Rogerian Argument Essay

B Y A . GRAHAM PEACE

Many universities do not consider the topic of

censorship important enough to address in a formal way— but the legal and ethical risks of neglecting this issue

should not be ignored.

BALANCING FREE SPEECH AND C E N S O R S H I P : Academia's Response to the Internet

U niversities are often hailed as bastions of free speech—places where censorship is reviled and students are given the freedom to explore dif- fering opinions. However, this attitude was

developed in a time when information flowed less freely than in today's world of the Internet. Technology has made available a wealth of knowledge, but along with Web sites dedicated to scholarship and harmless entertainment are sites that promote pornography, racism, and criminal activity. Should institutions of higher learning allow this material to be easily perused through on-campus labs and dormitory Internet con- nections? A conflict exists between the rights of stu- dents and faculty to free speech and privacy, and the obligation of universities, parents, and society to restrict access to information deemed unsuitable for our youth. The communal nature of academia further complicates matters. Controversial information is often accessed in computer labs, potentially exposing other computer users to information they may fmd offensive.

ILLUSTRATION BY TOMASZ W A L E N T A

COMMUNICATIONS OF THE ACM NovEmbcr 2003/Vol, 46, No, I I 105

Throughout history, politicians and philosophers hur is legal in some areas). A university must focus have debated the right of free speech. The rise of the its policy-making energy on these gray area.s. Internet has expanded tJie censorship debate to To date, limited research has addressed how, or include subject matter such as cyber-pornography, even if, colleges and universities are handling these andinthewakeofrheevemsofSepr. 11,2001,infor- issues. This article reports the results of an mation deemed politicilly undesirable. The results of exploratory survey of U.S.-based academic institu- these debates will have important legal and ethical tions, undertaken in an efFort to gauge the level and ramihcations for Internet service providers, including nature of Internet censorship in academia. A ques- the academic arena. Institutions relying on monetary tionnaire was developed, pilot tested, and distrib- support from outside sources Kiay also need to incor- uted to the heads of computer services departments porate awareness of the censorship debate into their (or equivalent) at 400 U.S.-based universities and fmancial aiid publicity considerations. colleges. Of the 400 surveys distributed, 133

In the workplace setting, a case can be made for (33.3%) were returned. Internet censorship, as employees are paid to per- The basic results, illustrated in Table 1, include form a task by a company that owns the technology the fmding that a significantly larger number of being utilized, although frequent Internet usage may institutions limit access to newsgroups as opposed to be linked to increased employee happiness [3] and learning [1). However, in academia, intellectual and social development take place as much outside of the class- room as in it, making it more difficult to find the limits that may need ro be set. Universi- ties have struggled with the issue of what. If any, Internet sites to censor, and how ro cur ofT access to unwanted infor- mation. Simply censoring sites or newsgroups based on

keywords, for example, could inadvertently restrict Table l. Basic results of the ' 3 3 responses).

Public or State Funded Universities

Private Universities

Average Number of Students

InternetAccess Provided in Computer Labs

Internet Access Provided in Some Classrooms

InternetAccess Provided in Some Dorm Rooms

Limit Access to Some Internet Sites

Limit Access to Some Newsgroups

Formal Internet Censorship Policy in Place

Informal Internet Censorship Policy in Place

N o Internet Censorship Policy in Place

N o Formal Policy In Place. But Developing One

47%

53%

11,000

9 9 %

9 9 %

8 9 %

4%

18%

34%

24%

41 %

12%

Web sites. This finding is not unexpected, since news- groups are technologically easier to censor. More sur- prising was the finding that institutions clearly do not see Inrernet censorship as a pressing issue on their cam- puses. When asked to rank the issues importance on a scale of 1 to 5, the mean response was 1.67. Simi- larly, most respondents do not think it is likely to be an issue at their institution in the future (1.78 mean

access to legitimate educational sources, such as *""'*y ^̂ 33 responses). response on a five-point medical sites. Efforrs ro censor have also met legal scale). However, over two- challenges, although not taking action may also ere- thirds of the respondents have received a complaint atelegalproblems. The University of Oklahoma ini- regarding information available on the Internet, tiated a policy of censoring pornographic Web sites mostly from students (45%), faculty (29%), and based partly on a concern over potential lawsuits, parents of students (20%).

Of those institutions with a formal policy in place, approximately 10% created that policy in response to an incident at the institution, and over 50% stated their policy is a factor in their opera- tions. Also, those that censor view the issue a.s more controversial than those that do not. Those that cen- sor also believe more strongly in the importance of developing a formal policy rhan those thac do not.

When it comes to the decision of limiting student access, 40% have no individual or committee desie-

^ , nated to make the final decision. Twenty-three percent exist where information is considered either illegal give this power to a committee, often made up of fac- but ethical (for example, gambling sites are illegal in ulty, computer services personnel, and university some areas, but considered ethical by many individ- administration. In 13% of rhe responses, the head of uals), or legal but unethical (for example, pornogra- computer services is lisred as the final decision maker. phy is considered unethical by many individuals. The respondents were asked to rate the impor-

but rhe policy was immediately challenged in court by a faculty member [4].

One of the few research papers specifically regard- ing university responses to these concerns argued that information under scrutiny can be viewed as either legal or illegal, and as either ethical or uneth- ical [2]. This approach allows for the formation of a two-by-two matrix. Legal and ethical information must clearly be uncensored and made available to all who have an interesr. Information that is unethical and illegal must be censored. However, gray areas

106 November 2003/Vol. 46, No I I COMMUNICATIONS OF THE ACM

Academic Concerns

Freedom of Speech

Legal Concerns

Privacy Concerns

Facuity Concerns

Administration Concerns

Student Coricerns

Parents' Concerns

Technological Concerns

Monetary Concerns

Public Concerns

tance of various facrors in the decision to limit Inter- net access, if such a policy exists {see Figure 1). Not surprisingly, the two factors rated as important in the decision to censor were the philosophical issues of freedom of speech and privacy. Legal and academic concerns were also highly ranked. The concerns of the individual stakeholder groups were considered slightly less important, but still significant, while the more practical issues of technology and money were viewed as far less important. These ftndings suggest institutions are trying to "do the right thing," as opposed to sim- ply considering the tech- nological and financial aspects of the decision.

In most cases, when the decision to censor is under consideration, institutions involve all members of the university community, with legal advisors play- ing an important role. However, the role of public groups is often viewed as unimportant, which is somewhat surprising, given that universities and col- leges play major roles in the community. Univer- sities may be ignoring an important constituent group.

With frequent discus- sions in the media and political arenas on the topic of Web pornography, one would expect this con- tent category to be a major consideration in the deci- sion to censor, but pornog- raphy was a clear third choice, behind criminal

Figure l . Importance of factors (rated on a scale of l = "very unimportant" to 5 = "very impor- tant") in the development of a formal censorship policy, and the decision to censor a Web site or newsgroup.

Computer Services

Administration

Faculty

Legal Advisors

Students

Public Groups

and racist content. However, pornography was a sig- nificantly more important consideration for institu- tions engaged in Web censorship than for those not currently censoring the Web. Universities choosing to censor pornography may be worried about poten- tial legal issues, which would be further complicated by the fact that many college students are under the legal age to view adult material. Sexist and anti-reli- gious content rated slightly below the three cate- gories mentioned here, while the general category of "non-academic content" ranked as significantly less

important than the oth- ers, suggesting the sur- veyed institutions are not averse to allowing students to use academic resources for entertain- ment purposes.

Worth noting are sev- eral differences between institutions with and without a formal policy in place. Not surpris- ingly, those with a pol- icy appear significantly better prepared to han- dle any possible contro- versies. While only 38% of those institu- tions with no formal policy had identified a specific decision maker, 84% of schools with a formal policy had iden-

tified a decision maker, reducing possible coiifiicts when decisions must be made. During policy devel- opment, schools must consider the practicalities of censoring; it is one thing to censor based on princi- ple, but the censorship decision must be imple-

mented. Institutions with policies appear more aware of practical issues that come into play in the decision to censor, such as techno- logical, legal, and mon- etary concerns. The study findings indicate that computer services personnel clearly played a major role in policy creation (see Figure 2). This is not surprising, as this group must han- dle the tricky techno- logical issue of cutting off access to censored information.

Fifty-seven respon- dents with a policy in place took the time to complete a second sec-

tion of the questionnaire. Of this subset of respon- dents, 79% developed the policy proactively, while 9% developed a policy in response to an incident at

InvoK Policy Creatic

Figure 2. involvement of stakeholder groups (rated on a scale of 1 = "not involved at all" to 5 = "extremely involved") in the development of a formal censorship policy, and the decision to censor a Web site or newsgroup.

COMMUNICATIONS OF THE ACM November 2003/Vol. 46, No. 107

the institution (11% did not respond or did not involvement in policy formation, the more effective know). The majority of the policies (53%) specifi- the policy was deemed. The degree of importance cally stated that Internet access must not be limited; assigned to legal concerns in policy development 13% limited access to certain information cate- also correlated positively with perceived efFective- gories, such as pornography; and 15% determined ness. A surprising negative correlation was found: access on a case-by-case basis. On average, the cur- the inclusion of parental concerns in policy creation rent policy had been in place for slightly over three was negatively correlated with the policy's perceived years. Respondents were generally plea.sed with the effectiveness. This may be due to the fact that parents policy effectiveness: the mean response on an effec- are only loosely tied to the academic community and tiveness scale of 1 to 5 was 4.22. Over 20% of are unfamiliar with its detailed workings. Many par-

ents live long distances from the academic institution and only visit on rare occasions. Unlike professors, administrators, and students, their views may not reflect the realities of the current state of the institution, and there- fore may not be useful in policy creation. Further research into this result is warranted.

Another surprising flndine: the

CLIMATE, STAKING CLAIM TO PRINCIPLES

SUCH AS PRIVACY AND FREEDOM OF SPEECH CAN

HAVE MAJOR CONSEQUENCES.

IN TODAY'S CHANGING LEGAL

AND TECHNOLOGICAL

respondents stated the policy is a constant factor in institutional operations, another 36% stated the pol- icy had been utilized in at least one situation involving possible Internet access limi- tation; and 40% stated they had never needed to utilize their policy.

As Figure I illustrates, many factors were considered important in policy develop- ment, with academic con- cerns and freedom of speech leading the way. As men- tioned earlier, technology and monetary concerns were gen- erally not considered impor- tant, and public concerns were considered unimpor- tant.

While private and public institutions were similar in many ways, several differ- ences existed. Private institutions were more likely to limit newsgroup access. Also, faculty and student concerns at private schools are more prevalent factors in the decision to censor than in public institutions, where legal advisors have much greater influence,

f

degree of involvement of computer services personnel in the decision to censor was also negatively corre- lated with perceived policy effec- tiveness. As previously stated, these individuals rated as the most involved group in formal policy development—probably due to the fact that formal policy develop- ment must take into account what can and cannot be censored, given the state of technology.

For institutions that wish to develop a formal policy, this

research yields several useful results. Since the roles of both administration and faculty correlate with the formulation of a policy perceived as effec- tive, it appears logical that inclusion of these two powerful groups in the process will lead to greater

These differences are expected. As private schools are support for the policy among university personnel, not subject to the same leg;U environment as public The study results also suggest the importance of schools, they may have more freedom to impose lim- involving other parties in the censorship decision its on Web access. Also, many private institutions besides computer services personnel, to counter- have religious affiliations. The display of porno- balance the negative effects of computer services' graphic or other offensive information would be involvement on perceived policy effectiveness, clearly deemed out of line with a religious organiza- tion's morals. Consequently, private institutions would be expected to be more likely to use censor- ship. Private schools may also have greater monetary and technological resources at their disposal to impose a censorship policy.

interestingly, the perceived effectiveness of the formal policy was correlated with several factors. The higher the level of administration and faculty

While those attempting to develop a policy can gain technical expertise from computer services personnel, who are likely to understand the hard- ware and software issues involved in formulating such policies, involving other parties ensures the consideration of legal, ethical, and social issues that may be neglected by computer services per- sonnel, who are more concerned with technologi- cal issues.

1 0 8 November 200VVol 46 No COMMUNICATIONSOFTHEACM

Conclusion The study findings give a clear impression that cen- sorship is not generally desirable to institutions. A majority of U.S. institutions do not censor in any way, and over 50% of those institutions with tormal policies in place specifically state that the Internet should not be censored. With the historic tendency of academia to be on the side of fteedom of speech, this is not unexpected. However, in todays changing legal and technological climate, staking claim to principles such as privacy and freedom of speech can have major consequences. Further investigation of this topic is warranted, both in the U.S. and globally.

The results of this survey also indicate that many U.S. insticutions do not consider this topic impor- tant. A significant percentage of academia has yet to develop policies regarding censorship, or to identify individuals or groups responsible for making these decisions. The prevailing view seems to be one of a lack of concern, but the legal and ethical tisks of neslecting the censorship issue cannot be ienored. Universities must examine the roles they play in the dissemination of information to our youth. The development of a thoughtful and enforceable policy will greatly reduce the occurrence of legal or ethical

dilemmas, and can be used to foster a constructive discussion of this important topic. Those institu- tions that have spent time and effort developing policies and identifying decision makers will be bet- ter prepared to face the uncertain future of academia and the Internet. B

REFERENCES 1. Relanget, F. and Van Slyke, C. Abuse or learning? Commun. ACM 45. 1

(Jan. 2002), 64-65. 2. Peace. A. Academia, censorship, and tht \\\KUKI. Journal of Infrrmation

Ethics 6, 2 (IW7).5'>^7. 3. Sranton, J. Company profik- of the frequent Internei user. Commun.

ACM 45, 1 U'ln- ^™2), 5'J-59. 4. Vail Gelder, L, Professor loses Internet suit. The New York Times, 146,

50687 (1997). Al'i.

A. G R A H A M P F A C E ([email protected]) is an

associate professor of management information syscenis in the College of Business and Economics at West Vii^inia University, Morgantown, WV.

Permission to make digital or hard topiirs of all or part of this work for (Personal or classroom use is granted without tet provided that copies are not madi- or distributed fur profit or conimercial advantajie and that copies bear this notice and the lull titaiion ijn [he first pa îe. To copy otherwise, to republish, to pose on scnfeis or to redistribute IO lists, requires prior specific permission and/or a fcc,

2(X)3 ACM ()()O2-O782/05/tlOO J5.()0

Stay on Top of ACM News with

MemberNet ACM launches new publications on entertainment technology, speech and language processing, and computational biology and bioinformatics. An interview with the new editor of Queue, ACM's practitioner magazine. A year's worth of USACM watchdog activities.

Also, celebrating the life and work of Anita Borg, a trailbtazer for women in computing. And an ACM Fellow is awarded Canada's highest honor.

Washington policy issues, ACM in the news, and real-life Portal user stories.

All online, in the current issue of MemberNet: www.acm.org/membernet.

COMMUNICATIONS OF THE ACM November 2003/Vol. 46, No I 1 0 9