Charles Ess
Director, Interdisciplinary Studies Center
Drury University
900, N. Benton Ave
Springfield, MO 65802
Fax: 417-873-7435
cmess@lib.drury.edu

Introduction*

he chapters included here emerge from a remarkable panel presentation organized for the Computer Ethics: Philosophical Enquiries (CEPE) conference held at Lancaster University, December 14-16, 2001. The panel was originally the inspiration of Helen Nissenbaum, who further energetically set about the business of writing the successful grant application to the United States' National Science Foundation.

Building on the efforts of the Ethics Working Committee of the Association of Internet Researchers (AoIR), we intended to bring together U.S.- and European-based researchers and ethicists who would help us move further along towards establishing an Internet research ethics that (a) recognizes both ethically-pertinent similarities and differences between traditional human subjects research and research into online behaviors, and (b) articulates values and guidelines for Internet research that are genuinely global in their validity (as required for a global medium) while acknowledging important cultural and national differences in values that might require specific ethical codes and guidelines for distinctive cultural groups. Thanks to the support of the NSF, we were able to bring to Lancaster three U.S. scholars well known not only for their research but also for their participation in ethical reflection: Amy Bruckman (Georgia Institute of Technology), Michele White (Bowling Green University), and Joe Walther (Cornell University). The grant provided further support for participation by European researchers and ethicists: Raphael Capurro (Hochschule der Medien, Stuttgart, Germany), Kate S. O'Riordan (University of Bristol, U.K.), and Dag Elgesem (University of Bergen, Norway). A number of unfortunate circumstances prevented Bruckman, Walther, and Elgesem from fully participating in the panel. Happily, we are able to present their written work here as a collection, thus fulfilling our intention of fostering an interdisciplinary and cross-cultural dialogue on Internet research ethics.

We begin with the philosophical overview developed by Capurro and Pingel, one that makes explicit a number of connections between the particular issues and dilemmas of Internet research ethics and the larger domains of computer ethics and Continental philosophy - specifically, Kant and Heidegger. They thereby introduce a central thematic for our contributors, i.e., the tensional difference (not, as they stress, a simple dichotomy) between online and offline existence: clearly, how we understand this tensional difference - and, in particular, what (offline) metaphors we use to conceptualize online interactions - directly determines our understanding of how far (offline) extant ethical codes and laws may be adequate to defining the ethics of (online) Internet research. We then turn to Dag Elgesem's analysis of the NESH (National Committee for Research Ethics in the Social Sciences and the Humanities, Norway, 2001) guidelines. Using a recent example of an online research project (Scharf), Elgesem argues that her specific ethical decisions - including seeking informed consent for citing sensitive material drawn from a listserv on breast cancer - are consistent with the NESH guidelines. In these ways, Capurro and Pengel, and then Elgesem announce and support my larger claim (discussed below) that more recent work in this field emphasizes the similarities between online and offline research ethics, and thus the general adequacy of extant research ethics codes and guidelines for online research. At the same time, both Elgesem and Capurro and Pingel recognize here a second central point: namely, the (remaining) limits of the analogies we must rely upon in order to apply extant ethical codes to the (perhaps) new domains of interaction made possible by CMC technologies, beginning with the common analogy between a public space and the "space" of public chatrooms.

Joe Walther and Amy Bruckman are singularly prominent and influential Internet researchers. Here, they both offer extensive reflections on the ethical obligations of researchers as illumined by both their own pioneering research experience and substantial engagement in ethical reflection (in part as both are members of the AoIR ethics working committee). Walther pursues the implications of the Code of Federal Regulations (CFR) that guide human subjects' research in the United States for online contexts and issues - thereby uncovering several limitations of the first major effort to develop guidelines for online research, namely, the landmark workshop co-sponsored by National Institutes of Health and the American Association for the Advancement of Science (Frankel and Siang, 1999). Bruckman extends this critique by more radically challenging its basic assumption - i.e., that the codes and guidelines for human subjects protection serve as the primary, if not exclusive starting point for thinking about Internet research ethics. Initiating a line of critique that first Kathleen O'Riordan and E. H. Bissett and then Michele White will extend, Bruckman argues that instead of viewing the human beings who produce Internet material primarily as human subjects, it is more appropriate to view them as amateur artists. O'Riordan and Bissett make clear through a case study that, paradoxically, well-intended human subjects' protections as applied to activist sites lead to unethical consequences as these protections (confidentiality, informed consent, etc.) work to reinforce a larger social marginalization, in their example, of lesbian communities. White, finally, re-orients us toward seeing Internet material as cultural production - and thus the provenance of humanities disciplines such as art history, literature, film and media studies, theatre and performance studies, etc. White thus points out a central danger in uncritically presuming that only human subjects models, as rooted in medicine and the social sciences, suffice for establishing both the ethical guidelines and methodological approaches needed for online research: doing so would thus force us to ignore the contributions of humanities disciplines and correlative ethical codes to online research ethics. White further extends the critique developed by O'Riordan and Bissett: using human subjects codes exclusively leads to the unethical consequence of suspending the critical analyses of, say, racist and homophobic hate sites - analyses central to the activist agendas of critical race studies, gay and lesbian studies, feminism, etc.

These essays, then, accomplish our original goal of bringing together a representative range of the diversity of methodologies, disciplines, and ethical approaches relevant to online research. At the same time, however, alongside the dialogues and debates that emerge between these distinct points of view - these essays, both individually and collectively, further contribute to and represent a larger, discernable convergence in the field of Internet research ethics. In this way, they likewise bring us closer to our goal articulating guidelines and values that are global while they also preserve and reflect irreducible differences between disciplines, ethical approaches, and national traditions and law.

To see this, I will first provide an overview of how this larger convergence can be seen both in the literatures devoted specifically to questions of Internet research ethics, and the larger literatures of CMC research and related discussion in philosophy of technology.

Internet Research Ethics: an Emerging Convergence?

As the range of views and arguments represented here suggests, the domain of Internet research ethics contains a considerable diversity of views and approaches. These include a host of particular ethical dilemmas faced by researchers in the often unique contexts made possible by CMC technologies - e.g., ranging from observing participants in public chatrooms to large-scale discourse analyses of USENET archives (i.e., postings to USENET groups now made public on the Web), etc. In attempting to address these issues, moreover, concerned researchers find no shortage of potentially relevant codes and guidelines. These include existing ethical codes of professional and disciplinary organizations such as the ACM (1993) and the American Psychological Association (1992); specific laws established within the boundaries of the United States and the European Union - including, as Walther reminds us, the CFR and the E.U. Data Privacy Protection laws (National Commission, 1995). Finally, there is now a commonly-cited literature of both individual efforts to articulate online research ethics (among others, King 1996; Reid 1996; Boehlefeld 1997; Schrum 1997) and collaborative national and international efforts. These include the AAAS Report (Frankel and Siang, 1999), the NESH Guidelines (2001), the Swedish Guidelines (HSFR, 1990), and the first efforts of the AoIR ethics working committee (2001, 2002).

Perhaps not surprisingly, these first efforts tended to emphasize important differences - and this on two levels, i.e., normative and meta-ethical. On the normative level, different researchers and ethicists would often draw contrasting conclusions as to the appropriate ethical responses to concerns. For example, should researchers first of all protect the identity of participants in a chatroom or contributors to a listserv by assigning them pseudonyms, developing group profiles, etc. (see Elgesem and Bruckman, this issue, for a list of such strategies)? Or should researchers fully identify participants and contributors, on the view that their writings amount to copyrighted material, and must be treated under provisions of copyright law, including "fair use" provisions for scholarly research (Walther, this issue, addresses this argument). On the meta-ethical level, a series of questions arise. First of all, what analogies may we productively use for applying extant codes and guidelines to the new venues made possible by CMC technologies? In particular, while we casually speak of cyberspace, as the discussion represented here by Walther, on the one hand, and Bruckman, O'Riordan and Bissett, and White, on the other, makes clear - the analogy of persons (=human subjects) in space leads us in quite different ethical directions than the analogies of textuality and persons as authors. More broadly, as our authors also raise forcefully: is online research best approached on the analogy with human subjects research as constrained by the guidelines developed in the decades since World War II in medicine and the social sciences - and/or can the disciplines and correlative ethical guidelines of the humanities also legitimately claim their purchase here? Finally, cross-cultural comparisons uncover important differences between more utilitarian approaches in the United States and more deontological approaches in the E.U. - both with regard to privacy laws (e.g., Reidenfeld 2000) and Internet research ethics (see AoIR 2001).

Nonetheless, more recent work appears to shift from these initial emphases on difference towards something like greater coherency and convergence. For example, a recent online consultation of the AAAS Committee for Scientific Freedom and Responsibility (CSFR) identified both a range of venues and list of ethical issues that appear to provide a reasonably complete inventory of the field of Internet research ethics. Beyond the venues discussed by our authors - listserves, chatrooms, and websites - the CSFR list further includes: USENET newsgroups; homepages on the World Wide Web; Weblogs; ICQ and other forms of "instant messaging"; MUDs and MOOs; various forms of audio- and video-teleconferencing (e.g., CUSeeMe); and, perhaps, some forms of Computer-Supported Cooperative Work (CSCW) systems.1 As well, the CSFR list of central ethical issues can be represented as follows:

1. Respect for persons (as foundational value for all the rest) [deontological value]
2. Privacy [deontological value]
3. Confidentiality [deontological value]
4. Informed Consent [deontological value]
5. Anonymity/pseudonymity as these - along with 1-4 are complicated in Internet venues
6. Risk/Benefit to Participants [consequentialist approach]
7. Risk/Benefit to Social Good [consequentialist approach]
8. Public vs. Private space
9. Subject compensation
10. Justice (i.e., the fair distribution of the benefits of research)
11. Cross-cultural issues
12. Special/ vulnerable populations
13. Deception (pro-active)
14. Non-disclosure (passive)
15. Conflict of interest
16. Research misconduct

This list not only includes the issues addressed by authors in this issue - most importantly, the range of values and acts dependent upon respect for persons, i.e., treating human beings as autonomous beings, whether as human subjects and/or as authors. Moreover, this list is consistent with the issues identified by the AoIR ethics working committee (2001, 2002), the NESH guidelines (2001), and the Swedish guidelines (HSFR, 1990).2

These initial agreements, moreover, are consistent with more recent agreement that despite noted differences between online and offline venues,3 researchers can more or less reliably apply the appropriate codes and law, as developed in the offline world, to online venues and issues. Malin Sveningsson (2001), for example, relies on a strong analogy between observation in a public space and her research on chatrooms to apply the ethical rules for the former to the latter. Likewise, in this issue, Capurro and Pingel use Kant and Heidegger to stress the importance of embodiment to our knowing and Being - and thereby appeal to a Council of Europe convention on the protection of human rights in medicine and its requirements for informed consent as a basis for endorsing informed consent as an ethical requirement for (at least some forms of) Internet research. Similarly, Elgesem demonstrates that the NESH guidelines would issue in the sorts of ethical decisions in fact characteristic of online researchers with regard to protecting privacy, etc. Walther likewise argues that the human subjects definitions and codes laid down in the CFR apply directly and appropriately to online research. While White, O'Riordan and Bisset, and Bruckman argue for different starting points (i.e., drawn from the humanities and the arts, in contrast with medicine and social sciences) - they nonetheless agree with Capurro and Pingel, Elgesem and Walther on the meta-ethical point that such analogies with offline methodologies and codes are strong enough to provide working ethical guidelines. In this way, they follow the example of Sharon Boehlefeld, who uses the ethical code of the Association of Computing Machinery (ACM, 1993), given her view that "doing ethical cyberspace research is not much different from doing any ethical research involving ethical subjects" (1996, 142). This view, finally, is likewise defended by Steve Jones (2003).

These examples of greater emphasis on the similarity between online and offline worlds are consistent, finally, with a larger shift in the scholarly and popular literature on CMC. Briefly, these literatures were dominated in the 1980s and 1990s by a range of Cartesian and postmodern dualisms. These included precisely the emphasis on a Cartesian opposition between body and mind - apparent in various celebrations of a "liberation in cyberspace" to be accomplished precisely by freeing the mind from the constraints of what John Perry Barlow contemptuously called "meatspace" (1996; cf. Haraway, 1990); a correlative call for "homesteading the virtual frontier" (Rheingold 1993) and fervor for virtual communities as replacing real-life communities; a postmodern celebration of new CMC technologies as realizing a revolutionary overturning of modernity and print (e.g., Landow, 1992, 1994); and, in the domain of computer ethics, the insistence that a more radical dichotomy between a contemporary computer ethics and earlier ethical systems (e.g., Gorniak-Kocikowska 1996). More recently, however, within the literature of CMC research (Nancy Baym), more popular representatives of 1990s postmodernism (Jay David Bolter, Howard Rheingold) and especially phenomenologically-oriented philosophy - there is a clear (re)turn to a (re)new(ed) emphasis on the role of embodiment in our sense of identity (Borgmann, 1999; Becker 2000, 2001) and learning (Dreyfus, 2001), and thereby on the importance of real-life communities (Rheingold 2000, Bolter 2001; Baym 1995, 2002).4 In short, the recent examples we have seen of greater reliance on analogies between online and offline worlds to develop Internet research ethics cohere with these larger turns from Cartesian/postmodern dualisms that emphasize the radical differences between online and offline domains towards a thematic interest in embodiment as emphasizing precisely the inextricable connections between the two.

On the Road to Convergence: Examples of Ethical Pluralism

Beyond these multiple emphases on the connection between online and offline domains, and thus the strength of analogies between the two for the sake of developing ethical guidelines for online research - recent work also provides a number of specific examples of an ethical pluralism that allows us to recognize a range of specific ethical positions as legitimate, rather than either insisting on a single ethical value (monolithic ethical dogmatism) or simply giving up on ethics altogether and embracing ethical relativism. The examples and models of such pluralism, as a middle ground between dogmatism and relativism, are consistent with the larger convergence I suggest is taking place - i.e., as these offer us specific instances and frameworks that encompass both agreement (e.g., on basic values or first principles) and irreducible differences (e.g., in the specific application of those basic values, principles, etc.)

As a first example, consider Bruckman's and Walther's arguments in this issue that materials posted on the Internet are by definition copyrighted, and thus are to be treated as authored material requiring permission for citation (except under the provisions of fair use), etc. On this line of argument, there is no absolute requirement that researchers protect the identity of the subjects whose work they study and may reproduce in varying degrees of disguise. By contrast, a number of researchers have argued that in light of the real harms that such disclosure may lead to, they should instead always protect the identity of their subjects (Bruckman in this issue, who cites Reid, 1996; see also King, 1996; Smith, 2003).

But as Bruckman repeatedly observes, this apparent dilemma may be just that - i.e., apparent. It may be more appropriate to see these choices as choices on a continuum, each of which is justified and legitimate. A first way of articulating such a continuum is to distinguish between those choices that represent a kind of ethical minimum - and thus to be mandated by codes and law - and those choices that represent a more ethically demanding perspective, but which remain precisely a matter of individual and/or institutional choice. Walther moves us in this direction as he offers important distinctions between the user's sense of fairness (including the expectation of privacy), the researcher's personal ethics, and "cross-media institutional guidelines that are reasonable and clear." In this light, he suggests the pluralistic compromise that includes both what we might characterize as the minimal but adequate ethical requirements that IRBs should sustain, and possibly more restrictive ethical choices that, whatever IRBs might do, researchers are free to impose upon themselves in light of their own ethical beliefs, precisely if they take seriously the importance of respecting the expectations and intentions of users. Both positions are justifiable choices: but as Judith Jarvis Thompson's discussion of abortion suggests (1971), respecting the diversity of our ethical views is well served by insisting on only a "minimally decent" ethic to be imposed by law - one that then leaves us perfectly free to impose more ethically demanding requirements on ourselves (as praiseworthy but not legally mandated "good Samaritans," for example). In this instance, researchers are thus left free to choose precisely as Reid (1996), King (1996), and others (e.g., Smith, 2003) have done: that is, to protect the identity of their participants and the listservs under study is certainly allowed as a more ethically demanding position - but not required under Walther's interpretation of the CFR (i.e., the listserv is a public space in which participants cannot reasonably expect privacy protection). In other words, the decision to protect privacy lies within a pluralistic range of allowable but not necessarily required possibilities - choices justified both by the precedent of previous researchers and the basic values of the CFR, including the primary obligation to avoid harm.

A second example addresses cross-cultural differences - and illustrates a second kind of ethical pluralism, i.e., one in which there is shared commitment to a fundamental norm, value, or guideline - but the interpretation or application of that norm differs in different contexts. A central issue for Internet researchers is whether, and if so, under what circumstances informed consent is required. On the one hand, both Elgesem (relying on the NESH guidelines) and Walther (with regard to the CFR) make clear that there are specific conditions under which informed consent is not required - i.e., when the research focuses on publicly-accessible archives and no link is made by the researcher to subjects' identity, or, where identity may be divulged, there is minimal-to-no risk to the subject (as in surveys of a non-sensitive nature, etc.). Both guidelines agree: insofar as a strong analogy can be made between a public space - in which observation does not require informed consent - and a public chatroom, there is likewise no requirement for obtaining informed consent (2001).

The additional question of recording activity, however, is answered differently by Walther and Elgesem in this issue. For Walther, such recording is ethically unproblematic. For Elgesem, by contrast, such recording (audio and/or video) requires informed consent. In both cases, however, the issue is one of expectations. For Elgesem and the NESH guidelines, it would seem, people in public places do not expect to be recorded without their knowledge and consent - and, in keeping with the strongly deontological cast of the NESH guidelines, these expectations are paramount. By contrast, Walther follows Jacobsen's argument that such expectations are misplaced:

Behavior in public settings is in fact not protected from recording for research; the only communication outside one¹s personal space that is protected by the CFR [Code of Federal Regulations] due to an expectation of privacy is that which occurs within very restricted contexts such as one's physician's, therapist's, or attorney's office.

Hence, while Elgesem and Walther reach different conclusions regarding the ethical propriety of recording behaviors in public spaces on the Net - they do so through a shared argument: in both cases, the expectations of the actors/agents involved are paramount. In this light, while a U.S. position and a Norwegian position may fundamentally differ - they do so only on a first level, i.e., in terms of the consequences drawn from the same starting point. On a second (meta-ethical level), they agree on the ethical importance of actors' expectations. Again, this is a pluralistic model - one that conjoins both important shared norms or values (i.e., on the importance of expectations in guiding our ethical responses) and differences (in terms of the interpretation or application of those norms or values). 5

This same sort of pluralistic structure, finally, is at work with regard the significant differences between U.S. and European Union approaches to computer ethics in general and Internet research ethics in particular. Broadly, there is an apparent contrast between both laws and guidelines. So, to begin with, European law (first of all, the E.U. Data Privacy Protection laws) and ethical codes for research (primarily, the NESH guidelines) more fully endorse a deontological insistence on protecting the rights of individuals, no matter the consequences. That is, these rights - including rights to autonomy, confidentiality, informed consent, etc. - and their protection are emphasized as (near) absolute values, ones that cannot (generally) be overridden by utilitarian considerations of possible benefits gained at the cost of compromising these rights, e.g., in the name of economic efficiency and/or research interests, including possible benefits to society. By contrast, U.S. law regarding data privacy, for example, appears to clearly favor the utilitarian interests of economic efficiency - first of all, the economic interests of corporations - over individual rights (see Aguilar 1999/2000 for an extensive comparison). This same contrast between European and U.S. approaches can be seen in the NESH guidelines: these provide more complete endorsement of human rights, even if these endorsements may lead to greater restrictions on research - and thus potential research benefits - than are likely to be found in the United States. As but one example, where U.S.-based research guidelines focus on the protection solely of the individual who participates in a research project - the NESH guidelines require researchers to respect not only the individual, but also "Š.his or her private life and close relationsŠ." (2001).6

At the same time, however, these large differences may again be seen as differences on a first level - i.e., with regard to interpretation, implementation, etc. - coupled with fundamental agreements on a second level, i.e., with regard to shared values, norms, commitments, etc. So, for example, Paul Reidenberg discerns a global convergence on what he calls the First Principles of Data Protection: the differences we have noted result from differences in implementation, i.e., through "either [current U.S.-style] liberal, market based governance or [current E.U.-style] socially-protective, rights-based governance" (2000, 1315).7 Similarly, Diane Michelfelder traces the ways in which both U.S. and European law are rooted in a shared conception of fundamental human rights - conceptions articulated both in the 1950 European Convention for the Protection of Human Rights and in the U.S. Constitution itself (2001, 132). In this light, a pluralist model again emerges - one that allows us to conjoin differences in interpretation and implementation with basic agreements on fundamental values.

Overall, these examples of ethical pluralism - ranging from particular interpretations of research codes to different implementations of privacy law - strongly support a pattern of convergence in Internet research ethics. At the same time, of course, such a convergence is not an excuse to dismiss or minimize important ethical differences and debates that remain. As Elgesem and Bruckman remind us, for example, there is still debate regarding the adequacy of any analogy between online communicative venues and "public spaces." Nonetheless, these examples of pluralism fit in with a larger coherency that I have argued may be discerned in especially more recent work in Internet research ethics

Concluding remarks: Beginnings of a coherent Internet research ethics?

We have seen, then, a pattern of convergence in the broader literature of Internet research ethics, the literatures of both CMC and philosophy of technology, and in the specific contributions collected here. These convergences, moreover, are supported by structures of ethical pluralism that help articulate both important differences and agreements in Internet research ethics.

These various points of convergence, finally, make it possible to begin articulating a series of guidelines for Internet research that appear to reflect both shared agreements on basic norms and values and a recognition of the ethical legitimacy of more than one interpretation or application of those norms and values. Indeed, Elgesem¹s "ethical rubric" (my term for his algorithm of ethical questions intended to help researchers decide whether or not a given research project fulfills the ethical requirements of the NESH guidelines) is but one example of several efforts to articulate the guiding ethical questions that researchers and their oversight boards should ask of a proposed research project. So, for example, the University of Bristol has published such a questionnaire (n.d.), based on the E.U. Data Privacy laws (Directive 1995; cf. Suler 2000). Similarly, Dr. Chris Mann (Cambridge / Oxford) is developing a list of guiding questions likewise designed to help researchers sort through the primary ethical concerns of online research (2002): her work, along with Elgesem's and Bruckman's (see also Bruckman, 2002) will be incorporated in the forthcoming report of the AoIR ethics working committee.

While a complete guide to Internet research ethics that reflects these convergences is beyond the scope of this introduction, it may be helpful to close with a brief indication of at least five elements such an ethics will likely include.

1. Insofar as the similarities between online and offline domains hold - what law(s) are applicable?

As we have seen, in the U.S., researchers in medicine and social science are bound first of all to the CFR, while their colleagues in the European Union must turn to the provisions of the Data Privacy Protection laws. While both, as Riedenberg and Michelfelder have argued, may be rooted in and converge on the same set of basic values and rights - researchers nonetheless need to first attend to the primary laws of their own country.

2. Insofar as the similarities between online and offline domains hold - which analogies do we use, as characteristic of which disciplines and ethical codes?

As we've seen elaborated here, a number of striking differences emerge between the ethics of human subjects research (as resting on analogies with persons as subjects and defined in the CFR for medicine and social sciences) and the ethics associated with humanities research (as resting on analogies with authors and textuality). Much depends here, of course, on the disciplinary background and approach of the specific researcher(s) involved. But especially as Internet research very frequently requires interdisciplinary approaches, it seems very likely that these diverse analogies and correlative ethical codes will likewise require some sort of amalgamation. But again, while their differences are irreducible - as the models of pluralism (including Bruckman¹s examples of ethical continua) explored above suggest, it may indeed be possible to craft an ethical code that conjoins these two approaches - echoing, in particular, O'Riordan and Bissett's argument that "Ša hybrid model of relational ethics that incorporates text, space and bodies needs to be applied to Internet research."

3. If cyberspace is a space - we are pressed to understand it as a public space.

That is, as Walther forcefully makes clear, the technologies of the Internet and the Web are intrinsically public in many ways. Whatever the expectations of privacy that users may hold, it is simply mistaken - unless one actively takes stringent precautions, e.g., by using various forms of encryption - to think that whatever one launches onto the Web and the Net enjoys any privacy whatsoever.

In light of these technological realities, we are first of all encouraged to pursue the analogies with public spaces that Walther and the NESH guidelines follow - with the consequence that observation in these domains without informed consent is ethically justified. (As we have seen, however, such a stance does not prevent concerned researchers from seeking informed consent, in light of possible harms - nor does it forbid the stricter position of the NESH guidelines, i.e., the prohibition against recording online behaviors without informed consent, as based on respect for expectations of privacy, even if these expectations are misplaced.)

At the same time, this public space, as Walther helps us understand, is protected by both U.S. and E.U. law. In particular, all other things being equal, anything appearing on the Web or the Net should be understood as copyrighted by the author and thus subject to copyright law and the provisions of fair use.

4. Expectations matter.

We have seen that a primary way of resolving ethical issues - especially if we take a more deontological rather than consequentialist approach - is to respect first of all the expectations of the persons involved (Elgesem). This attention to expectations, moreover, is supported here by Capurro and Pingel¹s call for an ethics of care and a specific practice of "Šrespect for the interests and values of the people subject to online research." More broadly, the importance of expectations is supported by strategies forwarded, for example, by Habermas and feminists that emphasize the primary importance of perspective-taking and solidarity, i.e., an effort to empathically understand and support, so far as possible, the perspectives and views of one's cohort (Ess 2002b). This is, more broadly, a form of the Golden Rule, which, whatever its complications in praxis, remains an important guideline for ethical behavior.8

5. Doing the right thing, for the right reason, in the right way, at the right time remains a matter of judgment or phronesis.9

Even if such guidelines and convergences regarding shared norms and values in place as I have sketched them here were to emerge as enjoying something of a consensus among researchers and ethicists - in the end, researchers and their oversight boards will have to rely on judgment to determine which guidelines apply to the specific details of a given project, and, in the case of conflict, which norms or values outweigh competing values and norms. For ethicists, this will come as no surprise. But some colleagues expect ethicists to be able to provide a sort of calculative algorithm or ethical recipe which will, in the face of specific dilemmas, result in clear, unambiguous, and final answers. Part of our task, then, is not only to assist as we can with clarifying choices and ways of making those choices: it is, further, to connect this process to the larger project of education - indeed, the pursuit of human excellence or virtue (arete) that makes the development of ethical judgment a central goal (cf. Hamelink 2000; Ess, 2002a)

On an optimistic note, these guidelines and their larger contexts - i.e., the convergences noted in the literatures of Internet research ethics and the broader literatures of CMC and philosophy of technology, especially as these entail a number of structures of ethical pluralism - would appear to stand as at least first steps toward fulfilling our initial goals. Again, these are the goals of establishing an Internet research ethics that (a) emphasizes both similarities and differences between offline research (now including both human subjects models and models drawn from the humanities) and research into online behaviors, and (b) articulates Internet research values and guidelines that are both global in their validity while acknowledging the cultural and national differences that require specific ethical codes and guidelines for distinctive cultural groups (in some measure, on the pluralist model, as distinctive applications and interpretations of globally shared values). In part, we will continue to fulfill the NSF mandate for collaborative, international discussion as the AoIR ethics working committee, for example, will further this discussion - most immediately, in anticipation of its next report to AoIR as part of the AoIR 3.0 Conference in October, 2002, in Maastricht, the Netherlands. More broadly, the work of the AAAS Committee for Scientific Freedom and Responsibility (also funded by the NSF) should result in not only the planned bibliography of all English-language literature pertinent to Internet research ethics: it is also likely that at least one international conference on Internet research ethics will emerge from this project as well.

At the same time, however, while we have seen here initial suggestions of coherence in an emerging literature and body of law and policy in the United States and the European Union - these are, of course, only tentative first steps. As the further work of the AoIR and CSFR suggest, whether - and if so, how far - these convergences, examples of pluralism, and suggested guidelines will emerge as indeed supported by an interdisciplinary and international consensus of researchers and ethicists very much remains to be seen. In addition, these literatures, laws, policies, and discussions still represent a limited framework - i.e., that of the North. Even within this framework, much remains to be done: for example, the NESH guidelines and their Swedish counterparts may reflect the greater influence of Protestant Christianities in northern Europe, in contrast with the greater influence of Roman Catholic tradition, including ethical tradition, further south.10 To further expand these frameworks into genuinely global ones clearly requires new dialogue and engagement with the traditions and value systems of Central and South America, Middle- and Eastern Europe, Francophone countries, the Middle East, the Islamic world, Africa, Asia, and the many indigenous peoples or First Nations who have survived the colonization of recent centuries - especially as these domains represent the most rapid growth of CMC and thus some of the most interesting domains for Internet research. While some recent research has helped articulate the diversity and agreements among the world¹s various cultural domains with regard to communication and technology11 - to my knowledge, discussion of research ethics, for example, with our counterparts in Japan and China are only at the very earliest stages.

Hence, there remains much to be done. We hope that this special issue will inspire some of the needed discussion and debate among the readers of Ethics and Information Technology, and thereby continue the discussion taken up by our delightful panel at CEPE'01 - a discussion that will perhaps one day indeed fulfill its goals of encompassing genuinely global perspectives on Internet research ethics.

References

Endnotes

* In addition to my gratitude to Helen Nissenbaum, the members of the CEPE panel, and the National Science Foundation (as acknowledged below) - I here wish to emphasize that the account I develop here of a growing convergence in Internet research ethics reflects the contributions and discussion of Dag Elgesem (University of Bergen), Chris Mann (Cambridge University/ Oxford University), Annette Markham (University of Illinois at Chicago) and the students and faculty participating in the "Making common ground: Methodological and Ethical issues in Internet-research," Department of Interdisciplinary Studies of Culture, Norwegian University of Science and Technology (NTNU), Trondheim, Norway. May 30 - June 2, 2002, and affiliated graduate course on Internet research ethics. My thanks especially to Janne Bromseth (Center for feminist- and gender studies/Department of Interdisciplinary Studies of Culture (NTNU) and May Thorseth (Programme for Applied Ethics, NTNU) for their insights and work in organizing these events. As well, I am deeply indebted to Gitte Stald (Film & Media Studies, University of Copenhagen), Anker Helms Jorgensen (Digital Aesthetics and Communication, The IT-University of Copenhagen) and Robin Cheesman (Centre for Net-based Collaboration and Learning, Roskilde University) for organizing a series of lectures in May, 2002, that further allowed me to present and critically evaluate many of these claims. BACK

1 Based on a draft document developed by the AAAS Committee for Scientific Freedom and Responsibility. Used by permission. BACK

2 In addition, this list survived the scrutiny of faculty (including ethicists and researchers) and students who participated in the recent series of lectures, conference, and graduate course on Internet research ethics noted in the acknowledgements, above. BACK

3 In the AoIR ethics working committee preliminary report, we summarized these differences as follows: online research entails