The Child Online Protection Act and the Canadian Charter of Rights and Freedoms:

A Hypothetical Charter Analysis of the 'Son of the CDA'

 

By Corey D. Steinberg

 

Owen Bird; Barristers and Solicitors

expectingrain@globalserve.net

 

 

 

April 6, 2000

 

 

 

 

 

 

 

 

 

 

I.       Background to the Child Online Protection Act

On October 21, 1998, American Congress passed the Child Online Protection Act[1] (COPA).  This federal criminal legislation targets Internet websites that post pornographic material.  The Act makes it a federal crime to use the World Wide Web to communicate "for commercial purposes"[2] material considered "harmful to minors,"[3] with severe pecuniary penalties for each day of violation and up to six months in prison.

            The passage of the Act met with immediate and severe resistance from civil liberties groups, 'cyber-rights' groups, and web publishers.  Various actions were commenced for temporary and permanent relief based on a constitutional challenge to the law. 

The American Civil Liberties Union (ACLU) was the central Plaintiff in these suits.  The ACLU presented testimony to the court from website operators who provide free information for artists, lesbian and gay men, and the disabled, who all feared that the new federal law would force them to shut down their websites.

"In an affidavit filed with the court, Mitchell Tepper of Sexual Health Network said that he feared prosecution under the law because his website provides graphic information on sexual pleasure for people with disabilities or illnesses."[4]

 

            The ACLU's main contention is that the law imposes a burden on speech that is protected for adults.  Essentially, the argument is that, in order for the law to effectively prevent children's access to material on the World Wide Web that is 'harmful to minors,' it must necessarily also impede adults' access to such material as well.  Furthermore, the prevention of anyone's access to such 'harmful' material will undoubtedly also result in the restriction of access to harmless information that is unarguably protected by the American Constitution.

"COPA threatens protected speech with civil and criminal sanctions, and effectively suppresses a large amount of speech that adults have a constitutional right to communicate and receive on the Web."[5]

 

This debate is not unfamiliar to the ACLU and the American courts.  An analogous battle took place several years ago when American Congress attempted to pass the Communications Decency Act[6] in 1996. This legislation was struck down as contrary to the First Amendment[7] of the American Constitution.  This was a result of the landmark decision by the United States Supreme Court in the case of Reno v. ACLU I.[8]

"COPA's ultimate constitutional flaws are identical to the flaws that led the Supreme Court to strike down CDA in ACLU I. While there are slight differences between the two laws, these differences are insignificant when compared to the fundamental and fatal constitutional defect of both laws: "In order to deny minors access to potentially harmful speech," COPA -- like CDA -- "effectively suppresses a large amount of speech that adults have a constitutional right to receive and to address to one another." ACLU I, 521 U.S. at 874; see ACLU II, 31 F. Supp.2d at 495. In passing both CDA and COPA, Congress made it a crime for adults to communicate expression that is clearly protected by the Constitution."[9]

 

The judgment in Reno I held that this legislation was an impermissible restriction on the First Amendment[10] rights of adults. The First Amendment[11] is roughly the American equivalent to Section 2(b)[12] of the Canadian Charter of Rights and Freedoms.[13]  It is the provision in the American constitution, which protects the right to free speech. 

Justice John Paul Stevens wrote the judgment in Reno I for the majority.  The ruling held that speech on the Internet must be afforded the highest level of constitutional protection.  Further, he likened the new medium to leaflets and books, which receive similar protection.  He compared the relative accessibility of chat groups, newsgroups, and the like to leaflets, and the relative inaccessibility of posting on the World Wide Web to book-publication.

He criticised and rejected the government's argument that:

"(E)ven though the CDA effectively censors discourse on many of the Internet's modalities--such as chat groups, newsgroups, and mail exploders--it is nonetheless constitutional because it provides a "reasonable opportunity" for speakers to engage in the restricted speech on the World Wide Web."[14]

 

In order to exemplify his criticism, he went on to characterise the Government's position in terms of a reductio ad absurdum argument:

"The Government's position is equivalent to arguing that a statute could ban leaflets on certain subjects as long as individuals are free to publish books."[15]

 

            In the wake of the demise of the CDA, COPA was an attempt by American Congress to meet the criticisms of the court in Reno I.  COPA has even been pejoratively dubbed "Son of the CDA"[16] or "CDA II".[17]  Elements of the CDA that attracted particular judicial commentary in Reno I were conspicuously not present in COPA; e.g. the CDA was sweeping in its breadth whereas COPA was restricted to 'commercial speech' only.[18]

Proponents of COPA extol its virtues as a constitutional law in contrast

with its unconstitutional predecessor the CDA:

"COPA is carefully limited in scope to deal only with this problem as it exists on the Web and only for commercial sellers of pornography. The technical capability of commercial WWW sites to use credit cards and PIN/codes was recognized by the Supreme Court in last year’s decision in Reno v. ACLU, 117 S. Ct. 2329 (1997). The Act applies only to Web sales sites and excludes other Internet, Usenet, email, BBS, chat, and online services. The Act applies only to commercial sellers of harmful pornography and excludes all non-commercial, non-profit, educational, governmental, and private communications. Finally, this Act adopts the constitutionally valid definition of "harmful to minors" to limit its reach to pornography that is not protected speech for juveniles. COPA’s intentionally narrow focus is a "least restrictive means" to control the availability to minors of harmful pornography on the front pages of the porn syndicate’s Web sites."[19]

 

Presumably American Congress, and other proponents of COPA, believed and anticipated that the courts, even in the face of a constitutional challenge, would uphold this 'modified CDA'.  As mentioned above, such constitutional challenges have been made.  In particular, injunctive relief has been sought and ordered.  Plaintiffs arguing against COPA have argued for a temporary injunction enjoining enforcement or prosecution under COPA.  The argument maintained that such an injunction would be prudent until such time that this contentious law could be tested at a trial on the merits to determine if it would survive First Amendment[20] muster.

            Reed J. of the United States District Court for the District of East Pennsylvania heard arguments for an injunction at two separate proceedings.  At the first action he awarded a ten-day injunction.  Subsequently, at the second proceeding on February 1, 1999, Reed J ordered a continuance of this temporary injunctive relief that continues to the present day.  These judgments made it clear that COPA is far from incontrovertibly constitutional.  The court found that the challengers had delineated a prima facie case that COPA could not necessarily meet a constitutional challenge.

            Reed J ruled:

"… it is not apparent … that the defendant can meet its burden to prove that COPA is the least restrictive means available to achieve the goal of restricting the access of minors to this material."[21]

 

Currently, the matter regarding COPA's constitutionality is still not conclusively settled in the United States.  Moreover, it would appear that the dispute is not progressing toward any rapid resolution.   A Commission on Online Child Protection has been appointed to investigate possible alternatives to COPA.  The object of this commission is to investigate alternatives that would provide the least restrictive means of protecting America's children from harmful material online. 

This commission held its first meeting March 7, 2000.  Reminiscent of a 'comedy of errors,' however, it has been determined that this commission does not fall under the administrative rubric of any existing American governing bodies that could fund such an inquiry.  For this reason the first issue discussed by the commission was to address the question of whether or not it (i.e. the commission itself) could afford to exist in its current economically tenuous state.

"The Child Online Protection Act (COPA) Commission held its first meeting on March 7.  The panel discussed whether it can be effective with no funding and no federal agency to act as its host."[22]

 

II.      Child Online Protection Act and Canada: Could it happen here?

It would appear that no definitive answer to this constitutional question should be expected from the American courts for some time.  At this juncture, however, it is clear that this legislation is of questionable constitutional validity based on the First Amendment[23] to the American constitution. 

            A more interesting question for Canadian constitutional scholars, however, is whether or not such a law could survive a S.2(b)[24] Charter challenge if enacted in Canada.  This section of the Charter is analogous to the First Amendment,[25] but differs in that it extends to "expression" and not merely to "speech".[26]  "(T)he word expression is very broad - broader than "speech"."[27] 

Perhaps even more significant, however, would be a consideration of the Charter provision in S.1[28] as it would be applied to COPA.  This is the sweeping section that does allow Canadian courts to uphold an otherwise unconstitutional limitation on fundamental rights and freedoms under particular circumstances.

It is notable that Canadian jurisprudence has, traditionally, been much less tolerant toward the protection of pornography than have the American courts.  "In short, freedom of expression interests have seldom been accorded much weight by Canadian judges in the context of pornography."[29]  It is for this reason that a detailed Charter analysis of COPA is a particularly worthwhile pursuit.  This paper will attempt to execute just such an analysis.

III.    COPA Held up to a Canadian Charter Analysis

(1)       General Framework

            Any S.2(b)[30] Charter analysis must begin by asking the following three questions derived from Canadian common law:

(i)               What is the benefit of the right or freedom in question?

(ii)            Are the interests at issue protected interests?

(iii)         Could it be said that the governmental action at issue would infringe those interests?

 

This framework for a Charter analysis pertaining to a limit on freedom of expression is delineated in the Supreme Court of Canada decision in Irwin Toy v. Quebec.[31]  This 1989 decision was not in respect of the censorship of pornography.  Rather, this case focused on legislation that restricted advertisements to Canadian children.  This general framework, however, applies to any S.2(b)[32] challenge.

(2)       COPA v. S.2(b)

In Irwin Toy, two challenges were made to the law in question.  The first challenge was made on federalist grounds.  This challenge was unsuccessful, and is, in any case, irrelevant to the current analysis.  The second challenge (which also failed on the facts) however, was a Charter challenge based on S.7[33] and S.2(b).[34]  It is from this challenge, particularly in respect of S.2(b),[35] that a schema has been developed for any Charter analysis of government action limiting Canadian freedom of expression.

(i)               What is the benefit of the right or freedom in question?

In terms of the first question outlined above the courts have responded by outlining three possible 'rationales' for, or 'benefits' that may be derived from, an expressive activity:

(a)     The activity helps the individual to discover truth as it contributes to the general 'marketplace of ideas''[36][37]

(b)     The activity contributes to the advancement of democracy[38]

(c)     The activity helps to actualize individual self-realization[39]

 

These possible benefits of any particular expressive activity are neatly outlined by Dickson C.J. writing on behalf of the majority of the Supreme Court of Canada in Keegstra[40]:

"(T)he court has attempted to articulate more precisely some of the  convictions fueling the freedom of expression, these being  summarized in Irwin Toy (at p. 612) as follows: (1) seeking  and attaining truth is an inherently good activity; (2)  participation in social and political decision-making is to  be fostered and encouraged, and (3) diversity in forms of  individual self-fulfillment and human flourishing ought to  be cultivated in a tolerant and welcoming environment for  the sake of both those who convey a meaning and those to whom meaning is conveyed."[41]

 

The benefits of freedom of expression, therefore, are characterized broadly.  It is important to remain aware of these benefits for consideration at later stages of our Charter analysis.  Having taken account of these, then, the analysis must move on to the next question articulated in Irwin Toy.

(ii)            Are the interests at issue protected interests?

This question is the first analytical stage of our framework.  This question is essentially asking: 'Is the act or speech at issue, in fact, 'expression' for the purpose of a Charter analysis?'  The interpretation at common law of this question has focused on the purpose of the expression.  The courts have found that if the purpose of the expression is to bestow any of the benefits (outlined in the section above) on an individual, then that expression is protected under the Charter.  This standard, however, is qualified by the fact that 'violent' expression does not receive Charter protection.

Such a standard is relatively simple to meet.  Any expressive activity is bound to effect the third, if not the first two of the benefits extolled in the section above.  If the expression does not serve the purpose of furthering the pursuit of 'truth' or the actualization of 'democracy', it is bound to contribute to the 'self-realization' of the individual engaging in the expression.  Moreover, this standard is at least as likely to encompass the act of publishing pornography as any expressive activity.

"(I)f a purposive approach is taken to the definition of "freedom of expression" in S.2(b), pornography should receive the protection that that provision affords.  In other words, in all cases in which a challenge is brought to legislation directed at pornography, it is appropriate to conclude that "freedom of expression" is at issue."[42]

 

In order to determine if speech or a particular expressive act qualifies under this loose standard, a framework consisting of a two-part analysis has been developed at common law.  This framework is outlined in the majority judgment in Irwin Toy.  It amounts to asking two questions of the act or speech seeking Charter protection:

(a)     Does this speech or act contemplate an attempt to convey meaning?

(b)     Is this speech or act violent in form?

 

(a)       Does this speech or act contemplate an attempt to convey meaning?

 

The Majority in Irwin Toy, comprised of Dickson C.J. and Lamer and Wilson J.J. delineated the first question of this two-part schema:

"(I)f the activity conveys or attempts to convey a meaning, it has expressive content and prima facie falls within the scope of the guarantee … It might be difficult to characterise certain day-to-day tasks, like parking a car, as having expressive content.  To bring such activity within the protected sphere, the plaintiff would have to show that it was performed to convey a meaning."[43] 

 

            It is apparent that this is not a difficult standard to meet.  Based on the above description of the standard, it appears that most activities could qualify as "expression".  In Irwin Toy the court found that the activity in question, i.e. advertising to children, clearly represented an attempt to convey meaning.  The message conveyed was, 'Buy this product.'  As an attempt to convey meaning, this form of expression cleared the first hurdle of the analysis. 

(b)       Is this speech or act violent in form?

 

            The principle espoused by the second question in this two-part schema is found concisely articulated in the Supreme Court decision in Dolphin Delivery.[44]  This case, like Irwin Toy, was not in respect of pornography.  Rather, the expressive activity at issue was civil disobedience in the form of picketing.

            McIntyre J. wrote on behalf of the majority of the court:

"All picketing involves some form of expression and enjoys Charter protection unless some action on the part of the picketers alters its nature and removes it from Charter protection. Charter protection of this freedom does not encompass violence, threats of violence or other unlawful acts."[45]

 

            McIntyre J's statement would logically lead one to believe that expressive activity is not protected under S.2(b)[46] if that activity is 'violent' in its form, or 'threatens violence'.  This standard, however, was modified in its later interpretation in Keegstra, wherein the supremes held that 'threats of violence' are not exempted from Charter protection.  Rather, only actual acts of violence are exempted.[47]  This modified interpretation was based on a distinction between the 'form' and the 'content' of an expressive act.

            Dickson C.J. wrote for the majority in Keegstra:

"While the line between form and content is not always easily drawn, in my opinion threats of violence can only be so classified by reference to the content of their meaning.  As such, they do not fall within the exception spoken of in Irwin Toy, and their suppression must be justified under s.1."[48]

 

            This distinction between an expression's 'form' and its 'content' is an important innovation in Canadian Charter jurisprudence.  Dickson C.J. in Keegstra also clarifies this tenet:

"When an activity conveys or attempts to convey a meaning, through a non-violent form of expression, it has expressive content and thus falls within the scope of the word "expression" as found in the guarantee.  The type of meaning conveyed is irrelevant.  Section 2(b) protects all content of expression."[49] 

 

            In the same judgment Dickson C.J. also dispels any persisting questions that might still arise regarding the form-content distinction as it applies to the violence exemption under S.2(b).[50]  He explains that, "This exception refers only to expression communicated directly through physical harm."[51]  'Violence', therefore, refers only to 'physical' violence.  Any form of expression falling short of physical violence, thus, necessarily falls short of this exemption from Charter protection.

Based on this two-stage analysis outlined in the cases discussed above, one can conclude that the act of posting pornography on the World Wide Web for commercial purposes falls under the protection of S.2(b).[52]  The act of posting such material clearly meets both criteria of such an analysis.

Regarding the first question of this analysis, such an act certainly attempts to convey meaning.  The message conveyed could be, as in Irwin Toy, 'Buy this product' (i.e. 'buy this or other pornographic material').  Alternatively, the message could simply be, 'The image depicted here is sexually arousing'.  In any event, the posting of such material certainly contemplates a conveyance of meaning. 

Regarding the second part of this schema, the act of posting such material could not be deemed at law a 'violent' act.  Although some feminist legal scholars might argue that this act of publication inherently does violence to the equality-rights of women or children,[53] this contention is denied any legal force when compared with the clear statement of Dickson C.J. in Keegstra.  That judgment unequivocally states that the act itself must be an "expression communicated directly through physical harm".[54]  Regardless of any sociological harm that such an act may bring to the feminist-legal movement for equality rights, it certainly falls short of communicating "directly through physical harm".[55]

            The judgment in Irwin Toy and a survey of related case law provides evidence, therefore, that the Supreme Court of Canada is predisposed toward recognising Charter protection for almost any activity.  The standard implemented merely requires that the act constitute some attempt at conveying some modicum of meaning without expressing itself through an act of physical violence.  The very fact that the act of "parking a car" could, theoretically and on particular facts, fall within this rubric indicates the court's willingness to err in favour of protecting the expressive activity.[56]    

            It is reasonable, therefore, to presume that the act of posting pornographic material to the World Wide Web passes the second stage of our Charter analysis.  It 'conveys meaning' and is 'non-violent' in its form.  Our analysis must, then, move on to the third question outlined above.

(iii)         Could it be said that the governmental action at issue would infringe those interests?

 

            This third question, is the second analytical stage of our framework.  In order to address this third question, the Court examines the alleged infringement on free expression to determine whether it is one of two alternative types.  The first possibility is a (a) purpose-based infringement.  The second possibility is an  (b) effects-based limitation on expression.[57]  The division of infringements into these two classes is found in the Supreme Court of Canada's decision in Big M Drug Mart.[58] 

(a)  Purpose-Based Infringements

            Big M Drug Mart, like Irwin Toy and Keegstra, did not deal with the publication of pornography.  In fact the case did not even litigate a S.2(b)[59] infringement.  Rather, Big M Drug Mart challenged The Lords Day Act,[60] a statute barring stores from opening on Sundays, based on S.2(a).[61]  This is the section of the Charter protecting 'freedom of religion'.

            Dickson C.J. wrote on behalf of the majority in Big M Drug Mart:

"The initial test of constitutionality must be whether or not the legislation's purpose is valid; the legislation's effects need only be considered when the law under review has passed the purpose test.  The effects test can never be relied on to save legislation with an invalid purpose."[62]

 

            Dickson C.J.'s comment on purpose-based infringements is actually a restatement of this principle which he initially made in the decision of Hunter v. Southam Inc.[63]  That case, as well, did not deal with a challenge based on 'freedom of expression.'  In that case a Charter challenge was made to a search of an individual's premises.  The challenge alleged an infringement of S.8;[64] the section of the Charter pertaining to a 'reasonable search and seizure'.

            Dickson J. (as he then was) wrote on behalf of the majority:

"Since the proper approach to the interpretation of the Charter of Rights and Freedoms is a purposive one, before it is possible to assess the reasonableness or unreasonableness of the impact of a search or of a statute authorising a search, it is first necessary to specify the purpose underlying s. 8: in other words, to delineate the nature of the interests it is meant to protect." [65]

 

            In the context of freedom of expression, a purpose-based infringement refers to any governmental action that can be said to restrict free expression in a manner directed at the content of that expression.  Essentially, if the governmental action restricts an individual's right to free expression based on the meaning conveyed by that expression, then it amounts to a purpose-based infringement.

(b) Effects-Based Infringements

An effects-based infringement focuses on the consequences of the expression.  In particular, as outlined in Irwin Toy and Keegstra, consequences that deserve, at law, such an effects-based limitation are those that manifest themselves in terms of a message conveyed "directly through physical harm".[66]

An analysis of an infringement grounded in the effect of the expression is limited to one possibility of two possible limits on freedom of expression.  This dimension of the limit on free expression constitutes an infringement of one's freedom to express oneself.  This type of infringement refers to an act of government that prevents an individual from expressing herself in a non-violent manner as she chooses.  This is in contrast to an infringement of ones freedom from expressing oneself.  Such an infringement occurs when an act of government compels an individual to express herself in a particular way.[67] 

            In order to prove an effects-based infringement of an individual's freedom to express herself, the challenger must show that the expression prohibited furthers one of the three rationales, i.e. 'benefits', listed earlier in this paper.  This means that the governmental act must interfere with (a) an individual's exploration for 'truth' through an examination of the general 'marketplace of ideas'; or, it must impede (b) the 'advancement of democracy'; or, it must interfere with (c) one's pursuit of 'self-realization'.[68]

            The majority in Irwin Toy explained this principle as follows:

"In showing that the effect of the government's action was to restrict her free expression, a plaintiff must demonstrate that her activity promotes at least one of these principles.  It is not enough that shouting, for example, has an expressive element.  If the plaintiff challenges the effect of government action to control noise, presuming that action to have a purpose neutral as to expression, she must show that her aim was to convey a meaning reflective of the principles underlying freedom of expression.  The precise and complete articulation of what kinds of activity promote these principles is, of course, a matter for judicial appreciation to be developed on a case by case basis.  But the plaintiff must at least identify the meaning being conveyed and how it relates to the pursuit of truth, participation in the community, or individual self-fulfilment and human flourishing."[69]

 

We must now return to the question above:

"Could it be said that the governmental action at issue would infringe those interests?"         

We must attempt to answer this question as it applies to COPA.  In order to do this, as we have learned from the relevant case law, we must first decide whether COPA is a purpose-based infringement on free expression or an effects-based infringement.  If this limitation on expression is purpose-based and that purpose is contrary to the principles of the Charter, then it will be unnecessary to determine the effects of the infringement.

This determination is not always easily discerned.  Dickson C.J. makes this point in Big M Drug Mart:

"Purpose and effect respectively, in the sense of the legislation's object and its ultimate impact, are clearly linked, if not indivisible. Intended and actual effects have often been looked to for guidance in assessing the legislation's object and thus, its validity."[70]

 

COPA provides an example of just such a situation.  It is not obvious whether or not the infringement in this case is purpose-based or effects-based.  In COPA the two are inextricably intertwined.

American Congress has always maintained that its object in enacting COPA was the protection of society's most vulnerable constituents.  Congress claimed that COPA was designed and enacted to prevent harm to children.

This objective was submitted by affidavit to a Pennsylvania court in defense to a motion for an injunction enjoining enforcement or prosecution under COPA:

"Defendant argues that COPA passes constitutional muster because it is narrowly tailored to the government's compelling interest in protecting minors from harmful materials."[71]

 

 The government's contention has always been that COPA was enacted with the interests of children in mind.  It would be prudent, however, to consider whether there is any other evidence of this objective.

Textually speaking, the title of this legislation is clearly intended to imply that the goal in enacting COPA was to meet the object of protecting children.  More significantly, the stipulations in the act also indicate that this was the government's intended purpose.  This is apparent in the first subsection of the act, which describes the breadth of the prohibition imposed:

"Prohibited Conduct.-Whoever knowingly and with knowledge of the character of the material, in interstate or foreign commerce by means of the World Wide Web, makes any communication for commercial purposes that is available to any minor and that includes any material that is harmful to minors shall be fined not more than $50,000, imprisoned not more than 6 months, or both."[72]

 

The American government's commitment to the protection of children from harm appears to resound clearly in the above excerpt from COPA.  The only prohibition listed explicitly mentions that the material most both "(be) available to any minor" and include "material that is harmful to minors". 

Presumably, material that is unavailable to minors but is harmful to minors would be permitted.  Likewise, material that is available to minors, but is not harmful to minors also would be permitted. Thus, it logically follows that the most likely purpose that this legislation could have been implemented to serve, is the protection of children.

One could argue that the protection of children was not the intended purpose.  One could suggest that the purpose of enacting the legislation was purely malicious in that it was to decrease the economic returns to those who post pornography on the World Wide Web as a business enterprise.  This could be maintained based on the inclusion of the stipulation that the communication must be for "commercial purposes".  This means that material posted that is "harmful to minors" but that is made available free of charge would not be prohibited. 

Logically, this could be an intended purpose of the legislation, but is not likely the main intended purpose.  The reason is that the stipulation included is narrowly focused in that the material prohibited must be "harmful to minors".  For this reason, even if the legislation does happen to serve the purpose of decreasing the profits that are returned to commercial pornographers, it only does so through a prohibition of material that is "harmful to minors" thereby also serving the goal of protecting children. 

The act could have prohibited material that is 'harmful to women and minors'[73] or 'harmful to society as a whole'. This would likely decrease the profits of professional pornographers to an even greater degree.  The act, however, is limited to material that is "harmful to minors". 

Giving American Congress the benefit of the doubt, it is likely that the criterion that the material be posted for a "commercial purpose" was only included as a pre-emptive concession to an anticipated constitutional challenge.  The government likely expected such a challenge based on 'overbreadth'[74] if COPA's application could encompass, for instance, non-profit charitable organisations. On a balance of probabilities, therefore, one could declare with some confidence that the act was enacted with the purpose of protecting children.

It is now more readily possible to characterise this limitation on free expression as purpose-based or effects-based.  The fact that the prohibition is aimed at a particular type of content, which is content that is "harmful to minors", leads one to believe that this is a purpose-based restriction on free expression.  After all, it is the message conveyed by the content that is being targeted by the legislation.  If the message is, 'Buy this or other pornography' or 'This material is sexually arousing' and that message is deemed "harmful to minors" and must therefore by prohibited, then it is the content of the message conveyed that is being abridged by this law. 

It must be noted, however, that COPA is not prohibiting material that is "harmful to minors" simply as a convenient means of defining the criterion by which one may determine if the material is prohibited.  This particular characteristic of the material is specifically included in the act as it represents the effect that this material has on children.  The effect that material "harmful to minors" evokes if made available to minors is, a priori, the effect of harming children. 

One might conclude, therefore, that it is the effect of the expression that caused American Congress to enact COPA and not the purpose of such expression.  It must, therefore, be an effects-based limitation.  The answer is not so simple, however.  Recall the criterion outlined in Keegstra:

"When an activity conveys or attempts to convey a meaning, through a non-violent form of expression, it has expressive content and thus falls within the scope of the word "expression" as found in the guarantee.  The type of meaning conveyed is irrelevant.  Section 2(b) protects all content of expression."[75] 

 

Is posting on the World Wide Web a "non-violent form of expression" or is it violent in form?  

One could argue that posting material on the World Wide Web might do indirect harm to individuals by creating an unfavourable societal environment for certain people.  This act might also encourage individuals to go out and commit physical acts of harm against others. 

Some academics, such as Catharine MacKinnon hold the view that the mere act of exposing pornography to others is a directly harmful act in and of itself.  She derisively makes this statement in sarcasm:

"Pornography in the marketplace of life where there are no equality laws - in the world of books, photographs, films, videos, phone-sex, and cable-television - has fallen into a reality-warp.  Harmless Fantasy it is called."[76]

 

A prevention of such free expression, however, would not be based on the direct effect of the expression as it is manifested in physical harm.  It would not meet the standard found in Keegstra that the act itself must be an "expression communicated directly through physical harm".[77]  Such an act of expression is communicated directly in a non-violent form and would only cause indirect physical harm.

Bettina Quistgaard comments:

"The second view of the harm that results from pornography focuses largely on men and their responses to pornographic messages.  This approach moves away from the individualistic conception of harm and considers men as a group who cause harm, and women as a group who are the victims of such harm.  It suggests that men "learn" from pornography; they learn that violence and degradation are acceptable and that women desire to be treated in this way.  In turn, men are seen to act on what they have learned to be true.  Those who take this view are also looking for an identifiable and objectively knowable causal link between pornography and harm to women, but here the causal link is between pornography and men's attitudes which, in turn, are acted on.  Attitudinal change becomes the crucial link between the message and the action."[78]

 

                Upon close and careful analysis, therefore, one must conclude that the goal behind enacting COPA, that of protecting children, actually constitutes a purpose-based infringement on free expression.  The act of posting pornography to the World Wide Web is not a directly physically harmful act and thus falls short of the standard espoused in Keegstra for an effects-based infringement. 

Some, like Catharine MacKinnon, would hold otherwise.  Upon close examination, however, such as the insight provided by Bettina Quistgaard, it grows obvious that the publication of pornography could never qualify as anything more than an 'indirect cause of harm'.  As such, a limit on this form of expression could not be targeting the 'directly harmful effect' of such expression, as the only harm caused could only be indirect.  Such a limit on posting material to the World Wide Web, therefore, must, necessarily be characterised as purpose-based, i.e. targeting the content of the expression itself.

As COPA can be characterised as a purpose-based infringement of S.2(b),[79] it is thereby unnecessary to examine its effects on free expression.  By way of brief comment, however, it is fairly simple to sum up these effects.  First we must decide if COPA is a limit on one's freedom to express oneself or one's freedom from expressing oneself.  This is a trite matter in this case.  COPA prevents one from expressing oneself freely, and does not compel one to express oneself in any particular way.  It is definitive, therefore, that this is an infringement on one's freedom to express oneself.

As a limit on one's freedom to express oneself, we must consider COPA's implications for the three rationales for free expression discussed earlier. 

Regarding the first rationale, the 'marketplace of ideas' rationale, this could possibly be infringed.  A child seeking knowledge of the truth about 'deviant' sexuality, for instance, could be barred from obtaining this information as a result of COPA.

Regarding the second rationale, the 'advancement of democracy', it is difficult to see how the prevention of material 'harmful to minors' from being posted on the World Wide Web could impede this benefit of free expression.  In all likelihood, this benefit would remain intact in Canada, even if COPA were implemented in this country. 

The third rationale, however, would certainly be infringed.  This rationale of free expression is that it benefits either the expresser or the recipient of the expressed material by contributing to that individual's self-realisation.  The open-ended subjectivity of this rationale, it would seem, makes it likely that the infringed form of expression here (and any infringement on free expression) would impede the actualisation of this benefit.  Those who seek to view such material, children or otherwise, likely perceive the 'enjoyment' of such material as contributing to full self-realisation.  The same can be said of those who make such material available.

It appears, thus, that certainly one and likely two of the three rationales for free expression would be infringed through an enactment of COPA in Canada.  It is also clear that, regardless of its effects, COPA's purpose of limiting non-violent free expression based on the content of that expression is invalid under S.2(b) of the Charter.  It is clear, therefore, that an enactment of COPA would limit a constitutionally protected right in Canada and on that basis should be struck down.  Now that this is clear, the next step of our analysis is to determine if this limitation on a Charter right would be justified and thereby upheld under S.1.[80]

III.    Section 1 Analysis: "Oakes[81] Framework"

(1)       Rational Connection Test

As the purpose of enacting COPA was to target the content of the expression being limited, it is necessary to decide if this is a valid purpose at law.  The purpose of this act, as has been discussed, is the protection of children.  Canadian common law has indicated that this goal can be a "sufficiently important" limit on free expression in Canada.  The courts have recognized that children are among the most vulnerable members of society.  They are the most susceptible to external attitudinal influences, as well as vulnerable to acts of adults with whom they have contact.

            This ground for limiting free expression was endorsed in the 1999 British Columbia Court of Appeals decision in R. v. Sharpe[82].  In that judgment, McEachern C.J. stated emphatically:

"The first step is to consider whether the objective to be served by the legislation in question is sufficiently important to warrant overriding a constitutionally protected right. I have no doubt that the protection of children in the context of this case is sufficiently important for such a purpose."[83]

 

            The "context" of Sharpe was a constitutional challenge to the Criminal Code of Canada provision prohibiting the possession of child pornography.  The case did not focus on the harm caused to children via the production of such material.  This fact was examined in the case and it was accepted that the utilization of minors in the sex industry was already prohibited in other sections of Canadian criminal law thus this matter was extraneous to the case. 

The narrow issue with which the court was concerned was the possession of such material.  As it was held that the potential harm brought to bear on children through the possession of a particular type of pornographic material warranted an abridgement of S.2(b)[84] as a valid purpose, it is reasonable to presume that the courts would rule similarly in the context of COPA.  It must be at least as important to prevent the proliferation of certain types of material based on a concern for the protection of Canadian children.

(2)       Importance of the Impugned Governmental Action

            The next stage of this S.1[85] analysis of COPA must examine how to characterise the role of the government in enacting this legislation.  The two possibilities found in Canadian common law are the state characterised as 'an agent that is merely attempting to mediate between competing claims' and the state characterised as a 'singular antagonist of the individual'. 

            This depiction of the state's role is extremely important.  Depending on this characterisation is the standard by which the Court will or will not choose to defer to the government's judgment.  If the state is portrayed as a 'mediator' then a reasonable amount of deference is given to the government's decision, on the facts.  If the state is painted an 'antagonist of the individual', however, deference is deemed inappropriate.

            The court makes this test explicit in Irwin Toy:

"Where the government is best characterized as the singular antagonist of the individual whose right has been infringed, the courts can assess with a high degree of certainty whether the least intrusive means have been chosen to achieve the government's objective.  On the other hand, where the government is best characterized as mediating between the claims of competing individuals and groups … This Court will not, in the name of minimal impairment, take a restrictive approach … and require legislatures to choose the least ambitious means to protect vulnerable groups."[86]  

 

            In order to characterise the state as a mediator; two factors must be present. First, there must be a vulnerable group within society whose interests are being advanced by the impugned governmental legislation.  If the court can prove that this legislation protects a particular vulnerable group then the degree of deference the court will adopt increases accordingly.

Dickson C.J. explains this tenet in the context of freedom of religion in Edwards Books v. The Queen:

"(A) legislative attempt to avoid economic coercion of one religious group may result in economic coercion of another religious group. How is a court able to second-guess the Legislature on such issues?" [87]

 

Second, if the court is satisfied that the balance to be struck between competing interests requires the government to rely on some type of social science data then the degree of deference shown increases still further. 

This is explained in the majority judgment in Irwin Toy:

"(W)here the government is best characterized as mediating between the claims of competing individuals and groups, the choice of means, like the choice of ends, frequently will require an assessment of conflicting scientific evidence and differing justified demands on scarce resources which cannot be evaluated by the courts with the same degree of certainty."[88]

 

S.1[89] arguments are often won by this stage of the analysis.  Essentially three criteria have been stipulated thus far:

(a)     The state is characterised as either a mediator or as an antagonist

(b)     Is there a vulnerable group whose interests are being protected through an infringement of a Charter right?

(c)     Have there been social science data relied upon by the government in making its decision to limit a Charter right?

 

If the government is able to prove that it was:

(a)               acting as a mediator between competing interests; and

(b)               acting in a manner that will protect the interests of a vulnerable group within society; and

(c)                relying on social science data in deciding to act

 

then the government's judgment is shown a fairly high degree of deference by the court.  The degree of deference shown by the court decreases if there are only two of these criteria met, and diminishes further if only one criterion is met.  Little, if any, deference will be shown by the court if none of these criteria are met.

            If, however, the challenger to the government action is able to prove that:

(a)    the state was acting as a singular antagonist to individual rights; and

(b)   the state was not protecting any vulnerable group through its actions; and

(c)    no social science data have been relied upon

 

then the challenger will almost certainly prevail.  If all three of these criteria are met, the odds are nearly certain that the impugned legislation will be struck down.

How would COPA stand up to these three criteria?

            It is reasonable to presume that the government could convince a court that it was acting as a mediator between opposing interests.  The interests competing would be the interest of civil liberty versus the interest of protecting children.  Even if it could not be proven that COPA would, in fact, protect children the gravity of this concern would likely convince the Court to accept that it was prudent and rational for the government to have attempted to protect this interest in this way.

            It is also apparent that the Court would accept that the government action does protect a vulnerable group in society.  Obviously, that group consists of all children in Canada.  Not only does a vulnerable group exist, therefore, but, the satisfaction of this criterion is only strengthened by the fact that the group is comprised of a very large number of very vulnerable individuals.   

            In order to decide whether or not social science data were relied upon would be a matter of fact, if this legislation were ever actually enacted in Canada.  We can, however, make an educated guess on whether or not such evidence would have likely been offered and accepted.  We can base our educated guess on history.

            In the precedent setting obscenity-case, Butler,[90] the Court was offered large amounts of evidence by concerned parties which indicated that there is no scientific reason to believe that men will act out in any physically harmful way based on exposure to pornography.  The court was also offered (less compelling) evidence, based on the socio-philosophical work of Andrea Dworkin and Catherine MacKinnon, that exposure to such material will, in fact, influence men to cause harm.  Interestingly enough, the court chose to accept the tenuous submissions based on Dworkin and MacKinnon's social theory.

David A. Crerar criticises this judgment:

The first fatal flaw in Butler is the lack of a scientific connection between pornography and the harms Sopinka J. hypothesizes. MacKinnon cites the work of Dr. Edward Donnerstein to support the harm connection, and indeed, the Butler defence team called him as an expert witness at the trial. Yet Donnerstein himself admits that his research is much more tentative than its citors would indicate: "We can show a causal link between exposure to porn and effects on attitudes; but no one can show a causal link between exposure to porn and effects on behaviour." The bulk of studies draw the exact opposite conclusion.  Indeed, in countries where government restrictions on pornography are most relaxed, as in Denmark and Sweden, the incidence of violence towards women is lowest. Some studies offer the theory that pornography curbs crime through the draining of otherwise dangerous sexual impulses.[91]

 

            This acceptance by Sopinka J. of relatively tenuous social data evidence over more sound scientific evidence seems to betray a bias of the court.  It would appear that the court is predisposed toward accepting evidence that will support a limit on free expression, which could protect children over evidence to the contrary.  It is apparent that the court would prefer to risk unnecessarily limiting free expression rather than risk harm befalling children.

            The fact that sociological evidence was offered into evidence and considered by the Court in Butler, indicates that such data do make an appearance when they are relevant to the development of Canadian freedom of expression rights.  Undoubtedly, submissions would be made by Dworkin, MacKinnon, and other concerned socio-philosophical theorists to be considered by the government during its deliberations on the decision to enact COPA. 

Furthermore, the government would be bound to publicly rely on such data, especially that of the nature of Dworkin's and MacKinnon's work.  Reliance on such data would be a politically prudent decision, if COPA actually were enacted.  Such data would give objective justification to the law in the eyes of the public.  Even if this data were not actually decisive, or even influential, in the actual decision to enact COPA, it would be a foolish government that did not at least use this data to justify its decision after the fact.

In order to recap, therefore, this is how the three criteria for a standard of deference would be satisfied in the event that a Canadian court litigated the constitutionality of COPA.  The government would maintain that:

(a)     it was acting as a mediator between the interests of free expression and protecting children; and

(b)    a vulnerable group in society, i.e. children, is protected by COPA; and

(c)     sociological data were relied upon in deciding to enact this law (which, this paper suggests, would be maintained by the government whether or not it was actually true)

                      

The challenger to COPA would be losing significantly at this point in the Charter analysis.  It would be nearly impossible to convince the court that any of the above criteria weighed in the challenger's favour.  The best a challenger could hope for would be that the government simply concedes that no sociological data were considered in its decision to enact this law.  In this way, the challenger could at least have one of these criteria in its favour.  This, however, as stated above, would be a very unlikely turn of events and would still only gain a marginal amount of deference from the Court.

(3)       Rational and Logical Connection Test

The next stage in our S.1[92] analysis requires us to determine if the ends pursued by the government action are proportionate and logically connected to the means of achieving them.  This test comes out of the Supreme Court of Canada decision in R.J.R. MacDonald v. Canada,[93] as well as a judgment by Dickson C.J. in Keegstra, and Sopinka J. in Butler.

Stevenson J. wrote in RJR MacDonald:

The first step in the proportionality analysis requires the government to demonstrate that the legislative means chosen under the Act are rationally connected to the objective.[94]

 

            The test for the proportionality of the means used by the government to achieve a particular end is that the means and ends must be logically connected.  This is a development from the original test outlined in Oakes wherein the standard was one of 'tightness of fit'.  The modified standard as outlined in R.J.R. MacDonald is considerably easier for the government to meet.

            This modified test is particularly conspicuous in Butler.  As has already been discussed briefly, the Court in Butler accepted evidence that the consumption of pornography will lead the consumers to act out in a physically harmful way. This evidence was accepted in spite of stronger evidence to the contrary. 

The court simply deduced for itself whether or not it was logical that preventing individuals from consuming certain types of pornography would be a means toward effecting the end of protecting women from harm.  The court deduced that it is, at least, rational to believe that the consumption of pornography could lead men to act out their violent fantasies to the physical detriment of women.  No hard evidence was necessary for this judicial finding.  All that is required by this 'logical connection standard' is a rational basis for believing in the finding.

Furthermore, the government's position is bolstered even more if it can show that the infringement it has made on one right is actually to protect a different right, thereby giving the ends a greater sense of importance.  In Butler, a law restricting the consumption of "obscene" material was upheld.  It satisfied the logical connection standard, but it also satisfied the standard of characterising the ends of the law as of the utmost importance.  The ends were defined in terms of protecting women's right to equality.

            How would COPA fair in regards to this rational and logical connection test?

            It would appear that COPA would likely pass this test as well.  It certainly seems logical that the consumption of material that is "harmful to minors" could lead to children being harmed.  This statement, in fact, is tautological.  Due to the wording of COPA, which describes the prohibited material as "harmful", no other conclusion could be reached than that the prohibited material will cause harm.

            This reasoning might seem somewhat suspect, but remember that the logical connection standard does not require any scientific facts.  All that is necessary is a logical connection between the means, i.e. the prohibition of posting material "harmful to minors' on the World Wide Web "for commercial purposes", and the ends, i.e. the protection of children from harm.  It is certainly logical that the prevention of children from being exposed to harmful material will prevent harm to children.

            Could this position be bolstered by characterising COPA as protecting other rights?

            COPA could be characterised as a reasonable limit on free expression, because it seeks to protect children's rights recognised both domestically and internationally.  This is the child's right to personal security.  This right is recognised in S.7[95] of the Charter in terms of "life, liberty, and security of the person".  This right is also recognised in the United Nations' Convention on the Rights of the Child[96].

            Article 19(1) of that international-rights instrument holds:

States Parties shall take all appropriate legislative, administrative, social and educational measures to protect the child from all forms of physical or mental violence, injury or abuse, neglect or negligent treatment, maltreatment or exploitation, including sexual abuse, while in the care of parent(s), legal guardian(s) or any other person who has the care of the child. [97]

 

            COPA could, therefore, have a very strong case in terms of the logical connection and relative importance criteria.  The challenger to COPA, however, could have a reasonable objection to the characterisation of the means of COPA being logically connected to its ends.

(4)       Technological Limitations and COPA

            In order for COPA to have any effect at all, there must be some means of preventing "harmful material" from reaching children online.  COPA does suggest certain means for doing so, which are valid defences to prosecution under the act.

S.231(c) lists possible due diligence defences to a charge under COPA:

DEFENSE.-It is an affirmative defense to prosecution under this section that the defendant, in good faith, has restricted access by minors to material that is harmful to minors-

(A) by requiring use of a credit card, debit account, adult access code, or adult personal identification number;

(B) by accepting a digital certificate that verifies age; or

(C) by any other reasonable measures that are feasible under available technology.[98]

 

COPA suggests such solutions as 'credit card verification processes' but this solution seems superficial at best.  Such a system would require a World Wide Web user to input personal information including a credit card number, before gaining access to pornographic material.  Through this means, the website operator could cross-reference information and determine the age of the credit-card holder.  The fatal flaw in this system,[99] however, is that any child with the ambition to sneak a look at an adult's credit card would be able to circumvent this type of 'security measure'. 

            Perhaps the second suggestion will be more effective.  This method requires a 'digital certificate verifying age'.  The only problem is that the World Wide Web touches every jurisdiction on the planet.  Who is to be indemnified with the authority to issue such digital certificates?  Furthermore, how should they be issued to a computer user?  Should they be issued based on information derived from a 'credit card verification process'? (I refer you to the paragraph immediately above!)

            So, the first two possible means of effecting the goals of COPA do not seem to be very effective.  Perhaps the third alternative will be more promising.  This open-ended defence allows 'any other means feasible under existing technology'.  In order to determine what this could mean, one must consider what technological means might accomplish such a feat, and whether they would be effective for the purposes of COPA. 

(5)        Limitations of Filtration Software

As it stands at present, there are two possible technological options for blocking content from reaching a particular computer or network from the World Wide Web.  These are known as 'filtration systems' and are subdivided into two possible methods.  These methods are (a) database filtration systems and (b) embedded filtration systems.  These would either be installed 'upstream' from one's computer, at the level of or before the Internet service provider, or 'downstream' at the level of the individual user (or user's parents as the case may be).[100]   

(a)   Database Filtration Systems and Their Limitations

Database systems of filters associate a particular rating with the content comprising a particular site on the Internet.  When a user requests a particular piece of information, the filtration software contacts the database.  The database then sends back the corresponding rating.  Sites rated at a level deemed 'inappropriate' by the filtration program are blocked and will be unavailable to the user. 

            Such a system, however, is limited to the point that one could say it really does not work, or, at best, that it works 'approximately'.  The inherent problem with such software is that such a system is static.  The Internet, on the other hand, is the most dynamic medium society has yet developed.  The idea of applying a static system to regulate a dynamic medium is inherently flawed and doomed to failure over the long term. 

            The content of the Internet has increased and evolved over the years and this evolution is bound to continue.  A regulation schema that is not suited to change will eventually fall short.  It is not possible for producers of database filtration software to anticipate any and all means of posting 'indecent' material.

            Carlin Meyer explains the shortcomings of these systems:

"(S)ophisticated hackers are able to hide obscene images and text within other seemingly innocent postings, and decoding techniques can be rapidly circulated among members of the computer underground."[101]

 

            Even if it were possible for database filtration programs to decode such 'encrypted'[102] messages, it would be impossible to implement such a system of watchwords in any practical way.  These filters rely on dictionaries of words and phrases in order to block particular material.  For this reason it would be impossible to adapt such a system such that it would be neither over-inclusive nor under-inclusive.  (The over-inclusive aspect of such systems will be discussed below in the section on "minimal impairment".)

Carlin Meyer expresses his feelings on the utter impossibility of practically implementing such a system:

"No matter how inclusive such a dictionary is, it cannot possibly embrace the enormous variety of words used to create obscene stories or to describe obscene images."[103]

 

            Database filtration systems would not be likely to block, for instance, a descriptor such as "Woman! Horse! Hot!"[104] Such a caption would be an obvious description of a bestiality site, to a human censor, but would pass entirely undetected by a database filtration program.

(b)   Embedded Filtration Systems and Their Limitations

Embedded filtration systems are tags that are 'attached' to websites themselves.  These are labels to be used by rating systems, embedded within Internet content itself that rate a site based on a number of specified criteria.  Such a system is far less haphazard and vague than a database filtration system as it necessarily involves the cooperation of the website producer in order to be implemented pursuant to that site. 

Cooperation is necessary, as it is the creator of the site that must add a line of encoded information to the HTML coding of the page.  This information describes the site in terms understandable to an external rating system.  This information acts as a tag that can be read by commercially available software.  The software program grades the site based on the tag.  According to this grading, the site is either filtered out, or allowed to pass on to the user's computer screen.

PICS: The Most Likely Candidate for an Embedded System

Currently, the most potentially viable embedded system is called The Platform for Internet Content Selection (PICS).   Paul Resnik of AT&T Research and James Miller of MIT developed this system as a means of regulating Internet content according to different standards as imposed by different communities.

R. Polk Wagner gives a concise explanation of how PICS works:

"PICS itself specifies little more than the syntax and protocols used to label content and transmit the labels; it does not itself specify a ratings system.  The creators of PICS intend to enable other groups (or even individuals) to develop their own rating schemes, using PICS as the underlying standard to ensure interoperability.  Thus, for example, any web browser that is PICS-enabled would be able to use any of the PICS-compatible ratings systems.  A market might then develop for such ratings systems, allowing a diversity of ratings systems as well as placing the choices regarding rating and viewing content in the hands of the producers and users, respectively."[105]

 

Such a system avoids many of the problems inherent in a database filtration system, which arise as a result of websites being judged and filtered by third parties.  An embedded system allows a site producer to rate her own site.  Further, the user may choose a particular rating system that will filter out particular types of sites with particular types of ratings.

The main flaw with this system, however, is that it is open to abuse.  Website operators could characterise their sites as falling within the standards of a particular rating when actually it should be rated much more harshly.  They would still, however, be able to plead the defence of having used  'an available technological means' for preventing children (whose parents use embedded filtration software) from accessing their sites' material.

(6)        The Implications of These Technological Limitations on COPA

            It appears, therefore, that none of the means currently available to implement COPA as a matter of practice are likely to be effective.  This seems to detract considerably from any possible argument pertaining to means and their connection to the ends of the government action.  If the government action calls for a limitation on free expression that is technologically impossible, this must be taken into account at this stage of our Charter analysis. 

It is difficult to anticipate how a court would view this practical impossibility of COPA.  It is possible, however, that the fact that the only 'means' available to serve COPA's 'ends' are ineffective would induce a Canadian court to strike down the law for the fact that the means do not, in fact, logically and rationally connect with the desired ends.

(7)       Minimal Impairment Test

            Leaving aside the practical problems of implementing COPA, there is one final stage to consider in our S.1[106] analysis.  This is the stage at which the court must determine whether or not the impugned government action infringes the right to free expression only to a minimal degree; i.e. only within the de minimis range.  A limit on free expression, if it is to be upheld, must not be a greater limit than necessary.

            Dickson C.J. wrote in Oakes:

"(T)he means, even if rationally connected to the objective ... should impair 'as little as possible' the right or freedom in question"[107]  

 

            This question of the range of a limit on free expression is subdivided into two other questions.  These questions are (a) Is this legislation overbroad? and (b) are there potentially alternative means to this legislation?  The courts have been far more detailed regarding the first of these two questions than the second.  It appears that the courts are somewhat expert on overbreadth, but not as confident when they must rule on alternative means.

(a)   Overbreadth

A statute is deemed 'overbroad' if the court finds that the drafter of the law went too far in defining the law's scope.  This means that the law ends up prohibiting more than necessary in order to meet its intended goal.  This type of over-extension, out of the de minimis range, arises as a result of imperfections in the drafting of the legislation.  It impugns the language of the statute as being over inclusive.

As overbreadth occurs as a result of sloppy drafting, the courts have responded by dealing with such an rights-infringement by narrowly construing the impugned statute.  In this way, they manage to uphold the legislation, while simultaneously rectifying the Charter breach.

Sopinka J. in Butler used such a method.[108]  In order to solve the problem of overbreadth, the Court redefined the common law understanding of 'obscenity'.  The definition was modified in that case from a test based in the 'standards of one's community' to an argument for promoting 'equal rights between the sexes.'

It is reasonable that the Canadian courts could deal with COPA in the same way.  Any problems of overbreadth would likely merely result in a redefinition of the terms "for commercial purposes" and "harmful to minors".  It would be simple enough for the court to read this law down so that it does not pertain to any number of activities that might be brought to the court's attention.

(8)       Filtration Software: A Problem Revisited

A problem for the court, however, is specific to COPA.  As we have already determined, filtration software is likely the most viable means of implementing COPA.

We examined two types of software, database systems and embedded filtration systems.  Embedded filtration systems, as we have noted, are susceptible to human error, either purposely or truly by accident.  These ratings systems are not a great concern for a discussion of overbreadth as an embedded system is more likely to be under-inclusive (thereby bringing more traffic to prohibited websites).

Database systems, however, are likely to be under-inclusive or overbroad.  We have already seen that such systems could legitimately fail to filter out a website depicting bestiality.  These systems are also a concern, however, because they will tend to filter too much content from reaching the user.

Rather than merely filtering content that could arguably be deemed obscene, such systems also have a tendency to block information relating to matters of genuine and pressing social concern.

Prof. Lawrence Lessig expresses his apprehension:

 

"Horror stories abound - sites opened to criticize blocking software (are) themselves included in the blocked list, sites opened to discuss AIDS, or gay rights, excluded because of "mistaken" associations with indecency, vegetarian pages excluded because of associations with animal rights movements.  Controversial sites are easily excluded, yet no one says who gets cut."[109]

 

This is the central concern for civil libertarians and other proponents of free expression.  There is a great concern, for instance, that a fourteen year old girl may be prohibited from viewing material "harmful to minors" and that the filtration software she is forced to use accidentally prevents her from accessing some time-sensitive and profoundly important information pertaining to 'breast cancer'.  This could cause the purpose of COPA, which is to 'protect children from harm', to entirely backfire and serve the opposite effect.

This concern would have to be weighed by a court should COPA ever be enacted and challenged in Canada.

 

(b)   Alternative Means

A statute is out of the de minimis range if a less restrictive alternative means could have been used.  The challenger in Keegstra made an argument of this nature.  In that case, it was alleged that criminal legislation was an excessive means for combating the harmful effects of hate propaganda.  

Douglas Christie, counsel for Mr. Keegstra suggested that 'human rights legislation would be more appropriate.  Alternatively, a system of mediation, or even simply education could have protected the public from the evils of hate propaganda.'

The court did not accept this argument.  Dickson C.J. wrote in Keegstra:

"As for the argument that other modes of combatting hate propaganda eclipse the need for a criminal provision, it is eminently reasonable to utilize more than one type of legislative tool in working to prevent the spread of racist expression and its resultant harm. [110]"

 

Dickson C.J. therefore did not reject the idea of using alternative means to the legislation at issue in Keegstra.  Rather, it would appear that the Court endorsed the idea of using alternative means to combat a particularly despicable form of expression.  Dickson C.J. did not go so far, however, as to say that only alternative means should be used.

It is difficult to say how the Court decides on the relevance of alternative means.  It suffices to say that it is unlikely for a challenger to win under such a heading.  Courts traditionally do not relish the task of determining alternatives.  Courts excel at determining whether or not an alternative chosen exceeds its due scope.

            Alternative means is such a vague and developing area of law that it is difficult to say with any certainty how it would apply to COPA.  The only evidence we have to base a prediction on, comes in short comments in judgments like the comment in Keegstra.  Based on such comments, it is likely that a court might find that other means could be used in conjunction with COPA in order to achieve the goal of protecting children.

IV.     Summary

 

            We have learned from the Canadian courts to characterise the purpose for S.2(b)[111] of the Canadian Charter in terms of three rationales or 'benefits' that this right serves to protect and foster.  These benefits from free expression are (i) the pursuit of truth (ii) the advancement of democracy and (iii) the actualisation of individual self-realisation. 

            We have also determined that the act of posting material on the World Wide Web, for commercial purposes, that is harmful to minors does advance at least one if not two of the above rationales.   This act is not 'violent' at law, and therefore deserves the protection of S.2(b)[112]

            It is apparent that if COPA were enacted in Canada, it would infringe S.2(b).[113]  It infringes this section for one or both of two possible reasons.  This statute is either a breach of the Charter because its purpose is to limit the content of expression on the World Wide Web, or it breaches the Charter because it limits the effects of this right express oneself.  It certainly limits the enjoyment of the third and likely the first rationale for this right.

Our analysis then progressed to the stage of performing a S.1[114] analysis on COPA.  This section determines whether or not an infringement of a Charter right is justified under the particular circumstances. 

Upon examination, it is revealed that this infringement of the Charter is for an 'important' and 'valid' purpose at law.  As well, the government could defend its actions in enacting this legislation based on three criteria: (a) the government was acting as a mediator between competing interests and (b) the government was acting to protect a vulnerable group in society; children and (c) the government would almost certainly rely on sociological data when deciding to enact this law.  Based on these three facts, we know that the Court is likely to pay a fair degree of deference to this governmental decision to act.

It was also determined that the enactment of COPA bore a reasonable and logical connection to the goal being sought; i.e. the protection of children from harm.  Furthermore, this action is further re-enforced in its validity as it can be characterised as a protection of children's rights to personal security.  This is a right recognised both in Canada and internationally.

            Upon examining the question of whether or not COPA could practically serve the ends it is logically meant to serve, however, we found reason to question the likelihood of this.  It seems that the technological means currently at the disposal of those who would seek to limit online proliferation of materials, are ineffective.  Either one must resort to easily circumnavigated systems such as credit card verification, or one must use an entirely infeasible system of age verification certificates. 

The only other viable possibilities are filtration software programs.  These we have learned are either too approximate in that they would fail to limit much of the content that COPA seeks to prohibit, or they are overbroad and would limit the proliferation of some unarguably constitutionally protected speech.  In this way, they would either fail the test of the means as they further the ends or they would be overbroad in their scope.

If the court considered the possibility that alternative means to COPA could be used to protect children, it is unlikely that the court would have accepted this.  More than likely, the court would have agreed that any various means could be used in conjunction.  It is not likely that COPA would be struck down based on such a contention.

V.               Conclusion: COPA Stands a Reasonable Chance of Survival if Enacted in Canada

 

There is no doubt that COPA would offend S.2(b)[115] of the Charter.  This is necessarily so, as it seeks to limit free expression that is non-violent in its form.  In order for such legislation to be upheld in Canada, therefore, it must qualify for a constitutional override based on S.1.[116]

In order to qualify for such an override, COPA would have to survive an Oakes analysis.  Upon performing this analysis, we arrive at a mixed result.  We find that COPA is enacted for a valid and important purpose.  We also find that the enactment of COPA is a logical and rational means of meeting this purpose.  Furthermore, the goal of COPA helps to advance the security rights of children.  These facts are in COPA's favour. 

These favourable characteristics lose much of their force, however, when one considers that the practical means available for meeting these laudable goals of COPA are essentially implausible at this date.  In order to implement COPA, new technology must be developed.[117]

It seems reasonable to conclude, therefore, that COPA could be upheld if it were enacted in Canada.  The courts would likely characterise COPA in such a manner that it would meet the theoretical tests outlined in Oakes.  If, however, the courts considered the practical obstacles to implementing COPA in a constitutional manner, it would likely be struck down.  This would occur either on the basis of overbreadth or due to the fact that it could not serve its intended end. 

In conclusion, therefore, COPA could survive a theoretical challenge under Canadian Charter law.  This legislation would be upheld in Canada in spite of the fact that it infringes our Charter provision protecting free expression.  The only stipulation, however, is that this law could not pass such Charter muster at this date for practical reasons.  Experience with the swift advancement of computer technology, however, leads this author to expect that such technicalities will soon be rectified.  If such technological advances are made that could make this law viable in the real world, it is reasonable to presume that COPA would be upheld if enacted and faced with a Canadian Charter challenge.

Bibliography

ACLU, Brief of Plaintiffs-Appellees, In the United States Court of Appeals for the Third Circuit, Case No. 99-1324, On Appeal from the United States District Court For the Eastern District of Pennsylvania http://www.epic.org/free_speech/copa/appellee_brief.html (last visited 1 April, 2000)  

 

"Child Protection Act Foes File Latest Attack" Newsbytes News Network (1 September 1999) http://www.newsbytes.com/ (last visited 1 April 2000)

 

Electronic Privacy Information Center, "New Cyber-Censorship Law Would Fracture Internet, Experts and Web Owners Testify" (20 January 1999) http://fed-soc.org/child-telecomv2i3.htm  (last visited 31 March, 2000)

 

Elliot R.M., "Freedom of Expression and Pornography: The Need for a Structured Approach to Charter Analysis" in J.M. Weiler & R.M. Elliot, eds., Litigating the Values of a Nation: The Canadian Charter of Rights and Freedoms (Vancouver: Carswell, 1986)

 

Hogg, P.W.,  Constitutional Law of Canada (Toronto: Carswell, 1999)

 

Internet Free Expression Alliance, "COPA Commission is Strapped for Cash" IFEA News (8 March 2000) http://www.ifea.net/news.html (last visited 2 April 2000)

 

Kuner, C., "Legal Aspects of Encryption in the Internet" International Business Lawyer (April, 1996)

 

Lahey, K., "The Charter and Pornography" in J.M. Weiler & R.M. Elliot, eds., Litigating the Values of a Nation (Vancouver: Carswell, 1986) 265

 

Lessig, L., "What Things Regulate Speech" (1998) 38 Jurimetrics J. 629

 

MacKinnon, C., "Vindication and Resistance: A Response to the Carnegie Mellon Study of Pornography in Cyberspace" (1995) 83 Georgetown L.J. 1959

 

Meyer, C., "Reclaiming Sex from the Pornographers: Cybersexual Possibilities", (1995) 83 Geo. L.J. 1969 at 1988

 

Mill, J.S., "Law of Libel and Liberty of the Press" (1825) in J.M. Robson, ed., Collected Works John Stuart Mill XXI: Essays on Equality, Law, and Education (Toronto: University of Toronto Press, 1984)

 

Quistgaard, B., "Pornography, Harm, and Censorship: A Feminist (Re)Vision of the Right to Freedom of Expression" 52 U.T. Fac. L. Rev.132

 

Resnik, P.  & Miller, J., "PICS: Internet Access Controls Without Censorship" (1996) 10 Comm. ACM 87

 

R.J.R. MacDonald Inc. v. Canada (Attorney General), [1995] 3 S.C.R. 199; 127 D.L.R. (4^th) 1

 

Taylor, B.A., "The Child Online Protection Act of 1998: A Good Old-Fashioned Harmful To Minors Law" National Law Center for Children and Families (1998) http://fed-soc.org/child-telecomv2i3.htm (last visited 3 March, 2000)

Table of Cases

 

ACLU v. Reno 31 F.Supp. 2d 473 (E.D. Pa. 1999)

 

R. v. Big M. Drug Mart Ltd., [1985] 1 S.C.R. 295

 

Boucher v. R., [1951] 2 S.C.R. 265

 

Donald Victor Butler v. H.M The Queen, [1992] S.C.C. D. 5840-01

 

Chernesky v. Armadale Publishers Ltd. (1979), 90 D.L.R. (3d) 321

 

Edwards Books and Art Ltd. v. The Queen, [1986] 2 S.C.R. 713

 

Ford v. Que. [1988] 2 S.C.R. 712

 

Hunter v. Southam Inc., [1984] 2 S.C.R. 145

 

Irwin Toy v. Que. [1989] 1 S.C.R. 927

 

R. v. Oakes, [1986] 1 S.C.R. 103; 26 D.L.R. (4^th) 200

 

Ref. Re Alta. Legislation, [1938] S.C.R. 100

 

Reno v. ACLU, 117 S.Ct. 2329 (1997)

 

RWDSU v. Dolphin Delivery [1986] 2 S.C.R. 573

 

Saumur v. City of Quebec, [1953] 2 S.C.R. 299

 

R. v. Sharpe (30 June 1999), Vancouver CA025488 (B.C.C.A.)

 

Smith & Rhuland Ltd. v. R., [1953] 2 S.C.R. 95

 

R. v. Zundel, [1992] 2 S.C.R. 731; 95 D.L.R. (4^th) 202

Legislation Cited

 

Canadian Charter of Rights and Freedoms, Part I of the Constitution Act, 1982, being Schedule B to the Canada Act 1982 (U.K.), 1982, c.11

 

Convention on the Rights of the Child, 12 December 1989, UNGA Doc A/RES/44/25

 

Child Online Protection Act, 47 U.S.C. 231 (1998)

 

Communications Decency Act 47 U.S.C.A. (1996)

 

Lord's Day Act, R.S.C. 1970, c. L-13

 

U.S. Const. amend. I.