"Censorship:  The Libraries, The Internet, and The University" 
			  Part I

		      Jeffrey Shallit
               Department of Computer Science
	           University of Waterloo

This article, the first in a series, will examine censorship at
the University of Waterloo, following the court-imposed ban on
details of the Teale/Homolka trial.


On July 5, 1993, Judge Francis Kovacs of the Ontario Court of Justice
issued a 77-page ruling imposing a ban on the publication of
information regarding the trial of Karla Homolka.  According to the
decision, "There will be no publication of the circumstances of the
deaths of any persons referred to during the trial."   The judge
stated, "A limited ban on publication must rest on the balancing of the
constitutional rights of freedom of the press and the right of a fair
trial by an impartial tribunal for Paul Bernardo Teale." 

This ban represents just one in a series of similar actions taken in
Canada this year.  As an article in the December 10 Toronto Star
reports, a New Brunswick judge imposed a publication ban in a civil
suit between two millionaire brothers, Harrison and Wallace McCain.  In
British Columbia, mining executive Robert Friedland went to court to
try to stop the CBC from airing a documentary that revealed an old drug
conviction.  A publication ban was imposed, but was lifted in
September.   As I write this, a trial is taking place somewhere near
Toronto, but nearly everything concerning the proceedings is being
withheld from the public, even the text of the judge's secrecy order.
(For more information, see the December 11 Globe and Mail article.)

Many newspapers have reacted negatively towards the ban.  The Toronto
Star called it "outdated and patronizing", and joined with several
other media outlets in a suit to overturn it.  The Washington Post
(which released details of the trial in a November 23 article) said,
"It's a bad idea to give judges anywhere the power to pick and choose
which facts can be the subject of commentary and which cannot."

The Story at Waterloo

On November 30, Provost Jim Kalbfleisch, following the lead taken by
other Canadian universities, issued a memo to librarian Murray Shepherd
to "take whatever steps are necessary to remove this material from
the Library ... on the basis that the University is not prepared
to risk being charged with contempt of court for violating a Court Order."

The library had received two foreign newspapers that contained details
of the case covered by the publication ban.  In each case, the page on
which the article appeared was removed, photocopied, and then locked in
a sealed archives box.  Photocopies of the other articles on the page
were then made available to the public.  According to a senior librarian,
this represents the first time in University history that materials
have been removed from public access because of a court order.

The broad questions of when publications bans are justified, and
whether Judge Kovacs' order was correct in the Teale/Homolka case, will
not be addressed here.  Instead, I will address the following
two questions:  If the university allows access to foreign newspapers
containing articles that could not be published in Canada,
is it breaking the law?  Was the Provost's order to remove 
access to certain articles in foreign newspapers really justified?

Libraries and "Publication"

It is important to note that Judge Kovacs' order was a limited
*publication* ban, and not a ban on *possession* of material
about the trial.  Is a library "publishing" something when
it continues to allow access to foreign newspapers?

The University's legal counsel may have thought so, but the
answer provided by the lawyers I have consulted is a resounding

For example, in a December 3 opinion, McGill Legal Advisor Raynald
Mercille stated, "The Ontario Court ban prohibits publication ...
There is no prohibition on the possession of banned material *per se*."
He went on to say, " ... merely placing the newspapers on the shelves
of the periodicals room would not constitute a prohibited act... 
I certainly would feel quite comfortable defending the University
against any complaint of criminal wrongdoing in such a case.  Our
defense is augmented by the fact that it is virtually impossible
for the University to monitor the content of each and every periodical
and newspaper it receives each day." Acting on this opinion, McGill
librarians decided not to remove the controversial articles.

The University of Western Ontario followed a similar course.
"I read the judgment," said UWO's Director of Libraries Catherine
Quinlan, "and it's a publication ban." When told of Waterloo's action,
she said, "I'm not really sure why people are doing this.  It
[what libraries are doing] is not publishing."

Was Waterloo's Library Ban Justified?

Was the order to remove material from the library correct?
It seems clear that the order was based on an overly-broad
interpretation of what constitutes "publication".  

Why did Waterloo's library act with such haste in removing the foreign
newspaper articles from public access?  I raised the question with a
senior librarian.  "I have a lot of more important things to do," the
librarian explained.  "No-one else is complaining.  You're the only one
who has called.  This is a non-issue for librarians."

The senior librarian told me, "We have no right to give access to
illegal materials."  When I pointed out that mere possession of the
article was not illegal, and that under Mercille's interpretation, what
the library was doing did not constitute publication, he replied, "The
law is not subject to interpretation.  This is Canada."

Not all library employees agree with the senior librarian's evaluation.
When I asked another librarian at Waterloo if he approved of the
Provost's order, he told me, "No, I don't.  I was kind of surprised
that it happened.  Academic freedom is a real concern for librarians."

It certainly is.  Consider the following statement adopted by CARL, the
Canadian Association of Research Libraries:  "All persons in Canada
have a fundamental right, as embodied in the Charter of Rights and
Freedoms and the Bill of Rights, to have access to all expressions of
knowledge, creativity, and intellectual activity."

"It is the responsibility of research libraries to facilitate access to
all expressions of knowledge, opinion, intellectual activity, and
creativity from all periods of history to the current era including
those which some may consider unconventional, unpopular, unorthodox, or

Waterloo's library was quick to censor foreign newspapers, but other
libraries in Canada have reacted differently.  On December 17, the
Halifax City Regional Library, based on the opinion of Mr. Mercille,
decided to ignore the advice of its own legal counsel, and place
the Washington Post article back on the shelves.  Laura Jantek said,
"Library staff cannot and do not read everything we put on our shelves...
We are neither publishing nor distributing anything by having this
newspaper in our library."  

The Nova Scotia Library Association also disagreed with Waterloo's
position.  According to a press release, "The [Nova Scotia library]
association...[is] disappointed at how eagerly public and university
libraries have censored US newspapers carrying coverage of the trial
...  Libraries are expected to be in the forefront of the struggle for
intellectual freedom, and these hasty decisions have tarnished this
reputation."  As a librarian in Halifax told me, "We're being hung out
to dry by the Universities.  Why aren't they taking the lead on this

The public library in Regina, Saskatchewan actually made available
a "reference kit" containing copies of the Washington Post and
other articles.  Ken Jensen, head of the Regina Public Library, told me,
"Removing access to material is not what a library should be all about.
If we're going to err, it's our job to err on the side of intellectual

Why was Provost Kalbfleisch in such a rush to issue his censorship
order?  In a phone conversation with a senior administration official,
I was told, "We didn't want to expose the University to the
embarrassment and cost of a contempt of court citation."

But is it really plausible that the University would be slapped with a
contempt-of-court citation?  Such an action would be highly publicized
and would be more likely to embarrass the Attorney General's office or
the OPP.  To date, not a single library in Canada has been cited.
In fact, the only person charged so far was a retired policeman
who distributed hundreds of copies of the "banned" article and then
demanded to be arrested.

Cost?  The University didn't flinch when it came to pursuing the Jack
Edmonds case, which exposed the University to tens of thousands of
dollars in legal costs.  It seems to be a matter of priorities.

Provost Kalbfleisch's order was unwise and unnecessary.  It was based
on shaky legal grounds, and it did not recognize the special role
of the University in our society as guarantor of intellectual
freedom.  What is most troubling, however, is that the Provost's
order placed the University in the peculiar position of being an agent of
the state, in deciding what can and cannot be read at Waterloo.

[Next:  Censorship and the Internet]

"Censorship:  The Libraries, The Internet, and The University"
			  Part II

		      Jeffrey Shallit
              Department of Computer Science
		   University of Waterloo

[This article, the second in a series, will examine the general
question of electronic censorship at Canadian universities.]

Once upon a time, a new technology was invented that allowed for the
recording and transmission of images more efficiently than ever
before.  Within just a few years of its invention, that technology was
being used to create "obscene" images that were widely distributed
throughout the world.  The new technology frightened and disturbed
those who did not use it or understand it.  Young people were major
users and consumers of that technology, and the proliferation of 
"obscene" images led to denunciations by university presidents and

That technology was photography.

More than a hundred years later, the new technology of the electronic
age again challenges us to fashion policy wisely so that legitimate users
and their fundamental democratic rights are protected.

Part of the problem in formulating good policy is that there is no one
single analogy that describes all the ways that a university computer
network is used.  Users may send electronic mail, which is a cross
between a telephone call and an ordinary letter.  They may access
information in a huge variety of databases and newsgroups, from cancer
information to geographical data to lists of extinct species.  In this
respect, using the network is like visiting a library.  Users may post
requests for information; this use is analogous to placing a
classified advertisement.   Users may subscribe to
electronic mailing lists, which play the role of a special-interest
newsletter.  Sometimes the network is like a newspaper:  the Gazette's
Chris Redmond posts daily summaries of news affecting the university to
"uw.campus-news".  School cancellations due to inclement weather,
traditionally announced on radio, are now also posted on "uw.general".
Users may `talk' to correspondents thousands of miles away; this is
like a telephone call.  Users may post messages to newsgroups; this is,
depending on the group and the audience, like a bulletin board, an
informal group discussion or a no-holds-barred public debate.

Thus, the network can be used the way one might ordinarily use
telephones, faxes, libraries, newspapers, newsletters, Canada Post,
bulletin boards, and radio.  The many faces of electronic
communication are a source of its strength, but they also make it more
challenging to decide what sort of usage policy is most appropriate.

While no one would confiscate and open a student's ordinary mail
without obtaining a court order, system administrators at some
universities routinely examine student computer files and e-mail
without going through any formal process.  There are legal precedents for
privacy in telephone and postal communication that can be used
as a basis for similar university policy governing electronic mail.

I contend that the best analogy for newsgroups is with books or
magazines in the library.  There is also legal support for this
view in the *Cubby* decision, mentioned below.  Users often
speak of "subscribing" to newsgroups, and some libraries use the
Net as a reference service the same way they use the reference books
on their shelves.  The Canadian Library Association and
the American Library Association have published statements on
library-related issues that university administrators should consult
when formulating policy affecting newsgroups and computer databases.
Administrative decisions need not be made in an ethical vacuum when
these extensive resources are available for guidance.

Offensive Material: What Should be Done?

In the past, the administration at Waterloo censored some newsgroups
merely because they carried offensive material.  In an incident in
1988, the newsgroup "rec.humor.funny" was banned in response to the
following joke posted there:

    A Jew and a Scotsman have dinner.  At the end of the dinner the
    Scotsman is heard to say, 'I'll pay.' The newspaper headline next
    morning says, 'Jewish ventriloquist found dead in alley.'

Ironically, during the ban, compilations of "rec.humor.funny"
material could be found on sale in the University's own bookstore.

About three years ago, in response to the appearance of "Cindy's
Torment", an offensive tale of sex and torture, on "alt.sex.bondage",
then-Provost Alan George banned the entire "alt.sex" hierarchy
(containing a dozen or so newsgroups), citing "financial reasons".
(Every MFCF (Math Faculty Computing Facility) employee with whom
I've spoken disputes this rationale, saying that in fact it was
far *more* costly to remove the newsgroups than to retain
the status quo.)

Waterloo and its newsgroup censorship became the international butt of
jokes.   The Computer Science Club distributed buttons that read, "It
doesn't have to make sense--it's University policy."

Censoring newsgroups merely because they carry offensive material is
contrary to the principle of academic freedom that supports
intellectual inquiry at a university.  For example, the Canadian
Association of University Teachers, in its Policy Statement on Academic
Freedom, says, "Academic members of the community are entitled,
regardless of prescribed doctrine, to ... freedom from institutional
censorship."  The American Library Association, in its Library Bill of
Rights, stated that "Materials should not be proscribed or removed
because of partisan or doctrinal disapproval."   Despite these powerful
statements, there has been little or no recognition by University
officials of these principles.

Illegal Material

The University's computer network constitutes one small piece in a
large web that now spans the globe.  Millions of users daily read
and post "news" on electronic bulletin boards on Usenet.  With this 
many users, it is inevitable that there will be those who use the
network in a way that will violate some Canadian law.  Material that may
potentially be defamatory, obscene, constitute mail fraud, or
violate copyright laws and publication bans is sometimes posted to
newsgroups that Waterloo automatically receives.

What, if anything, should the University do about material posted to
newsgroups that may violate some Canadian law?  Since none of the
network material deemed questionable at Waterloo has been explicitly
ruled illegal by a court, any decision to ban newsgroups necessarily
constitutes a *judgment* about legality by the administration.
Universities in Canada have historically made such judgments about
books in their libraries only very rarely, and then with great care;
yet the analogous decision to censor newsgroups is somehow considered
legitimate and even mandatory.

One proposed solution, the monitoring of newsgroups for illegal
material, is clearly unreasonable.  Not only does the present volume of
traffic makes monitoring infeasible, it is also legally unwise--because
taking on such a responsibility may make the University *more*
vulnerable to lawsuits.  This view is supported by US cases involving
university scrutiny of student newspapers.

Nor does the University have a legal responsibility to cleanse its
networks of all potentially illegal material.  In the only directly
relevant case in North America, *Cubby v. Compuserve*, a US software
developer brought suit against CompuServe, an electronic bulletin
board service, for allegedly defamatory comments posted by a third party.
The Federal judge granted summary judgment to CompuServe, observing
that "CompuServe's CIS product is in essence an electronic,
for-profit library... CompuServe has no more editorial control over such
a publication than does a public library, book store, or newsstand,
and it would be no more feasible for CompuServe to examine every
publication it carries for potentially defamatory statements than
it would be for any other distributor to do so."

The University cannot be held responsible for all traffic that passes
through its networks, and hence cannot be expected to monitor
newsgroups for illegal material.  But what, if anything, should it do
when illegal material is specifically brought to its attention?

The American Library Association has a position paper on challenged
materials that addresses precisely this issue in the case of
libraries.  It reads, in part,

	Challenged materials which meet the criteria for selection in
	the materials selection policy of the library should not be
	removed under any legal or extra-legal pressure...

	Freedom of expression is protected by the Constitution ...
	but constitutionally protected expression is
	often separated from unprotected expression only by a dim and
	uncertain line.  The Constitution requires a procedure designed
	to focus searchingly on challenged expression before it can be
	suppressed.  An adversary hearing is a part of this procedure.

	Therefore, any attempt, be it legal or extra-legal, to regulate
	or suppress materials in libraries must be closely scrutinized
	to the end that protected expression is not abridged.

An accompanying explanation states

	Particularly when sexually explicit materials are the object
	of censorship efforts, librarians and boards of trustees are
	often unaware of the legal procedures required to effect the
	removal of such items.  Many attorneys, even when employed
	by state and local governing bodies, are not aware of the
	procedures to determine whether or not a work is obscene 
	under the law.  According to U.S. Supreme Court decisions,
	a work is not obscene until found to be so by a court of law,
	and only after an adversary hearing to determine the question
	of obscenity.  Until a work is specifically found to be
	unprotected by the First Amendment, the title remains a legal
	library acquisition and need not be removed.

Although the policy is American, the idea behind it is important and
equally valid in Canada.  It emphasizes that the line between protected
and unprotected free expression is very murky; hence, universities must
tread with the greatest of care when they consider the removal of
material that is supposedly contrary to the law.

Universities have a sacred trust to support free intellectual inquiry.
In the next installment, I will show how the recent newsgroup bans have
violated this trust.

[Next:  analysis of the recent newsgroup censorship orders]

   "Censorship:  The Libraries, The Internet, and The University"
			  Part III

		      Jeffrey Shallit
	     Department of Computer Science
		   University of Waterloo

	"After all we have tasted of the age of totalitarianism, how
	can anyone suppose that the possible benefits of censorship
	could outweigh the probable dangers?"
			-- Irving Howe

[This, the final part of a three-part series, will examine the
administration's decision to ban "obscene" newsgroups and one that
allegedly violated the Teale/Kovacs publication ban.]

Judge Kovacs issued his publication ban on the transcript of the
Teale/Homolka trial on July 5, 1993.  By late November, copies of a
Washington Post article giving some trial details had appeared on
Waterloo's computer networks in the newsgroup "alt.fan.karla-homolka".

On November 30, Provost Jim Kalbfleisch issued a memo to Associate Provost 
for Computing Johnny Wong to "take whatever steps are necessary to remove
this material ... from the network on the basis that the University is not
prepared to risk being charged with contempt of court for violating a
Court Order."

Wong immediately issued an order to remove the newsgroup
"alt.fan.karla-homolka" from University networks.  Later, a single
article from the newsgroup "uw.general" was deleted.

More censorship followed in 1994.  On January 31, President Downey
issued a memo outlining a new process for handling "obscene"
newsgroups, involving submission of complaints to the Ethics
Committee.  The next day, Provost Kalbfleisch ordered five newsgroups
to be banned on grounds of "obscenity", citing legal advice and a
report from the Ethics Committee.

There are very serious problems with both the process used to set
policy and the announced policy itself.  Here are just a few:

1.  {Newsgroup censorship dramatically limits free expression at
Waterloo}:  99% of the messages on "alt.fan.karla-homolka" newsgroup
did not contain material that could not be legally published in Canada,
but rather, discussions about the wisdom of publication ban in general
and the Kovacs ban in particular.   These discussions are now open only
to the cognoscenti who know how to get around the ban.

Similarly, the ban on allegedly "obscene" newsgroups constitutes prior
restraint on the legally protected legitimate expression which
constitutes the vast majority of the messages posted.  Indeed, one of
the banned groups was explicitly labeled as a "discussion" group.  The
report from the Ethics Committee, used as justification for the
censorship, forthrightly admits that some of the banned newsgroups,
such as "alt.tasteless", "may" contain allegedly obscene material only
"from time to time".  Of course, if the "may from time to time"
standard were applied consistently, nearly every newsgroup could be
banned, not to mention magazines and other periodicals in the library.

2.  {The policy may increase, rather than decrease, legal liability}:
As discussed in Part II of this series, existing case law in North
America suggests that the University has no legal liability for
messages carried on its networks.  However, by selectively removing
some newsgroups and not others, the University leaves itself open to
the charge that it knew about other violations of the law
(copyright infringement, libel, mail fraud, etc.) and did not act
on them.  (Responsibility for copyright violation on University
networks was explicitly disclaimed in a 1991 Advisory Committee report.)

3.  {The policy focuses on the medium, rather than the message}:
President Downey cited the Supreme Court of Canada's _Butler_ decision
as grounds for his censorship order. (_Butler_ was handed down in
February 1992, not 1993 as incorrectly implied by Downey's memo.)

In _Butler_ it is held that material not otherwise obscene "does not
become so by reason of the person to whom it is or may be shown or
exposed nor by reason of the place or manner in which it is shown."
Therefore, the finding of obscenity is independent of the medium in
which the material appears.  Yet the administration has not begun to
review potentially obscene material in the University library.  Some
material contained there, such as Bret Easton Ellis' _American Psycho_,
a tale of sexual torture and murder, might very well be considered
obscene under _Butler_.  It is certainly no tamer than the content of
the banned newsgroups.    Why the inconsistency?

4.  {Censorship recommendations are placed in the wrong hands.} The ban
on "alt.fan.karla-homolka" was issued by executive fiat.  There was no
consultation or discussion.  Furthermore, as a February 7 letter 
from President Downey confirms, there is no provision for appeal.

Under the new newsgroups policy, it is the Ethics Committee who
determines which newsgroups are to be censored for reasons of
obscenity.  But the Ethics Committee currently consists of two faculty
members, two staff members, and two students, *none* of whom are experts
in obscenity law.  Although one member had legal training in Australia,
none are practicing Canadian lawyers.  Is it reasonable for President
Downey to expect them to make such decisions?

There is at the University a group of people familiar with responding to
legal and extra-legal challenges to archival materials:  the University
librarians.  Why were librarians not made part of the process?

5.  {Despite claims to the contrary, the newsgroup censorship is
inconsistent with earlier recommendations of a university committee.}
The Advisory Committee's 1991 report was in reaction to a previous
administration's newsgroup ban, and that report was the basis for the
*reversal* of the earlier ban.  How can such a report now, three years
later, be used by the administration to *justify* renewed censorship?

President Downey's claim that the new policy is "consistent with
recommendations put forward in the 1991 Report of the Advisory
Committee on Network News" simply defies credulity, The principles
enunciated in the Advisory Committee report are, in actuality,
diametrically opposed to the President's new policy.

For example, the Advisory Committee report stated quite unambiguously that the
university should "consult its user community when decisions must be
made about the use of resources which are committed to E-mail and news"
(no such consultation was done);  that "[t]he University of Waterloo
adopt and widely publicize the principle that, in ... posting an
article to a newsgroup, it is the user and not the University who
assumes responsibility for its contents" (the new policy implies it is
the responsibility of the University, not users); that "[t]he
University's primary news-server continue to receive all newsgroups
...  which arrive over the networks to which the University is
connected"; (recent administration decisions have actually banned six

When I asked Associate Provost for Computing Johnny Wong why existing
policy was not followed when "alt.fan.karla-homolka" was banned, he
replied that it was "a legal issue and not a computing issue".  But a
policy that may be sidestepped merely by asserting it is "a legal
issue" is neither wise nor binding, since nearly every action by the
administration could later be justified by this claim.

6.  {The legal advice should be released.}  Legal advice was cited as
the basis for action in all the newsgroup bans--yet it was not made
public.  Contrast this with the situation at McGill University, in
which all legal opinions used to set policy are public documents.  As a
McGill professor told me, "It would be absurd for an administrator to
say `we have to do this, it is the law', but then refuse to elaborate.
This is clearly a recipe for disaster."

Furthermore, if one reads the Ethics Committee memo of January 7, one
discovers that, contrary to the implication of Provost Kalbfleisch's
February 1 memo, the University's lawyer did *not* specifically advise
removal for three of the five banned "obscene" newsgroups.  Is it
any wonder that this Ethics Committee memo has not been widely

7.  {The new policy has unintended consequences.} Under the standard
enunciated by the Ethics Committee, a newsgroup may be banned simply
because it "may from time to time" contain "obscene" material.  This
standard may have the unintended consequence that those who disapprove
of a newsgroup (for example, one that discusses gay issues) might
contrive to have it censored by anonymously posting "obscene" material
two or three times, and then complaining to the Ethics Committee.  Such
sabotage is not hypothetical or farfetched--it has already occurred
on gay bulletin board systems elsewhere.

8.  {The bans place the University in the role of agent of the state as
interpreter and enforcer of murky laws.}  This last objection is the
most important.

No Canadian court has ever ruled that any message on any Usenet
newsgroup is obscene.  No Canadian court has ever ruled that a
university is liable for the passive reception and distribution of
"alt.fan.karla-homolka" over public networks.  Given that, the
University's claimed legal justifications are suspect.

In particular, the law of obscenity is notoriously difficult to
interpret and apply, and has historically been used to stifle
artistic expression and dissent (see, for example, Edward de 
Grazia's recent book).  It is not the University's role to make
determinations about what is and what is not obscene; that job
properly belongs to law enforcement and the courts.  The
University's duty is to guarantee free expression and free inquiry.

The administration's censorship makes the University an agent of the
state, in deciding what can and cannot be read at Waterloo.  It is a
role the administration seems to have no qualms about, but one that
fills me with dismay.