******************************************************************************* Please note: This is part of an on-line ASCII text version of the task force report. A full text and print-ready PostScript version is available via anonymous ftp from ftp.ucs.ubc.ca in /pub/info/ubc/reports/appropriate-use.ps. UNIXG account holders will find it on unixg.ubc.ca in the same file. Appendix C is post with permission of Dr. Danielson. ******************************************************************************* APPENDICES * APPENDIX A Terms of reference for Task Force and list of members * APPENDIX B Dr. Strangway's July 30, 1992 letter * APPENDIX C Submission to Task Force by Dr. Peter Danielson, Applied Ethics Ethics & Computer Mediated Communication(footnote 1) by (c) Copyright Peter Danielson (footnote 2) Senior Research Fellow Centre for Applied Ethics University of British Columbia Introduction The Task Force on Appropriate Use of Information Technology is charged with examining some of the most important and difficult questions facing an ethically sensitive university community committed to using computer mediated communication (CMC) (footnote 3). I amhappy that you asked me to appear before you, as these issues are central to my reseach, andvery interesting as well. I also realize that the features of the problem -- ethics and CMC -- suggest that you will likely encounter too much input. So I will try to be brief and focus on those facets of the problem about which I have something special to say as a moral philosopher and as a professional applied ethicist. I will organize my remarks under three headings. First, I sketch the structure of the moral problem at hand. Second, I shall sketch our experience at the Centre for Applied Ethics in ethics and computer mediated, communication indicating some practical risks of censorship to our work. Third, I sketch some practical proposals. 2. A Conflict of Values We face a conflict of two values: the value of freer communication due to computer mediated communication (especially networking) and the harms caused by offensive messages. Those unaffected by either value will not have our problem. (Those who do not wish to participate in the Internet's technological adventure do not share our ethical problem and those morally less sensitive to the harms of technology can let the technology take them where it may.) To focus my presentation, let me specify the conflicting value assumptions that I see structuring the problem: Assumption 1: CMC is a valuable new communication medium which comes under the prima facie case for mininal constraints on free speech within a university community. Assumption 2: Some find the content of some messages objectionable posing a prima facie case for censorship. 2.1 Specifying the Assumptions Notice two things missing from these Assumptions. First, `pornography' is not mentioned. I do not limit my discussion to pornographic messages because the terms of reference for the Task Force are not so limited, but refer generally to "materials from the Internet [others have] seen as objectionable." Since I have not seen the supplied examples, I cannot assume that they are all pornographic. Also, pornography is a strongly contested term, so an important policy issue will be defining what is to count as pornographic. It would seem more expeditious to focus directly on the more general problem of offensive communication. Second, I make no special mention of the technology involved. In most of my conversations about this issue, in classes, at conferences and over several Internet-based colloquia, I have found that many participants rush to the `technological gambit', taking the problem to be a technical one. For example, people often think that the solution is simply denying certain newsgroups disk space, "because this costs the University money". But there is nothing special about CMC technology -- paper also costs money. (footnote 4 ) We need a general solution to the problem of offensive communication, which we can then apply to various technologies: books, newspapers, newsletters, mail, posters, phone, e- mail, bulletin boards, IRC-chat, and so on. Finally, I note that we are raising the question in the context of a Canadian university. I stress Canadian because much of the discussion in the wider Internet community assumes the US. legal framework. (footnote 5) But more important, I stress university because quite different policies may be appropriate in institutions, such as businesses, or indeed the public realm at large, not predicated, as ours is, on the cen- trality of the value of open inquiry and communication of results, some of which are likely to be offensive to some of us and to the larger community. (footnote 6) 2.2 Resolving the Conflict? Having set out the conflict, what can be said about resolving it? My own inclination would be to develop a libertarian defense of the freest communication in the special setting of the (research) university campus. Several well-known lines of argument agree here. (footnote 7) On the one hand, Mill's famous utilitarian defense of free inquiry has its clearest application inour setting, as the university and CMC are settings which support intellectual and critical feedback. On the other hand, Scanlon's lesser known complementary account of the limitations on the rights of censors has special application in a university setting, where the authority of the censor presumably must be intellectual and moral, not political, and therefore the grounds for the decision must be accessible to the judgment of the university community, arguably the best informed community on matters intellectual and moral. But I do not propose that we should try to settle these issues, so central to our ethical tradition, here. I would contend that the wider university community must be engaged on these issues -- but this is simply to reiterate Mill's emphasis on the educational value of free speech, and Scanlon's on the right of autonomous agents to make up their own minds about moral and political matters. (See Proposals for one means to open up this discussion.) 2.3 Relieving the Conflict A more modest goal is to reduce the conflict between free communication and offensiveness. One established liberal technique to reduce this conflict is to distinguish private from public realms, and stress voluntary exposure to what consenting adults do in their private realms. Taking this perspective, it is surprising that CMC should be the focus of our current problem, since it tends to be governed by a high degree of voluntary exposure. Even the most offensive newsgroup requires that I log on to a computer network, execute a newsreader, select that newsgroup, and allow particular messages to appear on my screen long enough to read them. Even at that point they are still less invasive than the graphic dailies displayed in the newspaper box at my bus stop. This suggests a general recommendation. General Recommendation: Stress voluntariness of Exposure. There are many different ways to categorize the conflict between the harm of offensive content and communicative values. I propose that we stress the degree to which the exposure to offense is voluntary. This yields a ranking of some recent and hypothetical cases of offensive computer mediated communications in the following order: Best case for intervention: involuntarily receivable Unsolicited e-mail without informative header; posting to ubc.general (e.g. Burkholder's sex slave ad) Plan file Weaker case for intervention: publicly accessible to minors alt.sex; IRC-chat; tinysex M.U.D. Weaker case for intervention: publicly accessible to adults Weaker case for intervention: accessible to restricted group CBPENet; Thick/Thin Worst case for intervention: private files "The Japan that Could Say No" 2. Network Research Projects at the Centre for Applied Ethics The Centre for Applied Ethics is the home of three research projects using university computers and the Internet. I sketch them to indicate the range of our involvement and also to suggest the risks of a censorial policy to the type of research in which we engage. 2.1 Canadian Business and Professional Ethics Network (CBPENet) The Canadian Business and Professional Ethics Network, whose principal investigator is Michael McDonald, Director of the Centre for Applied Ethics, offers its members a convenient, private, and secure means of discussion, some topics of which may involve ethically sensitive material. CBPENet is a private network; the co-investigators and the initial invitees, representing a broad spectrum of the Canadian business and professional ethics community, are the original members. We have created democratic procedures for adding new members and spawning sub-networks.(footnote 8) This project has received some notice among granting agencies for its ability to draw together constituencies both within and without the university community. This trustbuilding task is not trivial when different disciplines and professions as well as a new technology are involved. I have studied it, both from the theoretical side (Danielson 1992) and in a review of the literature of the sociology of computerization (footnote 9) We included in the design of CBPENet an initial face-to-face meeting in Vancouver, both to build trust and personal relations prior to the introduction of CMC and also to make democratic decision on network governance the practical norm. What are the risks of policy change here? Privacy and -- of course -- freedom from censorship is part of the appeal of CBPENet, which has discussed sensitive issues effecting both the members and their clients. There are also broader questions of interfering with the democratic self- governance of a new virtual community. (Will the censors ask to join CBPENet or regulate it from without?). More legalistic questions arise whether the university, claiming `ownership' of equipment purchased with grant monies for a particular purpose, may interfere with that purpose, in this case a private, self-governing network. 2. 2 Computer Ethics through Thick and Thin Computer Ethics through Thick and Thin, of which I am the principal investigator, is an international and interdisciplinary experiment in the effects of different parameters of CMC on ethical discussion and decision-making. Participants from most continents, and from many disciplines and organizations meet in a pair of colloquia, one under conditions of full information and one under continuing pseudonymity. This experiment, which was launched at the First National Conference for Computers and Values, in August 1991, has been underway for almost a year, using software under development in my Group. (footnote 10) This project involves ethically sensitive material and is committed to protecting the privacy of its members, half of whom are shielded by their use of pseudonyms. This project indicates some of the difficulties facing censorship policies in a research institution. When I was searching the enormous resources of the Internet for software to implement pseudonymous communication, I was directed to alt.sex.bondage, which is a leader in this area. More recently, rec.games.mud has been the source of innovative software for constructing multi-user dialogues. Unfortunately the first of these groups is among the first to be removed as offensive pornography and the second might be at risk when someone reads about the possibilities of tinysex discussed there. 2.3 The Ethics of Environmental Modeling I have recently formed the Computer Modeling & Communication Group within the Centre to pursue research initiatives. We have recently been awarded a contract, in connection with a large NSF grant, to provide ethical consulting to the North Carolina Super-computer Group. which builds environmental models for the Environmental Protection Agency in the U.S.A. Part of my plan is to provide principled ways to decide what features of a model should be public and which should not. Since environmental models are one of the most controversial subjects in politics just now, we had to assure our client that we could provide the required privacy. Yet some will be offended by assumptions made by some of these models and some will charge that some of these assumptions should not be kept from the public. I also plan to develop a remoted whistle-blowers network for this client. All this information will be kept on university computers. 2.4 Risks The direct risks of censorship are fairly obvious. We have promised various constituencies privacy and anonymity. It is not clear that we will be able to provide this under new policies. It is not clear who the censors will be, what they will censor, who they will share the contents of which files with, and so on. Indirect risks are more difficult to foresee. Censorship "chill" is probably the most insidious. Will we be reluctant to take on projects on "poisonous" subjects. Will clients on the "wrong" side of these topics even approach us under the new policy? How will U.B.C's reputation be affected? I will probably be better able to answer this last question when I present my paper on Philosophical Research and Networks at the American Philosophical Association's panel on The Electronic Philosopher, as that audience has, in the past, been keenly aware of censorship as a philosophical issue. 3. Some Proposals You have asked for proposals that are practical in three respects: (1) they should implement the Sexual Harassment Policy, (2) they should be acceptable to the university community and (3) they should be readily communicated to the campus information user community. My proposals will fall short on (1), stress an educational version of (2) and satisfy (3) best. A leading thread of my research on ethics and technology -- and CMC technology in particular -- has been the inadequacy of attempts to regulate emerging technology by applying prior moral intuitions. To this somewhat skeptical and negative claim, I add a constructive and hopeful conjecture: CMC has potential to be an ethically progressive medium. This is not the place to defend a conception of ethical progress, but the point is simple. If we get better at solving our moral problems by wide, open, non-coercive intellectual discussion, then within the technological community, CMC should be morally progressive, since this plastic medium allows us better to model ideal moral communication. I am, however, not a technological utopian. For example, I stress within the technological community, because, of course, not everyone, not even every academic or intellectual, has access to the Internet. Nor is the medium always or even usually used in the critical, morally sensitive way that it might. But these are things that we can do something about. Therefore I suggest: Proposal 1. We should open a forum for discussion of ethics and sexual harassment on the UBC node of the Internet. Some of the best discussion of our problem I am aware of takes place on the Internet itself. For example, I found the discussion of President Strangway's letter of 30 July `92 in ubc.general to be quite enlightening. (footnote 11) And the Computer and Academic FreedomNews (in alt.censorship, ironically often one of the first groups to be censored) and the associated archives at the Electronic Freedom Foundation (ftp.eff.org) to be rich sources of primary materials, such as a transcript of the seminal CITY-TV broadcast and full text of various universities' policies in this area. I further propose that this be a professionally moderated newsgroup, staffed by someone with training in ethics and CMC. Comp.risks, under Peter C. Newman's extraordinary fine editorship, is a model that we might aim. Proposal 2. Open access to the Internet to a larger subset of the university community. At present, access is severely restricted. Even the privileged find access difficult. (My grants can't seem to buy me ethernet connectivity!) But it is a fairly obvious truism of moral sociology that the mores of a community excluding most humanists and social scientists will be less humanized and civil thereby. (A motto: Humanize the Internet -- Wire the Arts Faculty.) Proposal 3. Moral support for users Open access is not enough. We need to support users in the use of the critical and filtering tools that make CMC the progressive ethical instrument that I claim it might be.(footnote 12) Users need to know how to filter out offensive messages (footnote 13) and to answer postings appropriately. Perhaps they need anonymity or pseudonymity. I would also suggest that they need continuing network addresses to foster responsibility and community. Proposal 4. Encourage A Variety of Experiments Not everyone uses published media, mail, telephone, radio, or television in the same way. Surely we need to experiment even more with a less conventionally settled medium like CMC. Therefore it would be a mistake to apply restrictive policies at too broad a level. There are relevant differences between a widely available server in a teaching department and a research machine in a research group. Lumping these all together as "university computing equipment" seems to be a mistake that will curtail interesting experimentation. (footnote 14) Conclusion In sum, I propose that we deal with our problem as befits a university, that is by addressing it as a serious intellectual problem that it is. Further, I hope that we take this problem as an intellectual challenge rather than a purely political problem, and try to deal with it by the means that we deploy best: intelligence, discussion, trial and error mediated by open comment on results, playfulness and rigor. U.B.C., with its tradition of technological leadership, strong humanities and social sciences, and commitment to applied ethics, should be a leader in the task of civilizing this technological frontier. (footnote 15) References Danielson (1991) Review of M. David Ermann, Mary Williams and Claudio Gutierrez, eds. Computers, Ethics & Society. Canadian Philosophical Reviews. Danielson (1992) Artificial Morality. London: Routledge. Danielson (in press) Personal Responsibility, in T. Hurka and H. Coward, ed, The Ethics of Global Warming. Dunlop (1991) and Kling, ed. Computerization and Controversy, San Diego: Academic Press. Stewart (1986) Readings in Social and Political Philosophy New York: Oxford University Press. NOTES 1. Notes for a Presentation to the Task Force on Appropriate Use of Information Technology. 2. Comments welcome. Author's address: 1866 Main Mall, E-162, Vancouver, B.C. Canada, V6T 1Z1. E-mail: danielsnunixg.ubc.ca, FAX (604) 822-8627, Tel: (604) 822-4658. This document is available in machine readable form; contact artmoralunixg.ubc.ca. 3. I will use CMC as my most general term and the Internet as a narrower term to refer to the interconnected system of university computing equipment including those that provide access to many other computing sites. 4. This strategy can also backfire in several ways. The largest, and therefore more expensive newsgroups tend to be computer binary groups; smut is fairly compact, except in the form of images. Also, the cost objection can be met in ways that will not eliminate the offensive material. Pornographic images can be compressed, the info-mac archivists have recently raised money from the Internet community for a gigabyte disk drive, and alt.censorchip has a home financed by a millionaire patron. Finally, if full environmental cost accounting is used, newsprint may be more costly than disk space. 5. For example, the list "Banned Computer Material 1992" (archived at ftp.eff.org) of campuses practicing censorship, on which UBC appears, contains 7 Canadian and 13 US institutions, a disproportionate ratio that likely reflects the legal differences between the tro countries. 6. For a stronger statement, see "On Freedom of Expression and Campus Speech Codes of Expression", The American Association of University Professors, archived as academic/speech-codes.aaup at ftp.eff.ord. 7. The relevant selections from Mill and Scanlon are reprinted in (Stewart 1986). 8. A more detailed description of The Canadian Business and Professional Ethics Network is attached as Appendix A, with details on the first sub-network. 9. See (Danielson 1991); (Dunlop 1991) is an important collection of sources. 10. A more detailed description of Computer Ethics through Thick and Thin is attached as Appendix B. 11. Even though I was visiting UCLA at the time, I could read netnews remotely here. This should remind us that the Internet is technologically biased against effective censorship. 12. The Report of the Advisory Committee on Network News, The University of Waterloo, May 30, 1991, is a model. 13. CMC usually provides caller-ID automatically but the filter built into our mailer is neither supported nor currently working, so far as my assistants can determine. 14. Policies that put one political agency in charge of the means of communication have not historically been friendly to open inquiry and communication, because, even in the best case, that agency becomes responsible for what should, properly, be a matter of distributed responsibility, and acts more restrictively than would the varied set of distributed responsible agents. 15. I discuss the ethics of a frontier in (Danielson) in press.