Return to Jim Thomas's Homepage
This article is part of of a Special issue on ethics
in The Information Society (1996)
What's wrong with the "Golden Rule"? Conundrums of conducting
ethical research in cyberspace
Christina Allen
Northwestern University
Three key values for ethical cyberspace research practices are
evident in the papers by King and Waskul in this volume: (1)
protect the subjects from harm as a result of the research
fieldwork and the research practices; (2) produce good social
science research; and (3) do not unnecessarily perturb the
phenomena studied. Much of the argumentation in those papers
aims to negotiate the ethical conflicts that often emerge between
these goals.
These authors also offer suggestions for how to proceed ethically
in cyberspace research. They include: using "the golden rule"
when determining how to protect subjects; considering all
pertinent factors of particular situations when developing
research practices; tailoring ethical guidelines developed
regarding human studies conducted outside of cyberspace (e.g.,
those of the American Psychological Association) to cyberspace
research situations; considering those novel factors of
cyberspace social interaction that confound the ethical
guidelines developed for non-cyberspace situations; cultivating
the advice of experts on research ethics; and working with key
participants who are fully debriefed about the research process and
outcomes, risks and goals.
When carefully constructed and considered, these are laudable
goals to inform research approaches. However, reducing these
reasonable suggestions to research prescriptions (as does King)
risks eroding the responsibility for ethical action by the
researcher in every situation. This remark reveals my commitment
to a strain of ethical thinking represented in the
interdisciplinary writings of Mikhail Bakhtin (e.g., 1984a,
1984b, 1986, 1990, 1993; also see Clark & Holquist, 1984; Morson
& Emerson, 1990). Bakhtin argues that one can only strive for a
bottom-up ethical wisdom built upon concrete examples, and
disputes the possibility that rigid top-down application of
universal rules constitutes ethical action.
In this paper, I briefly review Bakhtin's perspective and its
implications for ethical research practices. To illustrate this
perspective at work through a concrete example, I consider my
recent fieldwork and writing practices in a virtual community. I
summarizes the methodological commitments made in light of my
research goals and concern for the protection of the research
participants. I then explain how "ethical work" with the
particularities of this research situation developed into these
choices.
ETHICS FROM THE BOTTOM UP
Mikhail Bakhtin (1895-1975), a Soviet scholar primarily known for
his early 20th century contributions to literary theory,
psychology and ethics, highlighted the responsibility of the
individual to engage critically and creatively in the ethical
"work" of everyday life. Bakhtin established his ethics in
contrast to both absolutism and relativism. According to
Bakhtin, these perspectives each deny the particularities of
everyday situations that are crucial to true ethical work.
Absolutism implies that abstract rules can be applied to any
situation without requiring any ethical work with the
particularities of that situation. Relativism removes the need
for situated ethical work by implying that all choices are
arbitrary, and hence that no aspect of the particular situation
can contribute to making one decision more ethically correct than
another (Morson & Emerson, 1990, pp. 25-29). Ethical action,
according to Bakhtin, is based on unsystemizable ethical wisdom
rather than abstract rules. Such wisdom is built from
understandings that develop out of previous situations, along
with an awareness of the need to bring these understandings to
bear flexibly and creatively in new situations.
Bakhtin acknowledged that "authoritative" language (e.g., codes
and rules) could be a beginning resource for ethical action.
However, an individual must not merely receive such language, but
rather must examine, assimilate and re-prioritize it. He called
those who live by identification with roles or abstract systems
of rules and thoughts, "pretenders" who acted as if they had an
"an alibi for being." Bakhtin described how in school two modes
of use for the voices of others are established, one "reciting by
heart" and the other "retelling in one's own words" (1981, p.
341), the former an authoritative language that does not allow
for play, the latter a richly generative opportunity to be what
he called "internally persuasive" (p. 342). He observes that one
cannot live ethically by reproducing established "rhythms"
(Bakhtin, 1990, p. 117f), but rather must always be receptive,
perceptive and struggle to act responsibly in particular
situations.
For Bakhtin, situated ethical action depends upon ethical wisdom
achieved through creative engagement in past, similar situations.
Yet cyberspace practices and contexts are not settled. Neither
participants nor researchers have much experience in cyberspace.
Given the size and demographic diversity of current and potential
Internet participants, and the generativity of Internet
programming paradigms, it is likely that they will never be
settled around any single set of central practices and norms.
However, as more participants gain experience in these now novel
modes of communication, diverse amalgams of values will begin to
emerge. Many cyberspace sites are already developing innovative
approaches to governance, conditions of membership and regulation
of patterns of use, partially in response to what researchers
"do" with the information that is available. In other words, the
cyberspace experiences that can contribute to ethical wisdom are
just now developing.
Beyond his insistence on the individual's responsibility for
creative ethical work in particular situations, central to
Bakhtin's later thought on ethics (e.g., Bakhtin, 1986) was the
concept of "dialogue." With this concept, he sought to replace
the centrality of the meaning of the "word" in linguistic and
literary theory with an emphasis on dialogic interaction. Every
utterance in a discourse, he noted, has a primordial dialogism.
It is part of a larger whole. How and what the utterance will
mean is only achieved at the moment of its production in the
social web of interpretation. Bakhtin's concept of dialogue
contrasts with structuralist accounts in which utterances are
thought to have meaning in and of themselves based on the meaning
of the words contained in the utterances. Bakhtin closely
attended to the dynamics of everyday living speech, to meaning
and truth made through utterances exchanged between specific
speakers in specific times and places. According to this
perspective, it is not up to the individual alone to perceive and
interpret "the meaning" of a word or situation, but rather to
actively engage interlocutors in making that meaning.
In dialogue, we creatively assimilate others' voices with our
own, making something new in terms of the uniqueness of our own
perspectives and histories. This dialogic perspective contrasts
with an ethical approach based on "the golden rule" . There are
few interpersonal, societal, cultural presumptions that one can
make about cyberspace groups or individuals at this time. Since
many forms of computer-mediated communication are new experiences
for many participants, and since they combine people and
perspectives from widely divergent social groups in novel
combinations, we cannot know much about the "perceptions" of
participants without engaging in dialogue with them. Moreover,
even within a group, the individuals represented may have very
different conceptions of what is going on, what is allowed, and
what "ought" to be the case. This is why the "golden rule"
breaks down particularly in cyberspace --currently, there is
little ground for presuming that one's own perceptions, values
and wishes as a researcher correspond to those of the other
participants involved. As time passes, the wisdom gained from
dialogic experience in cyberspace between participants and
researchers will serve as a basis for future ethical work
regarding the boundaries of participation and scholarly
observation and analysis.
In sum, since the ethical researcher cannot a priori adjudicate
what will be harmful, it is necessary to redevelop ethical
research practices by engaging in creative "ethical work" in
situ, in dialogue with participants and perhaps other
researchers, and throughout the research and publication
processes. In the next section, I summarize the choices made in
my current cyberspace research project, and then review the
ethical work that led to this approach. In doing so, I offer
readers one concrete example of choices and outcomes to consider
while developing their own ethical wisdom about research
practices in cyberspace.
AN EXAMPLE
I conducted a two-year ethnographic and rhetorical study of
identity practices in a virtual reality community of a type
called a "MOO" (MUD, Object Oriented). Before collecting data, I
met with the highest-level system administrator (the
"arch-wizard") face-to-face and explained my research intents and
practices. Once in the site, I spent time in the public rooms,
meeting participants and engaging in casual conversation.
Whenever my conversation with a participant became lengthy or
significant, I indicated that my purpose was to collect data and
write about my findings. In my own LambdaMOO self-descriptions,
there was always a reference to my role as researcher, or the
fact that I was collecting data, or that participants should
freely ask me about my purpose there. In the first six months,
along with general participation, I interviewed twenty-five
participants at length. From this group, I continued to
interview a total of ten for the next year.
Out of the ten, I selected the stories of four participants o
present in great detail in the final manuscript. I wrote
"ethnographies of the particular" (Abu-Lughod, 1991) that focus
on commonalties and differences across these individual
participants in how they negotiate common cyberspace
socio-technical factors as they construct and use virtual
identities in interpersonal relationships. Throughout the
fieldwork, I continually asked participants about their reactions
to my use of specific portions of data and conversation. One of
the original five participants I planned to write about dropped
out eventually, uncomfortable about what would need to be
revealed about him in the manuscript to meet my research goals.
I presented all text written about the remaining four
participants back to them at various times throughout the
fieldwork period, and afterward, while I continued writing. I
included the participants' responses to the text in the final
versions of the chapters, explicitly making changes that they had
requested and noting for the reader the changes made.
Besides the data collected through observations and interviews
with participants, I used other texts available in LambdaMOO in
the ethnography. For example, LambdaMOO is rife with bulletin
board-type forums that are accessible (as are their archives) by
all community members, as well as guests to the site. Texts from
these forums often provided critical background information for
situations experienced by the four participants. Along with
these forums, there are also the "identity texts" owned by
individual participants in LambdaMOO. Identity texts are textual
descriptions of participants' virtual homes, bodies, customizable
communication messages and objects. Any participant in LambdaMOO
can use simple programming code to look at another participant's
identity texts. Finally, there are "surveillance texts." These
are texts produced in "real time" about another participant
--including information such as that participant's most recent
connection time or current virtual location and compatriots. How
these texts were used in publications varied based on the nature
of the forum in which they were posted, and my research
relationship with the participants involved.
In the final manuscript, I identified LambdaMOO by name. With
their explicit permission, in the writing I called the four
participants by their true LambdaMOO names (i.e., their
pseudonyms). However, they were never identified by their "real
life" names. Unless specifically asked, other participants who
figure in the stories told by the four key participants were given
alternative pseudonyms. I used observations and second-hand
stories concerning such secondary participants when they
critically figured in telling the stories about experiences of
the primary participants.
All of these fieldwork and writing choices (see below) were keyed
to both theoretical and concrete circumstances of the research
goals, the nature of the site, and the particular situations of
the key participants. Now I will review briefly the details of the
ethical work that led to the above conclusions about how to
handle these forms of data.
REVIEWING THE ETHICAL WORK INVOLVED IN THIS PROJECT
While it may appear that developing ethical research and
publication practices for a participant/observer project might
not speak to the same issues present in King's consideration of
how to use postings to Usenet groups, the diverse forms of data
available in LambdaMOO, described above, nonetheless require
similar ethical considerations. I will briefly characterize
pertinent conditions in the sociohistorical context of LambdaMOO
and how they influenced the ethical choices I made. These
conditions included: (1) LambdaMOO's rhetoric and technical
structure; (2) the history of research and publication about the
site; (3) its text-based nature; (4) certain key characteristics
of its participants; and (5) the pronounced tendency of
participants to reveal intimate information about themselves and
to engage in esoteric interpersonal practices.
(1) Rhetoric and technical structure of the site
First, I considered the technical, rhetorical and administrative
structure of the LambdaMOO context. What messages did the site
itself signal about rights and responsibilities of participation
by members, guests and observers?
To begin, LambdaMOO is unrestricted to anyone with Internet
access. People may connect as guests, or apply for an account,
which has no membership conditions except ownership of a viable
computer account. This contrasts with other MOOs (e.g., MIT's
MediaMOO: Bruckman & Resnick, 1993) that have explicit conditions
of membership and conduct. The opening screen that appears when
participants connect includes a strong message of individual
responsibility in an open environment:
"LambdaMOO is a new kind of society, where thousands of people
voluntarily come together from all over the world. What these
people say or do may not always be to your liking; as when
visiting any international city, it is wise to be careful who you
associate with and what you say.
The operators of LambdaMOO have provided the materials for the
buildings of this community, but are not responsible for what is
said or done in them. In particular, you must assume
responsibility if you permit minors or others to access LambdaMOO
through your facilities. The statements and viewpoints expressed
here are not necessarily those of the wizards, Pavel Curtis, or
the Xerox Corporation and those parties disclaim any
responsibility for them."
Participants experience no other automatic messages prohibiting
or suggesting behavior before they are placed into the public
rooms. While many "help," "manners" and other systems for
guidance exist, it is up to the participant to choose to utilize
them.
As participants move through the virtual architecture of the
place, they come across two types of rooms -- "public" and
"private." It is noteworthy that "public" and "private" rooms
are technical categories in LambdaMOO. "Public rooms" are those
connected to the main architecture of the house, and may always
be accessed by all participants. Individual participants own
"private rooms," that may be locked or opened to other
participants at the desire of the owner.
There are also "public" and "private" communication channels in
LambdaMOO. When private, owners determine membership and
conditions of channel use. Some channels are topically oriented,
but when public, messages about intent and social censure from
other members are used to shape and enforce patterns of use.
Thus, the very architecture of LambdaMOO introduces participants
to public and private distinctions that are meaningful in the
local context, rather than abstractly defined and imposed.
In its five year history, LambdaMOO has had more than 25,000
accounts. Currently, there are more than 9,000 active accounts.
With such a large and transient population, and without
mechanisms (e.g., membership conditions) for establishing a
central, social order, LambdaMOO has become a heterogeneous site
comprising extremely diverse subgroups. LambdaMOO's rhetorical
and technical structure offer participants diverse contexts for
entering or avoiding certain social contexts and activities.
Therefore, participants who post to public bulletin boards, speak
on public channels or act in public spaces make a choice about
the scope of exposure of their words or actions, since there are
private or restricted forums for almost all types of activity and
discourse.
Given the existence of local public/private distinctions, the
heterogeneity of social practices of the site as a whole, and the
absence of restrictions, I regarded public words and actions to
be available for analysis without consent. However, it was no
great hardship to seek consent whenever possible. Using MOOmail,
I attempted to clear the use of verbatim transcript segments with
participants represented in them, offering them the option of a
pseudonymic replacement name if they desired it. However, given
the transient nature of the population, often I received no
response from participants who perhaps no longer frequented the
site, or who did not care to respond.
(2) History of research and publication about the site
Unlike many Internet sites, LambdaMOO has been studied
extensively throughout its history. Journalistic accounts of
the site have been broadly published, including Dibbell's famous
Village Voice article entitled "A Rape in Cyberspace" (1993), and
in Rheingold's (1993) book on virtual communities. Numerous
scholarly works are underway or have been presented in journals
or conferences. Many participants are aware through public
forums that research is a regular aspect of the site, including a
public bulletin board devoted to discussion of social issues and
research projects. When I began my research, there was already a
lengthy discussion underway by participants and other researchers
about feelings and issues of conducting research. While there
were a few posts that vehemently opposed any sort of research,
the majority of the posts argued against any rules regulating
research. One participant pointed out that, after all, LambdaMOO
was created as a research experiment by scientists at the Xerox
Palo Alto Research Center. Therefore, my research project did
not constitute a novel circumstance for most participants.
Two of my four key participants' stories are intimately tied to
the public history of LambdaMOO. Changing the name of the site
would not truly "protect the participants" since unique aspects
of their stories would be identifiable by many readers. Unlike
the sites devoted solely to sensitive topics that are described
by King and presumed by Denzel, LambdaMOO is a heterogeneous site
with many different social organizations and purposes active at
any time. It is more akin to a large city than to a private
support or discussion group. With its openness to research and
publication, and its lack of restrictions, participants could not
reasonably presume that all activities would be considered
private, protected or representative of their own LambdaMOO
experiences. Therefore, I chose to identify the LambdaMOO
research site by name. This methodological choice aligns with
anthropological strategies whereby a country, city or town is
identified by name, but perhaps a neighborhood or village has its
name changed in publications. Such local name changes allow the
researcher to include unique ethnographic details about a region,
while allowing any particular individual or social group
"plausible deniability" that they were participants in the
research. LambdaMOO's anonymity, combined with what many
LambdaMOO participants called the "typist problem," has always
offered participants such plausible deniability with which to
protect their real life identities. (The "typist problem" refers
to the fact that no matter what account is used, no one can ever
be really sure of who is typing at the keyboard at any time.)
Regarding protection of the reputations of participants' virtual
identities, I based my strategy on dialogue with them about their
wishes. Note that my choice was keyed to the particular nature
of the site, and to the other decisions I had made about how to
conduct research in a way that would not be harmful to
participants or the site as a whole.
By participating in interviews with journalists, and by
developing active voices in public forums, some participants had
intentionally become public figures in an unrestricted global
forum. I regarded information about these public figures in
LambdaMOO to hold different status than information about
participants who maintained entirely private lives in LambdaMOO.
Therefore, I made free use of the information about public
participants and situations that was generally available in
public forums.
(3) The fundamental textuality of LambdaMOO
LambdaMOO is a fundamentally textual place. Participants
construct identities through the production of virtual identity
texts and continually interpret one another's' texts. Moreover,
they write about their experiences reflexively in internal
LambdaMOO bulletin board forums, and read critically what others
write about them. Therefore, rhetorical criticism becomes
crucial to understanding the nature of identities and social
practices in LambdaMOO. This fact carries methodological
consequences.
To consider the topic of virtual identities, substantive content
analysis of each participant's identity texts was necessary.
Paraphrasing the content of these identity texts for these
detailed analyses would be akin to paraphrasing poetry for
literary critique. Since anyone with a text search mechanism
could simply search for unique text strings in LambdaMOO forum
archives and identity texts, replacing LambdaMOO pseudonyms in
the writing would constitute a facile protection, rather than a
real protection.
The conflict between my need to conduct detailed context analysis
and the requirement to "protect" participants from harm as a
result of my fieldwork and publications was a primary motivation
for my decision to conduct in-depth analysis of the LambdaMOO
lives of a few participants. Entering into an on-going dialogue
with key participants involved informing them up front about the
work and my research goals and practices. Throughout the
fieldwork, I reminded participants of my purpose and sought their
continued acceptance of the work. Exploiting the immediacy of
the medium, I regularly submitted drafts of chapters to the
participants for response, and I included their responses in the
later versions of the chapter, which were also submitted to
participants for review.
(4) The nature of LambdaMOO participants
LambdaMOO participants are typically "first world" literate and
computer-literate individuals. As such, often they have ready
access to what is written about the site. They also have the
skills to critically engage the material. The study of "first
world participants" makes relevant recent developments in
anthropological methodology and theory arguing for a more
reflexive, participant ethnographic method and dialogic research
ethics appropriate to these issues.
Members of such social groups both "write" themselves and read
what others write about them. Late twentieth century mobile,
literate populations are no longer at a spatial or cultural
divide from the representations produced about them. In
Brettell's edited volume When they read what we write: The
politics of ethnography (1993), authors struggle with the
politics of representation and interpretation in a context where
there is little divide between the researcher and the researched.
The authors review cases where the researched have the interest,
the literacy, the access and the critical skills to engage and
refute the representations that academics produce about them, and
that journalists and politicians reduce and reproduce in larger
legal and media forums that have direct practical effects on the
lives of those represented. Many of these authors argue that
researchers must not only re-interpret the texts already
produced, but include in the ethnographies the participants'
responses to the written fieldwork.
Beyond myself, some of the LambdaMOO participants were conducting
research about computer-mediated communication. Many
participants are computer programmers. I have also worked as a
programmer and participant in computational communities for more
than a decade. The first-worldness of the participants in
LambdaMOO contributed important factors that required a clear
articulation to participants and readers of my role and interests
in the site, and how participants responded to this project.
Thus, in the manuscript I included reflections on my own
experiences, explicitly exploring how my positionality mattered
for fieldwork experiences and the research outcomes.
In cyberspace, the sites of inquiry, reporting and critical
response have collapsed into one medium. This medium also brings
researchers closer to participants in time and space and in
sociocultural experience. This was a primary threat and an
excellent opportunity for ethical practice in the site. It was a
threat because my reports could have immediate impact on the
lives of participants. It was an opportunity because it allowed
us a close dialogue, clarifying for them how I was handling their
stories and my own.
(5) Other special conditions and situations of LambdaMOO
Underlying many concerns about research ethics in cyberspace is
the surprising degree of personal revelation and interpersonal
intimacy demonstrated by participants. If participants limited
their conversations to baseball statistics, issues of research
ethics might not seem as pressing. However, cyberspace
participants tend to reveal much about themselves, including
vulnerabilities of their self-conception, political flames, work
gripes, family histories, and the ups and downs of past and
contemporary relationships. King and Waskul allude to these
revelatory practices using the concepts of public and private
spaces and interactions. Drawing from anthropological and
clinical psychological literatures, I offer a different
terminology for understanding the ethics of cyberspace research
in these revelatory conditions -- the "liminal" and the
"liminoid."
The technical concept of "liminality" entered anthropological
discourse with French folklorist Arnold van Gennep's (Gennep,
1909/1960) description of the three phases of a rite of passage.
They are separation, transition (or liminal) and incorporation
(or aggregation). Separation marks the point at which initiates
move from ordinary, everyday space and time into ritual space and
time. This phase involves the creation of a social space clearly
marked as the location of ritual activities. The second phase,
called the liminal or the marginal phase, derives its name from
the Latin "limen," or "threshold." During this phase, initiates
are stripped of their old attributes and status and enter into a
state of ambiguity. It is in this phase that they receive
instruction and undergo the ordeals of transformation. The final
phase van Gennep called incorporation, in which initiates (or
liminars) return to everyday life with their new social location
and responsibilities.
Victor Turner's more recent and influential work focused on the
theoretical elaboration of the transition or liminal phase in
modern culture. According to Turner, van Gennep's formulation of
liminality offered a crucial anthropological and sociological
perspective on social structure (Turner, 1974, p. 250). It
provided an alternative to characterizing as chaotic the absence
or suspension of social structure. Van Gennep's work revealed
that the suspension of dominant social structure could reveal
smaller structures (liminal states) that had their own social
order and purpose. According to Turner, liminal states are:
"...suspensions of quotidian reality, occupying privileged spaces
where people are allowed to think about how they think, about the
terms in which they conduct their thinking, or to feel about how
they feel in daily life." (Turner, 1987, p. 102).
For liminal states:
"...the possibility exists of standing aside not only from one's
social position but from all social positions and of formulating
a potentially unlimited series of alternative social
arrangements" (Turner, 1974, p. 14).
In sum, for any social group, Turner argued that one may
distinguish those activities that are considered "normal," from
those that rely on the suspension of "normal" rules, expectations
and truth conditions. The former may be called the everyday
"what is," the latter called the liminal "what might be."
Whereas van Gennep highlighted tribal ritual processes, Turner
argued that the liminal could be institutionalized as ongoing
states in modern cultures. Turner went on to make a crucial
distinction between the "liminal" and the "liminoid." For him,
the term liminal described the obligatory and clearly marked
social dramas and rituals of pre-modern social groups. Liminoid
phenomena were found in post-industrial societies and represented
a "dismembering of the liminal" where:
"...various components that are joined in liminal situations
split off to pursue separate destinies as specialized genres --
for example, theater, ballet, film, the novel, poetry, music,
art, both popular and classical in every case, and pilgrimage"
(Turner & Turner, 1978, p. 253).
The liminoid is:
"...the successor of the liminal in complex large-scale
societies, where individuality and optation in art have in theory
supplanted collective and obligatory ritual performances"
(Turner, 1987, p. 29).
While Turner distinguished liminal and liminoid phenomena based
on the totality and comprehensiveness of ritual involvement in
the social system, Robert Moore (1991) argues for making the
distinction instead based on the establishment of boundaries,
responsible stewardship and conscious intentionality:
"One can participate in liminoid space without there being
present in any social actor conscious intentionality as to the
psychocultural purposes of the activities involved. Liminal
space cannot properly be said to exist without the existence of
such conscious intentionality on the part of its stewards"
(Moore, 1991, p. 24).
I find Moore's formulation to be particularly compelling and
useful for considering behavior in cyberspace communities. He
goes beyond Turner to consider both liminal and liminoid frames
of activity as existing alongside everyday frames in contemporary
societies. Moore designates spaces and times as liminal,
liminoid or everyday depending on the intentionality of the
actors, the availability of conscious guidance for activities,
and the purposeful establishment of ritual boundaries.
Contemporary liminal states -- Moore uses the therapist's office
as an example -- clearly signal and represent these conditions
to those involved and affected. In liminoid states, there may
still be a suspension of everyday norms, but without clearly
signaled intentionality, guidance or borders.
Liminal and liminoid frames are "meta-communicatively" signaled
as distinct from everyday frames. This signaling is crucial.
Bateson (1972), Eliade (1958), van Gennep (1960) and Turner
(1974) discuss the importance of marking the boundaries of sacred
or liminal space to signal that what occurs inside requires
special preparation in order to be beneficial, rather than
harmful to participants. Whether signaled with rock borders,
ritual preparation rites, or language cues, movement into liminal
or transitional space requires notification for participants that
the rules have changed. Otherwise, the experience can be
destabilizing and terrifying to the individual and the community.
"That this danger is recognized in all tolerably orderly
societies is made evident by the proliferation of taboos that
hedge in and constrain those on whom the normative structure
loosens its grip during such potent transitions as extended
initiation rites in "tribal" societies and by legislation against
those who in industrial societies utilize such "liminoid" genres
as literature, the film, and the higher journalism to subvert the
axioms and standards of the ancient regime -- both in general and
particular cases" (Turner, 1974, p. 14).
Liminal states and places traditionally have offered structured
end-runs around the action and response requirements of everyday
life. As Turner notes, liminal places offer the transformative
state of "reflexivity," where the values and norms of culture can
be interrogated and reflected upon.
Thus, some situations may clearly be designated as "everyday,"
and some may be understood as liminal or liminoid situations.
Nonetheless, those activities and attitudes that represent each
designation may vary across social groups:
"Every culture has a theory that certain "things" actually
happen, are "really true," that "have been" or "are" -- though,
of course, the frames within which these assertions are made may
vary from culture to culture. One culture's truth may be another
culture's fantasy" (Turner, 1987, p. 41).
Thus, we may see the terms "everyday" or "liminal" as potential
attributes of specific situations. For example, particular
virtual communities -- or subgroups within them -- may be defined
as liminal depending on how their internal and relational
organization conveys liminal status to participants, or how
participants agree upon these designations themselves.
With these distinctions in mind, we may ask: What is it about
computer-mediated communication that may engender in participants
a sense that they are operating in "everyday," "liminal" or
"liminoid" situations? Let me summarize briefly some reasons why
participants in my study found them so.
First, the cyberspace technical feature of anonymity is key to
participants' sense that "everyday life" has been overturned.
However, anonymity is not privacy, nor is it any type of social
compact between participants that they will not disclose what has
been revealed to them by others in these virtual community
practices. I have found that participants actively seek to erode
the anonymity of others, piecing together mosaics of information
casually revealed over time or sought out using surveillance
tools in the site to triangulate on the identity of others.
Waskul notes rightly that anonymity is developed
intersubjectively and in particular situations. Most
participants with even a little bit of experience are keenly
aware of the fragility of anonymity. How "anonymity" is used by
researchers depends on the perceptions and desires of the
participants themselves. Thus, I explicitly discussed with
participants over time whether they wanted me to protect their
IRL identities, their pseudonymic LambdaMOO identities, or
neither. Each participant explicitly allowed me to use their true
LambdaMOO pseudonyms. At some point, each also cited the
protection offered by "the typist problem," according to which
one cannot be certain of who is at the keyboard producing text at
any moment. As noted earlier, this property offers participants
"plausible deniability" regarding any instances of interaction.
Second, many participants connect to these communities from work
and home situations, because the technology of multiple "open
windows" onto different programs or systems affords what I call
multiplicity of attention and presence. While others (King, this
volume; Rheingold, 1993) have noted that connecting to public
forums from private homes and workplaces gives the impression of
privacy, I have found another reason. The computer offers an
imprimatur of legitimacy -- socially signaling to others in the
social space that the person is "at work." Often people are "at
work" --multitasking by interacting in many computer windows
nearly simultaneously. The interweaving of cyberspace activities
into other activities legitimated in the physical space where
they are working can lend the impression that LambdaMOO
activities are secret or illicit. Since participants often keep
their LambdaMOO participation secret from others in their
physical space, the cyberspace activities they participate in
attain the status of protected secrets, and secrets evoke a sense
of participating in a liminoid situation.
Third, LambdaMOO allows a seemingly infinite array of social
organizations and contexts. What I call thedesignability and
mutability of the computational paradigms allows experiments in
critique and regeneration of forms of everyday existence. For
individuals, many cyberspace communities offer the possibility of
re-making themselves in new bodies, genders and ethnicities.
Their practices in exploring this possibility typically upends
their everyday reality -- offering new perspectives. Yet just as
cyberspace frames are designable, they are also mutable. The
useful metaphors are as easily violated or erased as they are
produced and used. As a result, participants experience the
continual upending of personal identity and social organizations.
Their experience may be far removed from the relative stability
of their "everyday" lives.
In terms of personal, journalistic, scholarly and anecdotal
evidence, cyberspace sites regularly are designated as liminal or
liminoid spaces by individual participants or by groups. Yet the
sensitive content of the material does not promise that all
participants hold similar values about how such revelation should
be protected. In such situations, one can expect to see the
dangers to individuals and societal contexts discussed by
theorists like Moore and Turner. This means that extra caution
and sensitivity are require by researchers (and among
participants) because everyday boundaries and values (whatever
they may be for each individual) of interpersonal interactions
may be suspended in diverse ways by different participants.
Dialogism is particularly responsive to the sense of "danger"
participants can feel about what they are doing in cyberspace,
and how exposure of their actions is managed and communicated.
The fact of personal revelation in cyberspace contributed to my
decision to let key participants respond to my manuscript before
it is generally distributed. From the outset, I knew that some
participants might decide at any time to retract their
participation in the study, as one did. Therefore, I "overshot"
my mark, interviewing more participants at length than I planned
to include in the final work.
Finally, King uses an email written by an angry participant as an
example of how one participant's involvement in a Usenet
discussion group was "harmed" by the request of a researcher to
analyze postings. Considering my discussion of liminal and
liminoid spaces, and their ambiguities and dangers, it would seem
rather that this participant was helped to avoid a liminoid
situation that did not offer proper boundaries and stewardship
for the sensitive topics that were discussed. Without doubt,
early experiences of felt violation and vulnerability such as
these will provoke both forum creators and participants to
consider more explicitly the contexts for intimate revelation.
CONCLUSION
In conclusion, psychology research guidelines for the American
Psychological Association may not provide a broadly useful model
for studies of cyberspace communities. In many respects, the
portrayal of the role of the psychologist-as-scientist in the
guidelines used by King in his framework suggests that
psychologists have found out how to be dispassionate observers of
humans in naturalistic scenes. They do not disturb the behavior
they study, and they can predict whether fieldwork practice or
subsequent publications about the behavior may negatively affect
those that they study. Critical considerations of research
methodology and of methodological ethics in anthropology and
cultural studies have also examined these issues of studying
situated human activity, from a different perspective and leading
to different insights --which acknowledge and seek to understand
the ramifications of the positionality of the researcher about
the phenomena and individuals under study. These two
orientations suggest some interesting counterpoints for
consideration in future studies of cyberspace communications.
For example, both King and Waskul presume that research
intervention can only bring harm to the participants and the
phenomena of study. Yet, in my current study, participants have
reported positive gains from the process of interviewing and
reflecting on their cyberspace stories. In spite of the deeply
sensitive material represented in the final manuscript, the four
key participants in this study report that they experienced the
research process as a respectful dialogism between two equal
interlocutors. Thus, the "harm" that King and Waskul consider to
be the default for cyberspace communication research may have
more to do with the effects on cyberspace participants of the
experience of being reified as scientific "subjects" who can be
studied dispassionately without the participation of the
researcher than with the disclosures about their communications
per se. Faced with dialogic integrity and acknowledged
positionality by a researcher, research participants from
cyberspace communities may have a different attitude to research
concerning their practices in cyberspace. The site may be
different as a result of the research, but the crucial experience
is that the participants perceive the researcher as an active
ethical interlocutor, not as a clinically dispassionate
"collector" of behaviors as in the psychology model.
In conclusion, research ethics for cyberspace are like research
ethics for any other site. They should be situated, dialogic
agreements that develop over time between researchers and the
participants of the research study. It is unlikely that such
research ethics can be abstractly codified, because there will
often be differences between research sites, the situations of
participants, and researcher goals. Researchers can, however,
develop ethical wisdom that comes from experience with many
configurations of research in cyberspace, and report on the
conditions that grounded their ethical choices, and the results
that emerged from their work in the site. These results will
often be improved by feedback from participants about their
experiences with the fieldwork and the effects of publication.
BIBLIOGRAPHY
Abu-Lughod, L. (1991). "Writing against culture." In R. Fox
(Ed.), Recapturing anthropology: Working in the present. Santa
Fe, NM: School of American Research Press.
Bakhtin, M. M. (1981). The dialogic imagination: four essays. In
M. Holquist (Ed.), (Trans. Caryl Emerson & Michael Holquist).
Austin, TX: University of Texas Press.
Bakhtin, M. M. (1984a). Problems of Dostoyevsky's poetics.
(Trans., Ed. Caryl Emerson). Minneapolis, MN : University of
Minnesota Press. (Originally published, 1929).
Bakhtin, M. M. (1984b). Rabelais and his world. (Trans.
H=E9l=E9ne Iswols= ky.) Bloomington, IN: University of Indiana
Press.
Bakhtin, M. M. (1986). Speech genres and other late essays.
(Trans. V. W. McGee). In C. Emerson & M. Holquist (Eds.).
Austin, TX: University of Texas Press.
Bakhtin, M. M. (1990). Art and answerability: Early
philosophical essays by M.M. Bakhtin. (Trans. V. Liapunov & K.
Bronstrom.) Austin, TX: University of Texas Press.
Bakhtin. M. (1993). Toward a philosophy of the act. Austin, TX:
University of Texas Press.
Bateson, G. (1972). Minimal requirements for a theory of
schizophrenia. In G. Bateson, Steps to an ecology of mind (pp.
244-270). New York: Ballantine Books.
Brettell. C. B. (1993). (Ed.) When they read what we write: the
politics of ethnography. Westport, Conn.: Bergin & Garvey.
Bruckman, A. & Resnick, M. (1993, 15 May). Virtual professional
community: Results from the MediaMOO project. Austin, TX: Paper
presented at the Third International Conference on Cyberspace.
Clark, K., & Holquist, M. (1984). Mikhail Bakhtin. Cambridge,
MA: Harvard University Press.
Dibbell, J. (1993, December 21). A Rape in Cyberspace or How an
Evil Clown, a Haitian Trickster Spirit, Two Wizards, and a Cast
of Dozens Turned a Database into a Society,The Village Voice.
(pp. 36-42).
Eliade, M. (1958). Rites and symbols of initiation: The
mysteries of birth and rebirth. Dallas, TX: Spring Publications.
Gennep, A. van (1909/1960). The rites of passage. (Trans. by M.
B. Vizedom & G. L. Caffee. Chicago: University of Chicago Press.
Moore, R. L. (1991). Liminality and Transitional Phenomena.
Morson, G. S., & Emerson, C. (1990). Mikhail Bakhtin: Creation
of a prosaics. Stanford, CA: Stanford University Press.
Rheingold, H. (1993). The virtual community: Homesteading on the
electronic frontier. New York: Addison-Wesley.
Turner, V. (1974). Dramas, fields and metaphors: Symbolic action
in human society. Ithaca, NY: Cornell University Press.
Turner, V. (1987). The anthropology of performance. New York, NY:
PAJ Publications.
Turner, V. & Turner, E. (1978). Image and pilgramage in Christian
culture. New York, NY: Columbia University Press.
FOOTNOTE 1 Except, of course, a broad form of the golden rule;
"if you don't want me to make presumptions about what you would
like, don't make them about me."
Return to Jim Thomas's Homepage