Thorne, S. (1999). Chapter 3:
Educational and Foreign/Second Language Uses of Computer-mediation: A Review of
Research. In An activity theoretical analysis of foreign language electronic
discourse. Unpublished Doctoral Dissertation, University of California at
Berkeley.
Chapter
3:
Educational and Foreign/Second Language Uses of
Computer-mediation: A
Review of Research
Key words: Computer-mediated communication (CmC),
computer assisted language learning (CALL), computer-assisted classroom
discussion (CACD), drill-and-practice software, hypertext, human-computer
interaction (HCI), microworlds
“What
would “something better” look like?”
(Pavel Curtis, 1998)
This
chapter reviews a variety of research literatures and technologies concerning
the use of computers for the teaching and learning of foreign languages. A
comprehensive review of all research areas related to technology and FL/L2
learning is beyond the scope of this chapter. I have, however, tried to include
research which is representative of various fields, and those studies which, in
my opinion, are better examples of their genre and research focus. This chapter
provides a critical framework that illustrates what specific literatures both
highlight and put into shadow as a result of their research focus and
ideological orientation. Particular emphasis is placed on how research
frameworks and the traditions from which they emerge construe the object and/or
activity of study in specific forms.
Due to the volume and diversity of software environments put into use in foreign language classrooms over the past twenty-plus years, general categories for describing them are crucial, elusive, and, as Hawisher (1994) points out, inherently problematic. Standard classifications include: (1) drill-and-practice programs, (2) games, (3) hypertext books, (4) microworlds, and (5) computer-mediated communication and information. There are a growing number of web-based and video materials that do not necessarily fit into any of these restrictive categories. An example is the project on Quechua, reported on by Kramsch & Andersen (1999), that takes a linguistic anthropological approach to situating learners in the language-culture environment. For reasons of space and focus, web based instructional materials are not substantively discussed in this review, but I do tackle this rapidly emerging and important area in a separate manuscript (Thorne, in preparation). For historically framing primary forms of instructional technology, however, the five categories I list above offer a useful heuristic and show an overall temporal development leading up to the current preoccupation with internet-based communication.
In second and foreign language
education, drill-and-practice software provides untiring repetition of
particular language forms, with an emphasis on accuracy, fluency, and speed of
performance (Balajthy, 1986). Though offering self-paced advancement and
interactive in the sense of selecting between pre-set options, such software
often involves a student sitting alone at a computer trying to choose correct
answers from a list of multiple choices or supplying “correct
input” which is matched against a single possible answer. These
drill-based programs, some critics argue, often reduce educational activity to
a humdrum rehearsal of discrete subskills (Crook 1994).
The majority of software for
language learning falls into this first category. No doubt, part of the appeal
stems from a resemblance to the familiar language lab style Audio Lingual
Method activities of the past (e.g., structural patterns taught using
repetitive drills, structures taught in sequence one at a time, successful
responses immediately reinforced, taped dialogues). Faigley, in a compelling
book on postmodernity and the subject of composition, discusses Berlin’s
critique of cognitive rhetoric (specifically that of Flower and Hayes) and
notes that the compartmentalization of composition into discrete units
implicitly endorses “the structures of Fordist management” and the
representation of the mind “in terms of a rational hierarchy”
(Faigley, 1992:19). Such critiques apply to Audio Lingual Methods and are
pertinent to much drill and practice software as well.
From a software programming
perspective, drill-and-practice activities are attractive partly because
computers do such a good job at delivering such material. Expressed critiques
aside, drill and practice software has its value as a convenient, multiple
media tutorial resource, and the better designed programs can be productively
used as an adjunct activity or for self-study. In a Vygotskyan sense, a
tutorial program could act as a mediational resource and potentially create or
expand the learner’s zone of proximal development. With features such as
on-line help, L2-L2 translation (paraphrasing unknown lexical items and
constructions in the L2), and guidance for approaching a task or communicative
problem, a tutorial could support activity that the learner alone would be unable
to accomplish.
A second and not too distantly
related category of language software encompasses word games and puzzle
programs which, though still tending to focus on isolated skills, at least
offer a measure of entertainment. Game-like computer activities can make skill
instruction more palatable by increasing student motivation. In contrast to the
isolation imposed by drill-and-practice programs, game software at its best
can, like Brøderbund’s Carmen Sandiego series (available in several
languages), sometimes encourages whole or small group interactions around the
computer (Murphy-Judy, 1992). Focusing primarily on the acquisition of
vocabulary for Spanish and other more commonly taught languages, Syracuse
Language Systems (SLS) has developed a series of games based on sight and sound
propositions. Though SLS products were designed for solo use by students, in an
earlier, unpublished study I carried out in 1993, SLS games additionally
provided a common focus for younger students as they competed with one another
to achieve higher levels of proficiency. The subsequent competitive tensions
transformed what were designed as stand alone games into arenas of social
activity and in-class competitions (Thorne, unpublished manuscript). From an
activity theory perspective, the community and community rules that emerged
during the repeated use of this computer game produced an unexpected social
environment.
The useful but limited link
information employed in the Carmen Sandiego series, in which the students
physically look up detailed background information in a reference book in order
to make sense of the clues on the screen, anticipates a third category of
software relevant to second language education: hypertext. In a hypertext
information architecture, words or phrases on the screen can be electronically
linked to other texts and media representations in an electronic version of
what has described as inter-textuality (Kristeva, 1980; Lemke, 1995). While
reading, a user can choose to click on and follow a link and later return to
and continue reading the original passage. In addition, some hypertext systems
allow the student users to create new links themselves.[i]
Annotext is an example of software that keeps the main text central and
utilizes hyperlinks to supporting information, and users can add their own
exegesis and annotations.
This mechanism--the ability to juxtapose passages of text or text and other media--has the potential in second and foreign language education for making background information, vocabulary definitions, grammar explanations, inter-linear translation, commentary, or cross-references immediately and conveniently available to language learners. Moreover, recent research on the social nature of hypertext use in language arts classrooms suggests that the reading and creating of hypertext can be a social activity, involving the explicit connection of many social dialogues (Shen, 1995). The popularization and open access to hypertext afforded by the World Wide Web has created opportunities for a number of student foreign language projects. One of the earliest and still an excellent example is “Découvrir Berkeley!”, a web site designed and composed by 2nd semester UC Berkeley students of French in 1994. The Berkeley students were in e-mail contact with peers in France. Though these textual exchanges were interesting to the students, the recent availability of the web encouraged the idea of sharing sights and sounds in addition to text messages. For this project, students shot and digitized video, wrote and peer edited essays, and discussed issues of cultural representation as they decided what to portray to their French keypal classmates. Culturally and linguistically, students and the instructor reported the benefits to include heightened self-expression in the foreign language, large amounts of writing to and for a specific and authentic audience, and that the process of having to decide how to represent their home culture increased sensitivity to the cultural forms and stereotypes they had of France.[ii] The interaction and interpretive community that becomes possible when students not only view pre-existing hypertext, but create their own, provide foreign language activities which may enhance motivation and promote a diversity of language practices for language learners.
Seymour Papert coined the term
”microworld“ to describe a kind of computer-based learning
environment in which children don’t simply respond to predetermined
questions, but actually control when and how events will happen (Papert 1980).
In the view of Papert, who holds a constructivist perspective (in the Piagetian
sense), people learn by actively constructing new knowledge, and the learning
environments software designers create should allow a human learner to exercise
ideas or intellectual skills (Papert, 1980). Critics of Papert point out that
there are restrictive parameters within which students may be “active
builders of their own intellectual structures” (Papert, cited in Roszak,
1994:74). Roszak, a self-proclaimed neo-luddite who is openly critical of
instructional technologies, states that “as I read Papert’s words,
I found myself haunted by the image of a prisoner who has been granted complete
freedom to roam the “microworld” called jail: ‘Stay inside
the wall, follow the rules, and you can do whatever you want’”
(Roszak, 1994:75). Such contrary voices are important reminders as to the
limitations of microworld environments. Roszak’s fear of artificial and
parametered freedom, however, can also apply to the epistemology of formal
education at large.
The powerful normalizing
pressures and strict prescription of legitimate epistemologies and behavior
common to educational institutions have been the focus of much incisive
criticism (Bowles & Gintis, 1976; Bourdieu & Passeron, 1977; Apple,
1982; Giroux, 1988; McLaren, 1995). My own perspective is that technology
offers potential resources for certain types of cognitive and knowledge
specific development. The better designed instructional technologies offer
neither emancipation nor intellectual incarceration, but rather are implicated
in a broad set of institutional, historical, and discursive practices. As this
study will emphasize, the contexts created by users of technology, their prior
experiences and current uses of technology, and the social and academic
objectives of a given activity, both imposed and emergent, call for an analysis
that takes these contexts into account (see Thorne, 2000). Despite
Roszak’s perceptive sense of the limitations of computer-based
instructional software, missing is observation of how technologies are used by
individuals (and groups) in concrete circumstances, and the social and
developmental constraints and opportunities of concrete activity involving
specific tools and artifacts.
Of the growing number of
microworlds actually developed--the most famous of which being Papert’s Turtle
Logo--there is one strong example that has been used experimentally in
foreign language education. Developed at MIT for Project Athena, A la Rencontre
de Philippe is a microworld that incorporates users integrally into the
story line. Students’
activities and choices direct the movement of the narrative, and a number of
outcomes are possible, which makes for diverse story development over numerous
playings. Copious language reference material is available (access to help
material can be controlled by the instructor), as well as transcriptions of all
spoken texts. The French used in A la Rencontre de Philippe has been
termed at times unclear, native-like and difficult to understand by some
instructors, but it is well-suited to pedagogies that stress authentic language
use. The availability of grammatical and lexical references, an additional
voice rendition of all dialogues, and transcriptions of all spoken language
provide students the resources needed to parse initially incomprehensible
streams of phonemes. A major
strength of A La Rencontre de Philippe is the Gallic storyline that
users are exposed to (boyfriend (Philippe) gets dumped by girlfriend and has to
move out of the apartment they share, then must look for a place to live in the
horrifically impacted housing market of Paris. In some story endings, Philip
reconciles with his girlfriend at the end of the story), and the scrutiny they
must give to the dialogue to uncover its nuances.
Missing from Papert’s
influential treatise on microworlds--and from the Piagetian theories which
inspired it--is any in-depth discussion of the social/interactive aspects of
learning (Wood 1989). Not surprisingly, most educational constructed
microworlds present their electronic worlds full of virtual objects only to the
individual user (Crook, 1994).[iii]
Programs such as A La Rencontre de Philippe do in fact act as the focus
for collaborative group efforts, which illustrates that well designed tasks and
student orientations to the use of microworlds have the potential to make
social the interactive environment catalyzed by these programs.
Fürstenberg, the principal pedagogic designer of A La Rencontre de
Philippe, worked extensively to develop structured uses of this laser disc
program. Such careful attention to its use as a resource for collaborative
activity is responsible, in large part, for A La Rencontre de Philippe’s
positive reception and productive use in French foreign language courses.[iv]
In contrast to the isolation of
most constructed microworlds, a fifth and now booming category of instructional
technology use is the networked environment, and specifically forms of
computer-mediated communication (CmC). The impetus behind educational uses of
CmC is language development through social interaction. Over the past decade,
the ability to link students by networked computers has opened up a variety of
opportunities for language based social interaction in second and foreign
language education. Language use over networks provides a variety of
communicative situations, many of which are not readily available in foreign
language (Cononelos & Oliva, 1993) or composition (Colomb & Simutis,
1996) classrooms.
CmC comes in a number of forms.
Asynchronous forms of CmC include email, threaded discussion (also called
forums), and bulletin boards[v]
(now less frequent due to the ubiquity of the internet). Synchronous
communication, nicely described by Ferrara, Brunner, and Whittemore as
“interactive discourse” (1991), is dividable into four main types.
Werry (1996) documents the first three as 1) UNIX based TALK, used for
one-to-one conferencing; 2) MOOs and MUDS, text-based virtual environments such
as CafeMOOlano used to enable the computer-assisted classroom discussion (CACD) analyzed in this study. MOOs see
extensive use as role playing environments, as sites for educational activity,
and are commonly organized for theme based social interaction; and 3)
“Chat” systems such as Internet Relay Chat (IRC) used primarily for
thematic social engagement and by interest groups for topically specific
discussion. The key difference between MOOs and IRC is that MOOs are built upon
a spatial metaphor and IRC is not (MOOs will be described in technical detail
in the chapter 4).
There is a fourth and more recent
synchronous modality that combines a graphical interface with interactive
written discourse. Environments such as The Palace[vi]
and Worlds Alive allow users to move through three dimensionally rendered space
in the form of icons or virtual figures. Coming specifically to this study,
CACD indicates a form of interactive written discourse that occurs in formal
educational settings such as classrooms or computer labs. I first came across
this term in a paper by Ortega (1997), and find it useful as it juxtaposes the
use of real-time conferencing with the school/university setting as
institutional location. A primary difference between my work and that of
prominent cyber-theorists such as Sherry Turkle and Steve Jones is that my
study focuses on the use of internet based tools within the institutional
confines of classroom activity. Turkle and Jones provide illuminating and
important documentation of the digital cultures flourishing on the internet.
The existence of these digital cultures articulate in important ways with
contextual, social and discursive properties emergent of CACD. These issues
form the entirety of chapter 6, which provides an activity theoretical analysis
of the CACD data forming this dissertation.
CmC use in educational
settings--text-based
communication mediated by e-mail, electronic bulletin boards, local area
network conferencing programs like Daedalus InterChange and Aspects, and
real-time internet communication tools such as MOOs and forms of
“chat”--have become key technologies which make possible the
implementation of social constructivist pedagogies. Putting students in direct
communicative contact with one another, and distributing the printed transcript
of their on-line interaction as data for later written work, brings students
thinking and writing into the classroom as legitimate knowledge (Bruce, Peyton,
& Batson 1993). Network technologies have helped to initiate a significant
pedagogical shift, moving many language arts educators from cognitivist and
disembodied assumptions about knowledge and learning as a brain phenomena, to
contextual, collaborative, and social-interactive approaches to language
development and activity (Hawisher, 1994; Noblitt, 1995; Ortega, 1997).
The ‘electronic writing
space’, as Bolter (1992) describes computers for writing, and now
internet-based writing and communication, is coming to be accepted as as great
a revolution from the Gutenberg press as the Gutenberg press was from hand
copied text, and further back, from the invention of scripts and face to face
communication. Not only are
electronic texts in many ways more plastic than handwritten texts, but via
networked computers, the potential audience for such texts, whether as polished
written artifacts or a three sentence email message, is beyond the scope of any
communication or print technology that existed previously. This late modern
communication and information shift is resulting in two substantial, and
often-discussed changes. The first is the changing and multivalent structure of
‘texts’ as they are produced and consumed in digital and hypertext
environments. The second shift involves a reduction of the time and space
constraints which characterized pre-digital and pre-internet communication and
information practices.
The activity theory foundation
for this study makes the claim that the structure of texts, and textual,
literacy, and communicative practices, are tightly bound to the materiality of
their conveyance and representation (e.g., stone engravings, paper, computer
generated documents, and networked communication). Making an implicit
comparison between paper-mediated literacy practices and computer-mediated
hypertexts (though not from an activity theory perspective), Landow has argued
for the need for a new paradigm, saying “we must abandon the conceptual
systems founded upon ideas of center, margin, hierarchy, and linearity and
replace them with ones of multilinearity, nodes, links, and networks”
(1992:2). The relationship between medium, the structure of texts, and literacy
practices, is addressed by Lanham this way, “Sooner or later, …
electronic texts will redefine the writing, reading, and professing of
literature …” (Lanham, 1993:1). Lanham, concerned primarily with
rhetoric and the teaching of humanities, makes an even stronger statement later
on: “Electronic technology is full of promising avenues for language
instruction; it will be lunacy if we do not construct a sophisticated
comparative-literature pedagogy upon it” (1993:23, cited in Noblitt,
1995:269). By this, Lanham is emphasizing the need to adapt our pedagogies to
accommodate the changing social material conditions that electronic texts and
communication create. Like Landow, Lanham does not base his views specifically
on an activity theory approach, but his overall perception of strong
relationships between material conditions and social-cultural expression
(base-superstructure relations in the terminology of Marx), in this case
digital information and communicational artifacts and the social-linguistic
practices that they make possible, aligns closely to the theoretical approach I
develop in this dissertation. In this sense, the critical theory substrate which
implicitly evokes Marx’s base-superstructure relations can be seen to
permeate Landow and Lanham’s push to illustrate the substantive
aesthetic, communicative, and cultural shifts which are resulting from the
changing human practices that digital and networked technologies make possible.
The second aforementioned
dimension to the everyday practice of digital communication involves a
reconfiguration of time and space. Time is flexible as asynchronous and
synchronous communication tools (e.g., email, IRC, chat) support communication,
across any distance, almost instantly. Email can be composed during
non-traditional work hours, allowing individuals to personalize their work
schedules.[vii]
MOOs and IRC support multi-party real-time discussions in writing and can
synchronously accommodate individuals scattered around the globe. CmC supports
synchronous and asynchronous textual communication, as well as on-line
environments in which disparate participants can simultaneously gather and
communicate with one another through a variety of textual, visual and even
kinesthetic channels.[viii]
These two dimensions--the structural alterations accompanying communication,
the production of texts, information storage and retrieval on-line, and
increased options for customizing time and space, potentially constitute a new
age of community and communications practices.[ix]
These changes are rapidly becoming invisible to first world users as they enter
the doxa, or undiscussed background, of everyday practice. Barry Wellman, a
digital technologies researcher, notes that “New technologies get more
exciting when they get boring” (personal communication). Internet
communication tools are unproblematic for many of the participants in my study.
These students use email, chat, and instant messaging to communicate with
distant and local friends, siblings, and even parents and grandparents. The
educational uses of real-time conferencing and, not addressed in this study,
the production and consumption of web based texts, are indeed as or more interesting
now that they are “boring”. With the practices internet
communication tools support come new dimensions to existing and emergent social
relationships.
Below, I discuss some of the more
important empirical studies that have focused specifically on foreign language
uses of electronic conferencing. They represent primarily the socio-cognitive
and interactionist research traditions within SLA, though Kern, in particular,
develops a syncretic approach which combines productive elements from
socio-cognitive as well as sociocultural approaches. In a more recent article
on technology and foreign language literacy, Kern in fact moves more toward a
sociocultural perspective, stating that “Technologies of writing are
always tied to particular forms of social interaction and conceptions of
literacy” (1998:57). The bond between technology-as-artifacts and human
activity will provide a starting point for my later analysis. In this analysis,
I attempt to show a dialectic (though not necessarily symmetrical) relationship
between technology-artifacts and human uses of them, and the specific
characteristics of computer-mediated social interaction that form a
heterogenous and functional activity system which supports foreign language
development.
As I mention at the start of this
chapter, the studies I review below provide strong evidence for certain
positive claims about the use of synchronous CmC that I am in overall agreement
with. This section begins with an example of early research on the foreign
language use of a local area network program. In comparison to more recent
research, it is impressionistic and less rigorous. The overall conclusions,
however, are largely predictive of the findings that more empirical and
quasi-experimental research would later come to document.
To assess variable student
learning outcomes in a Spanish writing class, an experiment was conducted at
the University of Arizona comparing computer-conferencing, word processing, and
a traditional class setting.[x] Forty-four students used a CoSy
computer conferencing system to engage in writing activities that stressed
communication. A second group of
twenty-four students used word processing to write compositions, and a third
group of fifty students wrote compositions at home using pen and paper. Avots, citing Smith’s findings,
reports the following:
“The forty-four students
who used computer conferencing demonstrated creativity, progressed in written
expression, initiated cooperative learning ventures such as peer teaching and
tutoring, and actually spent more time writing than those who used only word
processing or wrote compositions at home as in a traditional classroom. Student outcomes demonstrate that an
environment which allows for interaction with others at any time beyond class
time and without pressure promotes learning as well as acquisition of
communication strategies.”
Though Smith’s conclusions
remain anecdotal, she formed four hypotheses based on her assessment of the Arizona
experiment:
§
Computer-based
interaction encourages increased time on task.
§
Computer-based
writing promotes creativity as well as accuracy.
§
Excessive
emphasis on accuracy can detract from the development of interactive
communication skills.
§
Development
of advanced organizer and subvocalization techniques affect students’
ability to communicate ideas orally as well as in writing (Smith, 1990:81, in
Avots, 1991:141)
More current studies on the use
of real-time electronic conferencing for foreign language learning (c.f., Kern
1995; Chun, 1994; Beauvois, 1998; Lee, 1998) begin to illustrate a more complex
and empirically substantiated picture of foreign language uses of sychronous
electronic communication. Foreign language educators need to know why electronic
discourse, in its many forms, would be a productive addition to their already
full curriculum. Researchers such as Kern (1995) and Beauvois (1998) address
specific dimensions of this question by focusing on the relationships between
the use of real-time electronic classroom discussions and oral proficiency
(Beauvois) and the differences between electronic discussion and face to face
classroom discussion (Kern).
Beauvois has carried out research
on foreign language uses of electronic conferencing since the early 1990s. In
her most recent work, she conducted a study of 4th semester French students
“to examine, under controlled conditions, whether a link between written
and oral communication can be established” (1998:94). Two classes participated
in Interchange (a local area network computer environment supporting real-time
conferencing) sessions once a week, while two control sections conducted an
oral discussion on the same topic. Students were randomly assigned. Beauvois
then tested both groups at the end of the semester by way of three oral
examinations. The Interchange groups achieved significantly better grades on
their oral exams than the control groups, which Beauvois suggests may indicate
a link between written real-time conferencing and the development of oral
foreign language proficiency. Beauvois notes that other researchers have shown
increased student motivation and participation, especially in composition and
L1 settings, while established linkages between electronic conferencing and
oral competency have been at best anecdotal. In her study[xi],
the “overall superiority of the oral expression in the exams of the
experimental group surprised the researchers and teachers” (1998:108).
Beauvois breaks down what she posits are contributing factors to her
study’s positive outcome. Electronic conferencing allows students to
“communicate at will” with “no turn taking (i.e., waiting for
others to respond to a given question)”, and to do so with other students
as well as the instructor (1998:103). Adopting SLA’s default
“input” metaphor, Beauvois notes that CmC sessions contribute to
“more contact with the target language than is possible in the
traditional classroom”, and that “more input, or more intake (Lee
and Van Patten, 1994), leads to more proficiency in the target language”
(1998:108).[xii] Though
grades on oral proficiency examinations are not an ideal indicator of language
development, Beauvois rightly asserts that the findings of her study provide a
starting point for evaluating the use of networked classroom discussion as a
tool for improving oral expression in a foreign or second language.
Kern’s students used the
electronic conferencing program Daedalus InterChange, the same local area
network software Beauvois used with her students.[xiii]
Putting numbers to the notion that foreign language students communicate more
in electronic conferencing environments than they do in large group
face-to-face classroom settings (in the quasi-experimental conditions he
analyzed), Kern compiled the following statistics from one of his second
semester French courses. From a 50 minute InterChange session on the topic of
the French RU 486 abortion pill, the total number of student messages, or turns
at talk for the period, was 172, teacher messages were zero, and each student
averaged 12.3 messages for the period. In comparison, an oral discussion by the
same students on the same topic produced 95 student turns at talk, 116
instructor turns, producing an average student production of 5.3 turns for the
face to face discussion. The students in Kern’s study produced between
twice and three times more turns, and more total number of sentences and words,
when they were interacting via InterChange when compared to the large-class
oral discussion on the same topic. In his conclusion, Kern states that
“compared to oral discussions, InterChange
was found to offer more frequent opportunities for student expression and to
lead to more language production” (Kern, 1995:470).
Beauvois (1998), Kern (1995;
1998), and Pratt and Sullivan (1994) all note either anecdotally, or in
Kern’s case, with empirical evidence, that students produce more
language, submit more turns at talk, and participate at high levels in
electronic conferencing sessions. Ortega (1997), in a thorough review article
of L2 computer-assisted classroom discussion brings forth some limitations to
research comparing computer-mediated classroom and whole-class oral
discussions. Ortega opens up the question of group size and communicative task
as variables not accounted for in Kern (1995) and others (e.g., Beauvois, 1992;
Chun, 1994; Kelm, 1992) research. She argues that “it is justified to
hypothesize that group size and equality of participation are negatively
related in traditional oral interactions and positively related in computer-assisted
interactions, and that the benefits of electronic over non-electronic
interactions will increase with the size of groups ... In other words, the
positive equalizing effect of the electronic mode will be accentuated when
comparing larger groups, as in the comparisons of teacher-fronted, whole-class
discussion with whole-class electronic discussion.” (Ortega, 1997:86)
Ortega also points out that communicatively oriented L2 pedagogies often
encourage small group work rather than whole-class discussions which may limit
the usefulness of a comparison between CmC and oral whole-class discussion. Her
issue here is intuitive but not empirically substantiated even as a
“hypothesis”, and I present my reasons for disagreeing with her
below.
Whole-class teacher led
discussions do indeed produce participation structures which differ from small
group and dyad work (e.g., Mehan, 1982; van Lier, 1996). Additionally, there
are well documented hypotheses about the benefits to group work (e.g., Long,
1989). But large group CmC does not always produce an overall increase in
full-group participation nor increased per participant language production.
These entailments seem to be an inherent, unexamined assumption on
Ortega’s part. In my own study and work with foreign language electronic
discussion over the past years, the production of groups over 12 or 14 in
number can require an overwhelming amount of reading for some students. Numbers
upward of 18-20 tend to produce a tier of prolific and experienced or more advanced
foreign language users who can read and post messages at the pace required to
keep up with the current development of the multiple conversational strands. A
second tier of students, however, may feel overwhelmed by the reading alone,
and their participation rates, measured by the number of posted messages, may
drop as the number of participants increase (this point is substantiated in
both my quantitative analysis in discussion with students in my interview
data).
To make these issues yet more
complex, coming again from my interview data, students who became overwhelmed
with the number of messages coming onto their screens did not necessarily feel
like they were not engaged. They rather felt as though reading and following
the various strands of conversation was already enough excitement and they did
not, or could not, participate directly as contributors (further discussion of
this point follows in a later analysis chapter). This latter point elicits the
potential need to disambiguate “participation” from
“engagement”, and calls for studies that would control for the
number of participants and activity type at the etic or systemic level of
analysis, while also explicating the emic, or user relative understanding of
the activity (via think-aloud protocols and/or in-depth interviews). I agree
with Ortega’s call at the end of her paper that “if task processes
and task conditions are left outside of the scope of inquiry”, the uses
of CmC for foreign language instruction “constitutes yet another
‘black box’ [in reference to Long, 1980] in L2 classroom-based
research” (1997:92).[xiv]
Close attention to the imposed organization of the activity under study, and
even closer attention to the dynamics and goals of the participants as the
activity is underway, are critical guidelines for the advancement of research
on L2 computer-mediated classroom discussion.
A necessary follow-up to the
issue of language production is the quality of language students produce. This
is an especially important question for institutional language classes since
there is only so much time in a foreign language curriculum for structured
language activities. How does the quality (and not mere quantity) of student
produced CmC language compare with oral discussions or written work that could be
taking place instead (the above discussion of task and activity types, and
participant numbers aside for the moment)?
Beauvois does not assess the
linguistic quality of her students’ computer-mediated foreign language
use (as this is not the focus of her study), but includes her impressions that
InterChange sessions were marked by “freedom of expression, full class
participation, openness and honesty …” (1998:107). Kern does
examine the linguistic quality of electronic conferencing discussions, and his
analysis reveals that “students’ language output was of an overall
greater level of sophistication in terms of the range of its morphosyntactic
features and in terms of the variety of discourse functions expressed”
(Kern, 1995:470). In a 1994 study of 4th
semester German foreign language students, Chun concludes that the students she
looked at demonstrated increased morphological complexity in their written work
over the course of the semester (greater ratio of complex sentences).
Additionally, and in agreement with Kern’s findings, Chun’s German
foreign language students used a wide array of discourse features in their use
of real-time networked computer discussions.
In a forthcoming study,
Pelleteiri has taken an interactionist SLA approach and applied it to foreign
language synchronous written discourse (Pelletieri, in press). I will briefly
spell out tenets of the interactionist paradigm in preparation for a discussion
of Pelletieri’s significant study. Negotiation of meaning is a core
component to interactionist approaches to foreign and second language
acquisition studies. The hypothesis is that non-native speakers may experience
(or be given tasks that precipitate) communicative breakdowns that require
negotiation to resolve. The negotiation process can include modifications on
both a linguistic and interactional level, and these modifications are presumed
to increase the comprehensibility of the talk at hand (e.g., Long, 1985; Pica,
1987; Varonis & Gass, 1985). Comprehension, or more specifically “comprehensible
input” (Krashen, 1982), has been argued to foster L2 acquisition (for
Krashen, in a way similar to L1 acquisition) and promote the development of a
learner’s interlanguage. The notion of communicative output, first
discussed by Swain (1985), offers a corrective to the emphasis on communicative
input by claiming that input is decipherable through largely semantic
processing while output requires the communicator to syntactically parse and
process the target language. These dimensions to communicative output are
hypothesized to promote a conscious attention to morphological and syntactic
form, and salience to form is in turn argued to foster interlanguage
development (e.g., Long, 1996; Schmidt, 1990).[xv]
Pelletieri adopts this understanding of SLA processes and profitably applies it
to the analysis of intermediate Spanish language learners using a Ytalk[xvi]
(Unix based) synchronous communication tool.
Pelletieri’s research
contrasts with that of Kern (1995), Chun (1994), and Beauvois (1998) in a few
important respects. She focused only on dyads and provided specific and often
closed tasks (e.g., jig-saws), whereas most research on FL CACD (including this
one) has examined larger group interaction in relatively open topical
discussions. Pelletieri concluded that her participants negotiated at all
levels of discourse, and that this prompted the learners to produce
form-focused modifications to their turns. She posits that dyad work encouraged
more corrective feedback than larger group discussions would have. This
corrective feedback “was demonstrated to result in the incorporation of
target language forms into subsequent turns” (Pelletieri, in press).
Additionally, task type--specifically goal oriented closed tasks--was
positively correlated to the quantity and type of negotiations produced.
Pelletieri, along with Kern (1995), notes that participants have more time to
monitor and produce their turns and that this may “play a significant
role in the development of grammatical competence among classroom language learners”
(Pelletieri, in press). The foundation for my own research diverges from the
interactionist approach taken by Pelletieri. This said, her work is useful as
it examines synchronous digital communication at the micro-interactional level
and descriptively shows that negotiation and feedback can encourage a focus on
form in CmC settings. Commensurate with Pelletieri’s conclusions, in my
interview data students discuss a heightened awareness of and concern for form
in CACD (see Chapter 6).
In her “linguistic
elements” discussion, Beauvois refers to the written communication in
electronic conferencing as “speaking” (in quotations). It is
unclear how she theorizes the language activity occuring in her data, and
“speaking” presumably indicates an implicit acceptance of the
“hybrid oral-written genre” approach describing electronic
discourse. With attention to the difficulty of He continues by making the
circumscribed claim that “during an InterChange session students may
operate essentially within an oral framework, even though the medium is
written” (1995:460). In a recent article on foreign language literacy,
Kern (1998) compares written communication by pen and paper to written
communication linguistically characterizing electronic discourse, Kern notes
that the language in his data represent an “outgrowth of students’
social
interaction, and its form
reflects that context” (1995:460). mediated by MOO technology and
provides the following contrastive table:
Table 3.1: Pen and Paper vs. MOO Environments
|
Pen and paper writing
|
MOO Environments |
|
• Writer supplies and
controls all text |
• Writer’s input is
transformed by MOO (i.e., not an exact input-output match) |
|
• Formatting and
punctuation are entirely writer’s responsibilities |
• Formatting and some punctuation
handled by the MOO • MOO specific commands
must be used, but are not visible in the language output |
|
• Imagined, anticipated
interaction with reader |
• Interactive environment
(even in the absence of an interlocutor, one’s writing allows one to
navigate through locales, meet robots, etc.) |
|
• Overt response from
reader (if any) is deferred well past the moment of writing |
• Expectation
of immediate response to writing (either from other participants or from the
MOO itself) |
|
• Allows one-to-one
or one-to-many communication |
• Allows one-to-one,
one-to-many-, and
many-to-many communication |
|
Table reproduced from Kern,
1998:80-81 |
|
This table addresses dimensions of MOO environments not taken up in this
study, namely the spatial metaphors upon which MOOs are based and the presence
of virtual objects and textually defined communication spaces. I do, however,
provide a description of these features in a later chapter (Chapter 5). Kern
accurately portrays MOOs as “electronically mediated social
environments” (1998:81), a gestalt description I endorse. The critical
features in the table above are that a writer’s input is reframed by the
computer, that co-presence creates an expectation of immediate response, and
that computer-mediation provides for one-to-one and one-to-many communication,
like conventional literacy media, and also many-to-many communication which is
unique to electronic discussion. Though Kern does not make this point,
synchronous many-to-many communication in writing has only been possible since
the advent of networked computer-mediated communication.
Addressing social-affective
dimensions to CACD, Beauvois notes that students generally have a positive
attitude toward computers and that this may increase their interest in CmC.
Over time, however, “students seemed less amused by the technology
… and more adept at using the [electronic conferencing] program as a tool
toward expressing their ideas” (Beauvois, 1998:108). This latter point is
substantiated in my own study as well. Electronic conferencing is a skill, and
as students become accustomed to CmC, and more skilled in using this
environment to express themselves and interact with their peers, they are able
to utilize and create social and material resources which support their on-line
activity. One of the features of extended on-line interactions with the same
group of people (classmates in Beauvois’, Kern’s, and my research)
is the development of communities of a sort. I will purposely close off a more
extended discussion of “community” in terms of my own research
since I address this in later chapters. For this review of the literature,
however, Beauvois and Kern both mention the development of a sense of community
that shares a “common language” (Beauvois, 1998:109), and provides
cross-cultural insights and celebration which potentially would not emerge in
whole class face-to-face discussion (Kern, 1995, 1998).
Available foreign language
research, such as that reviewed above, has contrasted the pedagogy, social and
linguistic dynamics, and potential learning effects of real-time CmC mediums on
language use and learning by FL students. Linguistic analysis of
computer-mediated language use provides additional insights into the linguistic
nature of on-line communication and helps to define the foundational question
of whether CmC, by its mediational qualities, changes human interaction, or
whether human patterns of communicative activity colonize digital media in
relatively stable, RL (real-life or “non-digital”) forms. Especially
for textual digital environments where extensive log file records exist,
ascertaining the linguistic dynamics of language use are empirical questions,
and such studies, currently few in number, add tentative pieces to the complex,
and like culture, ever evolving, CmC puzzle.
I report now on a large-scale
linguistic study of real-time CmC. In a corpus based empirical analysis, Yates
discusses a comparison between (presumably native speaker) English language
corpora of three distinct kinds: oral, written and electronic conferencing
discourse. For analysis, Yates used Halliday’s three-way analytic
framework focusing on textual, interpersonal, and ideational functions of
language. Yates found that the language used in his CoSy electronic conferencing
data illustrate an interesting mix of features that include considerably
greater lexical density than the speech corpora, and hence show the electronic
conferencing sample to be quite close to the written discourse corpora in this
textual aspect. An examination of the interpersonal linguistic features of
these three corpora showed that the electronic conferencing data included a
greater proportional use of 1st and 2nd person pronouns than either the speech
or writing corpora, and thus that in this regard the synchronous digital
corpora was more like oral speech than writing. The last measure focused on
Halliday’s ideational function of language and measured the frequency of
use of modal auxiliaries across the writing, speaking and digital conferencing
corpora. By this measure the digital conferencing data showed the greatest use
of modals from among all three corpora. He included in his analysis a count of
modals of obligation and necessity, ability and possibility, epistemic
possibility, volition and prediction, and hypothetics. Yates cites an earlier
research project of his and notes that qualitatively, the “contextual use
of modal auxiliaries within CMC is comparable with that of speech”
(Yates, 1996:45). In his concluding remarks, Yates revisits Halliday and states
that “an important difference between genres and modes of communication
lies in the semiotic field in which the communication takes place”
(1996:45). Yates gives the example of a doctor-patient interview. Extending
this notion to this project, the formal institutional qualities of higher
education, the societal context for this study if you will, is the semiotic
field within which genre and modes of communication come together in
communicative use. I will foreshadow one of the key findings of this study by
mentioning that “semiotic field”, in the discursive sense meant by
Halliday and also Bourdieu, is more complex in my study than the institutional
dimension of the local setting where CACD is taking place. I will touch on this
point again in Chapters 5 and 6. Returning to Yates, when the question is how
language use varies across media, Yates’ work demonstrates the usefulness
of a corpus-based empirical approach to the modeling of CmC discourse.
One of the advantages to this
prior research into foreign language uses of digital communication is that
these studies and their data and analysis can act as a base-line against which
to compare other data. The examples above serve these purposes. The research
forming the empirical portion of this dissertation project is approached from a
set of assumptions about language, interaction, cognition, and development,
which are not yet well represented in the second and foreign language research,
examples of which I have described above. The psycholinguistic and socio-cognitive
approaches, whether stated explicitly or implicitly represented by a set of
terms and assumptions about language development, have yielded productive
research results without which my current work would be impoverished. Such
research has begun to map out the interactional dynamics of CACD (Kern, 1995),
the relationship between CACD and oral examination performance (Beauvois,
1998), and the effects of CmC on the development of L2 morphology and syntax
(Pelletieri, forthcoming). At the same time, and explained in Chapters 1 and 2,
sociocultural and activity theory approaches to FL uses of CmC will bring out
dimensions of context which are still blank spots on the map for the
aforementioned research approaches.
Cognitive scientists were one of
the first academic groups to become seriously and experimentally interested in
human-computer interaction. Cognitive science encompasses diverse approaches to
the study of human mental functioning and development. As I discussed in
Chapter 2, there have been substantial critiques of cognitive science paradigms
based primarily on their lack of serious attention to variability, context, and
the historical-cultural specificity of situated cognitive activity. Norman
(1991), one of the foremost of this traditional cognitive science group, has
begun over the last decade to incorporate elements of Soviet psychology into
his model in quite productive ways. In part based on influences from
psychological models focusing on mediation, Norman develops the concept of
thinking about devices such as computers as cognitive artifacts, which he
describes as “artificial devices that maintain, display, or operate upon
information in order to serve a representational function and that affect human
cognitive performance.” (1991:17) Historically, a psychological and
anthropological focus on artifact mediation probably began with Wundt (1916),
and then was later taken up by the Soviet socio-historical movement (Vygotsky,
Luria, Leont’ev). This line of research has continued to the present
among post-Vygotskyan theorists (Engestrom, Cole and Wertsch for psychology;
Lantolf, Frawley, and Donato for SLA).
Though most current psychology
research focuses on the “unaided mind”, principally humans in
controlled-setting experimental conditions who are unassisted by external or
environmental artifacts, Norman seeks to “integrate artifacts into the
existing theory of human cognition” (1991:18). Norman discusses two views
of artifacts that are implicit to cognitive science. The first is the
“system” view of artifacts. The system view is often the
perspective of the analyst/research. The idea is that an agent, artifact, and
task are seen as a system, and the system can “accomplish more”
with the artifact in question (e.g., pencil and paper for math problems, a
computer for word processing, email communication for multi-party decision
making). The other is the “personal” or emic view. Through artifact
mediation, an actor will sense that “the task has changed: thus [from the
actor’s perspective], the artifact does not enhance cognition--it changes
the task” (1991:20). Norman makes clear the point that artifacts viewed
as cognitive “amplifiers” is a misleading metaphor. He instead
argues that artifacts “change the nature of the task being done by the
person and, in this way, enhance overall performance” (1991:19).
Bellamy begins a paper on
educational technology by discussing research which suggests that the introduction
of technology into schools can catalyze educational change (Papert, 1980;
David, 1991; McClintock, 1992; Dwyer, 1994). Her critique of this research is that the authors do not
explicate why technology instigates change, nor what technologies might most
benefit schools. Bellamy bases her thesis, that mediation can explain why
technology “has the potential to reform” the educational system, on
Cole & Engestrom’s work (1993), whose argument is that the actions of
individuals occur at the nexus of three factors; the tools and artifacts
available (computers, languages), the community and the understood laws of the
community, and the division of labor in these community-settings (roles,
interactional dynamics). Participatory design (Ehn, 1988) and situated action
approaches (Suchman, 1987) are pointed to as good design processes for
educational technology.
Bellamy (1996) draws on
Ringstaff, Kelley, and Dwyer (1993), who discuss four factors influencing how
and if technology changes the educational process: psychological factors
(teachers being comfortable), technical support, ease of access, and collegial
and institutional support. Out of
her own work and that of others, Bellamy suggests three Vygotskyan principles
for the design of educational technology: 1) cultivate authentic activities
(this notion of “authenticity” is problematic, see Chapter 6,
section 6.6), 2) encourage construction of artifacts, 3) collaborate in peer
and expert-novice groupings. Bellamy (1996) summarizes by saying that
technology can change educational processes since artifacts (including
literacy, computers, and chalk boards) mediate human activity. In short, change
the mediating artifacts and the nature of human activity will also change (the
degree of change in communicative processes due to mediation by CACD is
discussed at length in Chapters 5, 6, and 7). At the level of implementation,
the change of an artifact needs to also potentially include change at the level
of the explicit philosophy of education/learning held by the teachers,
administrators, and software designers.
The internet has transformed
earlier developments in computer-assisted language learning, moving students
from ‘stand-alone’-authoritative language learning software to
open, highly social, network-based communication environments. Small group
interaction, tutorials, and real-time class discussions take place in these
social spaces. There is nothing more or less “real” about
computer-mediated or face-to-face social interaction. Put another way, “cybersociety”
theorist Jones argues that face-to-face communication promotes no better or
more ideal a communicative process than do other communicative situations and
mediums (1995). Artifacts such as an internet conferencing environment used for
CACD, for example, take their form and significance from the human activities
they mediate, and cannot be fully apprehended from a positivist vantage point
as simply or generically “there” in the world.
The question of whether digital
mediation of L2 social interaction substantively alters the activity of
communication, and in what relevant ways for the learning of a foreign
language, is still unanswered in the SLA literature. In Chapter 5 of this
dissertation, I show that the mediational qualities of CACD do produce
demonstrable effects on the way communication is carried out. The issue of
which of these differences makes a difference (as the adage goes), is the focus
of Chapter 7 of this dissertation.
[i] Hypertext authoring environments such as StorySpace support
collaborative authoring and “open” hypertext documents. Internet
based hypertext webs began as largely closed to user-modification, but this is
changing as more open and collaborative web spaces are being experimented with,
and some of the technical impediments are being worked out.
[ii] The “Découvrir
Berkeley!” site is available at [www.itp.berkeley.edu/french/main.html]
[iii] Gaming technologies have become collaborative, multiparty
events due to internet based technologies. Due in part to significant
development budgets and the goals (to generate thrills versus set of learning
activities)
[iv] A La Rencontre de Philippe
was filmed by a French filmmaker, and therefore provides another valuable
attribute--a example of French cinematic style.
[v] Bulletin boards are accessed by phone lines. A user or
member accesses a particular machine by dialing in through a modem. Unlike
internet access, however, bulletin boards are isolated computers.
[vi] The Palace can be found at
[www.digitalspace.com/avatars/palace.html]. See also MIT’s Media Lab for
experiments such as “Chat Circles” [www.media.mit.edu]
[vii] Information is also more widely
available than it was during pre-digital days. Via the world wide web, access
to and transactions with information are not bound by the hours of the library,
book store, bank, or stock broker.
[viii] Kinesthetic dimensions to interface design, called
“haptic interfaces”, are currently still experimental. An example
of the kinesthetic dimension is resistance through touch using a roller or joy
stick that is linked to another user.
[ix] Claims for ‘global’
change elicit the many issues of equity and distribution of the material well
being necessary to support computer mediated communication. Though much of the world is in fact
networked, the most obvious examples of on-line discourse communities are at
present still largely from the Americas, Europe, and East Asia.
[x] This research is cited in Avots,
1991. The original study was
published by Smith (1990), “Collaborative and interactive writing for
increasing communication skills” in Hispania 73, 77-87.
[xi] Beauvois terms this a
pilot-study, presumably because the diagnostic tool was subjective instructor
grades of oral exams. She notes, however, that each instructor covered both an
experimental and control group section, which would potentially lessen
comparative bias in his or her grading practices. A future study will follow
the pilot-study in methodology, but additionally will include inter-rater and
intra-rater reliability checks.
[xii] I will offer a refinement of the
notion of turn taking and the nature of on-line communication in a later
chapter. For views on input-uptake metaphors in SLA, see Chapter 2. What is
important is that the ontological basis for foreign language studies are
predominantly rooted in the input-output framework, and that my research is
oriented in a different direction.
[xiii] As a reminder to readers,
Daedalus Interchange supports synchronous communication in writing, and in this
sense supports communicative activity that is comparable to the MOO interaction
data of this study.
[xiv] Ortega’s review article of
L2 computer-assisted classroom discussion is based on the interactionist
approach to second language acquisition. Despite the fact that I have critiqued
such a position (Thorne, forthcoming), I wish to acknowledge Ortega’s
incisive and well argued article. In her concluding comments, she pushes for
the very research questions and vantage points which this work attempts to
render, namely an illustration of the relationships between CmC data (log
files), self-reported data from participants, and observations by the analyst.
She states, and I agree, that an integrated analysis of these issues
“will enable a more accurate assessment of the ways in which
[computer-assisted classroom discussion] fosters self-regulatory learning and
communication strategies and of the impact of such strategies on SL performance
in the electronic mode” (1997:92).
[xv] Some of what I perceive to be
limitations to the interactionist paradigm can be found in Chapters 1 and 2 of
this dissertation. A diverse group of researchers and agendas are engaged in
the interactionist approach, but for the most part, they appear to retain the
terminological inertia of the engineering-computational metaphors common to
cognitive approaches.
[xvi] Ytalk provides two users with a
split screen. Unlike most synchronous communication tools (including the MOO
used in my study) where messages appear on the group screen in their entirety,
Ytalk shows messages as they are being constructed, key stroke by key stroke.
Though not discussed by Pellettieri, watching as messages are being typed
allows from real-time back-channeling that is not possible with MOO, IRC, or
other forms of multi-party CmC environments. Delayed back-channeling, as
documented by Cherny (1996), does occur among some MOO and IRC speech
communities however.
Balajthy, E.(1986).
Microcomputers in reading and language arts. Englewood Cliffs, N.J.:
Prentice-Hall.
Bays, H. (1999). Framing and
face in Internet exchanges: A socio-cognitive approach. Available at
[viadrina.euv-frankfurt-o.de/~wjournal/bays.html]
Beauvois, M.H. (1998). Write to
speak: The effects of electronic communication on the oral achievement of
fourth semester French students. In Judith A. Muyskens (ed.), New ways of
learning and teaching: Focus on technology and foreign language education.
Boston: Heinle & Heinle Publishers.
Bellamy, R. K. E. (1996).
Designing educational technology:
Computer-mediated change. In B.
A. Nardi (Ed.), Context and consciousness: Activity theory and human-computer
interaction (pp. 123-146). Cambridge, Mass.: MIT Press.
Bernhardt, E., Kamil, M.
(1998). Enhancing foreign culture learning through electronic discussion. In
Judith A. Muyskens (ed.), New ways of learning and teaching: Focus on
technology and foreign language education. Boston: Heinle & Heinle
Publishers.
Bohlke, O. (1999). Chatroom and face-to-face discussion
groups, and its potential implications for foreign language instruction. Paper
presented at the 1999 CALICO conference, Oxford, Ohio.
Bruce, B., Peyton, J. K., &
Batson, T. (Eds.). (1993).
Network-Based Classrooms. New York: Cambridge University Press.
Bruckman, A. (1992). Identity
workshop: Emergent social and psychological phenomena in text-based virtual
reality. Available at
[http://www.cc.gatech.edu/fac/Amy.Bruckman/papers/index.html].
Bryson, M., & de Castell,
S. (1993) En/gendering equity: On some paradoxical consequences of
institutionaled programs of emancipation. Educational Theory, 43/3:341-355.
Bump, J. (1990). Radical
changes in class discussion using networked computers. Computers and the
Humanities, 24:1-2, 49-65.
Bødker, S. (1997).
Computers in Mediated Human Activity. Mind, Culture and Activity, 4(3),
149-158.
Chun, D. M. (1994). Using
computer networking to facilitate the acquisition of interactive competence.
System, 22/1: 17-31.
Cole, M., & Engestrom, Y.
(1993). A cultural-historical approach to distributed cognition. In Gavriel
Salomon (Ed.), Distributed Cognitions: Psychological and Educational
Considerations. Cambridge: Cambridge University Press.
Colomb, G., & Simutis, J.
(1996). Visible conversation and academic inquiry. In Susan Herring (ed.), Computer-Mediated
Communication: Linguistic, Social, and Cross-Cultural Perspectives.
Philadelphia: John Benjamins Publishing Company.
Cononelos, T., & Oliva, M.
(1993). Using computer networks to enhance foreign language/culture education.
Foreign Language Annals 26: 525-534.
Cottam, M., Lafford, B., and
Lafford, P. (1999). Computer mediated communication in the language lab:
Foundation, implementation, and results. Paper presented at the 1999 CALICO
conference, Oxford, Ohio.
David, J. (1991). Partnerships for
change. Apple Classrooms of Tomorrow Report 12.
Day, M. & Batson, T. (1995)
The network-based writing classroom: The ENFI idea. Computer Mediated Communication and the On-line Classroom,
Vol. 2. Cresskill, NJ: Hampton
Press.
Dwyer, D. (1994). Apple
classrooms of tomorrow: What we’ve learned. Educational Leadership
(April).
Ehn, P. (1988). Work-oriented design of computer
artifacts. Stockholm: Swedish Center for Working Life.
Faigley, L. (1992). Fragments of rationality: Postmodernity
and the subject of composition.
Pittsburgh : University of Pittsburgh Press.
Ferrara, K., Brunner, H., and
Whittemore, G. (1991). Interactive written discourse as an emergent register.
Written Communication, 8/1:8-34.
Fornel, M. de (1989). Une
situation interactionnelle negligée. Reseaux x 33: 31-49.
Hawisher, G. (1994). Blinding
Insights: Classification schemes and software for literacy instrutcion. In C.
Selfe & S. Hilligoss (Eds.), Literacy and computers: The complications of
teaching and learning with technology. (pp. 37-55). New York: MLA.
Hawisher, G. & Selfe, C.
(1991). The rhetoric of technology and the electronic writing class. College
Composition and Communication 42/1: 55-65.
Herring, S. (1996). Two
variants of an electronic message schema. In Susan Herring (ed.),
Computer-mediated communication: Linguistic, social and cross-cultural
perspectives. Philadelphia: John Benjamins Publishing Company.
Herring, S. (1999).
Interactional coherence in CMC. Proceedings of the Thirty-Second Annual Hawaii
International Conference on Systems Sciences (CD-ROM). Los Alamitos, CA: IEEE
Computer Society.