Suppose one accepts MacKinnon and Dworkin's suggested statutory definition of pornography. How does one who generally accepts MacKinnon and Dworkin's views on the pervasively harmful effect of pornography, and who accepts a need for legal redress of the harms perpetrated by pornography, deal with pornographic material? The ordinance proposed by MacKinnon and Dworkin would deal with such material by enacting legislation which gives people adversely affected by the works, which clearly fit their definition of pornography, a cause of action against the producers, vendors, exhibitors or distributors for "trafficking", or for an assault "directly caused by the specific work. I do not think liberals, or others for that matter, should have much problem with the clause dealing with assault, since a causal connection to specific works is demanded by it. However, s. 3.2(iii) which deals with trafficking would be very problematic for liberals and legal conservatives because it creates a cause of action for a person contrary to the traditional conception of a rights holder's cause of action. This subsection reads: Any woman has a claim hereunder as a woman acting against the subordination of women. Any man, child or transsexual who alleges injury by pornography in the way women are injured by it also has a claim. [emphasis added] My goal in this paper is to suggest that a slight modification to this subsection of the ordinance would make it very difficult for liberals and legal conservatives to object to it. This modification would restrict the cause of action to the same persons as the other sections of the ordinance, namely, the particular victim of the specified injury. I shall argue that such a modification would largely cohere with the conception of harm already at work in Ontario law, would afford only a minor reduction in the potential efficacy of such legislation in curbing the harm of pornography, and would offer to empower the feminist camp which is behind such an ordinance with a mechanism for social and political change if a sufficiently organized feminist "vanguard" took hold of the opportunity to empower women. Adrian Howe argues that the concept of social injury which may be suggested by the ordinance recognizes the differential harm felt by women from pornography. Howe suggests this social notion of harm may be a necessary feature of any successful law reform which is to address the huge social problem of male domination and female oppression. The liberal notion of an individuated human right fails to capture, for MacKinnon and Howe, "the specificity of the harm to women." Thus, an ordinance which did not create a cause of action "for women as women" would fail to address the root of the social problem of which pornography is a manifestation. This conception of social harm, and thus subsection 3.2(iii), may offend liberals or legal conservatives in two ways. First, the notion of non-individuated harm is antithetical to the liberal conception of a rights holder claiming a cause of action. Fundamental to a liberal conception of harm is the notion of the individual who is autonomous, separate and fundamentally worthy of respect. Rawls and Kant exemplify this view in their analyses when they posit the undifferentiated self, free of any particular qualities save that of being an agent worthy of a fundamental, inviolable respect. This notion of the individual worthy of equal concern and respect in the eyes of the state permeates liberal conceptions of rights. It is also a fundamental, if not exclusive, tenet of the common law of torts: In tort litigation, the courts must decide whether to shift the loss suffered by one person, the plaintiff, to the shoulders of another person [emphasis added]. Clearly, on its face this conception of harm precludes the notion of a harm suffered collectively which cannot be delineated individually. While class actions are possible, and claims may be made on behalf of groups such as company shareholders, this is only by virtue of the fact that a legally recognized individual has suffered an identifiable particular harm. Thus, the conventional liberal notion of harm is radically distinct from that outlined by Howe and MacKinnon. Since on the liberal conception rights holders are autonomous, individual selves who are essentially distinct, harm to one is distinct from harm to another. It may be that a liberal conception of a rights holder simply renders the concept of a social harm, and thus a cause of action "for women as women" incoherent. I do not wish to discuss whether it is possible to develop a complete liberal notion of social harm. It is sufficient to note that the notion of harm to rights holders inherent in the dominant liberal legal discourse appears to preclude a cause of action by any individual simply by virtue of their membership in an oppressed social class. The problem for feminism is that the offence of trafficking in pornography, if the cause of action were limited to individuals who allege a direct harm stemming from this trafficking, may seldom if ever deliver a remedy. Consider the immense burden for a successful action: She must first prove that the relevant materials are pornography. They must be sexually explicit and they must contain one or more of the features listed in the definition. Second, she must prove that the materials sexually subordinated her. The materials have to be more than just offensive; this is not a law that worries about offending sensibilities, it is concerned with injuries to women. These injuries must be proven in court. Only then will the plaintiff be awarded damages or an injunction against the materials in question [emphasis added]. The harm which a particular woman suffers as a result of trafficking in pornography is not easily delineated. It is not the physical assault or forced viewing outlined in the other sections of the ordinance. Nor is it (for MacKinnon/Cole proponents) a tangible physical harm in the "John hits Mary" sense: [P]ornography causes attitudes and behaviours of violence and discrimination that define the treatment and status of half the population . [P]ornography institutionalizes the sexuality of male supremacy ... Since the harm caused by pornography is a social, collective harm to women, conventional liberal notions of tortious harm are seemingly unable to capture its seriousness (no single woman appears to have been grievously harmed). Thus, to limit the cause of action in the ordinance's trafficking provision to particular, individual women might seem futile for feminists in that a traditional liberal court would be unable to make sense of the claims of harm involved. The situation may not be quite so bleak. It will be useful to examine the notion of a social harm, a harm which cannot be tied directly to one victim, in the areas of criminal and tort law. I suggest that Ontario courts already have the basis for a framework of social harm in the federal statutory provisions on hate literature, and in the principles which can be adopted from the Bhadauria case. The Criminal Code in sections 318 and 319 prohibits the advocating or promoting of genocide and the incitement of hatred of identifiable groups respectively. It is noteworthy that "identifiable group" is defined as "any section of the public distinguished by colour, race, religion or ethnic origin", but does not include gender identification. These sections allow groups, rather than individuals, to seek redress for the dissemination of hateful or pro-genocidal material. Section 319 has been found to violate s.2(b) of the Charter of Rights and Freedoms, but to be justified under s.1 of the Charter. Thus, it is considered to be coherent in Canadian criminal law for a somewhat intangible social harm to have been suffered by a group through the publication of literature, and for a remedy to be appropriate. There are problems with this kind of legal protection from social harm if MacKinnon and Cole's assumptions about the legal system are accepted. The sections may take effect only on the initiative of the Attorney General; it is this feature which led to charges against Ernst Zundel [for the publication of literature denying the holocaust and claiming the existence of a Zionist conspiracy] being laid by Jewish activist groups under s.181 of the Code. Thus, Cole's claim that legal redress for the harm of pornography will not be effectively obtained through reliance on intervention by a male-dominated executive branch of government is supported by the failure of another identifiable victim group to have charges laid by the Attorney General in what appeared to many to be a clear case. In isolated cases like Keegstra, where children were the group to whom hateful information was being disseminated, the law recognizes social harms as actionable. It is clear though that the pragmatic barriers to criminal prosecutions for the harm pornography causes to women, as opposed to society's moral intolerance of the offensive content, are immense in a male dominated liberal society. What should not be lost in this pragmatic pessimism is the adequacy of the conceptual foundation of a social harm which arose in Keegstra. In this case, the social harm was seen not only to affect the "targets" of the information, in this case Jews, but to adversely affect "society at large". Furthermore, the type of harm caused to the target group is similar to that seen by feminists as suffered by women due to pornography: Disquiet caused by the existence of such material is not simply the product of its offensiveness, however, but stems from the very real harm which it causes. [E]motional damage caused by words may be of grave psychological and social consequence. [They] can constitute a serious attack on persons belonging to a racial or religious group, and in this regard the Cohen Committee noted that these persons are humiliated and degraded (p. 214). Referring then to a prominent liberal theorist, Dickson C.J. said: In my opinion, a response of humiliation and degradation from an individual targeted by hate propaganda is to be expected. A person's sense of human dignity and belonging to the community at large is closely linked to the concern and respect accorded the groups to which he or she belongs (see Isaiah Berlin, "Two Concepts of Liberty", in Four Essays on Liberty (1969), p. 118, at p. 155). Let us call the harm to a particular woman which is suffered as a result of trafficking in pornography a quasi-social harm. It is distinguished from a social harm in that the victim conceived as a member of a victimized class, but any action to redress this harm is brought solely on her own behalf for the harm personally suffered. Unlike the actions in the criminal cases previously cited, claims here are not on behalf of a group or on behalf of society as a whole, but are on behalf of an individual who has suffered as a member of a class. The modified ordinance I propose seeks to redress quasi-social harms. One may question whether this (as distinct from addressing social harm) is a tenable legal proposition or not. I suggest that it is, at least in Ontario, given our established legal categories and means of redress. The Ontario Human Rights Code provides an example of an attempt to redress quasi-social harms. It may be true that tort law is unable to address the "social injury that occurs at a personal level", but this is exactly the kind of injury the human rights codes of the country have been enacted to redress. While couched in the terminology of individual human rights, the OHRC's categories of protection indicate a necessary connection to the notion of a social harm. The OHRC does not promise equality, equal treatment, equal respect etc. of every person, its grandiose preamble notwithstanding. What it promises is that injurious discrimination to individuals due to membership in certain social categories will be redressed by damages or injunction. These social categories are those which are traditionally associated with social injury - race, ancestry, place of origin, colour, ethnic origin, citizenship, creed, sex, sexual orientation, age, marital or family status, or handicap. Notice that many categories are absent - foolhardiness, poverty, language group, education, etc. What this indicates is that the OHRC does not address an equality right per se, but addresses social harm as a result of being eg. black, female, Croatian, gay, blind, 25 yr. old, unmarried, etc. The remedies under s.40 of the OHRC are nearly identical to those in the modified ordinance - damages, including those for personal anguish, costs of the action, and injunction. The modified ordinance would thus be quite similar to the existing human rights legislation in Ontario in its recognition of social harm and its suggestion of remedies. Where it would differ is in its refusal to supplant the power of the victim to pursue their own action in court, rather than deal with a commission (and its discretionary powers) or board of inquiry to investigate matters. Thus the modified ordinance would remain "women-initiated and women-driven." It would also differ from the OHRC in that it would clearly specify an as yet unrecognized particular method of inflicting harm: trafficking in pornography. One well-known attempt to pursue a remedy for a quasi-social harm outside the administrative realm of the OHRC succeeded in the Ontario Court of Appeal, but failed at the Supreme Court of Canada. In Bhadauria, the plaintiff alleged that she had been discriminated against because of her race in applying for a teaching position, and brought an action on a common law tort basis of discrimination, and also cited a violation of the OHRC as giving a cause of action. Wilson J. in the Court of Appeal held that it was open to the court to allow the expansion of the common law to include the tort of discrimination, and would have allowed the action to proceed. The question of whether the OHRC gave rise to an independent civil action was not entertained given this finding. Laskin CJ. in the Supreme Court of Canada said that the OHRC was meant to supplant the attempt to seek a remedy at common law, not to supplement it, and thus barred the action from proceeding either at common law or directly from an alleged breach of the OHRC since Bhadauria had not attempted to invoke the procedures of the OHRC for redress. What is noteworthy from this case is that the question of whether this kind of harm was capable of judicial consideration was never at issue. For the Court of Appeal, the common law was fully capable of entertaining such a harm as a tort. For the Supreme Court, the OHRC was seen as the appropriate means of redressing such harm. What the examples from criminal and tort law demonstrate is that the notion of a quasi-social harm is tenable in our legal system, particularly if individuals are given a statutory right to pursue remedies for it. Thus, the modified ordinance would simply indicate to the court a category of social harm which has not previously been specifically addressed, the harm to women from the propagation of pornography. The relative success at achieving remedies from OHRC provisions, as compared to the reluctance of the government to permit the exercise of the Criminal Code provisions, indicates that retaining a civil right of action for individuals will be the strategically better move for feminists insofar as they are seeking redress. I shall leave discussion of whether this is a tenable feminist political strategy for dealing with pornography for a later part of the paper. It may be objected that the fact that our legal tradition is capable of making sense of the notion of a quasi-social harm, and thus could provide the judiciary with the conceptual tools to adjudicate on a modified version of the ordinance, does not imply that the modified ordinance and its conception of harm is acceptable in a liberal framework. A liberal framework may demand individuated harms, and the fact that our existing legal framework can work outside that limitation simply demonstrates that liberalism is not at the root of our legal framework's evolving notion of harm. Thus, the ordinance may still be seen by liberals as incoherent, or worse, to invoke an illegitimate conception of non-individuated rights and afford state enforced remedies for illegitimate purposes. This liberal argument may be theoretically tenable, and thus the "bleak" picture I painted may still apply insofar as we favour a liberal legal framework. Furthermore, the powerful liberal arguments concerning freedom of speech may override the concern for the kind of harm contained in the ordinance. Perhaps because the alleged harm has not been demonstrably linked to the propagation of pornography, or is not a harm in the liberal sense, but an expression of a preference, a liberal framework could not permit the ordinance since it is an undue restriction on free expression. My response to this is twofold. First, given that protection from harm is generally an acceptable justification for a restriction on liberty in a liberal framework, it is up to liberals to deliver a coherent rebuttal to MacKinnon et al.'s contention that pornography causes genuine physical and psychological harm to women, rather than just revulsion. To date I have not seen a liberal rebuttal which did not make the assumption that the root of the problem of pornography is simply moral offence, i.e. strongly held preferences against the propagation of pornography. I find the feminist claims about harm to be very persuasive, and until they are addressed by liberals in terms of a rebuttal of the harm, rather than by reference to the moral disvalue of pornography, the onus should rest on them. Second, the ordinance is not an attempt to arrive at a coherent theoretical position on pornography, but is an attempt to solve a social problem through the mechanism of law. If the attempt of the existing legal system to redress such problems is illegitimate simply on abstract liberal grounds, it need not be a fundamental practical concern of feminists to convince liberals that the ordinance is acceptable. From the feminist strategic perspective, it is enough to show, as I am attempting, that some form of the ordinance coheres well with the existing legal tradition whether that tradition is fundamentally liberal or otherwise. The problem of theoretical legitimacy of the legal system as a whole need not be of particular concern for proponents of the ordinance; what is important is redressing the harms done to women by the political and legal means at hand. Moreover, I am not convinced, given the comments of Dickson J. above, that liberal theories are committed to abandoning the notion of harm and the means of redress which we see in the existing legal framework. Perhaps then only certain categories of liberalism would take objection with the notion of harm addressed in Keegstra or the OHRC. The second major problem with the ordinance for our traditional liberal legal framework is the identification of the source of the harm. The liberal conception of autonomous individuals requires a particular victim and a particular perpetrator. MacKinnon and Cole extensively consider the notion of women as victims of a social harm, but give little consideration to the notion of the perpetrators of this harm beyond the simple definition of pornography. For them, it would seem that if we can identify pornography, we can identify the source of the harm. Clearly, identification of the perpetrators is required before an action for redress can be launched under the ordinance. Even though this is not a theoretical requirement of every system of redress for harm, it is both a theoretical and pragmatic requirement for launching a civil action. The frameworks of criminal law, tort law and the OHRC all presume an identifiable perpetrator of a harm can be identified. Even if it were not a legal requirement for a determination of entitlement to a remedy that one be capable of identifying the perpetrator, it would be rather pointless to launch an action for damages or injunction if there were no identifiable legal person from whom to collect or upon whom the injunction would act. The harm from pornography is not easily traced to a single source. MacKinnon et al. go to great lengths to point out the complexity of the problem of pornography, that harm ensues not just because of what the content of pornography is, but because of how the messages of pornography contribute to the social fabric of male hegemony. "Pornography institutionalizes the sexuality of male supremacy." If, as has been argued, pornography's harm is intimately connected to social practices, then perhaps blame for this harm cannot be pinpointed to pornography alone, or any particular source of pornography. It is beyond the scope of this paper to attempt an analysis of society which could offer insight into the distribution of responsibility for reparation of the harm of pornography across all members and institutions in society. Instead I shall attempt to offer insight into the smaller problem of distribution of responsibility among pornographers. Given the huge volume of pornography, in many cases it may be impossible to pinpoint the particular publishers, materials etc. which led to the quasi- social harm against a plaintiff. I suggest that a solution to the problem of perpetrator identity may be suggested by analysis of the California Supreme Court's treatment of the problem in a product liability case. The excerpt from Linden above indicates that traditionally the perpetrator of a tort must be clearly, individually identified as the cause of the harm suffered by the plaintiff. This traditional concept of causation in tort law is not sacrosanct. In Sindell, an action launched by a victim of a harmful drug succeeded against a multitude of pharmaceutical companies even though no one company could be causally linked to the harm suffered by the particular victim. The plaintiff's mother had consumed the drug DES during her pregnancy, and the plaintiff suffered birth defects as a result. Evidence of the particular supplier of this drug to her mother had long since vanished, but it was certain that some manufacturer out of a number producing it at the time of the pregnancy had promoted the drug without warning of the potential side effects. The California Supreme Court held that, in the absence of direct causal links to any particular supplier of the drug DES, the plaintiff could recover damages in proportion to the likelihood that any manufacturer was the one which provided the drug to her mother during pregnancy. This case has many obvious differences from a purported action for harm from trafficking in pornography. It was certain that the plaintiff had suffered a tangible physical harm from the product; the only question was whether manufacturer A, B, C etc. had been the perpetrator. What is interesting about the case for proponents of a modified ordinance is that if a woman could demonstrate to the court a harm from the propagation of pornography in general, this case would indicate that all pornographers or traffickers might be held liable in proportion to some measure of their market share. Of note is the fact that only "the producers of a substantial share of the market, that is, over 50 per cent" needed to be sued to invoke this "market share" liability notion. Thus, if a woman could demonstrate the relevant quasi-social harm from pornography, and name producers of at least 50% of the market share of the relevant material, she would meet the threshold for bringing an action. Of course, if a particular trafficker could show that theirs was not a harmful brand of pornography (or more accurately, was not harmful, and thus was not pornography), they would be immune from the action. One problem with this scheme is limiting the named defendants to those who produce an identifiable kind of pornography. I am not confident that in all or even most cases a woman would be able to identify any particular kind of pornography as that which caused the harm she experienced. This is again due to the complex social nature of the harm, its difficulty to pinpoint. There is a danger that an implausible or untenable number of publishers or traffickers of other sorts would be named in any given lawsuit. Furthermore, publishers might begin a "third party" frenzy in an attempt to draw in others to distribute the costs of the suit. However, it seems plausible in at least some cases that a particular class of material could be identified as the cause of the harm, and since (as I shall soon argue) the importance to feminists of the ordinance is not just its success at compensating particular women, but its political and social effects, if some cases succeed it will be a great victory. Thus, the problem of identification of a perpetrator is not insurmountable. There is at least some jurisprudence which would give judges the tools to offer redress where individual perpetrators cannot be identified. In particular cases there may simply be single or multiple defendants, or there may be an identifiable class of defendant where the particular perpetrators are unknowable. In either case, the Ontario courts have available to them the conceptual tools to deal with the matter. The addition of the indeterminate perpetrators doctrine from the DES case would be a welcome addition to the judicial treatment of a modified ordinance, but successful actions would not depend on it. It is not impossible to imagine the kind of material that would be claimed to be harmful - it would contain pictures or words where women in a sexual context are dehumanized, objectified, shown as enjoying pain, rape or humiliation, bruised, bleeding or hurt, etc. Once the identification of harmful material is accomplished, the publishers, distributors, etc. need to be identified and named. Then the major problem for a woman to overcome as plaintiff under s.3.2(iii) is to demonstrate that some genuine quasi-social harm to her came about from the propagation of pornography, although she was not assaulted or forced to view or participate in it. As the Ruth M. testimony indicates, this is not entirely implausible. To sum thus far, a modified version of the ordinance would give individual women a cause of action for quasi-social harms they have suffered as a result of trafficking in pornography. While the hate literature provisions of the criminal code suggest that our legal framework can deal with the notion of social harm, greater success can be expected if the modification is adopted. This modification would bring the feminist notion of harm suggested by MacKinnon and her proponents within a legal framework not unlike some of the existing legal schema in Ontario which give civil remedies for quasi-social harms. The problem of specifying a perpetrator, while great, is not insurmountable given the doctrine in Sindell and the accepted notion of multiple defendants in civil suits. Finally, though the ordinance may at first seem unworkable (as any new legal doctrine does until it has had judicial treatment), there are genuine fact situations in which redress seems just and plausible. I have mentioned feminist strategy in various contexts in this paper. Of course there is debate within feminist circles over the appropriate strategies for dealing with the problem of pornography. The ordinance, modified or not, will not satisfy every feminist. I think it would be a tenable proposition for MacKinnon and her proponents not only in its provision of a remedy for particular social harms suffered by individual women, but because it will serve to expose the harm of pornography to great public scrutiny, provided feminists devote substantial political effort to particular cases. MacKinnon et al. are concerned that the ordinance should be a mechanism for changing the power relations sustained by pornography. Since the harm of pornography is in a sense held collectively, is social, and since the modified ordinance restricts the cause of action to a single plaintiff on her own behalf as a woman, the modified ordinance has arguably created a law which is unlikely to be pursued. This is because the women most likely to succeed are the least likely to proceed - they either will not possess sufficient power in their situation of subjugation, or they will not recognize the harm since for them it is normalized, adopted, accepted. It is probably true that the ordinance will not turn upside- down the subjugation of women simply by offering remedies to individual women. The harm of pornography to women is social; individual remedies will not change that. However, the existence of the ordinance, and the existence of women like Ruth M. and Linda Marchiano who somehow break out from the bonds of a pornographic existence mean that some cases will come to light. If proponents of MacKinnon's ordinance adopt a suitable strategic posture, the ordinance will be effective in meeting their aim of limiting the harmful effect of pornography on women. The task for feminists, I would suggest, is twofold. First, organization of support mechanisms is needed to give women the resources to come forward and challenge those who harm them through trafficking in pornography is needed. The role of support groups, groups to provide legal resources, groups to provide personal support in a situation where one's established values, relationships etc. are shaken apart, is crucial to the success of actions brought under the ordinance. Individual women would be truly exceptional to successfully bring forth an action on their own. Second, feminists must try to contain and confront political opposition to the modified ordinance which can be expected. There is little doubt in my mind that cases brought under this ordinance would bring about much publicity, just as Keegstra and Zundel did. Opponents will be quick to point out the "censorship" involved, the restriction on freedom of expression, and cry for the invocation of the Charter of Rights to thwart efforts at redressing the harm to women. Feminists must strive to bring the harm to the attention of the public, show the public what it is that pornography does, as well as show the community what it contains. The campaigns, the publicity in both lobbying for enactment of the ordinance, and pursuing actions under it will no doubt rally a significant segment of the community to support women in their quest for freedom from harm. While it will no doubt also create controversies, polarizations, opposition, etc. (much as the Thomas hearings recently did on the issue of harassment), the exposure of the issue will, I suggest, be strategically beneficial. To conclude, a version of the ordinance which is modified to restrict the cause of action for trafficking in pornography to individuals would be a tenable proposition. It would not be an extreme departure from our liberal legal tradition, but would afford redress for individuals who suffer quasi-social harms in a manner consistent with existing legislation on discrimination and hate literature. The problem of identifying perpetrators is difficult, but existing doctrine in the sphere of negligence law provides some insight into dealing with it. Furthermore, the feminist goal of a large scale change in the power imbalance perpetuated by pornography will at least be advanced, though not fully attained, by the ordinance. I suggest that such a modified ordinance should be given serious consideration by feminists and our legislators.