a workshop in the

Bukowy Dworek, Poland, 27-29 April 2001

INTRODUCTION

by Wlodzimierz Sobkowiak

Guest editor

School of English, Adam Mickiewicz University,

Poznan, Poland

This is what appeared in the first circular for the CCAAL workshop in January 2001:

Applied linguistics is thriving in Europe and in Poland. The volume of foreign language teaching, translation or dictionary-making, to name just a few areas, is reaching unprecedented heights. Computer assistance is used more and more in all these activities. While there have been conferences devoted to applied linguistics and computer support in education, we believe that the issues concerning the theoretical and practical challenges currently encountered at the interface of the two fields have not so far been addressed properly. We feel that computer-assistance is either taken for granted or treated as a (yet to be deployed) panaceum for all applied linguistics problems and weaknesses.

We propose a different approach, one which is epitomized in the concept of 'challenges'. The workshop will be about the challenges of CAAL, both those which, while known, have not been faced and solved yet, and those which one may foresee in the future. We propose to identify and tackle these challenges here and now, in the workshop being part of the next Poznan Linguistic Meeting. The following is a short list of CAAL challenges which is meant to suggest and stimulate, but by no means exhaust the inventory:

• inflated promises and expectations,
• hardware and software problems,
• hardware and software availability,
• language barrier - software localization,
• user interface and effectiveness,
• learner and teacher training,
• classroom integration,
• Internet.

As far as CAAL itself is concerned, we propose a wide view in which 'computer-assistance' may be understood in many ways, from the minimum of word-processing to the maximum of systems fully equipped in artificial intelligence and working with no continuous human supervision. Applied linguistics, in turn, is equally liberally interpreted as encompassing practical linguistic applications of linguistics, socio- and psycholinguistics, computer-science, pedagogy, media studies and others. At the interface of these wide ranges of research will be, among others, such sciences and activities as:

• Computer-Assisted Language Learning/Teaching/Instruction (CALL/CALT/CAI)
• Computer-Assisted Translation (CAT)
• Computer-Assisted Lexicography (CAL)
• Applied Corpus Linguistics (ACL)
• Natural Language Processing (NLP)
• Computer-Mediated Communication (CMC)
• Human-Computer Interaction (HCI)

• Anatol Shevel showed how to teach the whole class with a multimedia program on one computer, using programs very much different from the ones accessible on the market. His programs are like a computer game, where there is only the content (dialogues, films, animations, etc.), but there are no language activities and the teacher is free to adapt each page of that multimedia book to present what is going to be taught.
• When talking about electronic dictionaries and learners of English, Wlodzimierz Sobkowiak presented the results of a survey investigating teachers' attitudes to e-dictionaries. The results show that although more than half of subjects know and use at least one computer dictionary, none of the teachers has used it in the classroom, even though most of them have the conditions to do that. The conclusion was that teachers react negatively to computer dictionaries and do not promote them, and that should be amended with proper teacher training.
• The next speaker, Krzysztof Jagielowicz, provided a theoretical background to other presentations by talking about Computer-Mediated Communication (CMC), and specifically about its advantages (equality, participation at one's own pace and without seeking permission, reduction of context clues relating to race, gender, handicap or status) and problems (flaming, abuse of power and control, lack of response, lack of purpose).
• Next, Jarek Krajka discussed the issue of training online teachers of English, giving the necessary qualities such a teacher should posses, as well as reasons why teachers of English should take their instruction online. The author presented the results of surveys conducted among students (pre-service training) and teachers (in-service training), and proposed specific training solutions for both of these environments.
• After that, Monika Tarantowicz-Gasiewicz, claiming that there are no established pedagogical standards for student model in CALL, wanted to establish some parameters and arrive at some objective standards. By working out a framework for standards, then choosing a pedagogical theory underlying them, deriving standards from it, she came up with a complex student model in a CALL environment.
• Pawel Topol, who could not make it to Bukowy Dworek, nevertheless contributed his paper on educational web-page design by language teachers. He wrote about the what (what is, and what can be, the content of educational web pages), the why (why is it advantageous to create private web-pages) and the how (how to design a page, possessing very little or no programming knowledge).

Altogether, then, in the six papers appearing here, the reader will find the discussion of a variety of CAAL issues: from the use of multimedia programs (Shevel), through electronic dictionaries (Sobkowiak), computer-mediated communication (Jagielowicz), teacher training in web skills (Krajka), intelligent CALL (Tarantowicz-Gasiewicz), to web-page design (Topol). It is easy to see that the focus of all six is the process of teaching and learning English as a foreign language. This is of course far from surprising: all contributors are practising teachers and/or teacher trainers who regularly use computer assistance in a variety of EFL tasks on the one hand, and critically reflect on the process on the other. Looked at from another angle, it is similarly not unexpected that the use of educational FL CD-ROMs, EFL electronic dictionaries, EFL-oriented e-mail or Internet is at the centre of attention of both teachers and researchers in widely conceived CALL, here and now, i.e. in the Polish FL classroom of 2001. This is not to deny that Polish schools have other, less technological, worries in the time of school reform and budget cuts. But there are telling signs that computer technology is rapidly entering the Polish educational scene, one of these signs being the enormous popularity of teacher-oriented journals like this one, "Teaching English with Technology", which has kindly devoted its special issue to the presentation of the content of the CCAAL workshop.

So, what is the bottom line? I would phrase it like this: yes, there is the challenge of bridging the technological gap, of overcoming teachers' reserved attitudes, of raising the computer/CALL literacy levels. Yes, it is true that disappointingly few (language) teachers use modern media in their day-to-day teaching practice. Yes, it is true that some (EFL) software is not very good in terms of didactic design or user interface. Sadly, enormous progress in language engineering has not yet made the quantum leap into speech understanding, which would open new, breathtaking vistas to CALL.

But all these challenges bring promise; the promise of more (artificially) intelligent CALL programming, the promise of even better access to the web with its alternative linguistic reality, the promise of language teachers who will not only know how to use computers to a good effect in class, but will also want to do it, for their own as well as their students' sake.

These promises will not come true by fiat; and they will not materialize overnight. Ultimately, it is us, teachers and teacher-trainers, who will bring the changes about. We will do it so that the educational potential inherent in the new generations of computer-savvy kids is not wasted. Nobody will do it for us.

Rzeszow Pedagogical University,

Rzeszow, Poland

When personal computers became affordable by a wide range of individual users, one of the programmers’ challenges was to offer virtual entertainment to our homes. First products were animated puzzles (like Tetris) or achievement games (like CD-man), which hardly communicated with the owner. One of the first most "talkative" was Larry - a game which could be properly run only through reading the messages.

The transition to present-day multimedia educational appliances was marked by awkward and expensive hardware attachments to a computer supplying school courses on discs the size of analogue LP.

Availability of a sound card and a CD-ROM drive opened new prospects for teaching practice by offering relatively small "talking" discs of high capacity. The state-of-the-art entertainment and educational programmes are now issued in a variety of titles and levels of sophistication.

The present article is an attempt to find a compromise between the attraction of multimedia facilities and teacher’s efforts to make their conversation practice most efficient.

The basic objective of the whole research is the ways of developing learner’s creative thinking which is regarded as an indispensable element in teacher training. This ability has proved to be most productive in triggering the fostering of cognitive processes crucial in the EFL teaching situation. The implementation of the communicative approach is regarded as development of the learner’s ability to take part in meaningful communication in different settings with special emphasis on adequate functional and situational language use.

The numerous articles and monographs which have recently been published on these problems have all tended to emphasise that creative thinking can be defined as a cognitive process which results in a number of alternative responses to a given task that are perceived in some way as novel or unusual. It is also very important to consider basic concepts inherent in creative thinking as including ideational, association, expression and word fluency, naming facility, figural fluency, figural flexibility, sensitivity to problems and originality (see Murray, 1991; Nonaka & Takeuchi, 1995; Wortham, 1994; Pask, 1976). These abilities are inevitably related to linguistic competence.

Practising language teachers realise that the situation of EFL framework shifts the focus onto other components of language proficiency in comparison with native speakers. The well-established four aspect communicative competence model which encompasses grammatical, sociocultural, discourse and strategic competence is not quite relevant for EFL teacher training in Poland, where one of the most critical limitations is scarcity of real world settings requiring the demonstration of proper language performance in terms of pragmatic strategies. As rightfully noted by Hrehovcik (2000), communicative theory puts emphasis on the fact that every speech act takes place in a specific social situation, and that the partners of communication have their specific sociological background. In consequence, this sociological situation is then reflected in the selection of formal or informal language, and the selection of lexical and grammatical units.

The crux of the matter here can be the teacher’s challenge to involve students in exercising special activities which further foster imagination and provide meaningful responses in typical life situations through the systematic analysis of ways and attitudes which prevail in familiar classroom situations. Among the most efficient ones I can mention elaboration on such notions as personal and emotional involvement in the work, peer assessment, self expression, co-operative learning, motivation, linguistic and communicative competence.

Furthermore, it is very important to realise the necessity of giving reasonable freedom to the students to work at their own rate, encouraging them to use their own knowledge.

Summing up the above theoretical implications I have been trying to work out a certain agreeable solution on how to combine the possibilities and attractions of multimedia equipment with the enormous role of human factor in efficient class language learning through conversation.

However technically updated a traditional computer language course is, it is always designed according to the same algorithm: the student is offered a set of vocabulary or grammar practice in variety of ways. The distinction between such programs basically lies in the scope of practice exercises and quality of graphics. They are intended to be practised by an individual student and are used as a multimedia manual of English. Without doubt, modern techniques of voice sampling and on-line grammar correction allow a learner sufficient independence from the teacher’s assistance. Nevertheless, judging by my personal teaching experience, such training can be treated as only a good basis for productive communication.

Considering presently most popular PC language courses (ET 4.0, Multimedia Language System, POP English, Euro++, English 1, Longman FCE Grammar ROM, Cambridge TOEFL, Days out in Britain, Doctor Watson, Novell Read With Me, Best Reading Programme Ever, etc.) we can see that they differ only in the degree of internal sophistication. Nevertheless, their structure remains typical: for vocabulary practice - supply the right word or phrase, for grammar - use the correct form, for phonetics - repeat after the speaker in recording mode. Of course, the programs have options of choosing the level, on/off timing mode or give the efficiency feedback, which, if done in class, enables the teacher to act only as a backseat driver.

To achieve the goal of providing a supportive environment to communicative learning I adopted another course of action. Apart from the mentioned drawback of classical linguistic programs, which minimise the human factor in education, there exists still more sizeable and rather common for Poland reason - scarcity of computers in schools. Though the situation is changing for better due to the governmental programs equipping schools in computer labs, it is still difficult for a language teacher to get access to the lab and have a lesson of English with computers.

The teacher’s creativity can be productively clued by a few principles and schemes of exercises used in communicative teaching. Basic grammar items on which the teacher can concentrate are: Article / Tenses / Passive voice / Conditionals / Modals / Reported speech / Adverb-Adjective / Gerund-Infinitive / Sentence structure (Emphasis) / Wordbuilding.

Types of exercises commonly used in designing communicative-type exercises include the following: Correct the statements, Complete the sentences (open-end sentences), Say the opposite of the sentence, Ask another student if statement is True or False, Ask another student for information, Guided questions (statement > ask "why", "where", etc).

Setting up a certain educational goal for each step of the game the teacher prepares exercises respectively. While running the game the teacher makes pauses (which is technically very easy) at certain steps of the programme to work with the students, who are usually absorbed by the development of the plot. Student’s involvement into what is happening on the screen creates a good opportunity to make a conversation going. The crux of the matter here is to adjust the tasks so that they fulfil their educational purpose, are entertaining in the given communicative situation and provoke students to language production.

School of English, Adam Mickiewicz University

Poznan, Poland

Twenty-five in-service teacher-trainees doing their one-semester post-graduate study of TEFL in the School of English, Adam Mickiewicz University, were asked about their experience of, and attitudes toward, electronic EFL dictionaries. Twelve yes-no-? questions were asked. The results of the questionnaire show that: (1) not a single teacher "has used a computer dictionary of English recently in class", even though most have access to one at home and at school, and few believe that it would disrupt the flow of the lesson, (2) teachers have rather superficial experience of MRDs, if at all, (3) those with no experience tend to have rather inflated expectations of MRDs' content and function.

There is some research on teachers' computer literacy and attitude to new educational technologies, both abroad (Simonson 1987, Galloway 1990, Massoud 1991, Pilus 1995, Brown 1999, Lam 2000, Tuzcuoglu 2000) and in Poland (Zemla 1998, Lis 1998, Gornikiewicz 2000, Gajek 2001, Krajka 2001). All results point to an urgent need for well-profiled teacher training, and emphasize that the lack of functional computer literacy is the greatest challenge to introducing CALL on a larger scale than is currently observed.

In none of the above studies, however, were respondents specifically queried about electronic dictionaries (even if other standard applications were touched upon: spreadsheets, text-editing, e-mail, etc.). And yet, the teachers' knowledge and experience of, as well as attitudes toward, electronic learners' dictionaries are among the most vital elements of the new situation in which foreign language (especially EFL) teachers and learners find themselves more and more often in Polish schools. FL dictionary is one of the crucial components of FL teaching and learning the world over, with many potential functions and uses actualized in practically all FL situations: from the formal setting of the FL classroom, through homework tasks, to more authentic contexts of multilingual communication, both in the real and virtual world (Internet). In all these situations EFL MRDs appear to be used more and more often; indeed sometimes to the practical exclusion of the more traditional lexicographic resources (e.g. in computer-mediated communication). Especially so, as many learners' MRDs take on new functions by offering substantial self-study components complete with exercises and tasks of all kinds, thus bridging the gulf between electronic dictionaries and CALL sensu stricto (e.g. Tschichold 2001).

I decided to probe some aspects of EFL teachers' MRD literacy, with particular attention paid to the changing attitudes relative to the teachers' experience, i.e. in addition to tabulating results globally I also grouped them according to the teachers' declared familiarity with MRDs.

Twenty-five in-service teacher-trainees doing their one-semester post-graduate study of TEFL in the School of English, Adam Mickiewicz University, were asked on 22^nd January 2001 about their experience and attitudes toward electronic EFL dictionaries. The questionnaire was run as (a short) part of a four-hour lecture crash-course on educational technology delivered by the author of this paper. Twelve yes-no-? questions were asked, eight of them duplicated from a much more comprehensive questionnaire circulated among over six hundred Poznan students of EFL in May 1998 (see Sobkowiak 1999). The questionnaire items ranged from rather technical issues of phonetic transcription to ones probing general attitudes to MRDs, as shown below. The four 'new' items, extending the coverage of the questionnaire to include the teacher perspective, are asterisked in Table 1 below. The ordering of all twelve is reverse-alphabetic, exactly as it was in the original questionnaire (see Appendix 1 for a facsimile).

The results have no pretense - from a small sample like this one - to reaching the full rigour of an empirical questionnaire study, but they are certainly suggestive to the point of proving it worth while to conduct a more thorough investigation. In the following table raw figures are presented for all respondents globally.

Table 1. Questionnaire results, raw figures (N=25)

The most significant observations derived from answers to the questionnaire appear to be the following:

Some interesting tendencies can be observed when data are grouped according to the professed respondents' experience of MRDs: I divided them, just like in my 1998 study, into 15 'experts', i.e. those who declared having "at least one computer dictionary" at home, and 10 'non-experts', who do not. In Table 2 below the eleven remaining questions of the original twelve appear in the same order as before, i.e. as they were originally presented to the respondents. With the original 1998 questions the top figure in each cell shows the students' response percentage, while the bottom one is the value scored by teachers in the present study. The other three questions, of course, only show the latter results.

Table 2. Questionnaire results, proportions, grouped into 'experts' and 'non-experts'

Expert teachers are more knowledgeable about some of the technical aspects of MRDs, such as their ability to show phonetic transcription on screen (47% <yes>, as opposed to 20% non-experts), but - surprisingly - not about some others: 53.3% experts, as opposed to 70% non-experts believe (correctly) that "one can search the whole text (including definitions and example sentences) in a computer dictionary" (experts may have older MRDs, and non-experts may have inflated expectations). There are also differences in attitude: one in three experts believes that "pictures are the main advantage of a multimedia computer dictionary", and not a single non-expert concurs. Non-experts are also staunch believers in the value of phonetic transcription in MRDs (80%, item #8).

Among the less technical issues, notice that some expert teachers (three persons) actually believe that "using computers with pupils in class must disrupt the flow of the lesson", while no non-experts do (experts know better?). And, as noticed before, not a single respondent, expert or non-expert alike, has "used a computer dictionary of English recently in class".

With the eight duplicate questions there are also some interesting differences between student and teacher responses. Take #4, for example: apparently students are more enthusiastic about MRD speed than teachers, but then it is students mostly who would function in dictionary lookup situations under time pressure... (for some reason there is a reversal of attitude as far as pictures are concerned, #5). However, another attitude item, #7, shows a more complex pattern in that more expert students than teachers believe that "it's easier to use a computer dictionary than a traditional one", but among non-experts, the proportion is reversed. With technical questions the knowledge of expert teachers vs. students does not differ very much, but non-expert students are characteristically more cautious than teachers in expressing their belief about the power of technology (e.g. questions #6 and #9).

To properly substantiate all these speculations a much more thorough study would be needed, of course, but some results are certainly perplexing. Specifically, notice that both among students and teachers there is complex correlation between knowledge and attitude. Experts tend to have different attitudes and expectations towards MRDs from non-experts. This is of course hardly unexpected; a similar pattern is common in other spheres of human life: experts after all know so much better the pros and cons of whatever they are experts in, and this usually fosters more realistic attitudes and motivations. It is here that the urgent need for computer training of non-expert ('computer-illiterate') teachers is most dramatically revealed: the growth of knowledge improves attitudes and motivations, these in turn are badly needed in conveying computer literacy to pupils/students, specifically in the foreign language context.

If it is indeed true that "student concerns about being able to cope with the demands of technology appear to cause some to choose less advanced language learning tasks", as Debski and Gruba (1998: 54) discovered, it is the obligation of the teachers to change these attitudes. If teachers themselves continue to have little knowledge, bad stereotypes and fears of educational technology, as appears from this study, no amount of computing machinery lavished upon schools will help. Specifically, from the point of view of MRD use, teacher trainers should think about it very seriously: the main challenge of electronic learners' dictionaries is in the head of the teacher.

School of English, Adam Mickiewicz University,

Poznan, Poland

Therefore, this paper will take a close look into both the advantages and disadvantages of the electronic mode of communication and, by drawing on the ‘critical perspective approach’, will report on the already-known empirical research in this particular field of computer-assisted applied linguistics. Consequently, the comparative study, investigating the extent to which Computer-Assisted Classroom Discussion seems to foster the development of students’ spoken language proficiency, will conclude the whole work.

The purpose of this paper is twofold. Firstly, it will briefly explore the current situation in the field of computer-mediated communication by taking a close look into both the pedagogically grounded advantages and disadvantages of this mode of electronic transmission and, subsequently, will report on the already-known empirical research in this particular field of computer-assisted applied linguistics. Secondly, as the majority of previous studies have been mainly preoccupied with the role of networking in improving learners’ written language proficiency and the thought processes involved in writing, the second part of this paper will venture to broaden this particular area of investigation by presenting the results of the author’s comparative study aimed at evaluating the computer-mediated potential in promoting learners’ acquisition of spoken language proficiency.

While referring to computer networks or computer-mediated communication (CMC) we are turning our attention to activities such as both the synchronous (e.g. real time) and asynchronous (e-mail, bulletin boards) modes of network-based computer exchanges. In a vein similar to Kelm’s (1996), familiarity with these networks will be assumed and no digressions will be made to describe them here.

One of the most pedagogically eminent features of CMC which distinguishes this particular mode of communication from the more traditional, oral one, is the greatly increased equality of participation in the electronic mode compared to the face-to-face discussion. Whereas in the latter discourse student participation ranged from 35% (Sullivan & Pratt 1996) to 37% (the first class in Kern 1995) to 60% (the second class in Kern 1995), the electronic mode actively engaged from 85% to 92% of its participants (85% in Sullivan & Pratt 1996; 86% and 88% in two classes examined by Kern 1995; 92% in Kelm 1992). Warschauer (1996a:7), in his experimental study comparing small group discussion online and face-to-face, observed that "the online groups were twice as balanced", which was the result of the fact that "the most silent students increased their participation many-fold online".

Similarly, McGuire, Kiesler & Siegel (1987) found in their study that in network-based environment women propose solutions to a problem as frequently as men do, not five times less often, as it was evidenced in the traditional, face-to-face discussions. Huff & King (1988) observed, that in CMC suggestions of both higher and lower status people are picked equally often as opposed to traditional exchanges where social position held by the participants virtually determined the weight of proposed arguments.

Such a great increase in students’ equality of participation in the electronic mode can be attributed to the fact that CMC reduces social context clues related to race, gender, handicap, accent and status, which normally reinforce unequal participation in other types of interaction (Sproull & Kiesler 1991), eliminates non-verbal cues, such as frowning and hesitating, which can intimidate those with less power and authority (Finholt, Kiesler & Sproull 1986, cited after Warschauer 1997), and, finally, enables everybody to contribute at their own time and pace, thus neutralizing those who tend to speak out loudest and interrupt the most (Sproull & Kiesler 1991).

Further, in the course of the computer-mediated exchanges students can initiate communication any time of a day (even outside the classroom) without seeking permission. In her study Wang (1993) stresses the importance of this particular feature of CMC to account for the fact that students conferencing via e-mail (when compared to traditional, paper-and-pencil dialogue journals) display the tendency to write more, generate longer stretches of text, ask as well as respond to more questions, and use a greater variety of language functions (Wang 1993).

Other pedagogically sound and potentially benefiting characteristics of CMC include:

• students’ chance to be engaged in intercultural communication conducted on a regular basis via a fast and inexpensive medium (see Warschauer (1995b) for the whole array of such examples),
• increased ability for reflection when compared to the traditional, oral mode (Lamy & Goodfellow (1999) argue that learners’ engagement in ‘reflective conversations’ in the asynchronous, computer-mediated learning environment fosters their language acquisition),
• numerous corollaries between theories concerning SLA and characteristics of electronic network communications, especially in terms of (cited after Kelm 1996: 21-23): (1) natural language environment (whereby the conversations focus almost entirely on content), (2) concrete referents (CMC topics usually relate to the here-and-now), (3) communication with peers (the process of SLA is believed to be more effective when learners’ model their speech with peers rather than teachers or parents (Ellis 1994)), (4) feedback (the large percentage of tags, questions, or requests for information that accompany the comments) and (5) affective factors ("students’ positive attitude toward using computers for writing and authentic, real-world communication in the language classroom, the feeling of personal empowerment and the enhancement of learning opportunities" (Warschauer 1996b: 41)).

Regrettably, however, the vast majority of the above-mentioned studies concentrate merely on noticing the potential of the electronic mode of communication and report on the, undeniably promising, corollaries between the theories concerning SLA and pedagogically salient characteristics of electronic network communications. As Warschauer & Kern (2000: 2) correctly observe, to date "there has been relatively little published and in-depth, data-based research that explores the relationship between the use of computer networks and language learning". Therefore, the above characteristics of CMC can only be perceived as the partially tapped potential of the great educational benefits which, in order to be explored to its full extent, requires a much more thorough examination of the network-based issues.

However, in spite of all the potential pedagogical benefits deriving from the very nature of CMC, some researchers set out to point our attention to the possible problems which can interfere with the concept of network-based collaborative learning.

To begin with, Weisband (1992) found that consensus-oriented electronic discussions differ substantially from their oral counterparts as the task of achieving an agreement is much more strenuous during computer-mediated sessions. Specifically, she comments that in face-to-face discussions the second member of a group was likely to agree with the first speaker, and the third even more so. As a result, by the time the third person took his turn, the group was on its way to reach a consensus. On the contrary, in online electronic discussion the third member’s proposal was equally far from the ultimate group decision as the first member’s was.

This intersection led Sproull & Kiesler (1991: 65) to conclude, that "electronic mail reduces conformity and convergence as compared with face-to-face group discussion". While Weisband (1992) focused her attention on the asynchronous mode of transmission (i.e. e-mail), it is a common preconception that in the much more vivid real-time interaction these results would only be strengthened. Obviously, this would characterise synchronous communication as being more appropriate for generating discussion and ideas rather than serving as a reliable means of solving decision-based tasks, the fact that has already been pointed out by Warschauer (1999) in the course of his excellent, 2-year ethnographic study.

Another factor of CMC that is capable of obstructing cooperative learning is the emergence of rude and belligerent language, called flaming. Flaming usually occurs as a side-effect of the same above-mentioned features of CMC which promote free expression, and can escalate to such an extreme degree that "in one electronic discussion participants had to be escorted individually out of the building" (Sproull & Kiesler 1991: 65).

Furthermore, Moran (1991) lists information overload as the other problem intervening with the proper flow of arguments during the computer-mediated discussion. Basically, participants can be so occupied with writing their own messages that they ignore the writing of others, making the conversation "a set of asocial monologues" (Moran 1991: 60).

Additionally, apart from its positive features, the shift of power (from teacher-centered, large-group sponsored teaching toward a more individualized and learner-centered working environment) can as well result in the abuse of power and control in computer-assisted environment (Janangelo 1991). Therefore, the very nature of CMC calls for the sudden urgency to train teacher professionals to effectively apply this new and powerful media into the classroom realia as, while not properly used, "new technologies can accomplish a great deal of good as well as a great deal of evil" (Janangelo 1991).

Other possible problems which can interfere with the concept of network-based collaborative learning include lack of response (featuring some students experiencing spirited international discussion whereas others are gazing at an empty screen) and lack of purpose (whereby the initial excitement quickly wears off) as the two major limitations that have been reported on by teachers involved in pen pal exchanges (Warschauer 1995a).

Undeniably, yet another potential source of problems preventing educators from trying to establish telecollaborative connections in their classrooms has its roots in administrative, technical and financial dilemmas, inequalities and barriers which most teachers of all kinds are confronted with on a daily basis in the course of their professional careers.

However, as Warschauer (1997) correctly observes, the peril of a language teacher losing control over the students conferencing in real-time on a computer network was metaphorically compared by Batson to riding the wild beast when he concluded: "Students on the network bring to bear their natural pleasure in social interaction; writing becomes more like talk. It seems an unruly beast at first, but... the way to deal with the beast is not to shoot it dead but to jump on its back and attempt to steer it" (Batson 1988).

Though the majority of the above-mentioned descriptive studies displayed some potentially positive characteristics of CMC, virtually all of the researchers were concentrated either on the role of networking in developing writing and the thought processes involved in writing, or merely on the general characteristics of both oral and electronic modes. Moreover, since the potential of CMC in improving students’ ability to express their ideas on paper, in spite of all the possibly hindering features described above, has already been noticed (for instance see Peyton 1990; St. John & Cash 1995; Tella 1991; 1992a; 1992b; Sullivan & Pratt 1996; Kern 1995), only a few (Chun 1996; partially Kern 1995) researchers ventured to investigate the role of synchronous Computer-Assisted Classroom Discussion (CACD) in increasing students’ spoken language proficiency.

Specifically, in her study Chun (1996) established that during the synchronous, electronic debate students displayed the ability to express a greater variety of linguistics functions in different contexts than was the case with the face-to-face discussion. This finding led her to determine the subsequent particular field of further research by remarking that "since these [computer] types of sentences strongly resemble what would be said in a spoken conversation, the hope is that the written competence gained from CACD can gradually be transferred to the students’ speaking competence as well" (Chun, 1996: 71).

And this is precisely the question of whether this newly-gained, electronic written proficiency is at all, and possibly - to what extent, transferred to students’ oral performance that determined the scope of the comparative study conducted in the computer laboratory in one of the Polish secondary schools.

Therefore, partially inspired by Chun (1996) study, the subsequent section of this paper will present both the procedure and the results of the experiment undertaken by the author in order to deepen our understanding of the processes that foster the potential learners’ acquisition of spoken language proficiency in the networked environment.

For the purpose of the present study two regular, intermediate classes of 3^rd year secondary school students (from V LO in Zielona Gora, Poland) were selected with each class additionally split up into two practice groups for the total of four (E1, E2, C1, C2) units formed (C1 and C2 stand for ‘Control 1’ and ‘Control 2’ groups, whereas E1 and E2 denote ‘Experiment 1’ and ‘Experiment 2’ groups). In the course of the study, groups E1 and E2 visited the networked computer laboratory once a week and held synchronous electronic discussions there whereas groups C1 and C2 were taught on the conventional basis, thus not being exposed to the electronic instruction at all. Basic typing skill was a requirement in both E1 and E2 formations and all the participants conformed with this demand.

The experiment was integrated into the school’s curriculum and, though the majority of the participants expressed their willingness to take part in the study, students were formally obliged to attend the classes regularly (critical attendance value was established at the 90% level so the allowances were made for one absence in a total of 10 sessions). All of the students were Polish, ranging from 18 to 19 years of age. Ten of them were male and 17 female (for gender distribution within groups see Figure 1).

Figure 1. Gender distribution within groups

Unfortunately, due to high level of absences among students enrolled in the control group C1 this particular formation had to be ruled out from the experiment and shall not be included in the further description of the study. For similar reasons, the number of subjects in groups E1 and E2 was reduced from 10 to 9 and 8, respectively, whereas group C2 remained intact throughout the study.

The general procedure was as follows: at the start of the study, during separate classes, groups E1, E2 and C2, each consisting of 10 students participated in traditional oral discussions with the teacher. The three sessions (pre-tests) were audio recorded and transcripts were produced and analysed with regard to the taxonomy proposed by van Ek & Trim (1991). The number of language functions used during the debates was calculated for each group separately and transcripts were kept for future reference. For the purpose of the study, in the course of the three-month long experiment, groups E1 and E2 visited the networked computer laboratory once a week and held synchronous electronic discussions there while group C2 was taught by an English teacher on the conventional, coursebook-based, basis. At the end of the study one additional, separate oral group session (post-test) was conducted with each unit, and again discussions were audio recorded, transcripts were coded on the basis of van Ek & Trim’s classification and the number of language functions used in each group was summed up. As a next step, the cumulative results of the orally held meetings were contrasted and evaluated (for the complete experiment schedule see Figure 2).

Figure 2. The experiment schedule

Consequently, the tenability of the claim that some kind of transfer between the learners’ electronic and oral production might have occurred, was tested. This hypothesis could be considered to have been substantiated if students from the experiment groups E1 and E2 had displayed the oral ability to express a greater variety of language functions in different contexts after the completion of the three-month long computer program. The claim was tested by employing the two-tailed, non-directional null hypothesis predicting no difference in the subjects’ oral performance after the electronic treatment. Subjects’ scores were examined at the 0.05 significance level (p<0.05).

The total scores on language functions for all the three groups have demonstrated, to various degrees, the improvement of subjects’ oral proficiency. In the case of the experiment group E2 the progress was the most evident with the subjects upswinging from the total number of language functions (214) uttered in the pre-test to the final score of 302 observed on the post-test (Figure 3).

Figure 3. Matched t-test. Experiment group 2

t value = 1.82 (critical t value = 2.306)

Improvement of groups E1 and C2, however, was less conspicuous with group E1 rising from the score of 175 to 258 (Figure 4),

Figure 4. Matched t-test. Experiment group 1

t value = 0.77 (critical t value = 2.262)

and group C2 remaining virtually at the same level (score of 204 on pre-test compared to 214 on post-test; Figure 5).

Figure 5. Matched t-test. Control group 2

t value = 0.26 (critical t value = 2.228)

Though, especially in the case of the experiment groups E1 and E2 (83 and 88 more language functions used after the three-month long treatment, compared to only 10 function improvement of the control group C2) the findings seemed to be more than promising.

Nonetheless, data were additionally checked for significance at alpha<0.05 in an attempt to reject the non-directional, two-tailed null hypothesis (H₀) and thus find the evidence in support of the alternative hypothesis of difference. Since the two means to be compared in every group came from the same subjects a matched t-test technique was used as the appropriate analysis tool suggested for verifying the significance of sets of paired data (Hatch & Farhady 1982; Butler 1985).

The critical t value for rejecting the H₀ hypothesis for groups E1, E2 and C2 was estimated at 2.262, 2.306 and 2.228 ratio, respectively (depending on the changing value of the degrees of freedom, i.e. the number of subjects enrolled in the study; the critical t values presented above were taken from Fisher & Yates 1974).

Nevertheless, the evidence in support of the transfer hypothesis was not found as none of the groups turned out to have a computed t value high enough to safely reject the null hypothesis with group E2 being the closest and missing the critical mark by a 0.442 value (for detailed results see Tables 1-3).

Therefore, though the students from the experiment groups E1 and E2 indeed displayed the oral ability to use a greater variety of different linguistic functions after the completion of the three-month long computer program (83 and 88 more language functions used compared to merely 10 function improvement of the control group C2), any findings of the study should not be seen as statistically significant results but rather considered to indicate trends in the specified direction while still lacking its measure of the essential empirical evidence.

Concluding, despite all the possibly hindering features, the potential benefits of computer-mediated interaction seem apparent and certainly capable of changing the face and customs of numerous collaborative language classes. However, as it has been suggested by Warschauer & Kern (2000), the carefully planned research effort, fostered by the easy access to the effective ways of electronic data analysis is demanded in order to help us determine WHAT exactly students are learning in CMC, multicultural environment. The very role of networking in developing writing and the thought processes involved in writing as well as the potential of synchronous CACD to play the role of the ‘bridge’ between speaking and writing and, consequently, contribute to the development of speaking ability is still unclear (though the above-cited experiment did display certain tendency in this direction). "The simple question to which everybody wants an answer" - Warschauer & Kern (2000: 2) observe - "Does the use of network-based language teaching lead to better language learning? - turns out not to be so simple", and still calls for the carefully planned, well-structured and data-based research effort.