Georgetown Law Journal

July, 1993

*2195 VIEWING AND DOING: COMPLICATING PRONOGRAPHY'S MEANING

Susan Etta Keller [FNa]

The First Amendment appears to be at the heart of the pornography debate, with the most ardent antipornography crusaders considering First Amendment concerns secondary to concerns about harm to women, and those opposing regulation of pornography insisting that First Amendment concerns should be paramount. Some commentators have attempted to bridge this gap, suggesting that pornography can be regulated while accommodating the First Amendment. As pornography regulation is proposed, this latter position is likely to appeal to many who are loath to choose between what seem to be two important concerns: preventing the degradation of women and free expression.

The most visible perspective on regulating pornography has been the feminist antipornography movement. The model ordinance proposed by Andrea Dworkin and Catharine MacKinnon [FN1] rejects the obscenity approach and defines pornography in terms of its harms to women. [FN2] An *2196 absolutist First Amendment approach to pornography regulation might describe the opinion in American Booksellers Association v. Hudnut, [FN3] which struck down an Indianapolis ordinance that was based largely on the feminist model ordinance. More importantly, an absolutist First Amendment approach exists as a counterpoint or foil within the writings of those supporting antipornography regulation, [FN4] and is clearly a position being accommodated by those seeking compromise. Cass Sunstein, who criticizes such a First Amendment position, characterizes it as holding "that all speech stands on the same ground and that government has no business censoring speech merely because some people, or some officials, are puritanical or offended by it." [FN5]

After the Hudnut decision and its affirmance by the United States Supreme Court, many assumed that the debate over regulation had ended, and that discussion of pornography would move to a different realm. [FN6] However, the emergence of new legislative proposals [FN7] makes the issue, and particularly the attempts at compromise, again relevant. Most compromise attempts, however, fail to address the deficiencies of the two opposing positions. Typically they suggest that the pornography proposed for regulation is not the sort of expression protected by the First Amendment, or that the method of regulation proposed does not reach expression, but in neither case do they question the meaning of pornography or the meaning of expression. Moreover, both the extreme and compromise arguments fail to consider the ways in which real life power arrangements and fantasy interact; they assume either a complete coherence or sharp disjunction between fantasy and reality.

In Part I of this article, I review some compromise attempts, as well as an alternative, in order to introduce a number of the difficulties in assessing the meaning of pornography. This alternative, a suggestion to enforce on its own the provision from the feminist antipornography ordinance addressing the coercion of models and actresses, raises important distinctions between the problems surrounding coercion in real life and those relating to depicted coercion. In Part II, I argue that the feminist antipornography and compromise theories have failed to acknowledge that the *2197 meanings and messages of pornography are variable and capable of producing variable effects. Finally, in Part III, I suggest ways in which the very complications involving pornography's meaning and its relationship to sexuality can be useful in a process of transforming sexuality.

Although I attack the coherence of the meaning ascribed to pornography and suggest that multiple interpretations are available, it will be noted that I continue to use the term "pornography" to refer to a range of sexually explicit material. In her discussion of the antipornography ordinance campaign, Mary Joe Frug warns of the difficulty in defining what constitutes pornography precisely because of the different meanings it may have for different people. [FN8] I believe, however, that the term remains helpful in evoking a particular if vaguely identified range of material‑sexually explicit material that has as at least one of its themes or possible interpretations the erotization of power‑even while I am intent on questioning the extent to which we can say with assurance what that range represents in the minds of consumers.

I. COMPROMISE ATTEMPTS

Attempts at a compromise between the harm‑to‑women position and the First Amendment position have taken various forms. In the U.S. Senate, a proposed bill giving the victims of pornography a cause of action limits its scope to material already criminally regulable: obscenity and child pornography. [FN9] The obscenity standard adopted by the U.S. Supreme Court, and incorporated in the bill, encompasses material outside the scope of the antipornography feminist definition of regulable pornography. However, it does not include materials that could successfully be shown to have "serious literary, artistic, political, or scientific value," [FN10] which would be covered by a feminist definition of pornography, [FN11] and which might otherwise deliver the message of pornography that the bill's authors deplore. [FN12] The political value of this compromise is, of course, that it endangers no material that is not already subject to regulation.

In Massachusetts, a bill has been proposed relying largely on the antipornography feminist definition of pornography. [FN13] Its rationale is that pornography is not only degrading to women, but produces concrete harm to *2198 women in the form of violence. [FN14] However, this legislation also appears to be attempting a compromise, in this case by limiting part of the cause of action to visual material only. [FN15] This limit may have been chosen because the case for causation is stronger, but it may also represent an attempt to appease First Amendment proponents by preserving some pornographic expression for protection. [FN16]

Cass Sunstein has offered another series of compromises. Although proposed six years ago, his framework continues to represent what will likely appear to be the most reasonable gap‑bridging position, one that attempts to accommodate each extreme, evoking simultaneously the appeals of both. He maintains that his analysis "supports the general position that pornography, narrowly defined, can be regulated consistently with the first amendment," [FN17] and he is confident that a narrower definition of pornography "can be framed so as to include only properly regulable materials." [FN18]

Under the feminist model ordinance, there are three major provisions that provide a civil cause of action against producers and distributors of pornography. These actions are available for coercion to perform or appear in pornography, [FN19] for assaults caused by pornography [FN20] and for " t rafficking in pornography." [FN21] While Sunstein is not specific about the form that he feels regulation should take, his far‑ranging assessment of the harmful impact of pornography‑on those who appear in pornography, on victims of sex crimes and on society as a whole [FN22]‑suggests that he would endorse the range of causes of action in the feminist model ordinance.

Another approach might be to examine the benefits and different rationales for the separate components of the feminist antipornography proposal, with an eye toward enacting some aspects but not others. All versions of these proposals, original as well as compromise, contain a provision giving a cause of action to models and actresses coerced into pornography. Such a provision allows coerced models and actresses to *2199 enjoin the distribution of the materials in which they appear. [FN23] No compromise version of which I am aware proposes saving the model and actress coercion provision and jettisoning the rest. The fact that this provision is not thought about separately from the others may reflect a belief on the part of both pornography regulation's proponents (in the original as well as compromise versions) and opponents that the coercion provision addresses material and harms in a manner substantially similar to that of the other proposed provisions. This belief, one that sees the model and actress provision simply as an easier version of the others, deserves closer scrutiny.

While it is important to acknowledge the connections between real‑life coercion and represented coercion, it is also crucial to recognize the differences: the very complicated array of correspondences and noncorrespondences between what is depicted and what is acted. The differences between the rationales for the coercion provision, in which the harms addressed are a result of lived abuses of power, and the rationales for the other provisions, which connect real‑life harms to depicted abuses of power, as well as the ways in which those differences are ignored, provide an important starting place for an analysis of these correspondences and noncorrespondences.

The model and actress coercion provision appears easy to justify because it is the least complicated provision. Giving models and actresses a cause of action to enjoin the distribution of material when they have been coerced into its production relies upon an immediate connection between the alleged source of the harm‑coercion‑and the harm itself‑the effects of coercion. If coercion and the material involved can properly be defined, proving the effects of coercion is unlikely to raise difficult issues of causality. All that is left is to weigh these harms against the harms of suppressing the expression involved. The rationale for allowing the cause of action should, in this respect, closely parallel the rationale for regulating child pornography.

The rationale for allowing regulation of child pornography that extends beyond the borders of otherwise regulable obscenity is the prevention of harm to the children who appear in the material. In New York v. Ferber, [FN24] the Supreme Court noted,

The Miller standard, like all general definitions of what may be banned as obscene, does not reflect the State's particular and more compelling interest in prosecuting those who promote the sexual exploitation of *2200 children. Thus, the question under the Miller test of whether a work, taken as a whole, appeals to the prurient interest of the average person bears no connection to the issue of whether a child has been physically or psychologically harmed in the production of the work. [FN25]

In Ferber, the statute under consideration only targeted material that used actual children in its production. [FN26] Presumably, written as well as animated material, or material using adults that appear to be children would fall outside its scope. [FN27] The Court noted a direct link between curtailing the distribution of the targeted material and the harm to children:

The distribution of photographs and films depicting sexual activity by juveniles is intrinsically related to the sexual abuse of children in at least two ways. First, the materials produced are a permanent record of the children's participation and the harm to the child is exacerbated by their circulation. Second, the distribution network for child pornography must be closed if the production of material which requires the sexual exploitation of children is to be effectively controlled. [FN28]

Cass Sunstein has drawn a parallel between the rationales for regulating child pornography and for the model and actress coercion provision in antipornography proposals: "The case for a ban on these materials depends on a conclusion that abusive practices are widespread and that elimination of financial incentives is the only way to control those practices." [FN29]

The case for the model and actress coercion provision, however, is more complicated than that for the regulation of child pornography. These complications primarily involve the differences between adult women and children. In evoking the parallel with child pornography and suggesting that the rationales for its regulation should apply as well to the regulation of material in which coerced women appear, Catharine MacKinnon also recognizes the differences: "Women are not children, but coerced women are effectively deprived of power over the expressive products of their coercion." [FN30]

*2201 The key complication in the comparison of model and actress coercion to child pornography comes in defining coercion. The overriding assumption in the discussion of child pornography, which mirrors assumptions underlying statutory rape laws and other laws affecting minors, is that no child could consent to its production. The fact that the child posed or acted is sufficient proof of coercion. [FN31]

It is problematic to apply the same assumption about consent to adult women that we apply to children, as Nan Hunter and Sylvia Law point out. [FN32] They critique the apparent refusal by the model antipornography ordinance to acknowledge that any pornography employment situation might be free of coercion. [FN33] Defining coercion in this context raises the difficult issues imported from areas like sexual harassment, rape, surrogate motherhood and domestic violence‑the conflict between a recognition that coercion extends beyond traditional liberal notions of choice and a desire to grant women a sense of empowerment and agency to make choices. [FN34] Instead of acknowledging the need for some tough balancing, the overriding response by advocates and opponents of the model and actress provision has been to ignore one or the other side of this debate; those favoring regulation focus on coercion and those opposing it focus on empowerment. [FN35]

I believe that the model and actress coercion provision has not been supported independent of other antipornography provisions because of the broad brush strokes used in the pornography debate. Those opposing regulation on the basis that it impermissibly limits expression see such a *2202 provision as an attack on expression like the others, [FN36] and therefore may overlook its potential value in attacking real‑life coercion. Those supporting regulation, including those proposing most compromise positions, see this provision as being an attack on the harms generated by pornography, indistinguishable from the other provisions. They therefore may overlook its potential to challenge the means by which pornography is produced, as well as pornography's content, without threatening its continued availability. [FN37]

Because compromise attempts have failed to challenge the monolithic character of either extreme position on antipornography regulation, the possibility of addressing some harms of pornography on their own has not presented itself. However, the regulatory provisions generally proposed alongside the model and actress coercion provision‑those that attack harm that is supposed to result from the viewing, listening or reading of pornographic materials‑require for their justification an additional analytical step. They depend on significant assumptions about pornography's meaning that are not raised by the coercion provision. In these other regulatory provisions, the effect or harm is seen to arise from the understanding of a particular message that the defined material generates, rather than from the coercion itself. The difficulties of defining coercion so as to get at the identified effect are complicated many times over when filtered through assumptions about the message of the material defined as pornography. Such provisions require not just interpretation of whether coercion occurred, but of what meaning to ascribe to depicted coercion. However, because antipornography advocates often see a seamless connection not only between the means of production and the product itself but continuing on from the product to the interpretation of it, a measure that only combatted working conditions while allowing pornography to continue to exist would be meaningless to them.

Recognizing the complications that arise from the differences between the results of viewing and the results of participating in pornography can *2203 lead to different conclusions. If we could overcome the difficulties of defining coercion, and if this segment of the proposed ordinances were independently enacted and effectively enforced, pornographers would ideally be faced with a situation in which the costs of making coerced pornography‑through insurance costs or direct liability‑exceeded its benefits‑presumably low wages, favorable contract terms for the producer, the economies of maintaining poor working conditions which would otherwise not attract truly willing performers, and less tangible personal benefits of power and satisfaction on the part of pornographers. Again through effective enforcement, this provision would create an incentive on the part of pornographers to reduce coercion and improve working conditions. While such enforcement might result in different pornography, and possibly open the industry up to more producers of alternative pornographic products, [FN38] it would not eliminate pornography altogether, unless it were impossible to make pornography without coercion. [FN39] Because I see pornography and sexuality itself as dynamic, and because I do not believe pornography is always produced by a wholesale replication of real‑life coercion, I feel confident that the goal of reducing coercion in the production of pornography may alter the nature of pornography without eliminating it. This confidence rests on a conviction that the same sorts of differences between viewing and participation that distinguish attacks on messages of coercion from attacks on real‑life coercion should also assert themselves in the interpretation and understanding of the messages of coercion and their effects. These differences can be seen through an examination of the assumptions that form the basis for the other key components of antipornography proposals.

II. PORNOGRAPHY'S MEANING: DEFINITION, MESSAGE, AND EFFECT

Discussions about pornography regulation will often assign meaning to pornography on three different levels. One level is the legal definition employed: the criteria used to recognize what is covered by the term "pornography." The rationales behind the choice of legal definition are usually seen as closely tied to the second level of meaning‑the message of pornography: what the material says about the world and how the material is perceived by those who consume it. Finally, there is the meaning of *2204 pornography that relates to the effects upon those viewing or reading the material. These effects are often seen as congruent with the message of the material that has been defined as pornography. Indeed, for those who believe that a class of pornography can be clearly defined for which the message produces certain undesirable effects, the layers of meaning are likely to appear identical to one another, and the distinctions between levels are likely to be ignored. [FN40]

A. THE FEMINIST ANTIPORNOGRAPHY CAMPAIGN

For Catharine MacKinnon and Andrea Dworkin, the leading proponents of the feminist movement to regulate pornography, the message of pornography is the systematic degradation of women. [FN41] The legal definition in their model ordinance is meant to encompass all material that contains this message. [FN42] They argue that the effect of pornography is also the systematic degradation of women. Indeed, they quite self‑consciously conflate the distinction between message and effect, arguing that the message is the effect and the effect the message. Dworkin, for example, writes: "Pornography is the institution of male dominance that sexualizes hierarchy, objectification, submission, and violence." [FN43]

Some who advocate the protection of expression of material defined as pornographic, while disputing the constitutionality of the feminist legal definition, often accept the arguments made about both the message and the effect of the material so defined. The court in Hudnut, for example, strikes down antipornography legislation, yet agrees with the legislation's drafters that "[d]epictions of subordination tend to perpetuate subordination. The subordinate status of women in turn leads to affront and lower pay at work, insult and injury at home, battery and rape on the streets." [FN44] While challenging the ordinance on First Amendment grounds, the court accepted the connection between depictions of coercion and real‑life coercion.

B. COMPROMISE ATTEMPTS

Those proposing various degrees of compromise also strive to keep definition, message, and effect congruent. For instance, the rationale for *2205 the choice by the Massachusetts bill's sponsors to limit part of the cause of action to visual material was likely prompted by an assessment that the purported effects, the harm stemming from consumption of pornography, would be more closely related to visual material. [FN45]

Similarly, Sunstein accepts that the message, as well as the effect of pornography, is that it is harmful and degrading to women: "The central concern is that pornography both sexualizes violence and defines women as sexually subordinate to men." [FN46] Although he recognizes the potential existence of hard cases, he believes that a determinable and regulable category of material can be accommodated through his definition, which requires that "regulable pornography must (a) be sexually explicit, (b) depict women as enjoying or deserving some form of physical abuse, and (c) have the purpose and effect of producing sexual arousal." [FN47] This definition partakes strongly, as he acknowledges, of attempts by radical feminists to define pornography. According to Sunstein, the rationale behind the definition is closely tied to the message and effects of the material:

In contrast to the vague basis of the obscenity doctrine, the reasoning behind antipornography legislation is found in three categories of concrete, gender‑related harms: harms to those who participate in the production of pornography, harms to the victims of sex crimes that would not have been committed in the absence of pornography, and harms to society through social conditioning that fosters discrimination and other unlawful activities. [FN48]

The harms to sex crime victims and society, which necessarily result from a processing of the message of pornography, are envisioned by Sunstein to be as direct an effect of the production of the material as the harms to the models and actresses who appear in pornography.

In constructing his definition, Sunstein uses two strategies to attempt to appease First Amendment critics of antipornography regulation. Although he argues that all pornography is low value speech, and therefore more easily regulable because subject to a lower level of judicial scrutiny, [FN49] he also seeks to limit the range of pornography that actually will be regulated so that his definition is narrower than the definition in the feminist model ordinance. [FN50] Not only does Sunstein believe that pornography*2206 in general has a message that is distinct from protected expression, he also believes that the pornography encompassed by his definition has a message that is the most closely related to the harms or effects associated with pornography. These include "a general increase in sexual violence directed against women, violence that would not have occurred but for the massive circulation of pornography." [FN51] While acknowledging methodological problems in the studies showing a direct relationship between message and effect, Sunstein maintains "that the existence of pornography increases the aggregate level of sexual violence." [FN52]

The key provision of Sunstein's proposal that actually narrows his definition of pornography requires that "regulable pornography ... depict women as enjoying or deserving some form of physical abuse." [FN53] This definition is narrower than the model ordinance's, which would grant a cause of action for graphic depictions of women as "sexual objects." [FN54] It is also narrower because depictions of a man receiving sexual pleasure from the physical abuse of a woman when she is portrayed as neither enjoying nor deserving such treatment would not be regulated. According to Sunstein, his definition is more directly related to the effects or the harms caused by what he sees as the message of pornography: "The approach proposed here excludes sexually explicit materials that do not sexualize violence against women, and it ties the definition closely to the principal harms caused by pornography." [FN55]

The expression "sexualize violence" as used in the above quotation plays an important role in reducing tension across the three categories of meaning: the legal definition, the message and the effects of pornography. Sunstein assumes that the definitional description‑the depiction of "women as enjoying or deserving some form of physical abuse"‑is the same thing as "sexualizing violence," and further, that the message of such material will be interpreted as sexualizing violence‑presumably making violence seem sexy‑and that the effect of that message will be to sexualize violence‑presumably to cause sexual violence to occur. [FN56] The basis he offers for this limiting definition, however, is a researcher's view that "the aggressive content of pornography ... is the main contributor to violence *2207 against women." [FN57] This does not explain why Sunstein's definition is limited to material in which women appear to enjoy or deserve physical violence. Surely unwanted or undeserved sexual violence depicted in pornography has aggressive content. The limitation also does not explain by what mechanism such material "sexualizes" violence. There is an unexplained difference in the way in which the expression "sexualize violence" operates on each level. The material has a shifting connection to violence: at the level of definition, the material must simply be recognized as having the subject matter of sexual violence; at the message level, it must be interpreted as making violence sexually arousing; and at the effect level, this message must be reconfigured into real‑life action of sexual violence.

C. COMPLICATING MEANING ACROSS THE DIFFERENT LEVELS

The failure to note the differences between definition and message, or between message and effect is not sloppiness but ideology. Each time a coherence is assumed between one level of pornography's meaning and another, it reflects a certain belief about sexually explicit expression and about sexuality itself. Sunstein's choice to narrow his definition to include only materials that "depict women as enjoying or deserving some form of physical abuse" [FN58] demonstrates his confidence that it is possible to determine in a depiction of abuse if the woman enjoys or deserves it (this assumes that there is one message), and also that there is something about this category that more specifically fosters harm (this assumes that the message will be translated into action in one way). Sunstein's definition further demonstrates confidence that the depiction of a woman enjoying abuse is a false depiction. This represents a confidence both in the unitary relationship between depiction and reality and in the nature of women's sexuality as well.

Sunstein is able to establish coherence across the three levels of meaning by: (1) a connection between his definition and the message of pornography; (2) his ability to ascribe one message for the material defined as pornography and segregate it from the message of expression generally; and (3) the ability to ascribe a particular set of effects related to that message. It is possible to challenge the coherence of Sunstein's argument at all three of these critical junctures by demonstrating his similarity as well as his divergence with the other theorists. Indeed, both *2208 the antipornography position and the free expression position present a potent and useful critique of the other that is either ignored or defined away by Sunstein's compromise.

1. Connection Between Definition and Message

One can see the disjunctions between definition and message by examining the basis for the narrower definition Sunstein chooses. His narrower definition of pornography does tap into a distinction between comfort with regulation of some materials and discomfort with regulation of others on the part of observers of the debate. Robin West, for example, suggests that she might feel comfortable regulating only violent pictorial pornography because it is more closely connected to harm. [FN59] Sunstein's definitional requirement of depicting women enjoying or deserving abuse is his attempt at accommodating the comfort and discomfort distinction. People might be more comfortable regulating material meeting Sunstein's definition because the idea of a woman actually enjoying or deserving abuse seems so false. Sunstein's attempt to narrow the definition to distinguish between the portrayal of violence and the portrayal of nonviolent subordination is thus designed to accommodate the concerns of both the antipornography and First Amendment positions. He agrees with the feminist antipornography advocates that the message of pornography is closely tied to the harms it causes, [FN60] while at the same time he agrees with the First Amendment advocates that not all pornography is regulable.

In creating a narrower definition focusing only on physical abuse, however, Sunstein ignores the important message of feminists that depicting women as sexual objects has as large a role in the erotization of the domination of women by men as does the more blatantly violent material. It is true that MacKinnon and Dworkin, as well as Sunstein, in their appeal for their proposals cite the most vicious examples of violent pornography. [FN61] However, the strength of the feminist message also lies in their assertion, made as a critique of positions seeking to protect expression, that the erotization of the domination of women by men is not limited to these *2209 most extreme examples. One of the more important contributions that MacKinnon and Dworkin have made to our understanding of sexuality is their explanation that traditional notions of consent, which draw a sharp line between violence and voluntarism, do not account for the role of power and therefore do not explain the reality of women's coercion. [FN62] In a world of unequal power between men and women, much that appears to be mutually enjoyed sex may in fact be governed by unequal power. The false enjoyment itself is a form of abuse. [FN63] This is an important issue, for instance, in determining when to draw the line with respect to real‑life coercion in the model and actress coercion provision.

The feminist antipornography activists have been able to ride the crest of that comfort and discomfort distinction without, as Sunstein does, giving up on an all‑encompassing definition. They have done this by convincing audiences that the pornography some might feel uncomfortable regulating actually shares important characteristics with the pornography that most would feel comfortable regulating. [FN64] In this respect, they teach an important lesson. Through our discomfort we can see the overlap, as well as the differences, between depicted and real‑life coercion.

Given this lesson, Sunstein's distinction between depictions of women enjoying or deserving abuse and those in which there is no abuse is particularly difficult to maintain. Depictions of what would be consensual, nonabusive acts under traditional notions of consent, may actually be depictions of power being inflicted, and therefore be depictions of abuse. The line may, in fact, be impossible to draw; the ability of his definition to capture these messages may be elusive.

In drawing the line as he does, however, presumably to assuage First Amendment concerns, Sunstein also ignores what is powerful about the First Amendment approach. Like the antipornography position, the First Amendment approach seeks to take advantage of the distinction between comfort and discomfort, but, in this instance, to argue that the material we feel comfortable regulating should be seen as similar to what we feel uncomfortable regulating. One of the hallmarks of First Amendment reasoning is to demonstrate the power of reprehensible ideas. For example, the court in Hudnut explains that it is precisely the ability of controversial *2210 expressions to rouse people that qualifies them as viewpoints deserving of protection. [FN65] Sunstein ignores this lesson in carefully segregating what he sees as the most controversial aspects from the rest of expression.

2. The Message of Pornography versus Expression

Sunstein can only maintain separate categories‑one subject to regulation, one not‑to the extent that he sees each category as coherent. By doing so, however, he privileges the barrier between categories over frank discussion about the meaning of the categories themselves. His ability to maintain this barrier is comprised of not only a confidence in the universality of the proper limits of regulation, but also a confidence in one agreed‑upon message of pornography. "Under any standard," he argues, "pornography is far afield from the kind of speech conventionally protected by the first amendment." [FN66] He further maintains that "traditional first amendment doctrine furnishes the basis for an argument in favor of restricting pornography." [FN67] For Sunstein, pornography has one meaning, and it is a meaning that can be taken and segregated from another conceptual realm that also has coherent meaning‑free expression.

Separating Pornography from Protected Expression. Sunstein makes the distinction between pornography and other expression in his attempt to address First Amendment arguments against regulation. Unlike MacKinnon and Dworkin, who often have been dismissive of First Amendment concerns, [FN68] Sunstein demonstrates a great deal of sympathy for the First Amendment position:

In one respect ... the feminist case for regulation of pornography might seem, quite paradoxically, to weaken the argument for regulation. The feminist argument is that pornography represents an ideology, one that has important consequences for social attitudes. Speech that amounts *2211 to an ideology, one might argue, cannot be considered low‑value, for such speech lies at the heart of politics. If pornography indeed does amount to an ideology of male supremacy, it might be thought to be entitled to the highest form of constitutional protection. [FN69]

He counters the argument for protection of pornographic expression by advancing the position that pornography is low value speech subject to lower protection. He argues that pornography communicates its message through a method different from regular expression, through noncognitive means: "Though comprised of words and pictures, pornography does not have the special properties that single out speech for special protection; it is more akin to a sexual aid than a communicative expression." [FN70] Sunstein assumes that because pornography has as its intent the production of sexual arousal, it must operate in a noncognitive manner. [FN71] He contends that we can discern such purpose by examining the material: "The pornographer's purpose in disseminating pornographic materials‑to produce sexual arousal‑can be determined by the nature of the material." [FN72] He is confident that a range of material can be defined, either by the feminist definition or by his narrower one, that has one message, one which is communicated in a unique and quickly identifiable manner. Because the rest of expression is thus identifiably different, it is still possible for Sunstein to maintain faith in the coherence of protectable expression.

Paul Chevigny has responded to Sunstein by arguing that no clear distinction can be maintained between the process by which we understand and engage in what Sunstein would characterize as rational or cognitive argumentation and the way in which we view pornography. [FN73] Chevigny maintains that so‑called rational thought is actually accomplished by intuition and comparing scenarios, much the way he asserts pornography works. [FN74] Indeed, he argues that pornography functions much like propaganda. [FN75] He critiques Sunstein's "sexual aid" hypothesis: "Pornographic scenes thus may be arousing, but the action or belief that they arouse depends on each viewer's imagination and beliefs. Pornography can be a 'sexual aid' for some viewers because their imagination enables them to use it as a sexual aid." [FN76] It is in this understanding of the method by which we receive *2212 pornography's message that he differs with Sunstein and challenges the distinction Sunstein sets up between protected expression and pornography.

An analogy to arguments about racist hate speech can also demonstrate that the categories of both protected expression and pornography are less coherent. [FN77] Mari Matsuda has argued, in a somewhat similar vein to Sunstein, that certain categories of speech, in her case racist hate speech, should be given reduced First Amendment protection. [FN78] While Sunstein has a two part argument‑first, pornography is low value because of its attributes, then, certain pornography deserves regulation because of the harm it causes [FN79]‑Matsuda rejects such an approach. She suggests that a category of speech can be defined as deserving of reduced protection, more easily and with fewer slippery‑slope problems, on the basis of its content. [FN80] She justifies making racist hate speech such a category because the message of racial supremacy presented by racist hate speech is universally condemned. [FN81] By refusing to separate racist hate speech from the rest of protected speech under any measure except for its reprehensibility, Matsuda challenges the coherence of the separate categories Sunstein sets up between protected and unprotected expression. Neither protected expression nor pornography are so easily contained as categories of meaning as Sunstein would have them.

Ascribing One Message to Pornography. Although Chevigny and Matsuda (by analogy) provide a challenge to Sunstein's position that pornography by its nature can be segregated from protected expression, they do not provide a challenge to Sunstein's implicit assertion that pornography has only one message. Like Sunstein, Chevigny accepts that there is one message to pornography‑the degradation of women. [FN82] While he maintains that the message, like any other argument, may or may not persuade, [FN83] Chevigny assumes that all those viewing pornography will read the same message and that there is consonance between what is viewed and a uniformly understood sexual practice. He also accepts the coherence of the message of pornography itself as uniform degradation, yet unlike Sunstein, he urges that it be protected. Matsuda's argument similarly rests on the assumption that there is a primary message for racist hate speech, *2213 although, unlike Chevigny, she argues for regulation. This assumption of a uniform message is apparent in the three criteria by which Matsuda defines the category she proposes for regulation:

(1) The message is of racial inferiority;

(2) The message is directed against a historically oppressed group; and

(3) The message is persecutorial, hateful, and degrading. [FN84]

A challenge to the idea that pornography can have only one message, however, is available through an examination of Sunstein's own arguments. Indeed, his position that pornography is reducible to one message most closely resembles positions on the other side of the debate that he critiques. For example, in a later article, Sunstein characterizes the First Amendment position as being one that seeks to protect "natural" impulses. [FN85] He argues that a position seeking to protect all speech including pornography "rests on the perceived naturalness of sexual drives, and it emphasizes the need to liberate those drives from the constraining arm of the state." [FN86] According to Sunstein, even when those holding the anticensorship position believe that sexuality is socially constructed, they may "still insist that sexuality is an important human good entitled to immunity from government." [FN87] He critiques this position as appealing to a false neutrality, one that ignores the harm and power‑laden quality of pornography. [FN88]

The problem Sunstein identifies represents only one of a range of anti‑ antipornography positions, [FN89] yet his critique of it is one that can be *2214 applied to the positions of antipornography feminists, and to Sunstein himself. Indeed, the force of the antipornography argument rests, in part, on the assumption that restraints on pornography will lead to restraints on at least certain types of imitative sexual activity. This assumption is demonstrated in Sunstein's contention that all pornography is low value speech because of its immediate impact on noncognitive sexual processes. [FN90] In making this suggestion, Sunstein engages in the same oversimplification as the First Amendment advocates he critiques; he assumes an unmediated connection between watching pornography and engaging in sexual behavior:

The effect and intent of pornography, as it is defined here, are to produce sexual arousal, not in any sense to affect the course of self‑government. Though comprised of words and pictures, pornography does not have the special properties that single out speech for special protection; it is more akin to a sexual aid than a communicative expression. [FN91]

In addition, the possibility that pornography can be understood in more than one way is demonstrated in Sunstein's own analysis. Through the very process of his argument, he does recognize that pornography can be approached and understood in a nonprurient manner, at least by the legal analyst. In suggesting that it is the purpose of speech that determines if it is low value, he contends that we can discern such purpose by examining the product: "The pornographer's purpose in disseminating pornographic materials‑to produce sexual arousal‑can be determined by the nature of the material." [FN92] Ironically, in the same sentence in which he reveals at least one alternative way of understanding pornography‑the analytical detachment of the decisionmaker‑he maintains that the material can have only one readily apparent "nature," an immediate appeal to the sex *2215 organs. Sunstein's own analysis demonstrates at least one way in which pornography can be seen to have multiple messages.

While those advocating acceptance of regulation of racist hate speech have drawn analogies to antipornography arguments, [FN93] perhaps on the basis of their broader acceptance, I believe Matsuda's position [FN94] makes an effective argument for why racist hate speech is more amenable to regulation than pornography. If we accept Matsuda's assumption that racist hate speech can be understood to carry one message, then the case for pornography must be distinguished, because it can be understood on many levels. If that is true, then it does not make sense to apply to those provisions that rely upon the delivery of pornography's message the same rationales applied to the provision that attacks the coercion of models and actresses. What seems like a straightforward problem of eliminating abuse of power in the model and actress provision becomes immeasurably more complicated when we have to consider how the message is defined and how it produces effects.

3. Message and Effect

Among many of those who have analyzed pornography, like Sunstein and MacKinnon, there is an assumption that those who enjoy pornography are not going to be sufficiently sophisticated about it to see a distinction between depictions and reality, suggesting a sharp divide between intellectual and nonintellectualpursuits. This perspective, while focusing on concerns like power and sexuality, is nonetheless a bit like the assumption that children are unable to differentiate the unscathed survival of the Roadrunner cartoon's Wile E. Coyote from the real‑life effects of violence. [FN95] Much has been done to counter assumptions about unitary messages and their effects with respect to the consumption of media in general. John Fiske, for example, suggests that there can be no unified television audience, nor unified television text. Rather, he argues not only that individuals watching television differ from one another, but also that the same *2216 individuals "constitute themselves quite differently as audience members at different times." [FN96] He similarly argues that a television program

is no unified whole delivering the same message in the same way to all its "audience."... What the set ... delivers is "television," visual and aural signifiers that are potential provokers of meaning and pleasure. This potential is its textuality which is mobilized differently in the variety of its moments of viewing. [FN97]

Pornography, like television, may not work by pure identification, and may produce a variety of different interpretations and effects under different circumstances.

Mary Joe Frug has critiqued the assumption of antipornography ordinance advocates that "pornography users mechanistically identify with same‑sex characters and mechanistically seek to reproduce pornography scenes in their own lives," noting that such an assumption would not be true of reactions to other media. [FN98] She argues that pornography consumers "may also use pornography to transform their lives by a more complicated reaction than simple imitation." [FN99] Frug suggests that it is not simply a question of how one reads the messages but what one does with them; the concern in the antipornography debate is that people viewing pornography will act upon the message they receive. However, if the message is not one, but many, determining how viewers will act is much more complicated. The problem is not so much one of noting the difference between message and effect, although that is important as well, but of noting that the two levels resonate together in different and complicated ways. They can neither be pulled easily apart nor fitted neatly together. Viewing is not an isolated activity divorced from the real world, but neither is the connection between viewing and doing a simple one.

D. THE DIFFERENCE BETWEEN VIEWING AND DOING: AN ANALOGY TO SADOMASOCHISM

Many analyses of pornography, such as Sunstein's, obscure the complicated differences and connections between viewing and doing. The failure *2217 to recognize this complex relationship arises from an assumption of an unmediated connection between pornographic expression and sexuality, as well as from a conflation of the definition of pornography with its message and its effects. When standard antipornography analysis does recognize the difference between viewing and doing, it dismisses its significance by asserting an uncomplicated causal connection between message and effect: viewing leads to doing. [FN100] This assertion is backed up by laboratory studies, [FN101] which have received no small measure of attack. [FN102] The best that can be said is that viewing leads laboratory volunteers to indicate that they are more likely to consider doing in a favorable light. [FN103] In other words, viewing may lead to thinking about doing, or at least saying you are thinking about doing. [FN104]

These differences between viewing and doing may be partly elucidated by considering, for the purpose of analysis, the concept of sadomasochism. Sadomasochism is both an apt and a difficult analogy to pornography because the concept has meaning both in fantasy and on the body. [FN105] In addition, consideration of sadomasochism often parallels and is intertwined with discussion of pornography. Sunstein's decision to narrow his definition of regulable pornography to cover works that "depict women as enjoying or deserving some form of physical abuse" [FN106] may very well arise from a belief that works depicting sadomasochism are the most harmful. However, like pornography, the concept of sadomasochism exists on many levels and, as with the distinction between viewing and doing, these levels have often been obscured in analytical presentations.

*2218 In the following discussion, I am attempting to explain the concept, not necessarily the reality, of sadomasochism; the reality is likely to be further complicated by the existence of power relationships that include but also extend beyond the sexual realm. Through this analysis it will become apparent, I hope, that as soon as the unity between viewing and doing is challenged, the notion that what is being viewed has only one message and produces a particular effect is challenged as well.

When people hear the term sadomasochism they think of at least four different levels of practice. At one extreme, people think of the sadomasochistic relationship in which one member, the sadist, truly threatens and terrifies the other, the masochist, who, despite the fear and terror inspired by the sadist, enjoys or at least is dependent upon and attracted to the sadist's ability to inflict pain and punishment. I believe this is what people envision when they characterize a relationship involving chronic domestic violence as a sadomasochistic one. While this understanding encompasses more than a sexual relationship, one would expect the sexual relationship of such a couple to have the same qualities as their general interaction. Such a relationship may be termed "lived s&m."

I have described the first level of practice of sadomasochism the way it is perceived, not necessarily the way it is practiced. As described, it requires that the masochist actively enjoys (or is dependent upon) the relationship and derives sexual pleasure from the punishment inflicted. Lenore Walker has done much to successfully challenge the notion that this perception accurately characterizes battered women's relationships. [FN107] While women in such relationships may, of course, derive sexual satisfaction, it may be despite the violence, or through a complicated channelling of the violence, by which they transform it to another level. In fact it is quite possible that lived s&m is not lived by anyone the way it has been described. Nonetheless, it reflects what people sometimes mean when they refer to sadomasochism.

On a second level, a sadomasochistic relationship often means a relationship in which a couple engages in a specific, ritualized sexual practice of inflicting and receiving pain for mutual pleasure. Such a sadomasochistic practice will involve specific rules and boundaries, i.e., certain agreed upon times when the activity will occur and specific code words that either partner can utter to bring the activity to an immediate stop. [FN108] Sadomasochistic practice at this level can involve real violence in the sense that it may *2219 involve lacerations, welts, and blood. But it differs from the first level in that both partners have control over whether and when the violence will occur; similarly, both partners know ahead of time the extent of the violence in which they are willing to engage. One could call this arrangement "ritual s&m."

At a third level, couples may engage in ritualized sexual practice that mimics the perceived activities of either of the first two levels. The key difference here is the absence of violence, intentional injury, or actual pain inflicted. In a manner similar to ritual s&m, couples may set up rules and boundaries. For instance, as in ritual s&m, they may choose a code word other than "stop" or "no" to signal the end of the game, thus allowing the words "stop" or "no" to be disregarded as part of the activity. Because pain occurs in the imagination rather than on the body, one could call this level "pretend s&m."

The differences among these three levels are sometimes obscured, because of the desire of those engaging in ritual or pretend levels to playact at lived s&m. For example, a personal ad in a Los Angeles alternative newspaper reads: "IMPERTINENT NAUGHTY GIRL wanted by firm‑handed man. Explore your interests with someone who understands." [FN109] The ad seems designed to be ambiguous, and for that reason intriguing, about which level of practice is meant. But when, through the miracle of 900 numbers, the voice mail number that follows the ad is called, one learns from the recorded message of the man who placed the ad that, "If you read 'firm‑handed' and guessed it meant erotic spanking and other activities, you guessed right." This indicates practice that, depending on the nature of the other activities, could probably be classified as pretend or ritual.

Finally a fourth level, "fantasy s&m," is one in which an individual, either alone or during sex with a partner, fantasizes about engaging in any of the three other levels, but does not act out the fantasy, not even to mimic.

The source of the pleasure experienced at these four levels will be different, though related. Whatever sexual pleasure experienced in lived s&m would be directly related to the pain or the power behind the acts. The reason that lived s&m seems less likely to be sexually fulfilling than ritual s&m, for the sadist as well as for the masochist, is due to the absence of rules and boundaries. For ritual s&m, the erotic quality may be enhanced by the fact that the violence is contained and controlled by the parties, allowing them to be aware of the ritual as well as the immediate experience. Because of the control, they are perhaps more able to identify *2220 not only with their own roles, but with those of the other. Jessica Benjamin suggests that much of the erotic appeal of sadomasochistic ritual comes from the simultaneous division of roles into control and dependence, and from the identification by each partner with both sets of roles. [FN110] This separation of roles creates distance from what I have called lived s&m, and, according to Benjamin, this distancing addresses the conflict within each individual between equal desires for independence and for dependence by creating extreme roles of power and powerlessness. [FN111]

For those engaging in pretend s&m, the act of mimicry creates even further distance from lived s&m. The pleasure at this level stems from identifying with, but not engaging in "real" sadomasochism; feeling what fear must feel like without feeling fear; saying no without meaning no, but (and this is key) knowing that a "real" no in the form of the code word will be understood; inflicting without actually causing pain. The "withouts" are likely to be just as important to the pleasure as the feelings, sayings, and inflictings.

Finally, for fantasy s&m, the amount of distance from lived s&m, as well as from the other levels of practice, allows great versatility; the fantasy may involve anything from imagining being tied up with silk stockings by a gentle lover to brutal rape. While the goal of fantasy is to create a feeling of reality, the concurrent unreality will always be part of the turn on. For instance, the ability to range further than one would in play acting is exemplified by the statement often seen in popular magazines: "I like to think about it, but I'd never want to do it." Indeed, the fact that the fantasizer would "never want to do it" is part of what makes the experience of fantasy exciting.

The fact that individuals practice at one of these levels does not necessarily and in fact is unlikely to mean that they will enjoy or feel comfortable at another level, because of either less or more distance involved. Robin West has noted a sharp division between submission based on trust and that based on fear in sadomasochistic novels such as The Story of O. [FN112] Such a division may correspond to the difference between ritual or pretend s&m and lived s&m; trust may be possible only when there is greater distance. The distance that separates ritual, pretend, or even fantasy s&m from lived s&m may also make it possible for couples or individuals to explore issues of power, and the attractions of submission and dominance, *2221 without directly living them. This process, by offering these individuals some mastery over these difficult issues, might even be empowering. [FN113]

The same layering of distances is likely to exist in the process of consuming pornography. If part of the appeal of fantasy is that it is fantasy, then part of the appeal of viewing may be that it is not doing. [FN114] When laboratory subjects report that they are more prone to consider violence after viewing pornography, it is possible that a layer of distance is still in effect. For example, one who engages in pretend s&m can be said to be more prone to consider "sadomasochism" than one who merely fantasizes, but there are still layers between that consideration and actually brutalizing someone.

While most of this discussion has focused on teasing out the differences among the levels, it is important also to note how the levels relate and intertwine with one another. Pretend s&m evokes the levels of ritual and lived s&m by the very process of its masquerade. Similarly, fantasy s&m cannot be seen as pure mind activity divorced from the body itself, and from the effects of the fantasy upon the body. An example of a practice that appears to blend the different levels is phone sex. For instance, an advertisement in an alternative paper offers the services of a company called "Fantastic Fantasy" as "Live Unrestricted Telephone Erotica: Bi, Lesbian, Kinky, Domination & Your Favorite Fetishes." [FN115] The domination interaction is likely to resemble fantasy s&m in some respects because of the anonymity and distance the telephone affords, yet it also is likely to resemble pretend s&m because there is an actual interaction. Similarly, some practitioners of ritual s&m have reported that the interaction of the ritual spilled over into their ongoing interactions with their partners. [FN116]

In order to properly recognize the spillovers between the different levels of sadomasochism, it is necessary to first examine each level individually. By separating the layers and defining the distance between each level, the complexity of the connections can be understood without collapsing the layers. In the pornography discussions, however, the layers and distances, which gain even more complexity as they are filtered through the viewing or reading process, often have not been separated.

Not only is there confusion regarding what is meant by the concept of sadomasochism, but there is confusion regarding the role sadomasochism plays in pornography. This confusion parallels the confusion over the *2222 meaning of pornography itself. Through his narrower definition, Sunstein may very well be targeting works that appear to be "about" sadomasochism. However, the idea that pornography is "about" sadomasochism can be interpreted to mean that a particular work: (1) contains sadomasochism as a plot element, (2) advocates sadomasochism, or (3) fosters sadomasochism. These connections correspond, respectively, with the three analytic levels of pornography: definition, message, and effect.

In his attempt to narrow the definition of regulable pornography and in his expression of confidence in the relationship of that definition, the message of pornography, and the effects of that message, Sunstein fails to make these distinctions. Not only does he assume that a particular piece of sexually explicit material will be susceptible to one interpretation‑a woman "enjoying or deserving some form of physical abuse" [FN117]‑but he assumes an equivalence between that interpretation and the infliction of physical harm. From this perspective, any portrayal of a woman enjoying abuse must be a false portrayal in the sense that no woman would ever enjoy such treatment. But the portrayal of abuse is not necessarily the same thing as abuse. Because Sunstein does not take into account the distance both separating the levels of sadomasochism and also created by the pornographic medium, he does not consider the possibility, for instance, that a woman might enjoy viewing a woman enjoying abuse without ever desiring to be physically maltreated herself. This omission precludes a discussion of what the implications are if our sexuality might include that aspect. Unlike the practitioners of, for example, pretend s&m, he does not display an understanding of the potential interconnection of fantasy and reality, but conflates the fantasy with reality.

III. PORNOGRAPHY AND SEXUALITY

If pornography cannot be defined and segregated the way Sunstein proposes, both because of the insight from antipornography feminists that the themes of pornography extend beyond his narrow category and because the meaning of pornography is susceptible to multiple interpretations, it remains to determine what to do about pornography. Precisely because of these feminist insights, which constitute one set of possible interpretations, an assertion that pornography is susceptible to multiple interpretations hardly means that as a medium it is unproblematic. Because the message of pornography is protean and eludes a legislative grasp, I suggest that we give up on attempts to regulate the message. Instead, we need to figure out a way to take the lessons ignored in analyses like Sunstein's and develop new interpretations and reinterpretations, *2223 while attacking real‑life coercion in the production of pornography in a way that recognizes its spillovers into the product itself. Just as the meaning given to the concept of sadomasochism changes on different levels, the interplay between the levels of message and effect in pornography is equally dynamic.

A. REALNESS AND REALITY

In the previous discussion, I suggested that it was important to consider not only how viewing and doing are distinct, but also the ways in which they are interrelated. One of the reasons this analytical process can be confusing is that pornography, like pretend s&m, often plays at being real. Linda Williams has pointed out that pornography strives for a sense of cinematic realism by attempting to convince the audience that what it sees has actually happened. [FN118] But as she also notes, this attempt is not the same as reality:

Our complicity as viewers of the act is different from what it would be if we were actually in the room with the 'object'; it is connected to the fact that we are watching (whether with fascination, pleasure, horror, or dread) an act that seems real but with which we have no physical connection ourselves. [FN119]

Successful pornography, like successful fantasy, will strive to create a feeling of realness for the consumer. But that realness is not the same thing, at all, as reality, although the two are related. It may be the combination of feeling real and not being real that makes the pornographic representation "work" much the way pretend s&m works‑by successfully imitating and not being at the same time. A comedian's impression of a famous figure creates a similar phenomenon. The more life‑like the imitation, the funnier it is. But no matter how life‑like, the imitation is funnier than if the actual person were present. The confusion arises not because pornography is unconnected to reality, but precisely because it is connected to reality in complicated and shifting ways.

Some of these complications are suggested in Williams's discussion of the movie Snuff and the reaction to it. [FN120] The movie, playing off rumors about films purporting to be documents of women murdered while reaching sexual climax, contains a scene at the end that appears to be "more *2224 real" than the rest of the film. After a slasher‑type murder, the picture widens and reveals all the trappings of a movie set. [FN121] Still being filmed, the director of the film that ostensibly has just been completed has sex with an assistant and murders her gruesomely. The sense of realness in this final scene is enhanced by the evidence, the other camera and movie crew, that the previous scenes were artifice, and by a number of other devices such as an off‑ camera voice worrying whether this final scene was successfully captured on film. [FN122] The sense of realness engendered by this presentation was so great that it took a law enforcement interview with the "murdered" actress to reveal that the scene was staged. [FN123]

This play with the sense of realness can make it dangerous to declare with certainty that what is depicted is totally unrelated to real violence or coercion, just as it is unwise to assume that what occurs on the screen is completely consonant with reality. Many pornographic films seem to get much of their kick from the viewer's uncertainty over whether they are documents of reality or staged performances (or a little of both). For example, one film, Bus Stop Tales, [FN124] purports to be a lived encounter between an amateur filmmaker and a woman he picks up at a bus stop. The following lines are flashed across the screen to convince the viewer: "You are about to witness a true real life situation that has been captured on video tape for all the world to see .... One woman's fantasy becomes a welcome reality." [FN125] If we accept this premise, the film becomes a disturbing eyewitness to the coercive powers of seduction as a naive woman is cajoled and persuaded to first disrobe, and then to engage in the stock variety of sexual numbers with the filmmaker. It therefore seems to be "about" coercion in a very direct sense; we are seeing actual coercion happen.

The film further links fantasy to reality by its opening declaration, which implies that nothing could be better than for our fantasies to become reality. This link is similar to the connection many analyses (including the most critical) make between real and depicted coercion. In Bus Stop Tales, however, the "reality" of the film becomes increasingly suspect, as the *2225 woman's inhibitions melt away and she becomes increasingly adept at performing in the variety of positions that she is asked to assume. At the end of the film, we are given additional names of films in which the actress has appeared. Of course, this does not mean coercion did not occur‑it may have, and it makes sense to give models and actresses every tool possible to attack coercion they experience. What it means is that a simple translation from coercion on film to coercion in real life is impossible; the connections are more complicated. The importance of working through the confusion lies not just in an investigative determination about the actual conditions of production, but in an understanding of the factors that account for pornography's appeal to the consumer.

It is perhaps the factor of distance, identified in the discussion on sadomasochism, [FN126] that accounts for the effect of the interplay between reality and unreality that I have called realness. Sunstein displays this distance in his ability to recognize cognitively, through analytical detachment, the supposedly noncognitive nature of pornography. [FN127] Similarly, many who critique pornography are also sophisticated enough to reject the "truth" of images of women enjoying abuse. Instead, our understanding may be that while the pornography depicts a relationship in which a woman enjoys terror and brutalization, we do not believe women would actually derive sexual pleasure from such a situation. I believe the same distance of viewing that allows us to stand back and make that political judgment is at work in making the material erotic in other circumstances, in the same way that distance makes the ritual or pretend levels of sadomasochism erotic to their practitioners. Despite the understanding of distance that we may experience in viewing material defined as pornography, there is also, sometimes, a feeling of horror. That is because although we may know it is not real, it still feels real. [FN128] Something may be utterly fantastic, and yet affect us (whether positively or negatively) because it feels real. Part of the horror in viewing the images of pornography may arise because they too closely resemble real‑life inflictions of power or pain.

This theory, however, does not account for what makes the realness sometimes erotic and sometimes horrific to different people. It is quite possible that pornography may be erotic and horrific at the same time; that it works through a combination of attraction, or identification, and revulsion. *2226 Indeed, it may be that the horror of doing is precisely what makes viewing, at a distance, appealing to some.

Our conclusions about the impact of pornography are bound to be more complicated if someone can be attracted and horrified by the material at the same time. The reason we may feel more comfortable about regulating certain material [FN129] may be because it seems too real and horrible at the same time. Because we are unable to, we assume no one (or at least no one who is not violent and deviant) can maintain the level of distance necessary to enjoy it. For example, the reason that the Massachusetts bill restricts part of its remedy to visual material [FN130] may also be related to the perceived lack of distance in visual material. What makes the horrible stuff disturbing is that its appeal requires not just distance, but distance in combination with breaking down distance, or creating that sense of realness. [FN131]

B. PERSPECTIVES ON SEXUALITY AND THE POTENTIAL FOR TRANSFORMATION

Examining the process of pornography is vital if one accepts both the idea that pornography in its multiple interpretations plays a serious role in the construction of sexuality, and the notion that because of those multiple interpretations, it can play a role in transforming sexuality. Sunstein's ambitions, on the one hand, are far more modest; he merely suggests that "antipornography legislation should produce important social benefits without posing significant threats to a well‑functioning system of free expression." [FN132] Although he sees pornography "as a filter through which men and women perceive gender roles and relationships between the sexes," [FN133] he plays down the impact of eliminating pornography on conditions of inequality because "pornography is only one of a number of conditioning factors, and others are of greater importance. If pornography were abolished, sexual inequality would hardly disappear." [FN134] The image *2227 he portrays is one of a static society in which adjustments in certain social factors can lead to helpful but not transformative reconditioning. This perspective partly explains why he narrows his definition of regulable pornography; for Sunstein, the portrayal of sadomasochism is neither central nor of a piece with other less violent portrayals, but is a particular form of deviance with specific though limited effects that need to be controlled.

MacKinnon and Dworkin, on the other hand, see power, and pornography, as central to sexuality and believe that transformation is possible. However, they do not see power, and, more particularly, the role it plays in pornography, as dynamic, nor as a vehicle for transforming women's sexuality within dominant culture. As a result, they create, ironically, a sharp disjunction between a reality of constructed unequal and power‑laden sexuality and a fantasy of unconstructed and equal sexuality. They account for sexuality by drawing a close connection between one message and one effect, and they use pornography as a vehicle for discussing the entire realm of heterosexuality, in which domination of women by men is the primary characteristic. For example, Andrea Dworkin has written:

It [pornography] is the conditioning of erection and orgasm in men to the powerlessness of women: our inferiority, humiliation, pain, torment; to us as objects, things, or commodities for use in sex as servants.

It sexualizes inequality and in doing so creates discrimination as a sex‑ based practice....

It is women, kept a sexual underclass, kept available for rape and battery and incest and prostitution. [FN135]

The benefit of such a totalizing theory is that it teaches how infused our media and our social interactions are with themes of the erotization of the domination of women by men.

Similarly, Dworkin and MacKinnon underscore the role of power in sexuality. Just as MacKinnon equates pornography with sexuality, she associates sexuality (and thus pornography) with power: "Inequality is what is sexualized through pornography; it is what is sexual about it. The more unequal, the more sexual." [FN136] And it is the process of social construction *2228 that makes power sexy: "Sexual meaning is not made only, or even primarily, by words and in texts. It is made in social relations of power in the world, through which process gender is also produced." [FN137] What's more, the taboos against open acknowledgement of the sexiness of power actually enhance it:

Taboo and crime may serve to eroticize what would otherwise feel about as much like dominance as taking candy from a baby. Assimilating actual powerlessness to male prohibition, to male power, provides the appearance of resistance, which makes overcoming possible, while never undermining the reality of power, or its dignity, by giving the powerless actual power. [FN138]

Sunstein, in an attempt at conciliation with the First Amendment, ignores these important lessons about the role pornography plays. When MacKinnon writes: "Take your foot off our necks, then we will hear in what tongue women speak," [FN139] she suggests that the nature of women's sexuality, as well as women's voice, as it is currently constructed is inextricably intertwined with power.

The problem with these lessons, however, is that they also teach that much of what we see around us and engage in is wrong, perhaps caused by the phenomenon of pornography itself. In doing so, the perspective of MacKinnon and Dworkin necessarily suggests that there is a different, better form of sexuality, entertainment, and romance outside of what we have been experiencing. The analysis relies on the belief, or at least the hope, that eliminating the bad will make way for the good. Indeed, MacKinnon maintains hope for change of a more transformative nature than Sunstein does. While she maintains that sexuality is socially constructed, and criticizes social constructionists who appear to assume that sexuality is something innate that is then conditioned by society, [FN140] she nonetheless believes that socially constructed dominance can be lifted. Referring to the antipornography ordinance she drafted, MacKinnon has maintained:

If this proposal were to become law and if it were to be used, if it were to be given the life in women's hands for which it is designed, there could come a day when she would speak in her own voice and you would hear *2229 her. And I think only then would we understand how unimaginable what she would say is for us now. [FN141]

This means that there will be something new outside of the social constructs that MacKinnon criticizes. Andrea Dworkin similarly envisions a better and different world: "[Equality] has to take the place of subordination in human experience: physically replace it. Equality does not co‑exist with subordination, as if it were a little pocket located somewhere within it. Equality has to win. Subordination has to lose." [FN142]

Judith Butler criticizes positions like those of MacKinnon and Dworkin that suggest a better and different world, maintaining that they divert energy from the more important task of working within the socially constructed world:

If sexuality is culturally constructed within existing power relations, then the postulation of a normative sexuality that is "before," "outside," or "beyond" power is a cultural impossibility and a politically impracticable dream, one that postpones the concrete and contemporary task of rethinking subversive possibilities for sexuality and identity within the terms of power itself. [FN143]

Indeed, the maintaining of two categories‑inside culture and outside‑each seemingly internally coherent, resembles Sunstein's attempt to maintain pornography as a separate category from protectable expression. [FN144] Kate Ellis has pointed out that the juxtaposition of this new social structure with the old exaggerates the coherence of patriarchy itself. [FN145] For those who see pornography's message and effect as static, what Judith Butler calls the "politically impracticable dream" [FN146] would seem like the only hope. However, for those who see message and effect as dynamic, accepting even more seriously than does MacKinnon the role of pornography in the construction of sexuality, the idea of working within the socially constructed world presents the exciting possibilities of deconstructing and reconstructing female and male sexuality, while challenging patriarchy.

MacKinnon's perspective, however, precludes those possibilities. First, she denies women any role in the social construction of their own or of *2230 men's sexuality. Women's sexuality is constructed for them by an invisible hand: "So many distinctive features of women's status as second class ... are made into the content of sex for women." [FN147] The invisible hand appears to be male: " T he interests of male sexuality construct what sexuality as such means, including the standard way it is allowed and recognized to be felt and expressed and experienced, in a way that determines women's biographies, including sexual ones." [FN148] Further, choice for women within this system is an illusion. Indeed, she argues that for lesbians as well as heterosexual women, "sexuality remains constructed under conditions of male supremacy." [FN149] She mocks those who might focus on women's sexual pleasure: "As if women choose sexuality as definitive of identity. As if it is as much a form of women's 'expression' as it is men's." [FN150] According to MacKinnon, analysis that suggests any role for women in the construction of their own sexuality only reinforces the patriarchal culture.

Similarly, in describing the role of power in determining what is sexy, she means only what is sexy to men. While women's sexuality clearly is constructed by this system as well, it is not seen by her as one that, like the man's, takes erotic pleasure in power: "Women also widely experience sexuality as a means to male approval; male approval translates into nearly all social goods. Violation can be sustained, even sought out, to this end." [FN151]

Denying women a role in the construction of their sexuality undermines the concept of social construction because it implies that if women could exercise their will freely, they would express some other version of sexuality, outside of society as we know it. "Take your foot off our necks, then we will hear in what tongue women speak," [FN152] may refer to a woman's voice or sexuality that is not constructed. This sexuality would be removed from and unaffected by pornography and other elements.

Neither MacKinnon nor Sunstein display any confidence that working within the socially constructed world can be transformative, and they share a static view of message and effect. While Sunstein does not share MacKinnon's vision about the possibilities for change nor about the centrality of media images of sexuality, his perspective on sadomasochistic portrayals as marginal to women's sexuality is similar to MacKinnon's confidence that pleasure from power is not part of women's true sexuality. Others have *2231 challenged this view and have suggested ways in which power is sexy to women too. For example, Robin West critiques MacKinnon and Dworkin for their failure to acknowledge the "meaning and the value, to women, of the pleasure we take in our fantasies of eroticized submission." [FN153] Some like Jessica Benjamin and Kate Ellis have attempted to explain psychologically why pleasure can be found by women, as well as men, in submission and power. [FN154] Of the various meanings pornography could have for a variety of audience members, one meaning for women could be pleasure in depictions of power. [FN155]

Neither Sunstein, who sees depictions of sadomasochism as aberrant, nor MacKinnon andDworkin, who see such depictions as endemic to heterosexuality, credit the role power and our understanding of it can play in reconstructing sexuality. Ellis suggests that denying the existence of these interpretations of pleasure in power denies women a role other than victim. [FN156] Similarly, West argues that such denial reflects a failure on the part of the antipornography feminists to recognize and seek to understand what women have said they find pleasurable. [FN157]

A challenge to MacKinnon's view of sexuality as male can be derived from a challenge to a perspective that sees cinema as exclusively male‑oriented. Feminist film critics have argued that the subjectivity of a film, or the viewer's perspective created by the film, is exclusively male because it is voyeuristic, and that women are left to identify only with the objects of the gaze. [FN158] Other feminist film critics have argued instead that members of both genders engage in multiple identifications with both the subject and the object of the gaze. [FN159] These multiple identifications are not equal and symmetrical for men and women, but, according to Tania Modleski, are skewed "by a patriarchal power structure." [FN160] Teresa de Lauretis suggests this interpretation may be further influenced by political and *2232 other contexts:

If gender is a representation subject to social and ideological coding, there can be no simple one‑to‑one relationship between the image of woman inscribed in a film and its female spectator. On the contrary, the spectator's reading of the film ... is mediated by her existence in, and experience of, a particular universe of social discourses and practices in daily life. [FN161]

The analogy to film suggests that pornography can be as dynamic as sexuality. If that is true, then the impact of working within society can seem to be transformative. Not only may the meanings of pornography shift by reference to the shifting role of power in sexuality, but by adjusting and readjusting the images of power in pornography we may help transform sexuality.

C. THE ROLE OF PORNOGRAPHY IN A STRATEGY FOR TRANSFORMING SEXUALITY

Imagine that the totalizing message of antipornography analysis was truly totalizing. Imagine that, by describing what appeals in both pornography and the rest of the media, it described the available terrain for sexuality. For those who accept the MacKinnon and Dworkin dichotomy‑debilitating socially constructed sexuality of the present contrasted with the hint of another, better sexuality awaiting‑that prospect would be a demoralizing and demobilizing dystopia.

The dystopia of a world in which sexuality is defined by and itself defines the elements of pornography need not be demoralizing. Dominant culture is not as uniformly evil as portrayed. If the hope of a yet unknown other quality of sexuality is a source of pleasure for many, it is formed, after all, in response to dominant culture. In other words, the source of the hope can be found within dominant culture itself. Of course, if that is true, an additional risk is revealed‑that the hope I hold out for working within the socially constructed culture is equally constructed and illusive. The only response to that possibility, that I can think of, is to keep constantly in mind the hazards as well as the potential benefits of the material we are trying to reconstruct.

While pornography is an important factor (both in its hard core and softer resonances) in this dominant culture, it is, like the dominant culture itself, not the uniform evil portrayed. Pornography's variable meanings are the building material from which we can form a feminist sexuality‑the *2233 "subversive possibilities for sexuality" Butler talks about. [FN162] This is possible because, as I have argued, we do not have to (and we don't) use pornography at face value. The key factor that distinguishes my more optimistic view about being "trapped" in dominant culture from MacKinnon's position is my belief that women have and will continue to have a role in shaping their sexuality, both heterosexual and lesbian, within dominant culture. There is no need to wait. It is possible to redefine pornography, with a goal toward serving feminist ends. This process can take the form of either using the medium itself to challenge gender roles and gender power, or providing interpretations of existing pornography that expose and exploit such challenges.

Although pornography's variable meanings offer a source of transformation, we cannot just impose on pornography any meaning we want. The existing field of pornography, along with the lessons we have learned from it, composes our entire repertoire of material from which to fashion a transformative feminist pornographic process, one that can take the form of either critique or actual production of pornography. [FN163] If sexuality is made out of (and in turn makes) pornography, and if there is nothing existing outside of our socially constructed culture to turn to, then maybe the only way to transform sexuality is to exploit the many meanings of pornography. Claude Levi‑Strauss's use of the concept of bricolage to represent the constraints imposed on mythmaking can be employed as an analogy for the process of sexuality‑making. Bricolage, as Levi‑Strauss explains, is the activity engaged in by the French bricoleur, a type of handyman who is a combination of craftsman and packrat:

His universe of instruments is closed and the rules of his game are always to make do with 'whatever is at hand', that is to say with a set of tools and materials which is always finite and is also heterogeneous because what it contains bears no relation to the current project, or indeed to any particular project, but is the contingent result of all the occasions there have been to renew or enrich the stock or to maintain it with the remains of previous constructions or destructions. [FN164]

*2234 The process of approaching the project under these conditions would be considered bricolage. Levi‑Strauss characterizes the process of constructing myths as "a kind of intellectual 'bricolage,"' because it similarly "expresses itself by means of a heterogeneous repertoire which, even if extensive, is nevertheless limited." [FN165] Moreover, one cannot opt out of this limited state: " Mythical thought has to use this repertoire, however, whatever the task in hand because it has nothing else at its disposal." [FN166]

Sexual thought, I believe, works the same way. In redefining pornography, we need to perform the bricoleur's task of gathering and retrofitting, simply because we cannot opt out of the constraints. [FN167] However, like the bricoleur or the mythmaker, there is still great creative potential. [FN168] I have previously compared this phenomenon to the Watts Towers in Los Angeles County, an artistic structure built by a man named Simon Rodia who, using shells, tiles, bottles, pipes, and other scavenged material, constructed a series of airy, colorful towers. [FN169] New culture is constructed only out of cultural detritus. While Levi‑Strauss does not address the issues of power and subversion, I believe that one can similarly use the images of power in pornography and other places to subvert power itself. Indeed, we must. Andrew Ross has noted that pleasure must play a role in any attempt to reform the erotic; [FN170] if pornography communicates through pleasure the harmful messages that have been perceived, then it is through pleasure that these messages must be challenged.

Judith Butler's description of how we construct gender strongly resembles Levi‑Strauss's theory of the bricoleur. For her, gender is a socially constructed performance, rather than a set of innate characteristics. She describes a process of performance that allows for a versatility that is similar, both in freedom and constraint, to the process of film watching described by de Lauretis. [FN171] Butler expresses optimism about using such gender performance to undermine current constructs of gender, because *2235 some performances, by playing with the expected gender roles, challenge the notion that our sexuality is innately determined:

Even if heterosexist constructs circulate as the available sites of power/discourse from which to do gender at all, the question remains: What possibilities of recirculation exist? Which possibilities of doing gender repeat and displace through hyperbole, dissonance, internal confusion, and proliferation the very constructs by which they are mobilized? [FN172]

Performance of men in drag is an example she provides of this type of subversive gender performance. [FN173]

Pornography is related to gender performance in a variety of ways. For one, pornography is gender performance, and can be a source both for examining our current repertoire of performances and for expanding that repertoire. In addition, our versatility in gender performance allows us to both enjoy and critique pornography at the same time. Either way, if gender performance is, as Butler suggests, a site for subverting dominant culture, pornography can be an important tool.

Opportunities for gender performance exist in everyday life. For example, a Los Angeles Times article comments upon the exclusive attention paid to men's interests in the automobile industry. These interests are identified as power, speed and handling, that is, the feel of a sports car. [FN174] Ignoring female interests, however, has had apparently little impact on women's willingness to buy cars; indeed, women‑directed efforts like the Dodge LaFemme have all failed commercially. [FN175] My explanation for this disparity is that car driving is one of the more socially acceptable ways that women put on male drag. There is often great pleasure to be found in assuming the attributes of a gender that you typically feel discouraged from assuming. I for one appreciate power, speed, and handling when I drive. [FN176] The newspaper article suggests that car manufacturers avoid directing advertising to women because of fear that men will not buy women's cars. [FN177] I think the same holds true for women customers; they *2236 may often eschew women's cars because of a fondness for putting on male drag when they drive. [FN178]

Of course, there is also great pleasure to be found in assuming the attributes of a gender that you are typically encouraged to assume. For a woman, dressing up for a fancy occasion can also feel like putting on drag. [FN179] Mary Ann Doane has described this process of flaunting femininity as masquerade, which she contrasts with transvestism or the assumption by a woman of the male point of view. [FN180] The strategy of masquerade serves to create distance between a woman and the objectified representation of female gender, thereby avoiding what feminist film criticism has described as an uncomfortable closeness between the female viewer and herself as representation. [FN181] But masquerade also may have a subversive role. It can appropriate a subjective (or viewing at a distance) position for the female spectator, " a s long as the masquerading woman does not forget that the masquerades of femininity and masculinity are not totally unreal or totally a joke but have a social effectivity we cannot ignore." [FN182]

This subjective view is even more complicated: if I were to truly plumb my car driving experience I would have to admit that my pleasure derives from assuming the drag of a woman assuming the drag of a man‑that I feel like the tough, somewhat racy (as in race car) woman that I rarely am on other occasions. Women, probably more than men, have had the experience of gender versatility, either because their choices to experiment with gender are less threatening to patriarchy, or because women are typically faced with the double bind of being bereft of female subjects as role models while still being expected to be actors in the world. Versatility with gender may be an appealing method of negotiating this double bind. [FN183] Nonetheless there are dangers in this versatility. Jane Gallop points out one difficult possibility in academic discourse:

*2237 Postmodernist thinkers are defending against the downfall of patriarchy by trying to be not male. In drag, they are aping the feminine rather than thinking their place as men in an obsolescent patriarchy. The female postmodernist thinker finds herself in the dilemma of trying to be like Daddy who is trying to be a woman. The double‑cross is intriguing and even fun, but also troubling if one suspects that it is the father's last ruse to seduce the daughter and retain her respect, the very respect that legitimized the father's rule. [FN184]

Pornography is yet another site, both in terms of viewing and producing, where we can pick and choose among sexual roles, try them on for size, critique or turn them inside out. But unlike a trip through the vintage clothing store where the previous wearer is irrelevant, [FN185] the meaning of the sex roles we select can be multiple and fraught with ambiguity based on their previous and continuing uses. Historical and ongoing gender power relations, for example, provide a constraint that can inform the meaning of the sex roles assumed, although these same power relations can provide the basis for reinterpretation as well. [FN186]

To the extent that effective enforcement of the model and actress coercion provision can help reduce coercion in the production of pornography, it can play an important role in helping to open up pornography to competing interpretations. Duncan Kennedy has argued, "In the extreme case, abolishing real life male sexual abuse of women would reduce the dangerousness of" the process of using costume and fantasies for both pleasure and resistance, "by cutting the connections between rape fantasies and real life rape, between incest fantasies and incest." [FN187] To the extent that coercion in the production of pornography is not reduced, it forces us to confront the spillovers from real life coercion into the product being viewed; for example, knowledge about the coercion experienced by actress Linda Marchiano in the making of Deep Throat [FN188] can affect the meaning and interpretation of the images from that film. Attempts to change these types of working conditions may be valuable not only for the crucial impact they have on the lives of models and actresses, but also *2238 because these efforts may allow interpretations of what is viewed to be more easily separated from what is done.

Carl Stychin argues that gay male pornography is already a locus in which traditional gender roles are being challenged through these types of interpretations and reinterpretations. He argues that "gay male pornography undermines rather than reinforces male supremacy because it questions the coherence of sexual subjectivity." [FN189] In other words, if a prime characteristic of heterosexual pornography is male subjectivity and female objectivity, gay male pornography, by presenting men in both dominant and submissive positions, challenges the idea that subjectivity is tied to gender. He refutes MacKinnon's suggestion that dominance and submission reinforce the gendered nature of sexuality, no matter what the gender of the participants, as well as the contention that valorization of masculinity in gay male pornography may also only reinforce the valorization of masculinity generally. [FN190] In this respect, he presents alternative messages and alternative effects for the material he considers.

Stychin, with Judith Butler, argues that gay male pornography, by disconnecting the gender of the participants from their socially ascribed gender, reveals gender as performance: "The replication of heterosexual constructs in non‑heterosexual frames brings into relief the utterly constructed status of the so‑called heterosexual original." [FN191] Further, " i f performance reveals the artificiality of gender identity, it also undermines hierarchical gendered arrangements." [FN192] In contrast, MacKinnon accepts the performance aspect of switching gender roles but disagrees about its potential.

The capacity of gender reversals ... to stimulate sexual excitementis derived precisely from their mimicry or parody or negation or reversal of the standard arrangement. This affirms rather than undermines or qualifies the standard sexual arrangement as the standard sexual arrangement, the definition of sex, the standard from which all else is defined, that in which sexuality as such inheres. [FN193]

*2239 Butler disputes this notion:

If the anatomy of the performer is already distinct from the gender of the performer, and both of those are distinct from the gender of the performance, then the performance suggests a dissonance not only between sex and performance, but sex and gender, and gender and performance. As much as drag creates a unified picture of "woman" (what its critics often oppose), it also reveals the distinctness of those aspects of gendered experience which are falsely naturalized. [FN194]

Playing with parody, however, runs the risk of reinforcing stereotypes at the same time that it creates the possibility of transformation. [FN195] While these parodies, like the practice of sadomasochism itself, can be seen to be empowering explorations of power, they may also serve to legitimate roles of power and powerlessness. It would be wrong to simply argue for another unitary set of messages and effects; the alternatives Stychin proposes most likely exist in combination and tension with the readings MacKinnon offers.

While recognizing the potential for more than just gay male parodies of gendered sex, Stychin focuses on the uniqueness of the gay male challenge. [FN196] However, heterosexual as well as lesbian pornography may also make these challenges. The potential has been recognized in mainstream cinema. Bette Gordon argues that her film Variety [FN197] manipulates conventions of pornography to give a woman regarding pornography the position of subject and voyeur. In so doing, " t he film suggests that women, even in patriarchal culture, are active agents who interpret and utilize cultural symbols on their own behalf." [FN198] Similarly, at least one of Madonna's videos makes these types of challenges by manipulating the images of traditional gender roles and distributions of power. [FN199]

CONCLUSION: A FEMINIST PORNOGRAPHIC PROCESS

A willingness to consider the tools pornography has to offer does not preclude sustained critique. John Fiske sees television as a potentially *2240 subversive medium because it maintains tension between its ostensible control by a dominant ideology and openness to multiple meanings. [FN200] Pornography similarly appears to be controlled by a dominant ideology of sexual subordination, yet is open to multiple meanings. Mariana Valverde has argued that pornography is too open to interpretation to be amenable to classification, but "is a process because it necessarily establishes, and is established by, a particular set of relations between producer and consumer, between consumer and his or her social context, and between the social context and the producer." [FN201]

In the piece left unfinished at her death, Mary Joe Frug expresses concern over the implications of a critique of the movement to regulate pornography: "I do not want to be understood as a pornography apologist, for I believe that the proliferation and character of the pornography genre is one of the most complicated cultural events of our time, an event whose meanings are still quite indeterminate." [FN202] It is often necessary to make this kind of emphasis because, as Frug points out, arguments around pornography have tended to polarize; either you find pornographic depictions of sex degrading, or you find them liberatory. [FN203] This polarization is similar to the all or nothing positions surrounding the model and actress coercion provision. Such division is apparent not only in the debate between those supporting the model ordinance and those supporting the First Amendment position, but also among those who reject both sides of that debate. Feminist discourse critics, those who according to Mary Joe Frug seek "to disrupt, reverse, or transform" pornography by analyzing its ambiguities and "points of resistance," [FN204] need to integrate ideas of both the potential degradation and potential liberation in pornography in order to account fully for its meanings and effects. [FN205]

*2241 While discussing the transformative potential of pornographic images I have not only warned of the dangerous implications of those images, but also have noted that their transformative potential lies in the very elements that make them dangerous. [FN206] Although MacKinnon does recognize the role of power in the construction of sexuality, she disagrees about the extent to which power is an inseparable but ever‑shifting part of sexuality. She dismisses the possibility of an analysis that recognizes the power structure of sexuality while considering its positive possibilities. [FN207] The transformative potential is also dismissed by Sunstein, because he fails to consider the role of power in pornography as MacKinnon does‑central to sexuality rather than a deviant aspect subject to minor tinkering.

The connections between reality and image do need to be explored, but it is precisely by treating them differently that we can explore these connections. Because image and fantasy‑in a variable way‑do help form and are formed by real life power relationships, focusing on the model and actress provision as a means of manipulating both reality and image (by altering the working conditions under which pornographic images are made) seems necessary. Pornography as a process, the opportunities the medium affords for interpretation and reinterpretation, cannot be divorced from the context in which the material is produced. Videos like Madonna's may be so susceptible to empowering interpretations precisely because she is perceived as a powerful media mogul. In a very different way, my use of an analogy to the Watts Towers takes on meanings from their connection with the community of Watts; this may be one of empowerment and community building within Watts, [FN208] or one that taps into lingering outside stereotypes about the Watts community. [FN209]

At the same time, the interpretations affect the context; Madonna's public image outside her videos is affected by her image in the videos, public perception of Watts is affected by the towers themselves. This *2242 suggests that addressing coercion in the production of pornography must work together with a process of readjusting pornography's images. On the one hand, empowering interpretations of pornography's meaning may be made more easily when disempowerment is not part of its production. On the other hand, coercion in the production of pornography, and how we define such coercion, may be affected by how we interpret the meaning of the product.

This last point is problematic. Those advocating regulation of pornography's message as well as the means of its production have warned that if we delude ourselves into believing that images of disempowerment can be reread as empowering, we will also be deluded into believing that real life coercion has not occurred. [FN210] It is true, however, that our definition of coercion is affected by how we perceive the message of pornography; those who see the message as unitary and negative may very well believe that no one can be said to have consented to the production of such material, no matter what her indication. Those who see pornography's message as more variable may be more likely to credit those who indicate a willingness to appear in such material with having made a similar interpretation, while recognizing that "willingness" is not merely a matter of signing a consent form. [FN211] Finally, reinterpretation of the images of pornography should be part of a program to improve actual working conditions.

Instead of seeking meaning for pornography by establishing an ill‑founded coherence among definition, message, and effect, "pornography" should instead be understood as a process by which images of sexuality are pasted and repasted together, both in the creation of the product and in the process of viewing and critiquing, so that we create and recreate meaning about human sexuality and gender. By assuming that pornography, once produced, arrives on the screen and stays in the mind with one message, we cede the process uncontested (at least with respect to those who produce visual material) to an industry that often maintains conditions deplorable for the women who work there. The response to pornography's images does not have to be systematic, but may be variable depending on the context. By seizing the opportunity to interpret and reinterpret the many levels on which pornography's meaning operates, while attacking coercion, we can have a role in redefining sexuality through what may be the only means available.

[FNa]. Assistant Professor of Law, Western State University College of Law. Copyright reserved. J.D., Harvard Law School. Many thanks to Linda Chapin, Karen Engle, Neil Gotanda, Judith Greenberg, Mark Levine, and Gary Peller for helpful comments, suggestions, and assistance. Thanks also to the other participants in the Pornography Panel at the 1992 Critical Networks Conference‑ Kathryn Abrams, Dianne Brooks, Lynne Henderson, Hester Lessard, and Carl Stychin‑for inspiration, ideas, and questions. And many thanks again to Mary Joe Frug for her continuing influence and to Duncan Kennedy for all his help.

[FN1]. Model Anti‑pornography Law § 3.1, in Andrea Dworkin, Against the Male Flood: Censorship, Pornography, and Equality, 8 HARV. WOMEN'S L.J. 1, 25‑ 26 (1985) (generally referred to as the model ordinance).

[FN2]. Pornography is defined as

the graphic sexually explicit subordination of women through pictures and/or words that also includes one or more of the following: (i) women are presented dehumanized as sexual objects, things, or commodities; or (ii) women are presented as sexual objects who enjoy pain or humiliation; or (iii) women are presented as sexual objects who experience sexual pleasure in being raped; or (iv) women are presented as sexual objects tied up or cut up or mutilated or bruised or physically hurt; or (v) women are presented in postures or positions of sexual submission, servility, or display; or (vi) women's body parts‑including but not limited to vaginas, breasts, or buttocks‑are exhibited such that women are reduced to those parts; or (vii) women are presented as whores by nature; or (viii) women are presented being penetrated by objects or animals; or (ix) women are presented in scenarios of degradation, injury, torture, shown as filthy or inferior, bleeding, bruised, or hurt in a context that makes these conditions sexual.

Id. § 2.1, at 25.

[FN3]. 771 F.2d 323 (7th Cir. 1985), aff'd per curiam, 475 U.S. 1001 (1986).

[FN4]. See Carole Vance, More Danger, More Pleasure: A Decade After the Barnard Sexuality Conference, in PLEASURE AND DANGER: EXPLORING FEMALE SEXUALITY xxii‑xxiii (Carole Vance ed., 1984, new introduction 1992) (criticizing scholars for presenting feminist positions on sexuality as more polarized than they are).

[FN5]. Cass R. Sunstein, Neutrality in Constitutional Law (with Special Reference to Pornography, Abortion, and Surrogacy), 92 COLUM. L. REV. 1, 18 (1992).

[FN6]. Mary Joe Frug, A Postmodern Feminist Legal Manifesto (An Unfinished Draft), 105 HARV. L. REV. 1045, 1067 (1992).

[FN7]. See infra text accompanying notes 9‑16.

[FN8]. Frug, supra note 6, at 1068‑69.

[FN9]. S. 1521, 102d Cong., 2d Sess. (1992).

[FN10]. Miller v. California, 413 U.S. 15, 24 (1972).

[FN11]. American Booksellers Ass'n v. Hudnut, 771 F.2d 323, 325 (7th Cir. 1985), aff'd per curiam, 475 U.S. 1001 (1986).

[FN12]. The stated purpose of the legislation is: "To provide a cause of action for victims of sexual abuse, rape, and murder, against producers and distributors of hardcore pornographic material." S. 1521, supra note 9, pmbl.

[FN13]. H.B. 5194, 177th General Court, § 1, Mass. (1992).

[FN14]. Id.

[FN15]. Id. § 2.

[FN16]. See Cass R. Sunstein, Pornography and the First Amendment, 1986 DUKE L.J. 589, 625 (suggesting that an effective strategy to guard antipornography legislation from overbreadth might be to limit its reach to visual material because research indicates film and photographs are most closely linked to violence).

[FN17]. Id. at 591.

[FN18]. Id. at 591‑92.

[FN19]. Model Anti‑pornography Law § 3.1, in Dworkin, supra note 1, at 25‑ 26.

[FN20]. Id. § 3.4, at 27.

[FN21]. Id. § 3.2, at 26. There is also a cause of action available against a person who forces pornography on another. Id. § 3.3, at 26‑27.

[FN22]. Sunstein, supra note 16, at 595.

[FN23]. Model Anti‑pornography Law § 3.1, in Dworkin, supra note 1, at 25‑ 26.

[FN24]. 458 U.S. 747 (1982).

[FN25]. Id. at 761 (citing Miller v. California, 413 U.S. 1, 15 (1972)).

[FN26]. Id. at 751; see also United States v. X‑Citement Video, Inc., 982 F.2d 1285, 1290‑92 (9th Cir. 1992) (holding child pornography law unconstitutional when it contains no requirement for a finding that distributors are aware that the actor is underage).

[FN27]. See Ferber, 458 U.S. at 759 (suggesting that older persons or simulations could be used to avoid the statute's reach).

[FN28]. Id. (footnote omitted).

[FN29]. Sunstein, supra note 16, at 596.

[FN30]. CATHARINE MACKINNON, FEMINISM UNMODIFIED 182‑83 (1987).

[FN31]. As the Court in Ferber noted, "The legislative judgment, as well as the judgment found in the relevant literature, is that the use of children as subjects of pornographic materials is harmful to the physiological, emotional, and mental health of the child." 458 U.S. at 758.

[FN32]. Nan Hunter & Sylvia A. Law, Brief Amici Curiae of Feminist Anti‑ Censorship Taskforce et al., in American Booksellers Ass'n v. Hudnut, 21 U. MICH. J.L. REF. 69, 128 (1987) (arguing that treating women like children in terms of consent undermines women's ability to enter into agreements and negotiate for favorable terms).

[FN33]. Id. at 127; see Model Anti‑pornography Law § 3.1, in Dworkin, supra note 1, at 25‑26 (defining coercion and setting forth a number of circumstances, including direct manifestations of consent, that "shall not, without more, negate a finding of coercion").

[FN34]. See Elizabeth Schneider, Describing and Changing: Women's Self‑ Defense Work and the Problem of Expert Testimony on Battering, 9 WOMEN'S RTS. L. REP. 195, 198‑200 (1986) (suggesting that homicide cases in which expert testimony is offered to support a battered woman's self‑defense claim "pose the dilemma of how we describe both victimization and agency in women's lives").

[FN35]. See, e.g., MACKINNON, supra note 30, at 179‑83; Hunter & Law, supra note 32, at 127; see also Hester Lessard, Obscenity and the Politics of Meaning (unpublished manuscript, on file with The Georgetown Law Journal) (criticizing the failure of civil liberties groups to recognize the coercive aspects of the pornography industry in their opposition to Canada's obscenity law, as well as the failure of proponents of the law to recognize the oppressive aspects of government enforcement).

[FN36]. "The ban on distribution of works containing coerced performance is limited to pornography; coercion is irrelevant if the work is not 'pornography,' and we have held the definition of 'pornography' to be defective root and branch." American Booksellers Ass'n v. Hudnut, 771 F.2d 323, 333 (7th Cir. 1985).

[FN37]. See, e.g., Sunstein, supra note 16, at 602. Sunstein does recognize a difference between the provisions:

It is important to recognize that the various different harms point to different avenues for legal regulation. If the harm to women who participate in pornography is emphasized, regulation will depend on whether such harm has occurred. If the causal connection is emphasized, the question will be whether the material at issue is likely to cause sexual violence and subordination.

Id. However, he sees the questions and the rationales behind them as being closely related.

[FN38]. For a discussion of self‑identified feminist pornographers, see Laura Fraser, Nasty Girls, MOTHER JONES, Feb.‑Mar. 1990, at 32.

[FN39]. An obvious example of pornography made without the coercion of actual individuals is an interactive computer program called MacPlaymate. See generally Patt Morrison, Fun and Games or Compu‑Porn, L.A. TIMES, Nov. 14, 1988, § 5, at 1 (describing mixed reactions to "interactive erotica," which permits user to disrobe an animated woman named Maxie, cause her to engage in sex acts with another animated woman or deploy various paraphernalia, like handcuffs, upon her).

[FN40]. The image I have in mind is of three translucent layers, like three identical sheets of plexiglass. When viewed from one perspective, straight down from the top for instance, they seem to compose only one layer. My aim is to shift the perspective.

[FN41]. MACKINNON, supra note 30, at 172 ("[Pornography] institutionalizes the sexuality of male supremacy, fusing the erotization of dominance and submission with the social construction of male and female."); Dworkin, supra note 1, at 17.

[FN42]. Dworkin, supra note 1, at 13‑14.

[FN43]. Id. at 17.

[FN44]. 771 F.2d at 329.

[FN45]. See H.B. 5194, supra note 13, § 2(a).

[FN46]. Sunstein, supra note 16, at 592.

[FN47]. Id.

[FN48]. Id. at 595.

[FN49]. Id. at 606‑08.

[FN50]. Id. at 592.

[FN51]. Id. at 597.

[FN52]. Id.

[FN53]. Id. at 592.

[FN54]. Id.

[FN55]. Id.

[FN56]. Dworkin makes a similar connection. See supra text accompanying note 43.

[FN57]. Sunstein, supra note 16, at 592 n.27 (quoting Edward Donnerstein, Pornography: Its Effect on Violence Against Women, in PORNOGRAPHY AND SEXUAL AGGRESSION 53, 78‑79 (Neil M. Malamuth & Edward Donnerstein eds., 1984)).

[FN58]. Id. at 592.

[FN59]. Robin West, The Difference in Women's Hedonic Lives: A Phenomenonological Critique of Feminist Legal Theory, 3 WIS. WOMEN'S L.J. 81, 137‑38 (1987).

[FN60]. As Andrea Dworkin has written: "[Pornography] creates systematic harm to women in the form of discrimination and physical hurt. It creates harm inevitably by its nature because of what it is and what it does. The harm will occur as long as it is made and used." Dworkin, supra note 1, at 11.

[FN61]. See, e.g., ANDREA DWORKIN, Introduction, in PORNOGRAPHY: MEN POSSESSING WOMEN xxxvi (1979, new introduction 1992) (describing pornography depicting Jewish victims and Nazi torturers); Catharine A. MacKinnon, Pornography as Defamation and Discrimination, 71 B.U. L. REV. 793, 797 (1991) (describing pornography as replete with torture, insults, and racial victimization); Sunstein, supra note 16, at 593 (describing advertisement that depicts a naked woman as a hunting trophy ready to be "stuffed and mounted").

[FN62]. See MACKINNON, supra note 30, at 181.

[FN63]. Id. at 149; West, supra note 59, at 118.

[FN64]. See MACKINNON, supra note 30, at 137 (explaining the connections between the images in Playboy and erotization of forced sex). After this article was already on its way to press, I discovered Jeanne Schroeder's recent and interesting critique of Sunstein's version of MacKinnon's pornography analysis, Jeanne L. Schroeder, The Taming of the Shrew: The Liberal Attempt to Mainstream Radical Feminist Theory, 5 YALE J.L. & FEMINISM 123 (1992). She raises issues similar to those contained in this section by arguing that Sunstein trivializes MacKinnon's perspective through his focus on violent pornography only. See id. at 126‑35.

[FN65]. 771 F.2d at 329.

[FN66]. Sunstein, supra note 16, at 608. Elsewhere he expresses confidence that a legal definition can capture this distinction: "[T]here is little reason to doubt that a carefully worded statute, posing no greater threat to free expression than the other categories of regulable speech, could be drawn." Cass R. Sunstein, Feminism and Legal Theory, 101 HARV. L. REV. 826, 845 (1988) (reviewing MACKINNON, FEMINISM UNMODIFIED, supra note 30).

[FN67]. Id.

[FN68]. Dworkin has written that protecting the speech of pornographers "means protecting what they do to us, how they do it. It means protecting their sadism on our bodies, because that is how they write: not like a writer at all; like a torturer." Dworkin, supra note 1, at 19.

MacKinnon similarly argues: "Pornography is historically defended in the name of freedom of speech. I am here to speak for those, particularly women and children, upon whose silence the law, including the law of the First Amendment, has been built." MACKINNON, supra note 30, at 204. But cf. id. at 192‑94 (arguing that regulating pornography is consistent with exceptions to First Amendment protection).

[FN69]. Sunstein, supra note 16, at 607 (footnote omitted).

[FN70]. Id. at 606.

[FN71]. Id.

[FN72]. Id. at 607.

[FN73]. Paul Chevigny, Pornography and Cognition: A Reply to Cass Sunstein, 1989 DUKE L.J. 420, 422.

[FN74]. Id. at 425.

[FN75]. Id. at 426.

[FN76]. Id. at 430.

[FN77]. In her casebook, Mary Joe Frug uses the analogy to racist hate speech in part 'to dilute the power of First Amendment arguments in the pornography debates." MARY JOE FRUG, WOMEN AND THE LAW 685 (1992).

[FN78]. Mari J. Matsuda, Legal Storytelling: Public Response to Racist Speech: Considering the Victim's Story, 87 MICH. L. REV. 2320, 2357 (1989).

[FN79]. Sunstein, supra note 16, at 608.

[FN80]. Matsuda, supra note 78, at 2357.

[FN81]. Id.

[FN82]. Chevigny, supra note 73, at 431.

[FN83]. Id. at 426.

[FN84]. Matsuda, supra note 78, at 2357 (emphasis added).

[FN85]. Sunstein, supra note 5, at 13‑14.

[FN86]. Id. at 19.

[FN87]. Id.

[FN88]. Id. at 26. Sunstein launches a similar critique against an antipornography position that is based on religious convictions about the purity of sexual relations. Id. at 19. Interestingly, he depicts these as the two extreme positions and posits the antipornography feminist position, with which he sees himself aligned, as the third alternative. Id.

[FN89]. Sunstein's critique does not apply to all of the positions taken in opposition to pornography regulation. These other positions tend to argue that the messages of pornography may in fact be mixed. For example, the amicus curiae brief opposing the regulation at issue in Hudnut maintains:

Depictions of ways of living and acting that are radically different from our own can enlarge the range of human possibilities open to us and help us grasp the potentialities of human behavior, both good and bad.... Such an enlarged vision of possible realities enhances our human potential and is highly relevant to our decision‑making as citizens on a wide range of social and ethical issues.

Hunter & Law, supra note 32, at 120. Yet in other places these authors recognize that sexually explicit speech can be interpreted as misogynistic, as well as liberatory. See, e.g., id. at 121 (recognizing misogyny as one among many messages that sexually explicit material may deliver, but suggesting that "[e]ven pornography which is problematic for women can be experienced as affirming of women's desires and of women's equality").

Indeed, a position most similar to the one Sunstein critiques can best be found in some pornography. In the film Deep Throat II, the views of antipornography commissioners change when they are encouraged to engage in a wide variety of sexual activities. The movie suggests, in a closing fantasy scene of a copulating couple being arrested by "sex police," that putting restraints on pornography will result in restraints on sex itself. DEEP THROAT II (Arrow 1987). The problem with this position, as Sunstein points out, is that it assumes an unmediated connection between pornography and sexual activity. Sunstein, supra note 5, at 19.

[FN90]. Sunstein, supra note 16, at 606.

[FN91]. Id.

[FN92]. Id. at 607.

[FN93]. See, e.g., Matsuda, supra note 78, at 2376.

[FN94]. See supra text accompanying note 84.

[FN95]. See THE LOONEY TUNES VIDEO SHOW #1 (Warner Bros. 1982). A more difficult example is PEPE LE PEW, GOLDEN JUBILEE 24 KARAT COLLECTION: PEPE LE PEW'S SKUNK TALES (Warner Bros. 1986), the cartoon skunk who in each episode not only mistakes a black cat (who has accidentally acquired a white stripe) for a female skunk, but interprets her terror and efforts to escape him as desire. It might, however, be wrong to assume that the cartoon's only message is to trivialize sexual harassment and rape, because the humor involves a process in which the audience recognizes the error Pepe makes both as to the cat's identity as well as to her consent.

[FN96]. John Fiske, Moments of Television: Neither the Text nor the Audience, in REMOTE CONTROL: TELEVISION, AUDIENCES, AND CULTURAL POWER 56, 56 (Ellen Seiter et al. eds., 1989).

[FN97]. Id.

[FN98]. Frug, supra note 6, at 1074.

[FN99]. Id.; see also Hunter & Law, supra note 32, at 106 (arguing that messages from sexually explicit material "are often multiple, contradictory, layered and highly contextual"); Kate Ellis, I'm Black and Blue from the Rolling Stones and I'm Not Sure How I Feel About It: Pornography and the Feminist Imagination, SOCIALIST REV., May‑Aug. 1984, at 103 (arguing for a feminist sexuality that recognizes a distinction between fantasies or role playing and real violence).

[FN100]. See MACKINNON, supra note 30, at 184.

[FN101]. For summaries of research results, see MacKinnon, supra note 61, at 799‑800; Sunstein, supra note 16, at 598.

[FN102]. For instance, the Feminist Anti‑Censorship Taskforce brief attacks these studies on a methodological basis, for their focus on only violent imagery and for their lack of uniformity. Hunter & Law, supra note 32, at 112‑ 13. Sunstein also acknowledges that the studies are imperfect in establishing the connection between pornography and violence, Sunstein, supra note 16, at 598, but believes that the probabilities of harm are serious enough to justify action. Id. at 601‑02.

[FN103]. For example, Sunstein summarizes one set of results as follows: "Men questioned after [pornography] exposure seem more prepared to accept rape and other forms of violence against women, to believe that women derive pleasure from violence, and to associate sex with violence; they also report a greater likelihood of committing rape themselves." Sunstein, supra note 16, at 598.

[FN104]. And, of course, it should be added, viewing that leads to thinking of doing involves only a selected repertoire‑the more violent material‑ available for action under the model ordinance. Hunter & Law, supra note 32, at 113.

[FN105]. LINDA WILLIAMS, HARD CORE: POWER, PLEASURE, AND THE "FRENZY OF THE VISIBLE" 195 (1989) (suggesting the complications that arise in viewing representations of sadomasochism because of their nature as acts as well as illusions); see also infra text accompanying notes 115‑116.

[FN106]. Sunstein, supra note 16, at 592.

[FN107]. See generally Lenore E. Walker, Battered Women and Learned Helplessness, 2 VICTIMOLOGY 525 (1977‑78) (maintaining that a socialization process she terms "learned helplessness," rather than sadomasochism, explains the continuation of relationships marked by chronic domestic violence).

[FN108]. BARBARA EHRENREICH ET AL., REMAKING LOVE 126 (1986).

[FN109]. L.A. WEEKLY, June 5, 1992, at 117.

[FN110]. Jessica Benjamin, Master and Slave: The Fantasy of Erotic Domination, in POWERS OF DESIRE 280, 291‑92 (Ann Snitow et al. eds., 1983) [hereinafter POWERS OF DESIRE].

[FN111]. Id. at 283.

[FN112]. West, supra note 59, at 129 (discussing PAULINE REAGE, THE STORY OF O (1954)).

[FN113]. EHRENREICH ET AL., supra note 108, at 129.

[FN114]. Ellis, supra note 99, at 20‑21 (arguing that in pornography or fantasy "the 'at a distance' factor puts erotic stimuli in a different category from stimuli that elicit violent behavior").

[FN115]. L.A. WEEKLY, Jan. 29, 1993, at 141.

[FN116]. See Marissa Jonel, Letter from a Former Masochist, in AGAINST SADOMASOCHISM: A RADICAL FEMINIST ANALYSIS 16, 18 (Robin R. Linden et al. eds., 1982).

[FN117]. Sunstein, supra note 16, at 592.

[FN118]. See, e.g., WILLIAMS, supra note 105, at 101 (describing the "money shot" as an attempt to prove sexual pleasure); id. at 203 (suggesting that depictions of inflicted pain may work the same way).

[FN119]. Id. at 188; see also id. at 147 (explaining the combination of authenticity and phoniness at work in many pornographic sexual numbers).

[FN120]. Id. at 189.

[FN121]. Id. at 191.

[FN122]. Id. at 191‑92.

[FN123]. Id. at 193. For a send‑up of this type of framing device, consider the Monty Python television comedy sketch in which four lost explorers are relieved to discover they are not truly alone because of the presence of the camera crew following them. As the camera that forms the viewer's point of view pans back to reveal the first camera crew, one explorer exclaims: "If this is the crew who were filming us ... who's filming us now?" We cut, of course, to reveal two sets of camera crews, and presumably a third one still behind the camera through which we see the scene. GRAHAM CHAPMAN ET AL., THE COMPLETE MONTY PYTHON'S FLYING CIRCUS: ALL THE WORDS 85 (1989).

[FN124]. BUS STOP TALES (Privert Video 1989).

[FN125]. Id.

[FN126]. See supra text accompanying notes 110‑114.

[FN127]. See supra text accompanying note 92.

[FN128]. See WILLIAMS, supra note 105, at 188 (arguing that the problem in analyzing pornography is not "that the image of violence is the same as if it were happening before our very eyes; rather, it is that the spectacle seems both so real and yet so distant from us, both temporally and spatially").

[FN129]. See supra text accompanying note 59.

[FN130]. See supra text accompanying note 15.

[FN131]. WILLIAMS, supra note 105, at 188.

[FN132]. Sunstein, supra note 16, at 627.

[FN133]. Id. at 601.

[FN134]. Id. As Sunstein explains:

One need not believe that the elimination of violent pornography would bring about sexual equality, eliminate sexual violence, or change social attitudes in a fundamental way, in order to agree that a regulatory effort would have some effect in reducing violence and in diminishing views that contribute to existing inequalities.

Sunstein, supra note 5, at 26. Cf. Sunstein, supra note 66, at 847 (suggesting that "the relationship between sex discrimination and sexual sadism is hardly simple and clear cut"). Jeanne Schroeder argues that Sunstein presents in his book review a weakened version of MacKinnon's pornography analysis, one that avoids a threat to male power. Schroeder, supra note 64, at 135.

[FN135]. Dworkin, supra note 1, at 10.

[FN136]. CATHARINE MACKINNON, TOWARD A FEMINIST THEORY OF THE STATE 143 (1989). Andrea Dworkin similarly maintains that "pornography, by sexualizing [hierarchy,] makes it dynamic, almost carnivorous, so that men keep imposing it for the sake of their own sexual pleasure‑for the sexual pleasure it gives them to impose it." Dworkin, supra note 1, at 16.

[FN137]. Id. at 129.

[FN138]. Id. at 133.

[FN139]. MACKINNON, supra note 30, at 45.

[FN140]. MACKINNON, supra note 136, at 131‑32.

[FN141]. Id. at 196.

[FN142]. Dworkin, supra note 1, at 22.

[FN143]. JUDITH BUTLER, GENDER TROUBLE 30 (1990).

[FN144]. See supra Part II.C.2.

[FN145]. Ellis, supra note 99, at 115 ("By carrying it to an extreme [antipornography feminists] surround us with ... a patriarchal system of which all the parts are equally developed and fit perfectly together.").

[FN146]. BUTLER, supra note 143, at 30.

[FN147]. MACKINNON, supra note 136, at 130 (emphasis added).

[FN148]. Id. at 129. See Schroeder, supra note 64, at 174‑75 (suggesting that this emphasis on male agency in constructing sexuality is inconsistent with a theory of social construction).

[FN149]. Id. at 141‑42.

[FN150]. Id. at 135.

[FN151]. Id. at 147.

[FN152]. MACKINNON, supra note 30, at 45; see supra text accompanying note 139.

[FN153]. West, supra note 59, at 118. West, however, believes that sexual submission has value only "when it is an expression of trust." Id. at 129.

[FN154]. Benjamin, supra note 110, at 281; Ellis, supra note 99, at 120‑21.

[FN155]. Hunter & Law, supra note 32, at 129.

[FN156]. Ellis, supra note 99, at 114; see also Hunter & Law, supra note 32, at 102.

[FN157]. West, supra note 59, at 117‑18.

[FN158]. See Laura Mulvey, Visual Pleasure and Narrative Cinema, in ISSUES IN FEMINIST FILM CRITICISM 28, 33‑34 (Patricia Erens ed., 1990) [hereinafter ISSUES IN FEMINIST FILM CRITICISM] (discussing the "active/male" and "passive/female" roles in films in which women are "simultaneously looked at and displayed," while the man represents "the main controlling figure with whom the spectator can identify").

[FN159]. Tania Modleski, Hitchcock, Feminism, and the Patriarchal Unconscious, in ISSUES IN FEMINIST FILM CRITICISM, supra note 158, at 66‑69. This multiple identification has been called a "bisexuality" of viewing. Id. at 62‑63.

[FN160]. Id. at 67.

[FN161]. TERESA DE LAURETIS, TECHNOLOGIES OF GENDER 96 (1987).

[FN162]. BUTLER, supra note 143, at 30‑31.

[FN163]. One of the antipornography campaign's insights has been that these qualities of pornography extend as well to the full range of media artifacts. See Dworkin, supra note 61, at 23 ("The celebration of rape in story, song, and science is the paradigmatic articulation of male sexual power as a cultural absolute."); see also MARIANA VALVERDE, SEX, POWER AND PLEASURE 143‑44 (1987) (stressing importance of connecting pornography to its larger social context).

[FN164]. CLAUDE LEVI‑STRAUSS, THE SAVAGE MIND 17 (George Weidenfeld trans., 1966).

[FN165]. Id.

[FN166]. Id.

[FN167]. Judith Butler has stated: "There is no self that is prior to the convergence or who maintains 'integrity' prior to its entrance into this conflicted cultural field. There is only a taking up of the tools where they lie, where the very 'taking up' is enabled by the tool lying there." BUTLER, supra note 143, at 145.

[FN168]. Similar observations have been made about blues music performance. There are constraints in structure and available chords, as well as traditional topics, but great creativity is possible. SAMUEL CHARTERS, THE LEGACY OF THE BLUES 56 (1975); PETER GURALNICK, FEEL LIKE GOING HOME: PORTRAITS IN BLUES AND ROCK AND ROLL 38 (1971).

[FN169]. Susan Keller, Review Essay: Justify My Love, 18 W. ST. U. L. REV. 463, 468 (1990); LEON WHITESON, THE WATTS TOWERS 23 (1989) (describing the composition of the towers).

[FN170]. ANDREW ROSS, NO RESPECT: INTELLECTUALS AND POPULAR CULTURE 199 (1989).

[FN171]. See supra text accompanying note 161.

[FN172]. BUTLER, supra note 143, at 31.

[FN173]. Id. at 137.

[FN174]. Amy Harmon, U.S. Auto Firms Wonder: "What Do Women Want?," L.A. TIMES, June 19, 1992, at A1.

[FN175]. Id.

[FN176]. Of course for me to fully appreciate these attributes, my Honda Civic must put on the drag of a dragster. To the extent it can, and to the extent Honda advertising encourages me to think of my car that way, I am the happier.

[FN177]. Harmon, supra note 174, at A1. This point highlights that gender performance remains subject to a variety of constraints, including existing gender power structures. For example, cars are sufficiently connected to male power to make unlikely a mirror image of men assuming female drag when they drive.

[FN178]. Interestingly, women have expressed preference for practical details like control buttons that accommodate long nails and floor mats that do not scuff high heels. Nails and heels seem like some of the more obvious ways in which women assume the drag of female gender. As a result of the complaints along these lines, the mostly male design‑engineers of one car company pasted artificial nails on themselves in order to test their buttons. Greg Bowens, Mimi Vandermolen: Women Drivers Have a Friend at Ford, BUS. WK., Nov. 16, 1992, at 66.

[FN179]. Duncan Kennedy, Sexual Abuse, Sexy Dressing and the Eroticization of Domination, 26 NEW ENG. L. REV. 1309, 1344 (1992). Similarly, men may feel like they are assuming male drag when they drive sports cars.

[FN180]. MARY ANN DOANE, FEMMES FATALES 25‑26 (1991). Doane associates this flaunting of femininity with the femme fatale of cinema. Id.

[FN181]. Id.

[FN182]. Id. at 43.

[FN183]. See Modleski, supra note 159, at 61‑63. Interestingly, Cynthia Epstein has suggested that women in the legal profession are less practiced than their male colleagues in the process of what she calls "role alternation," or the easy shifting from the role of adversary to buddy. CYNTHIA FUCHS EPSTEIN, WOMEN IN LAW 288 (1981).

[FN184]. JANE GALLOP, THINKING THROUGH THE BODY 100 (1988) (footnote omitted).

[FN185]. An exception to the general rule that the original owner of secondhand clothing is unimportant occurs when celebrities have owned the garments. For example, stores like "Star Wares on Main," in Santa Monica, California, market clothing previously worn by movie stars.

[FN186]. See Keller, supra note 169, at 468 (comparing alternative meanings of Madonna's Justify My Love video); Kennedy, supra note 179, at 1376‑77 (comparing alternative meanings of Madonna's Open Your Heart video).

[FN187]. Kennedy, supra note 179, at 1392‑93.

[FN188]. See MACKINNON, supra note 30, at 127‑29.

[FN189]. Carl F. Stychin, Exploring the Limits: Feminism and the Legal Regulation of Gay Male Pornography, 16 VT. L. REV. 857, 883 (1992).

[FN190]. Id.; see also MACKINNON, supra note 136, at 142. While MacKinnon and Dworkin's model ordinance defines pornography in terms of the depiction of women, it also states that "[the] use of men ... in the place of women ... is pornography for purposes of this law." Model Anti‑pornography Law § 2.2, in Dworkin, supra note 1, at 25.

[FN191]. BUTLER, supra note 143, at 31; see also Stychin, supra note 189, at 884‑85 (citing Butler).

[FN192]. Stychin, supra note 189, at 884.

[FN193]. MACKINNON, supra note 136, at 144.

[FN194]. BUTLER, supra note 143, at 137.

[FN195]. Keller, supra note 169, at 468.

[FN196]. Stychin critiques "the feminists' failure to acknowledge the unique position of gay men at the margins of dominant culture." Stychin, supra note 189, at 881. This emphasis probably results from his goal of presenting a free expression argument to exempt gay male material from the ambit of an antipornography ordinance; he maintains that such material is political speech because of its subversive potential. Id. at 896.

[FN197]. VARIETY (Media Home Entertainment/Heron Communications 1985).

[FN198]. Bette Gordon, Variety: The Pleasure in Looking, in PLEASURE AND DANGER 189, 196 (Carole S. Vance ed., 1984).

[FN199]. Keller, supra note 169, at 468. These manipulations include stylized crossdressing and unexpected gender roles in bondage scenes. Id.

[FN200]. Fiske, supra note 96, at 75.

[FN201]. VALVERDE, supra note 163, at 125.

[FN202]. Frug, supra note 6, at 1074.

[FN203]. Id. at 1074‑75.

[FN204]. FRUG, supra note 77, at 684.

[FN205]. Susan Keller, Powerless to Please: Candida Royalle's Pornography for Women, 26 NEW ENG. L. REV. 1297, 1299‑1300 (1992). Kimberle Crenshaw has made a similar argument in her discussion of the 2 Live Crew controversy, in which she seeks to integrate the positions of both those who found the material degrading to women and those who found that repression of the material was founded on racism and a denial of African‑American cultural forms. Kimberle Crenshaw, Beyond Racism and Misogyny: Black Feminism and 2 Live Crew, 16 BOSTON REV., Dec. 1991, at 5. See Vance, supra note 4, at xxii‑ xxiii (arguing that the positions of those whom she terms "feminist sex radicals" have been mischaracterized as "pro‑pornography" or "pro‑sex" when in fact their purpose was to initiate a more expansive agenda on sexuality, including considerations of power in sexuality).

While it is true that analyses of pornography have tended to emphasize either degrading aspects or liberatory aspects, rather than both, the divergence has been to a certain degree overstated. The editors of the volume Powers of Desire, which MacKinnon uses as an example of "other contemporary explorations that purport to be feminist" that "lack comprehension either of gender as a form of social power or of the realities of sexual violence," MACKINNON, supra note 136, at 135, actually display more sophistication about the role of power than she gives them credit for. They maintain:

Sexual experience with men or women can be abusive, objectifying, and degrading, but it can also be ecstatic, inspiring, illuminating. It can also be‑and here the inadequacy of a polarized discourse becomes clear‑a peculiar mixture of all these things: objectifying and pleasurable, degrading and inspiriting. We must bring together the complexities and contradictions: we must integrate what we know with what we do not want to know.

Ann Snitow et al., Introduction to POWERS OF DESIRE, supra note 110, at 9, 42.

[FN206]. Keller, supra note 169, at 470.

[FN207]. See, e.g., MACKINNON, supra note 136, at 135.

[FN208]. See, e.g., WHITESON, supra note 169, at 40 (describing the role of the Watts Towers Art Center in promoting local art and culture).

[FN209]. See id. at 30‑31 (connecting the historical neglect of the towers to the perception of Watts as riot‑torn).

[FN210]. See MACKINNON, supra note 30, at 171.

[FN211]. In the context of the response by international women's rights advocates to the practice of clitoridectomy, Karen Engle argues that the women subject to the practice are marginalized both by those who ignore these women's professed desires to maintain the practice and by those who are sensitive to them. Karen Engle, Female Subjects of Public International Law: Human Rights and the Exotic Other Female, 26 NEW ENG. L. REV. 1509, 1525 (1992). The same sorts of complications regarding consent she suggests would be relevant in this context as well.

[Back to Module 5]