? 'What?' as in example 6.

Example 4

0x01 graphic


Example 5

0x01 graphic

Example 6

0x01 graphic

Example 7

0x01 graphic

Students are exposed to formulas of the type "what is X?" in the instructional setting. They learn these chunks in beginning language courses and continue to use them regularly in subsequent ones. It is possible that the classroom context in which the task took place favored the use of this formula as a way to indicate nonunderstanding.

The predominance of explicit ways of expressing nonunderstanding and the infrequency (or absence) of other types of indicators in our data may be due to the use of the written medium as mediated by the OT Chat. It should be noted that there were only two instances of echoes, and no instance of inappropriate response.9

The low incidence of echoes in our data can be explained by comparing how the echo functions in the oral versus the written medium. In face-to-face oral interactions, an echo may indicate that the interlocutor is not sure about what s/he has heard or if s/he understands. In this type of interaction, suprasegmental features (e.g., intonation) and paralinguistic features (e.g., gestures, facial expressions, and head and eye movements) are part of the message and can help to clarify the source of the nonunderstanding. In addition, in an oral exchange the immediate pressure to keep the conversation going would favor the use of brief and less elaborated ways (e.g., echoes) to indicate non-understanding. In contrast, in the electronic medium one would expect echoes to be infrequent. On the one hand, they cannot emerge as a result of perceptual difficulties in the decoding of the message; on the other, the unavailability of the suprasegmental and paralinguistic features of oral face-to-face interaction mitigates their effectiveness. The additional time that the electronic medium


affords for interactants to code and decode messages would explain their preference for more explicit ways of indicating nonunderstanding.

Some of the factors already mentioned that relate to the nature of the medium can explain why inappropriate responses do not appear in our data. The exchange in Example 8, taken from Varonis and Gass's oral data, is not likely to be found in a written discussion.

Example 8

0x01 graphic

In this example, it seems that the presence of a certain type of noise— accent—interfered with the message itself, and the listener replied to an erroneously perceived utterance. Furthermore, in the oral medium, the pressure to keep the conversation going may prompt the interlocutor to respond in an unsuitable way. This pressure does not exist in written discussions. Thus, participants may choose to remain silent when they do not understand or are not sure whether they understand a previous turn.10

The response is a turn that acknowledges in some way the request for additional information that is implicit or explicit in the indicator. The types of responses found in Varonis and Gass' study were: repetition, expansion, rephrasing, acknowledgment, and reduction. Given that the majority of the indicators in the routines of this study are of the type ¿qué es X?, there are restrictions on the types of responses that might logically follow. Interactants might have chosen to rephrase their utterance to try to clarify the meaning of the unknown word. Nevertheless, they overwhelmingly preferred to use the native language equivalent. This type of response was not present in Varonis and Gass's study. We must take into account that the nonnative speakers of their study were second language learners of English who lived and studied in the country where the target language was spoken. In addition, not all of the learners shared a common L1. In contrast, we are examining here a situation of foreign language learning in which students share the L1 among themselves, with the instructor, and with the community at large. Consequently, the tendency to rely on the L1 should be expected.

Acknowledgments and reductions are not logical options after the types of indicator present in the routines of this study. Acknowledgment could follow an echo indicating that the interlocutor is unsure about what s/he heard, but this type of echo cannot occur in written interaction. Recall that the echoes in the present study were equivalent to "what does X mean?" The data contain only one occurrence of a response in the form of a repetition (see Example 9).


Example 9

0x01 graphic

NNS1 typed the definite article la and the following noun mucama together lamucama. Her turn functions as the trigger. The next turn with the indefinite article una preceding the sequence lamucama clearly shows that NNS2 interpreted lamucama as one word. Then, NNS1 self-corrects and separates the article LA from the noun mucama. It is interesting to note that a typing error resulted in a breakdown in communication similar to those that occur in oral interactions because of certain types of missegmentation errors (Peters, 1985).

Most of the breakdowns in the conversations were successfully resolved by providing a translation into English of the unknown word in the trigger. Still, in two exchanges the provision of an English equivalent in the response was not effective and brought about a series of embeddings in which the interactants tried to clarify the appropriateness of the use of a given term in the context of the classroom discussion. The excerpt in example 10 illustrates what we have labeled "pragmatic negotiation."

Example 10

0x01 graphic

This exchange starts with a negotiation routine to clarify the meaning of the word chulo. The response Chulo es como "pimp" triggers a first embedded routine in which the interactant attempts to justify the appropriateness


of the word in the context of the Spanish class. NNS3's comment (camilo es un "pimp"?! que dice! and saben que la profesora lee lo que escribimos, si?) and NNS4's turn (Carmen tu estas extrano!) suggest—or at least seem to suggest to NNS1—that NNS3 and NNS4 assume that chulo has the exact connotative value as 'pimp.'11 NNS1's response (Pienso que la profesora sabe que es chulo. No te preocupes! Chulo no es una palabrota) attempts to excuse the use of the word chulo in this context.

To sum up, the examples presented above show that the OT Chat allows participants to engage in the negotiation of meaning, that is, to indicate a breakdown in communication and to work towards its resolution at different stages in the unfolding of the discussion. As the examples illustrate, learners negotiated the meaning of a lexical item in the majority of the routines. Nonunderstandings that are likely to occur in the oral medium due to noise, accent, etc. cannot emerge in the written electronic environment. This situation explains the absence of certain type of primes (that are medium dependent) in the negotiation routines of the electronic medium. Participants showed a preference for certain types of indicators and responses. This preference may be related to the academic and the foreign language learning context in which the interactions took place.


The second language literature has identified negotiation routines in oral interactions by which learners give and receive feedback and help each other to modify output and obtain more comprehensible input. This process is said to contribute to second language development.

The present study provides evidence that learners of Spanish as a foreign language engage in negotiation of meaning in computer-mediated discussions. Nevertheless, some of the types of primes used in the electronic medium differed from those documented in the oral medium. In the overwhelming majority of routines, learners indicated a breakdown in communication by means of an explicit statement of nonunderstanding in the form of a classroom learned formula learned. Other types of indicators were either absent or very infrequent. The types of responses documented in previous studies were not present in this study, with the exception of one instance of repetition in the form of self-correction. The learners in this study resorted to their native language to resolve instances of nonunderstanding.12 While recourse to the L1 was an efficient and fast means to return to the horizontal line of the conversation, it did not push learners to modify their output. It remains to be seen whether the use of the L1 is a characteristic of a typical foreign language learning situation in which learners share the native language or whether there are other factors that might account for this use.


The model adopted to analyze the data of the study allowed the researchers to observe how each learner contributed and benefited from group interaction, both to indicate/resolve misunderstandings and to continue the discussion in the horizontal line. The electronic medium not only seems to afford more opportunities for active participation in a group discussion (Kern, 1995), but it also provides a forum where participants can engage in the negotiation of meaning at their own pace.


1 It does not imply that interaction involving teachers does not have any impact on interlanguage development. Studies such as Tanaka's (1991, cited in Ellis, 1994) offered evidence that interactionally modified input (through teacher-learner interaction) resulted in better comprehension and in more words being learned and retained over time than either baseline input or premodified input.

2 A possible objection that might be raised against group work is that learners are exposed to and might incorporate defective forms. Several studies (Gass & Varonis, 1989; Bruton & Samuda, 1980) have shown that learners do not generally incorporate errors of a nonnative speaker peer. On the contrary, there are numerous examples of modifications in the direction of the correct target language forms.

3 While writing the results of this study, it came to our attention that another investigation was addressing this issue though in a slightly different approach (Pellettieri, 2000).

4 The questions were on the content of an adapted short novel, Rosaura a las diez by Marco Denevi, that students were reading for this class.

5 OT Chat, a networking protocol used by Macintosh computers, allows users to engage in written synchronous discussions. Chatters sit at individual computer terminals linked together electronically. With OT Chat, it is possible to open chat rooms in which participants can work in groups. Each participant can compose and send messages which appear on all participants' screens. Participants can respond to whichever messages they choose.

6 Numbers between turns represent other utterances that are not part of the nonunderstanding routine. These utterances usually move the discourse forward in a linear fashion (represented by Varonis and Gass's model by a horizontal line).

7 Note that question marks and other orthographic conventions are sometimes omitted. Due to the online nature of the task, participants may have been more focused on the content of the messages than on the conventions of the written language.

8 A turn that moves forward the discussion of the content question.

9 No (verbal) response, one of the indicators that appears in Varonis and Gass's data, is difficult to trace in transcripts from computer-mediated group interaction due to the flexibility in turn taking that the combination of medium and group size affords.

10 Although there is less pressure in the written medium to respond than in the


oral medium, the number of conversational participants also contributes to lighten the turn-taking pressure.

11 There is no basis in the novel to think that the character Camilo is a pimp. The Spanish word chulo has several meanings, one of them being 'prepotent, arrogant' and another one being 'pimp.' It is not clear in which of the two senses the student is using the word chulo.

12 The L1 emerged, even though participants were specifically told to use Spanish only throughout the activity.


The authors wish to thank Dennie Hoopingarner and Michael Kramizeh for answering their many questions about technical issues.


Beauvois, M. (1992). Computer-assisted classroom discussion in the foreign language classroom: Conversation in slow motion. Foreign Language Annals, 25, 455-464.

Bruton, A., & Samuda, V. (1980). Learner and teacher roles in the treatment of error in group work. RELC Journal, 11, 49-63.

Chun, D. M. (1994). Using computer networking to facilitate the acquisition of interactive competence. System, 22 (1), 17-31.

Denevi, M. (1964). Rosaura a las diez (D. A. Yates, Ed.). New Jersey: Prentice Hall.

Ehrlich, S., Avery, P., & Yorio, C. (1989). Discourse structure and the negotiation of comprehensible input. Studies in Second Language Acquisition, 11, 397-414.

Ellis, R. (1994). The study of second language acquisition. Oxford: Oxford University Press.

Gass, S. M., & Varonis, E. M. (1985a). Variation in native speaker speech modification to non-native speakers. Studies in Second Language Acquisition, 7 (1), 37-58.

Gass, S. M., & Varonis E. M. (1985b). Task variation and nonnative/nonnative negotiation of meaning. In S. M. Gass and C. G. Madden (Eds.), Input in second language acquisition (pp. 149-161). Rowley, MA: Newbury House.

Gass, S. M., & Varonis, E. M. (1986). Sex differences in non-native speaker/nonnative speaker interactions. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (pp. 327-351). Cambridge, MA: Newbury House.

Gass, S. M., & Varonis, E. M. (1989). Incorporated repairs in nonnative discourse. In M. Eisenstein (Ed.), The dynamic interlanguage: Empirical studies in second language variation (pp. 71-86). New York: Plenum Press.


Kelm, O. R. (1992). The use of synchronous computer networks in second language instruction: A preliminary report. Foreign Language Annals, 25 (5), 441-453.

Kern, R. G. (1995). Restructuring classroom interaction with network computers: Effects on quantity and characteristics of language production. The Modern Language Journal, 79, 457-476.

Long, M. (1983). Native speaker/non-native speaker conversation and the negotiation of comprehensible input. Applied Linguistics, 4, 126-141.

Long, M., & Porter, P. A. (1985). Group work, interlanguage talk and second language acquisition. TESOL Quarterly, 19 (2), 207-227.

Long, M., Adams, L., McLean, M., & Castaños, F. (1976). Doing things with words: Verbal interaction in lockstep and small group classroom situations. In J. Fanselow & R. Crymes (Eds.), On TESOL '76 (pp. 137-153). Washington, D.C.: TESOL.

Nunan, D. (1988). The learner-centred curriculum. New York: Cambridge University Press.

Pellettieri, J. (2000). Negotiation in cyberspace: The role of chatting in the development of grammatical competence. In M. Warschauer & R. Kern (Eds.), Network-based language teaching: Concepts and practice (pp. 59-86). New York: Cambridge University Press.

Peters, A. M. (1985). Language segmentation: Operating principles for the perception and analysis of language. In D. Slobin (Ed.), The crosslinguistic study of language acquisition: Vol. 2. Theoretical issues (pp. 1029-1067). Hillsdale, NJ: Lawrence Erlbaum.

Pica, T. (1994). Research on negotiation: What does it reveal about second-language learning conditions, processes, and outcomes? Language Learning, 44 (3), 493-527.

Pica, T., & Doughty, C. (1985). Input and interaction in the communicative language classroom: A comparison of teacher-fronted and group activities. In S. M. Gass & C. Madden (Eds.), Input in second language acquisition (pp. 115-132). Rowley, MA: Newbury House.

Pica, T., Kanady, R., & Falodun, J. (1993). Choosing and using communication tasks for second language instruction and research. In G. Crookes and S. M. Gass (Eds.), Tasks and second language learning: Integrating theory and practice (pp. 9-34). Clevedon: Multilingual Matters, Ltd.

Porter, P. (1986). How learners talk to each other: Input and interaction in task-centered discussions. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (200-224). Rowley, MA: Newbury House.

Rulon, K. A., & McCreary, J. (1986). Negotiation of content: Teacher-fronted and small-group interaction. In R. R. Day (Ed.), Talking to learn: Conversation in second language acquisition (pp. 182-199). Rowley, MA: Newbury House.

Scarcella, R., & Higa, C. (1981). Input, negotiation, and age differences in second language acquisition. Language Learning, 31 (2), 409-438.


Sullivan, N., & Pratt, E. (1996). A comparative study of two ESL writing environments: A computer-assisted classroom and a traditional oral classroom. System, 29, 491-501.

Swain, M. (1985). Communicative competence: Some roles of comprehensible input and comprehensible output in its development. In S. M. Gass & C. G. Madden (Eds.), Input in second language acquisition (pp. 235-253). Rosley, MA: Newbury House.

Varonis, E. M., & Gass, S. (1985a). Miscommunication in native/non-native conversation. Language in Society, 14 (3), 327-343.

Varonis, E., & Gass, S. (1985b). Non-native/non-native conversations: A model for the negotiation of meaning. Applied Linguistics, 6, 71-90.

Warschauer, M. (1996). Comparing face-to-face and electronic discussion in the second language classroom. CALICO Journal, 13, 7-25.

Warschauer, M. (1997). Computer-mediated collaborative learning: Theory and practice. Modern Language Journal, 81, 470-481.


Marisol Fernández-García is an Assistant Professor in the Department of Modern Languages at Northeastern University. She teaches courses in Spanish language, linguistics, and applied linguistics. Her research focuses on input, interaction, and second language acquisition.

Asunción Martínez-Arbelaiz is an Assistant Professor in the Department of Romance and Classical Languages at Michigan State University where she teaches graduate and undergraduate courses in Spanish linguistics. Her research interests are Spanish syntax and second language acquisition.


Marisol Fernández-García

Department of Modern Languages

400 Meserve Hall

Northeastern University

Boston, MA 02115-5000

Phone: 617/373-3659

Fax: 617/373-2298


Asunción Martínez-Arbelaiz

Department of Romance and Classical Languages

314 Old Horticulture Building

Michigan State University

East Lansing, MI 48824-1112

Phone: 517/353-0769, ext. 130

Fax: 517/432-3844