Report of the Surgeon General's Workshop on PORNOGRAPHY and PUBLIC HEALTH June 22-24, 1986 Arlington, Virginia Prepared by: Edward P. Mulvey, PhD Western Psychiatric Institute and Clinic University of Pittsburgh Jeffrey L. Haugaard, MA Department of Psychology University of Virginia For Release to the Public on Monday, August 4, 1986 ' by the Office of the Surgeon General U. S. Public Health Service U.S. Department of Health and Human Services Washington, D.C. Tables of Contents -- Letter of Transmittal i Final Printed Program List of Participants Overview ii vi 1 Basic Research Problems 5 Consensus Statements 13 Future Research Agenda 39 Prevention and Intervention 47 Summary 56 References 58 Background Papers (ln the order they were discussed in the program) 67 "Psychological Research and Public Policy: Taking a Long, Hard Look Before We Leap" Donn Byrne and Kathryn Kelley "Effects of Prolonged Consumption of Pornography" Dolf Zillman "Do Sexually Violent Media Indirectly Contribute to Antisocial Behavior?" Neil WI. Malamuth "Effects on Juveniles of Being Used for Prostitution and Pornography" Mimi Halper Silbert "Techniques Designed to Mitigate the Impact of Mass Media Sexual Violence on Adolescents and Adults" Edward Donnerstein DEPARTMENT OF HEALTH 8a HUMAN SERVICES The Surgeon General of the Public Health Service Washington DC 20201 The Hon. Edwin Meese, III Attorney General U:S. Department of Justice Constitution Avenue and 10th Street, NW Washington, D.C. 20530 August 1,1986 Dear Mr. Attorney General: I am pleased to transmit to you the report of the "Surgeon General's Workshop on Pornography and Public Health," which was held in Arlington, Virginia, on June 22-24. The Workshop and this report are in respose to a request from the Chair, Mr. Henry E. Hudson, and the Members of the Attorney General's Commission on Pornography that the U.S. Public Health Service provide information to the Commission on the possible effects pornography may have upon public health. I hope the enclosed report is helpful to the work of your Department. May I add that we were extremely fortunate to have had participants at the Workshop, each of whom demonstrated throughout the intensive day and evening sessions an extraordinary degree of interest, knowledge, and collegiatity. They deserve great credit and the sincere thanks of their government. In addition, a great many persons from both our Departments contributed to the planning and the ultimate success of this Workshop; however, I would like to give special acknowledgment to the following: Jo AM Gasper, Deputy Assistant Secretary for Population Affairs, OASH Saleem Shah, PhD, Chief of the Antisocial and Violent Behaviore Branch, National Institute of Mental Health, ADAMHA; David Heppel, MD, Chief, Child/Adolescent Primary Care Services Branch, Division of Maternal and Child Health, HRSA; Theodore 0. Cron, Special Assistant to the Assistant Secretary for Health and the Surgeon General, OASH; Stephanie J. Stein, Special Assistant to the Surgeon General, OASH; Jacqueline Friedewalde, proprietor of Conference Management Associates. If the U.S. Public Health can be of any further assistance to the work of the Department of Justice in this or other matters, please let me know. Sincerely yours, C. Everett Koop, MD, DSc Surgeon General Enclosure ii SURGEON .GENERAL'S WORKSHOP JUNE 22;23, 24, .1986 ARLINGTON, VIRGINIA Tire Surgeon General's Workshop on Pomographg and Public Health June 22-23-24. 1986 Stoufk Co~~cowsc Hotel Ad.ington. Virginia AGENDA Workshop Chairprron: AJbato Scmno. MD Director. Philadelphia Child Guidance Clinic Professor. Child Psychology, Univenity of Pcnns+%ia Sunday, June 22, 1986 3:oo PM REGISTRATION (Plaza level) 6:OO PM RECEpllON. Cub bar (Dewey I. Plaza level) 730 PM DIMIER (Dewey 1. Plaza Iwet) o ???*??? ?*? ??????? ?? ??? ???????*??? C. Evcrert Koop. MD surgeon General US Public Health Service * .`Psychologial Research and public Policy: Taking a Long. HuL Look Before we Lap" Aurhorr: Don Byrne. PhD Professor & Chairman. Depanmenr of Psychology State University of New York at Albany Kachryn Kellcy. PhD Associarc Professor. Dcpvrmenc of Psychology Strw Univenity of New York II Albany Drmrumr: Jon It. Come. PbD Assistam Professor School of Social Service Administtarion Universiry of Chicago Monday, June 23, 1986 7-30 AM BREAKFAST (Dewy I. Plaza level) . "Rc&" Henry Hudson Charmtan Attorney General's Commission on Pornography Mondav , June 23, 1986, tcont'd) 9+Q AM 10 30 AM IO:45 AM 12.15 PM 1:30 PM 3:oo PM 3:15 PM 4.45 PM 6.30 PM OPENING PLENARY SESSION i.Jamcs Room. Lobby Icvcl) o "Effects of Prolonged Consumption of Pornography" Anhr Dolf tillman. PhD. Professor Institute for Communications Research. Indram Umv DNCYIIPA~. Alben Bandun. PhD Davrd Starr Jordan Professor of Socral Scicncc u-r Psychology Stanford University COFFEE BREAK RECONVENE PLENARY SESSION (James Room. Lobby Level) o "Do ScnsaUy Violent Media Lodirectly Conuibure IO Antisocial Behavior?" An1bor. Neil M Malarnurh. PhD Chairman. Commurucarions Srudrcs Univcrrrry of Cahfornia at Los Angeles Dmur~m~. Carol Nadclson. MD Professor & Vice Chainnan. Dcpartmem of Psychiatry Tufts University School of Medicine New England Medical Ccnrer LUNCH (Dewey 1. Plaza kwl) RECOhwNE PLENARY SESSION (James Room. Lbby level) o "EfTccrs on Juvenile of Being Used for Prosrirution and Pomognphy" AUlk~: Mimi H. Silbcn. PhD President & Chief Executive Officer D&nccy Strm Foundation Drrcnssant Robcn W. D&her. MD Professor of Pediatrics Director. Division of Adoksccnr Medicrnc Univeniry of Washington School of Medicine COFFEE BREAK RECONVEY PLENARY SESSION (lames Room. Lobby Icvcl) . "Techniques Dcsigacd to Mitigate the Lsnpacr of Mass Media Sexual Violence oo ~ddacaws and AduJu" Andor: Edward Donnemein. PhD Professor: Center for Communications Research Depanmcnc of Communications Arts University of Wisconsin at Madison Dircnuaa~c Gnc G. Abel. MD Professor of Psychiatry Emory Univcrsiry School of Medicine ADJOURN AFTERNOON SESSIONS DINNER (Dewey 1. Plaza kvel) o "FBI Rma& and Violent Criminal Behavior" Robcn Resslcr. Supervisory Special Agent National Center for the Analysis of Violcnr Crime FBI Academy Mondav. lune 23. 1986. ~orr~.d,~ 8:oo PM RECONVENE PLENARY SESSION uama Room. Lobby Icvel) o "00 Showing Sea to Childrrn: Video ad Remacka" Bndly Greenberg. PhD Chairman. Department of Telecommunicatrlons Michigan Sure Univcraicy Tuesday, June 24, 1986 7:30 AM BREAKFAST (Dewy 1. Plaar level) 8:30 AM CONVEM IN TWO CONCURRENT CONSENSUS SESSIONS Participanta will be assigned to one of the following scsaiooa: o Cemcom S&on I (Room 104. Finr floor) "E&co d porwlnpby on childrco ad Youtag Addo" F- Joyce N. Thomas, RN. MPH D&rot. Division of Child Pmceccion Clddsen'r Hospinl Na&mal Medical Center o CanaawSeaakII(Roomz0(.%condfIoor) "Ncr Paibilitia for Prcven&o and Trmcomu" Fsrdhror GrolHamnan. RN. DNSc Profasor. Boston College S&cd of Nuning IO:00 AM COFFEE BREAK IO:15 AM RECONf'ENE CONSENSUS SESSIONS I AND II 11:OO AM CONVENE IN CONSENSUS SESSION III (lames Room. Lobby level) o The Raadt Agenda la the Future" Fditaror: Mumy A. Srnua. PhD Director. Family Rcacarch I&oratory Univcniry of New Hampshire 12:1> PM LUNCH (Dewey I. PIaaa kvel) I:15 PM RECONVENE IN PLENARY SESSION-WORKSHOP SUMMATION o placnndoatorbcsargeoO~Rcsulofmm~kpioarI. *? O o ????????*?????????? 2.30 PM WORKSHOP IS ADJOURNED AUpitmq ICSS,O~, d be recordrd. vi List of Participants -a Gene G. Abel, MD Professor of Psychiatry Emory University School of Medicine Atlanta, Georgia Jeffrey J. Haugaard, MA Research Assistant in Clinical Psychology University of Virginia Charlottesville, Virginia Albert Bandura, PhD Professor of Social Science in Psychology Kathryn Kelley, PhD Stanford University Associate Professor of Psychology Stanford, California State University of New York Albany, New York Ann W. Burgess, RN, DNSc Professor of Nursing, School of Nursing University of Pennsylvania Philadelphia, Pennsylvania C. Everett Koop, MD, DSc Surgeon General, U.S. Public Health Service Washington, D.C. Donn Byrne, PhD Professor and Chair, Department of Psychology State University of New York Albany, New York Neil M. Malamuth, PhD Chair, Communications Studies University of California Los Angeles, California Jon Conte, PhD Edward P. Mulvey, PhD Assistant Professor of Child Psychiatry, Assistant Professor, School of Social Service Administration, University of Chicago Chicago, Illinois Western Psychiatric Institute and Clinic University of Pittsburgh Pittsburgh, Pennsylvania Jessica Henderson Daniel, PhD Supervising Staff Psychologist, Judge Baker Guidance Center, Children's Hospital Boston, Massachusetts Carol Nadelson, MD Professor and Vice Chair, Department of Psychiatry, Tufts University School of Medicine, New England Medical Center Boston, Massachusetts William Daniel, Jr., MD Emeritus Professor of Pediatrics Emeritus Professor of Pediatrics and Chief of Adolescent of Adolescent Medicine University of Alabama scnooi 01 University of Alabama School of Medicine Montgo Montgomery, Alabama Albert Serrano, MD Professor of Child Psychiatry and Director, Philadelphia Child Guidance Center University of Pennsylvania Philadelphia, Pennsylvania Robert W. Deisher, MD Professor of Pediatrics and Director of Adolescent Medicine University of Washington School of Medicine Seattle, Washington Edward Donnerstein, PhD Professor, Center for Communications Research, University of Wisconsin Madison, Wisconsin Bradley Greenberg, PhD Chair, Department of Telecommunications Michigan State University East Lansing, Michigan Mimi Halper Silbert, PhD President and Chief Executive Officer Delancey Street Foundation San Francisco, California Murray A. Straus, PhD Director, Family Research Laboratory University of New Hampshire Durham, New Hampshire Joyce N. Thomas, RN, MPH Director, Division of Child Protection Children's Hospital NatL Medical Center Washington, D.C. Carol Hartman, RN, DNSc Professor, Boston College School of Nursing Chestnut Hill, Massachusetts Dolf Zillman, PhD Professor, Institute for Communications Research, Indiana University Bloomington, Indiana OVERVIEW Purpose of the Report From certain perspectives, the question of restraining the availability and use of pornography is a relatively straightforward issue. For the active moralist, for example, it is an instance in which the government should fulfill its obligation to provide a healthy public atmosphere for the development of its citizens. For the ardent feminist, government restrictions can combat a system that promotes the active subjugation of women and the fostering of the public perception that women actually desire subjugation. For the avid civil libertarian, restrictions in this area should be resisted as dangerous indicators of a government's capacity to control infor- mation and thus undercut our basic freedoms. Although the particulars of govern- mental action are not resolved by any of these general positions, the limits of acceptable action are clearly set by them. This report does not attempt to side with, reconcile, or even comment upon the general positions that can be taken on restricting pornography. The aim is much more modest. This report summarizes the consensus of a group of social scientists and mental health professionals regarding the nature and extent of evidence about the effects of pornography, especially the effects on children and adolescents. The pur- pose of this assessment is not to prescribe public policy. To assume that social science or clinical practice offers such dispositive wisdom would be to greatly over- estimate both the scope and accuracy of professional knowledge. Public policy decisions are better decided by the balancing of tensions engaged in by judges, legislators, and social critics. An examination like the present one informs those more central to the policy process about the views of a group of social scientists regarding what is known in an area and what yet needs to be known in order to make more defmitive statements. In performing this function, social scientists help to clarify the evidence that supports or refutes arguments for certain courses of social action. Expert commentary on the validity of assumptions of fact or the adequacy of theory is valuable to the policy formation process because it highlights the line between accepted-social science evi- dence and values or morality. Ideally, it can thus force discussion of these latter issues to be done more openly. 2 Previous Investigations Into the Effects of Pornography The Surgeon General's workshop that generated this report was not the first govem- mentally sponsored effort to examine the question of the social effects of pomog- raphy. In the late 1960's, the Commission on Obscenity and Pornography was formed, and after funding much research and holding many hearings, released its report in 1970. Its findings were basically summarized in its introduction: "Empirical research designed to clarify the issue has found no reliable evidence that exposure to sexually explicit materials plays a significant role in the causation of delinquent or criminal sexual behavior among youth or adults." The report was criticized by both Commis- sion members (Cline, 1970) and others (see Krafka, 1985) for flaws in the design of some commissioned research, misinterpretations of some of the research, limitations of the scope of much of the research done. Despite critisisms, however, the general conclusion of the Commission-that pornography has no marked so&l effects-has continued to be the generally accepted and often cited wisdom in this area. In response to a growing concern over child pornography in particular, both houses of Congress held hearings into the issue of child pornography in 1977. During these hearings, witnesses estimated that between 300,000 and 600,000 children were involved in the production of pornography and that more than 260 child pornography publications were being produced (Pierce, 1984). Subsequent to these hearings, the federal government and nearly all state governments enacted laws against the pro- duction, distribution, and possession of child pornography. In 1979, the British government sponsored the Committee on Obscenitp and Film Censorship. The conclusions of this Committee, based in large part on a research review by Yaffe and Nelson (1979) were similar to those of the 1970 United States Commission on obscenity and Pornography. Like its predecessor, however, this com- mittee report was also criticized for its failure to include what were considered relevant research studies (Court, 1980). The Canadian government convened the Special Committee on Pornography and Prostitution, which delivered its report in 1985. The Committee found that while individual research projects had shown some of the effects of pornography, as a whole, the research was contradictory and inconclusive, and could not be relied upon as a guide for policy formation. The United States Attorney General's Commission on Pornography was formed in 1985 in response to a number of concerns: (a) continuing criticism of the fmdings of previous commissions; fb) recent technological advances and changes in social standards that had allowed both youth and adults easier access to pornography; (c) recent research (e.g., Malamuth & Spinner, 1980; Smith, 1976) indicating that the 3 content of both "hardcore" and traditional "softcore" pornography had changed in the 1970's and 1980% to include considerably more sexually violent material; and (d) preliminary research suggesting that these new forms of pornography had different effects on viewers' attitudes and behaviors than the pornography studied earlier. These developments called for a reassessment of the possible impact of the changes in the content and patterns of use. Moreover, a new research approach had emerged, emphasizing that the message communicated by the pornography might have at least as much effect as the explicitness of the sexual content, and an evaluation of new findings thus seemed in order. The mandate of the Commission was in part to "determine the nature, extent, and impact on society of pornography in the United States, and to make specific recommendations to the Attorney General concerning more effective ways in which the spread of pornography could be contained, consistent with constitutional guar- an tees." Although no new research was funded by this Commission, it did undertake public hearings in which current research was explained and information about the current manufacture and distribution of pornography was gathered. Numerous wit- nesses who viewed themselves as victims of pornography also testified. The Surgeon General of the United States was asked to testify before the Attorney General's Commission, and subsequently agreed to provide the Attorney General with a report summarizing the evidence that was available from the scientific community regarding the effects of pornography on public health. In order to provide this information, the Surgeon General's Workshop on Pornography was convened, con- sisting of 20 members from the communications, medical, mental health, and social science fields. The goal of the Workshop was to provide the Surgeon General and the U.S Public Health Service, as well as the Attorney General's Commission, with infor- mation regarding three general questions: (1) what is known, with a reasonable degree of certainty, about the effects of pornography on the mental and physical health of those in the United States, especially on children and young people; (2) what additional effects are justifiably suspected and how these posited effects could be verified or refuted ; and (3) what actions could be taken by those in the medical, mental health, and public health fields in order to combat any negative effects of pornography. Papers were prepared by five experts who had either conducted research into the effect of pornography or were familiar with the effects on youth who had par- ticipated in the production of pornography. These papers and discussant reactions to the papers served as focal points for the first part of the Workshop. Three consensus sessions were then held: one to summarize current knowledge about the effects of por- nography on children and adolescents, a second to suggest prevention and intervention measures for those who might be or have been affected by pornography (either as con- 4 sumers or as participants in its production), and the third to outline a course of future research. This report captures the essence of the discussions held during the two days of the Surgeon General's Workshop on Pornography. A general discussion of the process of research in this area is presented first. Then, the conclusions of the Workshop participants regarding the effects of pornography are presented, as well as the evi- dence upon which these conclusions were based. FinalIy, recommmendations for prevention and treatment and for future research are given. 5 BASIC RESEARCH PROBLEMS The Underlying Issues Designing and executing research to isolate the unique effects that exposure to pornography may have on children and adolescents is a formidable task because of the methodological and ethical barriers inherent in doing such investigations. Exposure to pornography is only one of many interdependent factors potentially affecting a child, and attempting to isolate its unique effects forces the researcher to oversimplify the complex process of development. Moreover, conducting investigations regarding ex- posure to pornography while protecting subjects from potential harm creates a number of vexing ethical dilemmas. Given these constraints, it is not too surprising that the amount of directly relevant information that can be gathered from the social sciences regarding this question is limited. It is important to note first that the effect on children of any particular influ- ence like exposure to pornography is the result of many interacting factors. Most obviously, the effects of exposure are likely to differ depending on the age of the child. Because of new cognitive and social skills that develop as a child matures, the effect that an experience with pornography has at one age does not always correspond to the effect that it has at another. In addition, recently it has become clear that there is considerable variation between children in their proclivities for certain activities or susceptibility to particular influences. Children bring dispositional tendencies and early learning patterns into any interaction with the environment, and these may also produce wide variation in the impact that any particular influence or incident has on the child. Finally, exposure to pornography does not occur in a vacuum. Like any other factors that might influence child development, exposure to pornography is unlikely to exert an infIuence totally distinct from the setting and social context in which it is experienced. For example, a short exposure and subsequent discussion with a parent about the factual accuracy or value message of the material would in all likelihood have a quite different effect than repeated solitary exposure. How to test systematically the effects of exposure to pornography in the face of this myriad of potential influences is obviously quite a methodological challenge. 6 Even if one is willing to chip away systematically at these multiple interactions through circumscrtbed investigations, however, there are ethical constraints that inevitably limit the types of research designs that can be used. Because the exact short term effects on children of exposure to pornography are unknown, and because exposing youth to pornography for experimental purposes could conceivably produce actual ill effects, this tactic violates a basic moral sense of the limits of acceptable social scientific practice. The researcher, therefore, is caught in a Catch-22 situa- tion. It is impossible to know exactly what effects different types of materials might have on children without investigating the phenomenon systematically, but it is impos- sible to do the investigations as long as adverse effects are hypothesized. Moreover, showing an effect would guarantee that future investigations would be even further restricted, since there would then be proof of the risk. Aside from the simple fact that showing pornography to children is a morally distasteful and at present an ethically unsupportable manipulation, the lack of clear knowledge or theory regarding expected long term outcomes of such exposure also presents an ethical difficulty. Even if it could be demonstrated that negative short- term effects could be controlled by debriefing, it would still be important to be able to assert that long term effects of a child's involvement in such research would be negligible. Because of the possible multiple interactions between exposure to pornog- raphy and other contextual and dispositional variables, however, it is impossible to predict what the ultimate effects of such exposure might be for a youth taking part in such research. Without adequate theory to guide an assessment of the risks of this research for children, investigators are unable to present a balanced (even if only theoretical) argument regarding methods to protect against ill effects. In the end, the individual risk factors predis- posing children to various influences are too unclear, the potential negative outcomes are too diffuse, and the theory too sparse to protect children taking part in the direct research that would need to be done. Because of these ethical constraints, we are unable to scientifically test many of the propositions related to the effects of pornography on children. Thus, the field of available data upon which any general conclusions can be based is bound to be of questionable scientific rigor. Instead of examining direct tests of hypotheses, we must extrapolate and speculate regarding the effects of pornography on children and ado- lescents from related research. There are three basic types of investigations- clinical, laboratory, and correlational-that are relevant in this effort, and each one has its advantages and limitations. It is useful to be aware of the qualities of each approach while considering the applicability of their findings to children and ado- lescen ts. Clinical Studies One source of information about the effects of pornography on children and adoles- cents are clinical studies involving youth who have come to the attention of juvenile justice or mental health agencies. These reports are valuable because they provide a glimpse at the casualties of youthful exposure to harmful influences. Unlike statisti- cal analyses of patterns within large groups of subjects,: clinical studies frame problems in human terms. Devastating subjective experiences do not become diluted by numbers representing differences between group averages. The rich texture of clinical data is valuable for the development of theory, especially in an area Iike pornography where ethical considerations prohibit empirical investigations with larger, more representative samples. Clinical studies highlight relevant issues to be considered in any theoretical formulation and pose potential hypotheses regarding the mechanisms behind certain observed effects. Clinical ac- counts provide direction for more controlled investigations and promote theory that is linked to the subjective experiences of the practitioners and the youth involved. Clinical studies are limited in value, however, when testing hypotheses and theories. Their greatest general limitation is the inability to isolate the specific effects of the variable being considered (such as exposure to pornography) from other potentially influential variables. Using clinical data alone, it is impossible to attribute a particular amount of an observed effect to a variable in question. For that matter, it is impossible to conclusively attribute any effect to a specific variable. In order to make such causal statements, it is necessary to compare the effects seen in a clinical sample with those seen in an appropriately matched control group. This cannot be ac- complished through the use of clinical samples alone. Clinical studies have another problem resulting from their focus only on youth with identified problems. The effects of the circumstances that form the basis for the identification (e.g., sexual abuse) cannot be separated accurately from the effects of the discovery of those circumstances. For instance, it is impossible to tell with any rigor how much of the distress shown by a child in therapy results from actual parti- cipation in pornography and how much is precipitated by the reactions of others such as parents, friends, and teachers to the discovery of the child's involvement. For example, Burgess, Hartman, MacCausland and Powers (1984) found that behavioral problems of some children involved in sex and pornography rings increased after their participation was discovered. The identification process itself provides an influence on behavior that is very difficult to gauge. Perhaps the most subtle problem with clinical reports is the inevitable tendency 8 for individuals who report on clinical phenomenon to interpret and report their obser- vations in a way that supports their beliefs about the phenomenon in question. Those writing clinical reports have beliefs that necessarily influence the general approach taken to the subject matter, the types of questions asked, and the interpretations of the results. The lack of a broadly accepted method in this area is both a strength and a weakness. It allows for insightful and broad analyses and condones bias at the same time, often resulting in contradictory findings and conclusions that produce more careful debate than resolution. Clinical reports thus often illustrate, but rarely irrefutably substantiate, different perspectives on a question. All of these concerns make it troublesome to rely soley upon clinical reports to gain a clear picture. of the potential problems connected with exposure to or involve- ment in pornography for children and adolescents in generaL Children who have run away from home and are encouraged to produce pornography by being shown porno- graphic pictures do not necessarily represent all children, and their reactions to being shown pornography do not necessarily represent the inevitable response of children to exposure. Similarly, the effects on delinquent youth of viewing pornography do not necessarily mirror those of nondelinquent youth. The reactions of these select samples to participation in or exposure to pornography may differ significantly from those of children from less troubled backgrounds. It is important, therefore, to remember the values and limits of clinical re- search as the literature in this area is reviewed. Rather than interpreting clinical accounts as proof, we should instead view them as rich reminders of how devastating certain influences can be to victimized children. In addition, they can clearly point toward the issues that must be addressed more rigorously and make us feel why it is so important to pursue those investigations. Laboratory Studies Laboratory research provides another source of potentially valuable information about the effects of exposure to pornography. Using controlled settings and consistent materials, this type of research can systematically test for individual differences produced by altering either the types of pornographic materials seen or the conditions under which they are seen. The attraction of this approach is that it is "scientific" in the sense that hypotheses can be directly tested and findings can be amassed to support global theories. The drawback of such an approach is its inherent artificiality; phenomena in the lab are not always what they may be in the real world. 9 The advantages of this approach for increasing our understanding of the effects of exposure to pornography are considerable. First, it allows for control over the types and amount of materials seen. As a result, the effects of sexually explicit materials can be compared to the effects of sexually violent materials or the effects of singIe exposures can be compared to those obtained with repeated exposures. Second, this approach allows for control over possible extraneous variables that might affect the results, allowing heightened confidence that any observed differences were caused by the factors that were manipulated. For instance, individuals can be ran- domly assigned to different conditions of exposure to control for self-selection factors, or anger levels can be manipulated before testings for the display of an out- come behavior of interest. Third, it promotes the systematic construction of theory regarding the mechanisms by which any effects resulting from exposure occur. Results from one series of experiments can be further refined in later studies until a clearer picture.emerges. Laboratory aggression against women in these studies, for example, has been shown to be affected by the content of sexually violent materials, thus pro- moting a theory of a synergistic effect between the presentation of violence and sex in promoting targeted aggression against women. There are also many limitations of laboratory approaches applied to studies of exposure to pornography. First, as mentioned earlier, there is the nagging question of whether the results obtained in the hb actually reflect the world as it is outside of the lab. The vast majority of lab studies in this area use college students as subjects, create manipulations that are usually contrived and less complex than the influences occurring in real life, and produce effects that are only analogues of the actual behav- iors of interest. Regarding pornography, the concern about the applicability to the everday world centers on whether the attitudes and behaviors of college students regarding sex are adequately representative of the general population, whether the types and patterns of exposure in the lab are comparable to regular or sporadic por- nography use, and whether the short term effects produced can be interpreted as hav- ing any substantial relationship to behavior patterns outside the lab. There is a particular related concern that those individuals who volunteer for laboratory studies (especially those studies involving penile tumescence measures) may not be represen- tative of the general pool of potential subjects, possibly being more sexually liberal and having more varied sexual experiences (e.g., Malamuth & Check, 1983). Second, laboratory studies operate primarily from a hypothesis testing model. In this approach, demonstration of differences between groups exposed to different treatments is taken as evidence that the manipulation has been sufficiently powerful to rule out the possibility that the observed group differences were produced merely by chance. A lack of differences is taken to mean that the manipulation did not have an effect substantially greater than what might have resulted from random fluctua- tions, but it does not necessarily rule out the possibility that there might be an effect under slightly different circumstances. Thus, this model for experimentation nearly guarantees that only significant results wiIl be reported in the literature, since "no effect" is an inconclusive statement. As a result, it is difficult to know how many studies were done that were unable to produce any observed result. `Ihis generally unavailable information is important because the presence of a large number of these unreported results may indicate that the regularity and strength of a particular effect is really rather low even though it has been reported to have occurred under controlled conditions (Rosenthal, 1978). This drawback makes it impossible to merely compare the number of published studies showing a particular effect and those not showing the effect, when assessing whether the effect exists. In a related vein, it is difficult to judge whether the strength of an effect that will produce statistically significant group differences in a controlled study reflects a difference that is of significant magnitude in the real world A difference of 2.5 on a `I-point rating scale completed by a large number of college students, for instance, may produce a highly statistically significant effect in a controlled study. Whether this difference really translates into an attitude difference of noticeable proportions in the real world, however, is another question. In short, lab studies offer the opportunity to contruct theory but can only rarely test how well that theory goes beyond the controlled setting of the lab. Correlational Studies The third type of social scientific evidence commonly considered when at- tempting to assess the effects of pornography is that examining relationships between or changes in social indicators. This strategy usually involves examining the correl- ation between two indicators under different conditions (e.g., the incidence of rape and the distribution of sexually oriented magazines in different states) or changes in the frequency or magnitude of an indicator over a given time period (e.g., changes in rape rates after a loosening of the restrictions on pornography). Examinations of social indicators are particularly useful for generating hypotheses about how people might be acting or for predicting how a social system might operate. Because of their general nature, investigations using these designs are open to several problems. The major problem is their limited capacity to control for the numerous variables that may be contributing to any observed correlation or shift. For 11 example, a shift in demographic composition or reporting practices over time can af- fect the relationships seen in aggregate population data, or other unmeasured vari- ables can be mediating a relationship between the two variables being examined. These influences may cause the change in one variable to be mistakenly attributed to a second variable. Researchers attempt to guard against these influences by examining changes in other variables that might be logical causes of the observed changes, but it is always an open question whether enough of these extraneous factors have been con- sidered or if the Researcher has examined the right ones. Somewhat like clinical studies, correlational research offers us a glimpse of the real world, but at an aggregate level that is highly dependent upon the measures chosen. Studies of this sort also come under considerable criticism regarding the ac- curacy with which the social indicators chosen actually reflect the particular theo- retical notion in question. Variables are usually taken from existing data bases and are assumed to be reasonable proxies for more abstract notions. The number of "adults only" movie theaters in an area, for examph, might be used as an indicator of a community's acceptance of pornography, or salary differential between sexes might be used as an indicator of the status of women. How well these measures actually capture the concept under consideration, however, is often debatable, and there is often the real possibility that different results would have been obtained using different indicators. Finally, interpretation problems arise with studies of social indicators. It is often difficult to keep in mind that associations seen in aggregate data may or may not reflect processes operating within or among individuals. For example, although the availability of sexually explicit materials may be correlated with the incidence of rape in particular areas (Baron & Straus, 1984), this association does not mean that the same individuals are accounting for both rates. This observed relationship could be present even if two totally nonoverkpping groups of individuals were accounting for each indicator. As a result, investigations of this sort that rely on aggregate data provide no conclusive information about the extent of the relationship of the variables of interest in the individuals who compose the groups examined. In the end, the best that can be offered are demonstrations of an observed correlation, but no real information about whether the results indicate a causal relationship. Because of the open-ended nature of the arguments that can usually be raised against any observed correlation, studies of this sort are usually viewed as adjunct sources of support for particular arguments or as indicators of potentially fruitful places to begin more controlled studies. 12 Summary Given the problems with each of these approaches to research, one might be left with the impression that all social science data is flawed to the point of being unconvincing to the discerning reader. This interpretation, however, would be overly harsh. Instead, what should be taken from this critical review is the simple idea that no one study or set of findings using only one method should be taken as definitive. While certain questions are best suited to particular forms of investigation, one piece of evidence standing alone or without considerable theoretical support should be viewed with skepticism. When more than one approach is used to investigate the same specific question or hypothesis and consistent findii emerge, however, it is evident that something has been found that warrants consideration by policymakers and researchers alike. There are ways that each method outlined above compensates for the inadequacies of the other methods and consistent findings therefore argue for little likelihood that the particular approach taken has skewed the results in a particular way. Each of the above methods has been used in examining the effects of pornography, and an assessment of what this evidence ultimately says requires a careful weighing of the methods used and the consistency of fmdings across different methods. There are some statements that can be made with scientific confidence because they are validly demonstrated and/or theoretically sound. 13 CONSENSUS STATEMENTS Introduction Workshop participants were given the charge of reaching some consensus regarding the effects of pornography on the public health of citizens, expecially children and adolescents. The list of consensus statements that emerged was intended to provide a reasoned, rather than a partisan, summary of what social science can say with con- fidence about the effects of pornography. Each of the five consensus statements that follows relates to a circumscribed effect of pornography that is supported by directly relevant social science data and is tenable in light of demonstrated theory in related areas of inquiry. There have been many claims other than those listed here regarding the effects of various forms of pornography. What are listed here are conclusions that have, in the opinion of the participants, been demonstrated with a required degree of social science accuracy. This is not to say that other purported effects of pornography have been examined and found to be false. Presently, however, the state of the evidence in this area appears to substantiate only the following limited conclusions. Consensus Statements Children and adolescents who participate in the production of pornography expkknce adverse, enduring effects. Involvement of children in the production of pornography is a form of sexual exploitation, victimizing vulnerable children and leaving them with the aftermath of this involvement. Sexual exploitation has been linked to a variety of adverse emotional, behavioral, and somatic consequences in children as well as adults who were exploited as children. Incidents of sexual exploitation are not easily put aside by a child, but instead appear to re-emerge as a variety of difficulties (for a review, see Finkelhor & Browne, 1986). 14 Several aspects of the process of sexual exploitation are theorized to be im- portant contributors to the emergence of adverse effects. Finkelhor and Browne (1985) have theorized that these effects can be caused by several factors: traumatic sexualization, betrayal, powerlessness, and stigmatization. Traumatic sexualization is a result of the child being involved in, and rewarded for, developmentally inappro- priate sexual behavior. This can produce an obsession with or adversion to sexuality, and may be seen behaviorally in childhood and/or adulthood as a preoccupation with sexual activity, sexual dysfunction, or phobic reactions to intimacy. Betrayal results from the child being manipulated by a trusted adult. Effects can be depression, de- pendency, mistrust, and hostility, and behavioral consequences can include isolation, vulnerability to other abuse, and conduct problems. Powerlessness results from a sense of vulnerability felt because of repeated invasion of the child's body and the inability of the child to stop the abuse. Effects include anxiety, fear, self-perception as a victim, and identification with the abuser. Behavioral consequences include somatic difficulties, sleep disturbances, school phobias, and delinquency. Finally, stigmatization can occur as a result of the child blaming his or herself for the abuse or being blamed for the sexual activity by the abuser, the child's family, or others. Effects of this process are thought to be guilt and lowered self-esteem. Possible behavioral consequences include isolation, drug use, criminal activity and self- destructive behavior. Involvement with pornography does seem to have a place in the dynamics of sexually exploiting children. Pornography has been used by adults to teach children how to perform sexual acts and to legitimize the children's participation by showing pictures of other children who are "enjoying" the activity. In some cases involvement in the production of pornography has led to other sexual activity; in others, pornog- raphy involvement has followed sexual activity. Since most children and adolescents involved with pornography are also involved in prostitution or other sexual activity with adults (Burgess et al., 19841, and since many come from homes where they experienced neglect and abuse (Silbert & Pines, 1984), the specific effects of involvement with pornography cannot be isolated cleanly. Involvement with pornography is one of many influences operating in the lives of these youth, and discussion of the effects of participation in the production of pornography must be understood in terms of the combined effect of this influence with previous negative life experiences. There is no reason to believe, however, that involvement with pornography is less traumatizing than other forms of sexual exploitation of children. There is even some suggestion that it may produce a unique form of trauma since the child knows that there is a permanent record of his or her participation. Evidence about the role and effects of pornography in sexually exploiting children necessarily comes exclusively from clinical studies. Burgess et al. (1984) completed extensive interviews with 62 children referred by law enforcement agencies when their participation in sex and pornography rings was discovered. Although one of the most extensive investigations of its kind, the authors noted several possible limitations to the generalizability of the results of the interviews to all children involved in such activity. Most notably, the sample might not have been representa- tive of all such children since only those referred by law enforcement agencies were interviewed and 14 children who could have participated were not permitted by their parents to do so. Also, the children's emotional and behavioral patterns may not reflect those found in children whose participation had not been disclosed. Difficulties reported by the children as occurring during their association with the sex rings included somatic complaints, sleep disturbances, withdrawal from peer and adult contact, and acting-out behaviors. Disclosure of the children's involvement in the rings produced new symptoms in 49 of the children and caused the exacerbation and solidification of existing symptoms in many others. Data analysis indicated that those involved in a sex ring for more than one year and those involved in the produc- tion of pornography (34 children) were more likely to exhibit more severe symptomol- ogy. Unfortunately, the specific contribution that each of these two conditions made to the symptomology is unclear. From structured interviews with hundreds of children, adolescents, and young adults, Silbert (1986) compiled a list of symptoms exhibited by those involved with pornography. Those involved with pornography at the time of the interview were seen as uncommunicative, withdrawn, inattentive, and fearfuL Long-term consequences of those who had been involved earlier were characterized by the internalization of those behaviors. That is, the adolescent personalities of previously involved youth were best characterized as withdrawn, anxious, and paranoid. Only five of the subjects inter- viewed had previously disclosed their involvement with pornography. For the two whose disclosure met' with support by those in their environment, a time of initial heightened turmoil was followed by a more healthy resolution. For the others who were met with disbelief or punitive reactions, the disclosure meant an exacerbation of their negative reactions. In a contradictory report, Inciardi (1984) interviewed nine girls between the ages of 8 and 12 who were involved in prostitution and pornography and found few overt signs of disorder. All of the girls had been introduced into the sexual activity by parental figures, all of whom were already involved in prostitution or pornography. None of the girls' activities had been disclosed to law enforcement or social service 16 agencies, and their involvement in sexual activity was ongoing. Four of the girls re- ported that they did not use drugs, and Inciardi stated that the others' drug use did not seem to be directly related to their sexual activities. While these findings are counter to the previously cited investigations and the more generally accepted notion of the grave impact of such activities on children, several aspects of the group of girls interviewed must be considered as possible fac- tors producing these results. First, the sample was small and not `necessarily repre- sentative of all girls in similar circumstances. The girls' parents encouraged their sexual activity and the appearance of the activity as more normal may have mitigated the effects reported by other children. Second, the girls were still quite young. As Inciardi noted, these girls had the same trouble expressing affect as do other children their age, and this limitation may have produced an underreporting of symptoms in the short interview used. Moreover, many effects of sexual exploitation may not appear until the occurrence of significant life events involving sexuality or intimacy (Gelinas, 1983). `The lack of current symptoms, therefore, does not preclude later problems around dating, intimate relationships, or childbearing. Finally, the sexual activity had not been disclosed. Recalling that symptoms can increase in severity after disclosure (Burgess et al., 19841, there is the possibility that these girls may have presented (or may yet present) a different clinical picture after the possibly traumatic experience of disclosure. Although it is impossible to isolate the effects of participation in the production of pornography from the effects of the other forms of sexual exploitation that are often experienced by those involved in pornography, clinical evidence indicates almost unequivocally that the effects of such involvement are adverse and enduring. Whether or not we can generalize from the reported effects to all those involved in the production of pornography has not been shown, yet it is clear that harmful effects are experienced by many. While some evidence is available of minor effects when participation is encouraged by parental figures, this evidence is not overwhelming. In the end, it is fair to conclude that pornography plays a clear role in the sexual exploitation of children, and that this exploitation produces damaged children and adults. 17 Prolonged use of pornography inereasas belie& that less common sexual practices are more common. The basis for this conclusion comes mainly from the work of Zillmann and Bryant (1982) and the application of recent theoretically relevant findings from cognitive psychology research. The theme of this conclusion is that repeated exposure to por- nography is likely to alter one's estimates of the frequency of people engaging in behaviors similar to those depicted. While there is only one direct experimental test of this premise, these results match other studies of human judgment processes using materials other than pornography as content. The research study can thus be seen as a demonstration that an observed, theoretically sound psychological mechanism appears to be operating when individuals are repeatedly exposed to pornography. Zillmann and Bryant (1982) exposed a group of male and female undergraduates to varying amounts of pornographic ftims. `The massive exposure group watched six different eight-minute sexually explicit films during each of six weekly sessions, the intermediate exposure group watched three erotic and three nonerotic films, and the no exposure group watched six nonerotic films. All of the erotic films contained only consenting heterosexual activities of fellatio, cunnilingus, coition, and anal inter- course. Three weeks after the end of the film presentations the subjects completed several questionnaires, one of which asked them to estimate the percentages of adults in the United States that engaged in various common and uncommon sexual practices. Those in the intermediate and massive exposure groups, compared to the no exposure and no prior treatment groups, estimated that significantly higher numbers of adults engaged in fellatio, cunnilingus, and anal intercourse. Comparison with data from broadly based sexual surveys indicated that the estimates of the intermediate and massive exposure groups were actually more accurate than the no exposure and no prior treatment groups, which underestimated the prevalence of these behaviors. Of particular interest was the additional finding that estimates of the prevalence of group sex, sadomasochism, and bestiality were higher for the massively exposed group than the other groups, even though none of the pornographic material to which they were exposed included these types of activities. Those in the massive exposure groups significantly overestimated the reported prevalence of these activities. 18 This study indicated that those viewing intermediate or massive amounts of more commonly accepted pornography over a six-week period believed that these forms of sexual behavior occur more frequently in the general population than those with no exposure to pornography during the same time. Also, those who viewed more massive amounts of pornography believed that less common forms of sexual behavior which were not included in the materials seen also occur more frequently. In short, the perceptions of the subjects regarding the prevalence of sexual practices were affected by the amount of pornography that they viewed. However, the implications of this change in perception toward attitudes of tolerance for the material or the less common sexual practices were not and have not been investigated. This single study would not be as convicing if its results were not so predictable in light of other research. Recent work in cognitive psychology has consistently found a set of processes that appear to drive human judgment under conditions of uncertain- ty (see Kahneman, Slavic & Tversky, 1982). When confronted with problems requiring estimates of the likelihood of an event occurring, people appear to perform a selective search of the outcomes of representative situations that they have encountered, and rely heavily on their recollection of the outcome of the most cognitively available incidents for representations. The availability of these cognitive reference points is influenced by the recency of a person's contact with the given material or the dra- matic salience of the materiaL Since pornography is a primary means by which many people (especially children and adolescents) learn about the sexual behavior of couples, and since there is seldom any corrective, more accurate information available, it is easy to see how this material can affect perceptions about how common certain behaviors are. Exposure could logically provide a heuristic of couple's sexual behavior in which less common acts are regular occurrences. Moreover, regular exposure could help to make these representations highly available as reference points. While more research is obviously required to gain a full picture of how repeated exposure to pornography affects perceptions, it presently appears that exposure to this content material operates in a way consistent with what would be expected. 19 Pomogmphy that portrays serual aggmsion as pleasnrable for the victim incm the acceptance of fhe use of coercion in SexnaI rehtions This statement is based on experimental findings, the fact that these findings are congruent with theoretical notions of attitude formation, and clinical reports from samples of sex offenders. None of the available data sources taken alone are suffi- cient to posit an effect of violent pornography of a particular type on attitudes. Taken together, however, there appears to be a convincingly clear picture of attitudes toward the acceptability of sexual coercion being substantially altered by exposure to particular types of violent pornography. Since the 1970 Commission on Obscenity and Pornography, a number of re- searchers have explored the possibility of distinct effects related to exposure to vio- lent pornography. These studies have been undertaken primarily because of the per- ceived increase in the availability of violent sexual materials (MaIamuth & Spinner, 1980; Dietz & Evans, 1982) and the fact that the Commission only tangentially con- sidered the possibility that these materials could exert an influence unique from erotic pornography. As part of this line of laboratory studies, attitudes have been investi- gated as possible factors potentially increasing or decreasing any behavioral effects produced by exposure to violent sexual materials. Attitudes in these studies have been measured using a variety of paper and pencil scales, with measures of the same type also used to gauge the individual's pre-existing, self-assessed "proclivity" toward rape. These investigations have been and continue to be classical, social psychological laboratory studies, almost exclusively using undergraduates as subjects and question- naires or simulated tasks as outcome measures. Two studies using this approach (Malamuth, Haber & Feshbach, 1980; Malamuth & Check, 1980) have shown that exposure to pornographic rape scenes in which the as- sault resulted in the female victim's sexual arousal altered males' assessments of a later rape depiction. These studies compared the effects of materials in which the victim was aroused by the attack, the victim abhorred the attack, or two people were involved in a mutually consenting act. Those viewing a scene in which the victim was aroused by the attack saw the victim of the later rape depiction as having suffered less. 20 Another study (Malamuth, Reisen h Spinner, 1979) failed to show similar ef- fects regarding a later rape depiction, however. In this study, exposure to aggressive pornography films showed no effects on an immediate assessment of an interview with a rape victim or on attitude survey responses regarding the seriousness of rape ob- tained three weeks later. What is striking about this study, however, is that the materials used did not show the victim becoming aroused by the attack, but were in- stead merely pictures of sadomasochism and rape. Unwittingly, these researchers provided verification of one of the conditions of, the other studies, testing for the presence of an effect using violent materials without depictions of victim arousaL The fact that there were no effects thus lends credence to the supposition that the portrayal of victim enjoyment or repulsion is an important determinant of any attitude effects produced by violent pornographic materials. Another investigation using a slighly different methodology provides confirming evidence. Malamuth and Check (1981) compared three groups of male and female un- dergraduates regarding changes in their acceptance of rape myths (e.g., that women actually want to be attacked) and their attitudes toward the acceptability of violence in sexual relations (as measured using three scales developed by Burt, 1980) as the result of their exposure to violent sexual materials. Under the guise of examining the movie rating process, these researchers had one group of students see two movies showing a favorable victim response to rape and another group see two neutral movies without violent sexual content. Both groups saw the movies in a theater as part of a regularly scheduled film series rather than being exposed to them in a laboratory setting. Males, but not females, showed significantly higher acceptance of inter personal violence and an increase in rape myths when tested several days later as part of an apparently unrelated general attitude survey given in class. An interesting and important secondary finding of this study was that the greater proportion of the shift in attitudes in the group that was exposed to the materials showing the aroused victim was produced by individuals who reported a higher likelihood to rape. These resarchers asked subjects to indicate the likelihood that they would rape a woman if it could be guaranteed that they would not be caught. This "likelihood to rape" (LR) scale had been used in previous research and shown to have adequate validity for distinguishing individuals with pre-existing favorable attitudes toward violent sexuality (see Malamuth, 1984, for a discussion of the development of this scale). Subjects scoring high on this scale were more likely to show attitude changes as the result of exposure to materials in which the victim showed arousal. Whether the victim becomes aroused as the result of a violent sexual attack has also been shown to be a significant factor related to the arousal patterns of males. As might be expected, nonrapists have shown higher arousal (as measured by penile tumescence) to depictions of consenting sex than rape. Convicted rapists, meanwhile, have shown high and about equal arousal to depictions of consenting sex and depictions of rape. However, rapists have also shown higher arousal to rape depictions that con- tained greater aggression than to rape depictions with less aggression (Abel, Barlow, Blanchard & Guild, 1977; Abel, Blanchard, Becker & Djenderedjian, 1978; Barbaree, Marshall & Lanthier, 1979; Abel, Becker 13 Skinner, 1980; Quinsey, Chaplin h Varney, 1981; Quinsey, Chaplin & Upfold, 1984). An interesting further finding of this line of research is that the sexual arousal of nonrapists to rape depictions can be inhibited by previous exposure to depictions that clearly emphasize the victim's suffering and repulsion (Malmuth h Check, 1980; Malamuth, Heim & Feshbach, 1980). When nonrapists were presented with depictions that differed on the victim's reaction, their self reported arousal and arousal measured by penile tumescence to later rape depictions differed significantly. Those who orig- inally viewed the repulsed victim showed lower arousal than those who viewed the aroused victim. It is also worth noting that, judging from clinical reports, the perception of the victim as having been willing or eventually sexually aroused by their assault appears to be a common theme among sex offenders regarding incidents of sexual assault which they committed. Gager and Shurr (1976) and Clark and Lewis (1977) have reported that many rapists have justifications or excuses for their actions, and are often able to portray their actions as understandable in light of the situations in which they found themselves. Investigating this justification process further, Scully and Marolla (1984) interviewed 114 convicted, incarcerated rapists and found that a sizable proportion maintained that the victim enjoyed being raped, even in cases where considerable doc- umented harm to the victim could be found. Obviously, these reports must be inter- preted with a great deal of caution, especially given the retrospective, qualitative research designs used. Whether the theme of victim arousal is an indicator of a pos- sibly socially acceptable post-hoc justification or an attitudinal precursor to action cannot be determined. These reports are worth noting here only because they show the consistency with which the theme of victim arousal arises as a key factor qualifying the vicious nature of sexual assault. Finally, it is important to emphasize that the findings regarding the salience of victim arousal for sexual arousal and the demonstrated effect of this factor in altering attitudes toward rape corresponds to psychological theory. Relying on theories of judgmental processes (Higgins, Rholes & Jones, 1977; Wyer & Sroll, 1981; Kahneman, Slavic EC Tversky, 19821, it is possible to see how perceptions of victim's reactions 22 could become highly influential in the formation of thinking patterns about the acceptability of sexual coercion. Presentation of sexual violence as a precipitant of victim arousal could either serve as the groundwork for a reorientation of sexual attitudes or as a reference belief that becomes influential in certain situations. `Ihe first potential process of attitude reorientation is rather straightforward. Depictions of sexually coercive scenes have the capacity to be arousing to a variety of individuals, but this arousal sequence can be inhibited in many by the perception that such incidents are really more harmful than sexually arousing to the victim. The perceived outcome of any social action is a powerful factor in determining whether an individual might consider himself likely to engage in that behavior. As a result, portrayal of sexual coercion as producing positive outcomes (in the form of sexual satisfaction for both parties) could negate possible inhibitory influences and instead produce a powerful conditioned pairing of sexual aggressive content and sexual arousal Once this pattern of arousal has been established repeatedly, it is likely that an individual wlV alter the inhibitory attitudes to justify the existence of the link between his own perceived behavior (his arousal) and his processing of sexually violent materiaL In short, he will begin to believe that women actually get enjoyment out of sexual victimization because it is consistent with his own newly acquired arousal patterns, A second possibility is that the portrayal of victims becoming aroused as the result of sexually violent coercion serves as a reference point that is easily accessed when situations similar to the ones portrayed are depicted or occur. In this formu- lation, an individual uses this general rule about victim reaction as an economical way of assessing the likely outcome of a situation in which there are several possibilities of uncertain probability. The scenes of sexual victims receiving pleasure may serve as an available orientation point when judging sexually coercive situations. Regardless of the factual or logical accuracy, what has been seen can become something to be weighed in any formulation of the likely outcome of future events. This easily accessible piece of memory could thus become influential in the expression of attitudes about these situations. The line of research on attitude change regarding the acceptability of sexual coercion is a convincing one when place in its theoretical context. It is important to remain aware, however, that the observed attitude changes are generally restricted to exposure using depictions of sexually violent incidents in which the victim becomes aroused as a result of the attack. Attitude changes from exposure to violence or sexually elrplicit behavior alone are not consistently observed. In addition, the effects of such materials on women's attitudes is relatively unexplored, and the few studies 23 that do exist (e.g., Krafka, 1985) seem to point toward potentially different patterns of effects. For reasons mentioned earlier, the effects of exposure to these materials on the attitudes of children and adolescents is an empirically open question. It is certainly reasonable to speculate, however, that the results of such exposure on less socially mature individuals with less real world experience to counteract any influ- ences of this material would be equally (or more) powerful than those seen in college students. Attitude formation in childhood is a matter of exploration and "trying on" of potential world views. Being exposed to one in which sexual coercion produces posi- tive results could likely influence adoption of attitudes condoning the use of sexual force as a reasonable alternative in intimate relationships. Acceptance of coercive sexuality appears to be reIated to sexual aggredon `Ihis conchsion is based on primarily correlational (but some laboratory based and longitudinal) evidence of a link between attitudes indicating an acceptance of sexual coercion and the presence of sexually aggressive behavior patterns. `Ihe association between attitudes and behavior in this area is one of the most difficult to comment upon conclusively from social science evidence, but it is obviously a lynchpin of informed policy regarding the effects of exposure to pornography. At the present time, it appears that there is evidence that attitudes indicating acceptance of coer- cive sexuality are, along with a number of other variables, related to sexually ag- gressive behavior. While there is the possibility that these attitudes are causally related to the development of this behavior, it cannot be said presently that these attitudes are causally related to this behavior. Moreover, it is not clear that exposure to pornography is the most significant factor in the development of these attitudes. A review of the types of available evidence on this question should illustrate why this is a question that requires such cautious interpretation. There are a limited number of ways that one can examine the relationship of certain attitudes to certain behaviors. First, people who have demonstrated a certain behavior can be assessed regarding their attitudes relative to other individuals who have not demonstrated the behavior. In the case of attitudes and behavior regarding 24 sexual aggression, this has mainly meant comparing the attitudes of rapists to those of nonrapists. An alternative strategy along this line has been to compare the attitudes of those who report having engaged in sexually coercive behavior to those who report no such involvement, regardless of whether such involvement resulted in an arrest. A second approach is to test whether these attitudes are related to specific expressions of aggression toward women in a controlled, Laboratory setting. Third, and ideally, one can look at the development of attitudes and behaviors related to sexual aggres- sion in the same individuals over time. In this way, the question otwhether attitudes precede and/or help predict sexually aggressive behavior can be addressed. The largest amount of information about the relationship of sexually coercive attitudes to sexually aggressive behavior has been collected using the first strategy; that of comparing the attitudes of groups with reported or demonstrated differences in behavior. As alluded to earlier, considerable clinical evidence exists that convicted rapists express beliefs that women are at least partially responsible for being raped or that a large number of women actually enjoy the experience (Gebhard, Gagnon, Pom- eroy & Christensen, 1965; Gager & Schurr, 1976; Clark & Lewis, 1977; Wolf & Baker, 1980; Scully & Marolla, 1984). This difference is apparently not totally consistent, however, with some investigators reporting that rapists in their samples expressed views that were different from, but not dramatically at odds with, those expressed by convicted felons or members of the general population (Feild, 1978; Burt, 1978). This lack of congruence of the above findings can probably be attributed to the different methodologies used in these investigations (qualitative analyses of interviews in the first set versus differences on questionnaires in the second set) and the effect of sample bias concerning the characteristics of incarcerated rapists. There is little reason to think that arrest for a sexual offense necessarily implies the presence of a homogeneous set of attitudes, and it is equally likely that such status implies a set of other social and personality characteristics (Koss & Leonard, 1984). Moreover, given the underreporting of sexual coercion, it is likely that any sample of normal subjects would be likely to contain a number of individuals who had engaged in some sexual aggression. As a result, differences in attitudes about the acceptability of sexual coercion could be masked in any examination of groups drawn with arrest as the pri- mary selection criterion. Other researchers, therefore, have looked at the relationship of attitudes about the acceptability of sexual coercion to self reported sexual aggression. Generally, these studies have administered scales assessing attitudes regarding rape myths or 25 sexual coercion to males (usually college students) and related these scores to either self reported likelihood to rape or self reported involvement in sexual aggression. A series of investigations (Malamuth, Haber & Feshbach, 1980; Malamuth, 1981; Tieger, 1981; Briere & Malamuth, 1983,) have shown a consistent relationship between a sub- ject's attitudes toward the acceptability of the use of coercion in sexual relationships and the self reported likelihood to commit rape. These studies have thus demon- strated that the measures used to assess attitudes regarding acceptance of sexual coercion appear to have some validity when compared to another measure of an indi- vidual's overall view of the acceptability of sexual coercion. Of more direct interest, however, are studies that have shown a relationship between' attitudes about the acceptability of sexual coercion and self reported behavior regarding involvement in sexual aggression. For example, Koss, Leonard, Bee&y & Oros (1985) found that attitudes toward sexual coercion and demographics provided a statistically significant model that predicted group membership between subjects classified as sexually assaultive, sexually abusive, sexually coercive, and sexually nonaggressive based on a sexual experiences survey. Similarly, Mosher & Anderson (1986) found significant correlations between components of a scale meas- uring what they termed a "macho" personality orientation (e.g., seeing danger as exciting or woman as submissive) and self reported activities of the use of force in sexual relations. Finally, in an elaborate test of several possible precursors to sexual aggression, Malamuth (1986) has shown that attitudes toward the acceptability of interpersonal violence toward women, hostility, and dominance interact with arousal variables and sexual experience variables to predict a subject's self reported sexual aggressiveness. Each measure alone did an unimpressive job of predicting the self reported activity, but a statistically significant model was constructed using the variables in an interactive fashion. Given the consistency of the results in these studies, it see'ms that favorable attitudes toward the use of sexual coercion are related to self reported likelihood of engaging in or having engaged in sexually aggressive behavior. As mentioned earlier, however, an unresolved issue is whether these attitudes led to different behavior patterns or whether the attitudes were adopted after the subject's behavior patterns were already established. Another bit of evidence that bears on the relationship of attitudes and behavior regarding sexual aggression comes from laboratory studies that have examined the be- havioral effects of exposure to sexually violent materials. Malamuth (1983) admin- istered both the Rape Myth Acceptance Scale and the Attitude toward Interpersonal Violence Scale (Burt, 1978) and measured the sexual arousal to a rape depiction in a group of 42 male undergraduates. He later tested the aggressive behavior of these 26 males (using an aversive noise task) toward a female confederate as part of an inde- pendent experiment ostensibly related to ESP transmission. In the later phase of the experiment, subjects were angered deliberately by the confederate in order to pre- cipitate aggression. Structural modeling of the results showed that a model predicting separate contributions for both arousal and attitudes best reflected the regularities seen in the data. Taken together, arousal and attitude scores accounted for 43% of the variance in the aggression measure. Malamuth and Check 0982) report that these results have been replicated and that a General Attitude toward Violence scale did not contribute significantly to the overall solution, leading these investigators to posit that attitudes specific to the use of coercion with women accounts for the observed Iaboratory aggression. Another study that has shown an association between attitudes about the ac- ceptability of violence toward women and the expression of aggressive behavior in a laboratory has been done by Donnerstein (reported in Malamuth & Donnerstein, 1984). In this investigation, male undergraduates were a.1 either angered or treated in a neutral manner by a female accomplice, b.) exposed to either erotic, aggressive, or sexually violent films, c.) tested for their attitudes regarding their willingness to use force in sexual relations, their willingness to commit rape if they would not get caught, and their acceptance of mpe myths (using the Bape Myth Acceptance scale), and d.) finally given the opportunity to aggress against the female confederate in a laboratory task involving the administration of shock to promote learning. Both atti- tudes and aggressive behavior were shown to be most affected by the exposure to the sexually violent film, less affected by the aggressive film, and least affected by the solely erotic film. This investigation demonstrated that similar effects for attitudes and behavior appear to occur from exposure to films of a sexual and violent nature. Up to this point, evidence has been presented that attitudes about the accept- ability of sexual coercion appear to be rather prevalant in individuals who have either been convicted of rape or admit to incidents of sexual aggression, and that these attitudes appear to be positively related to statements about the likelihood of engag- ing in rape and the expression of aggressive behavior toward women in a laboratory setting. What is missing from this picture is clear evidence that these attitudes cause the behaviors of interest. In order to be confident that attitudes are precursors rather than results of sexually aggressive behaviors, we would have to see the development of these attitudes precede the expression of these behaviors in the same individuals. Longitudinal research on this issue is sparse, however, and offers little in the way of clarification of this critical issue. In a study done by Alder (19851, attitudes legitimizing sexual agression toward women were found to be one of three factors 27 (including presence of sexually aggressive friends and service in Vietnam) that significantly predicted self reported sexual aggression. These data were taken from a longitudinal study of 239 randomly sampled youth in one county in the Pacific Northwest, but the data linking attitudes and self reported aggression were all taken at the last sampling time when the subjects were 31 years old. Although this study is from a longitudinal investigation, the results regarding the relationship of attitudes and behavior are contemporaneous. As a result, it must be stated that these findings only show that attitudes and behavior tend to coexist; a verification of earlier research but not a causal finding. The only study that appears to have been able to actually assess the relation- ship of attitudes and naturalistic sexual aggression over time is that of Ageton (1983). Using several hundred youth from the National Youth Survey of self reported delinquency, Ageton isolated a sample of males who reported incidents of sexual ag- gression and analyzed responses to particular questions-regarding sexual attitudes and behavior over the five year time frame of data collection. In these analyses, in- volvement in a delinquent peer group appeared consistently as the most powerful factor related to reported sexual aggression. In one analysis in which previous attitudes, behaviors, and reported group involvement were tested for their ability to discriminate later sexually assaultive behavior, attitudes toward rape and acceptance of interpersonal violence toward women did emerge as a statistically significant addi- tion to the discriminant function. In other words, presence of these attitudes did appear somewhat related to the expression of sexually aggressive behavior two years later. It is important to note,however, how marginally distinctive the contribution of this attitude factor really was. Involvement with delinquent peers alone was able to account for 76% of the variance in the self reported later sexual aggression, and inclusion of the three other factors (of which attitudes was one) accounted for an additional 1% of the variance. As Ageton (1983, p.119) was forced to conclude... "The idea that sexual-assault offenders are influenced strongly by stereotypic views of rape and sexual assault, traditional sex-role attitudes, and liberal beliefs about the use of violence was not borne out in these analyses. None of these variables consistently differentiated the offenders and the nonoffenders in the annual comparisons. Furthermore, in the initial discriminant analyses, only the measure of rape attitudes contributed to the separation of the groups. Clearly it is not a critical factor, however, since its absence in the discriminant analysis involving just delinquency variables had little effect on the accuracy of 28 the classification. Although we do not deny that the constellation of beliefs about male and female roles, behavior, and sexuality may influence sexual acts, the data do not indicate that such attitudes play a major role in predicting sexual assault." In the end, the clarity of the causal Iink between pre-existing attitudes and later sexual aggression has yet to be demonstrated. The fact that this direct link has not yet been demonstrated can be seen as at least partially the result of the difficulty isolating the effects of attitudes clearly in any longitudinal or cross-sectional research design. In the Ageton study, for example, it is very difficult to determine where the effect for peer association ends and the one for attitude begins. Attitudes are obviously related to one's choice of friends inan intricate fashion, and disentangling these two constructs is difficult within one study. There is some longitudinal evidence from related areas that attitudes can be impor- tant interactive variables combining with identification with television characters in producing pee-elated aggression in children (e.g., Huesmann, Eron, Klein, Brice, & Fischer, 1983) or with personality and behavioral characteristics in producing anti- social behavior in adolescents (e.g., Jessor h Jessor, 1977). However, in these cases, as in the case of the effects of pornography, the independent effect of attitudes has not been shown to be large or direct. Although there is an association between acceptance of coercive sexuality and sexual aggression, the difficulty comes when we attempt to assign a strict causal interpretation to this association. From one perspective, it is quite logical that favorable attitudes toward sexual coercion should be more prevalant in individuals judged to be more sexually aggressive if we expect one's attitudes to be consistent with one's behavior. At present, however, it cannot be stated conclusively that attitudes favorable to sexual coercion produce sexual aggressiveness. These two appear together, and the exact mechanisms of causation and strength of relationship between them has yet to be determined. hlabomto?ystlldks fnw short term effe!cts, fzxpoeetoviolent~ incream punitive behavior toward women An increase in aggressive behavior toward women has been proposed often as one like- 29 ly effect of exposure to pornography, but there does not seem to be sufficient sci- entific support for a generalized statement regarding the presence of this effect. There is no paucity of hypotheses about how exposure to pornogmphy and aggression against women may be related. For example, men predisposed to aggression who are aroused by pornography may translate this increased arousal into targeted aggression; attitudes or restraints toward aggression might be changed in pornography viewers, increasing the chance that they would act aggressively; or, certain behaviors of the film victims (e.g., saying "no") might take on cue properties for aggression, promoting aggressive behavior in men toward women who display these behaviors. The limited statement above, however, reflects the circumscribed links between exposure to por- nography and aggression toward women that have been scientifically demonstrated. Testing for this effect in a natural setting is clearly impossible. It would be unethical to expose persons to pornography and then observe their level of aggression toward those encountered outside of the laboratory. A series of investigations have been done in the laboratory, however, that examined this effect in analogue situa- tions. The first experiments found that sexually aggressive films caused more punitive behavior than sexual nonaggressive films. Further investigations indicated that the outcome depicted in the sexually aggressive films created different levels of punitive behavior. In order to tease apart the effects that the aggressive and sexual themes of the films had, investigations into the effects of nonsexual aggressive films were then pursued. Aggression in recent laboratory studies is usually represented by the "Buss paradigm" (Burs, 1961). Subjects am told that they and a second subject (actually a confederate of the experimenter) will be involved in a learning experiment. After exposure to the pornographic material, the subject is instructed to choose and administer one of several possible levels of shock or aversive noise when the con- federate (who is out of the sight of the subject) makes an incorrect answer. Higher levels of shock are taken to represent higher levels of punitiveness or aggressiveness. In some experiments, the confederate angers the subject before exposure to the por- nogmphy by the reacting derisively to an opinion expressed by the subject. This pamdigm as an analogue for aggression outside of the labomtory has been criticized on several grounds. Ihe most general criticism has been that the lab task is not an appropriate representation of aggression. However, there has been some (albeit circuitous) evidence for behavior in the Buss paradigm as an adequate analogue of sexual aggression: Subjects who report higher levels of sexually coercive behaviors also score higher on scales designed to measure beliefs in rape myths and acceptance of sexually coercive behavior (e.g., Briere & Malamuth, 1983; Tieger, 19811, and sub- jects scoring higher on these scales have provided higher levels of shock In the Buss 30 pamdigm (Malamuth, 1983). These results indicate that there may be a link between aggression on the Buss paradigm and reported likelihood to act in a sexually coercive manner. How well this lab task and actual aggression, rather than attitudes, are related is still an open question. Two other criticisms regarding this design are common to many lab studies. The first is that the subjects provide shocks because they believe that the experi- menter wants them to do so. The assumed effect of this belief is that the subjects appear more aggressive in the laboratory than they are outside. A final criticism is that the responses allowed the subject are more limited than, and not representative of, those that are available outside of the labomtory. Some experimenters allow other responses (such as rewarding the other person), yet even then the subject's choices are unrealistically constrained. While these two criticisms may be valid when genemlizing from behavior within to that outside .of the laboratory, they do not appear to account for differences between groups within the laboratory (see Kmfka, 1985, for a more detailed critique). Several studies have used this general approach to aggression research to examine the effects of sexualXy aggressive fiIms, nonsexual aggressive films, and sexual nonaggressive films on laboratory behavior. In the first of two experiments, Donnerstein and Berkowitz (1981) first had a male or female confedemte anger a male undergraduate, and then showed these subjects a film of either a neutral talk show, an erotic nonaggressive scene, a sexually aggressive scene in which the female finally enjoys being aggressed against, or the same sexually aggressive scene with the woman continuing to abhor the aggression she experiences. Immediately after the films, the subjects administered shocks or rewards to the confederate who had angered them earlier. The amount of punitive behavior exhibited by those paired with a male con- federate did not change significantly regardless of the type of film seen. Those paired with a female confederate, however, provided more punishment after viewing either of the sexually aggressive films compared to the erotic or the neutral film, even though the erotic and sexuatiy aggressive films were equally arousing. There was no difference in punitive levels between those viewing the erotic film and the neutral film. The results paralleled an earlier study by Donnerstein (1980). This experiment showed that punitive behavior is increased in angered subjects after watching sexually aggressive films. `The authors concluded that it was the modeling effects of the association of the female confederate with the female victim in the film, rather than heightened arousal, that caused aggression against the female and not the male confederate. This conclusion is strengthened by an earlier study by Donnerstein and Hallam (1978) in which nonsexual violent films that consisted of male 31 to male violence caused more punitive behavior toward male than female confeder- ates. The authors also concluded that the aggression depicted in the film lowered the subjects' inhibitions against aggressive responses. Since the erotic nonaggressive film did not show the aggression toward a female, it did not lower inhibitions and therefore did not increase punitive responses after viewing. The finding of no difference in punitive behavior for the different types of sex- ually aggressive films did not seem logical to these investigators. Previous attitude research (see Malamuth, 1984) had shown a differential effect for depictions in which the victim was aroused versus those in which the victim was not aroused, and it seemed that this factor might also be influential in experiments with punitive behavior as an outcome. A second study was thus undertaken to explore this possibility further. The procedures were the same as in the first experiment, except that all of the confederates in this experiment were female and only half of the subjects were angered. This allowed for a test of the intemction between the anger manipulation and the type of film seen. As before, the neutral and erotic films were followed by similar levels of punishment. Those previously angered provided significantly higher punishment following both sexually aggressive films, while those not previously ang- ered provided significantly higher punishment only to the sexually aggressive film in which the woman portrayed a positive outcome. However, this experiment also showed that the outcome of the sexually aggres- sive films (i.e., whether the victim is aroused) interacts with the emotional state of the viewer in the production of punitive behavior. For angered subjects, both sexually violent films had the same (increased) effect on punitive behavior. For the non- angered subjects, only the film with the aroused subject resulted in higher punitive behavior. The authors concluded that all those watching the sexually aggressive film with the aroused victim had their inhibitions lowered by seeing the males' sexually aggressive behavior in the fiim eventually result in a positive outcome for both the attacker and the victim. The aggression of those watching the aggressive film with the repulsed victim, however, was affected by their previous angering. Those who were previously angered were predisposed to hurt timeone and consequently reacted to the pain cues of the victim with heightened aggression; those who had not been angered did not react to the pain cues of the victim, nor were they less inhibited since the outcome of the film indicated a negative outcome of the male's aggression. An untested implication of this conclusion may be that those who fiid the negative outcome of these films (i.e., a victim's pain) to be pleasurable would be more aggressive following the film. In two later studies, Donnerstein (1983,1984) investigated the unique effects of the aggressive and sexual components of sexually aggressive films using a nonsexual aggressive film, a sexually aggressive film, an erotic nonaggressive film, and a neu- 32 tral film. As had been shown earlier, with male confederates, the type of film did not affect the intensity of the shocks that previously angered male undergraduates deliv- ered. The levels of shock intensity delivered to the female confederates were similar following the neutral and nonagressive erotic films, were significantly higher following the nonsexual aggressive film, and were highest following the sexually aggressive film. The study shed light on the role of arousal in producing laboratory aggression. The nonsexual aggressive film was less arousing than either of the sexual films, but pro- duced aggressive behavior. The highest level of aggressive behavior, however, was produced by the sexually aggressive film. Arousal, therefore, did not appear to be a necessary condition for the production of aggression, but when paired with exposure to aggression (as in the sexually aggressive film), the arousal appeared to add signficantly to the level of lab aggression produced. As a group, these studies consistently indicate a relationship between type of film and the intensity of shocks delivered immediately after the film is viewed. Films that contain aggression or sexual aggression are associated with higher levels of shock to a confederate of the same sex as the person aggressed against in the film. These results consistently show that nonaggressive sexual films are associated with similar levels of shock as nonsexual-nonaggressive neutral films. It should be noted, however, that the fiidings about sexual nonaggressive films in these studies are not consistent with earlier studies where previously angered subjects reacted punitively foIlowing sexually explict, nonaggressive films (e.g., Meyer, 1972; Jaffe, Malalmuth, Feingold h Feshbach, 1974; Donnerstein & Barrett, 1978; Donnerstein & Hallam, 1978). Donnerstein and Hallam (1978), however, found that male subjects rated erotic fflms chased for their nonaggressive content as 2.5 on a S-point aggression scale. This suggested that the subjugation of the women in the film may have been interpreted as aggression, facilitating later aggression from the subjects. It may be, then, that the perceived aggressive content of the nonaggressive erotic films was varied across these studies, thus producing contradictory results. Other studies have investigated the effects of variables that might mitigate the effects that sexually aggressive fllms have on punitive behavior, and the length of the effects that sexually aggressive films may have on punitive behavior. Malamuth (1978), for example, had male undergraduates read passages containing either aggres- sive pornography, nonaggressive pornography, or neutral stimuli. After being angered by a female confederate, subjects were given the chance to aggress against her using electric shocks. Before delivering the shocks, half of the subjects read a passage designed to make them more self-conscious about delivering the shocks (an inhibiting communication), while the other half read a passage suggesting that it was permissible to give as strong shocks as they wished (a disinhbitlng communication). Shock inten- sities delivered were similar for all males who read the inhibiting communication, 33 regardless of the pornographic passage read. More intense shocks were given fol- lowing the aggressive than the nonagressive pornography in the group who read the disinhibiting communication. In contrast to other studies, however, the levels of shock intensity given after the neutral film were between those of the two pornographic films, and were not significantly different from either. Malamuth offered no explana- tion for this conflicting result, but did state that the results indicated that internal states or external situational variables can have a significant effect on behavior following exposure to pornography. One important influence may be previous exposure to pornography. As part of a larger project, Zillmann and Bryant (1984) assigned male and female undergraduates to see varying amounts of nonaggressive pornography one day a week for six weeks. At the end of the exposure period, members of each group saw one 8 minute film of either nonagggressive sexually explicit material, sadomasochistic material, bestiality, or were included as a no-exposure control subject. Subjects were then allowed the opportunity to inflict pain via a blood pressure check, upon a confederate who had just inflicted pain upon them in the same way. Those watching the sadomasochistic and bestiality material inflicted more pain than those watching the erotic matrial, who inflicted more pain than those in the no-exposure control group. However, for those with the massive previous exposure to nonaggressive pornography there was no differ ence in aggression across the four final stimulus conditions. It appeared that the prior exposure mitigated the effects of viewing the fiial films. Malamuth and Ceniti (1986) investigated the length of these behavioral effects. They exposed male undergraduates to either sexually aggressive pornography, nonag- gressive pornography, or a control condition of no exposure. Each of the exposure subjects saw two full-length films each week for three weeks, and were then given two book chapters portraying sexual activity to read during the fourth week. One week after the exposure period the subjects were involved in an ostensibly separate experi- ment purporting to measure the effects of punishment on an ESP task. The results indicated that there were no significant differences in punitive behavior related to type of film exposure. These findings may indicate that the behavioral effects found by Donnerstein following sexually aggressive films do not last for a week. However, study differences regarding type of learning task presented to induce aggressive behavior may aIso ex- plain the differences in results. While Donnerstein's experiments involved the confederate learning material independently of the subject, Malamuth and Ceniti had the subject "send" the message to the confederate. A lack of punitive behavior on the part of the subject may have been at least partially influenced by doubt that he had been successful in his part of the task. Also, Malamuth and Ceniti did not question the 34 no exposure control group about any sexually aggressive or erotic films that they saw on their own during the exposure phase of the experiment, and the overall film viewing of the three groups may have actually been quite similar even though the experimental exposure conditions were different. Because of these differences, it is difficult to draw conclusions about the length of any effect on behavior produced by these experi- mental manipulations. At present, it is known that these effects can be produced in the lab, but how long they last is still in question. In summary, published studies have consistently indicated that punitive behav- ior can be produced within the confines of the laboratory by the viewing of sexually aggressive and aggressive films. This observed pattern of results fits well with previously tested psychological theory concerning the effects of modeling, with heightened aggressive behavior seen foLIowing the modeling of such behavior in films. The more closely the confederate in the experiment is aligned with the victim in the aggressive film, the more aggressive the behavior toward the confederate is. Of equal interest is the fact that arousal also appears to be an influential element in producing this behavior, having both a separate and combined influence with modeling. Sexual aggressive films produce more aggression than do less arousing nonsexual aggressive films. In addition, previously angered subjects (i.e., more aroused) show more aggres- sive behavior than nonangered subjects. Several questions about the widespread nature of this phenomenon remain. For example, the relative effects of modeling and arousal across individuals has yet to be explored. Also, the duration of these effects for producing aggressive behavior is unknown and it is important to remember that only immediate effects have been shown. Similarly, the effects on behavior outside of the laboratory, where behavioral choices are much more varied and where there are additional inhibiting and facili- tating factors on behavior, are unknown. ' On one hand, these data could be viewed with concern, since they show changes from LittIe exposure; after one exposure to a sexualry violent film, men are more aggressive to women. On the other hand, there is no evidence that those viewing sexually aggressive films under more normal circumstances have become more aggres- sive immediately after the films are viewed. Reports of this causal relationship being a noticeable one in the real world have not emerged consistently (Byrne & Kelly, 1984). In sum, these experiments should heighten concern that aggressive behavior toward women may be increased by viewing aggressive and sexually aggressive films, but presently this effect has only been seen in controlled and potentially artificial laboratory settings. 35 Summary of Consensus Statements The preceding statements reflect what the Workshop participants believe can con- fidently be said regarding the effects of pornography. Pornography has been consis- tently linked to changes in some perceptions, attitudes, and behaviors. These links, however, are circumscribed, few in number, and generally laboratory-based. To say that this means that any observed effects are artifactual, however, would be an error. Pornography does have effects; it is just not yet known how widespread or powerful they really are. `There is a clear lack of extensive knowledge or unifying theory, and global statements about the effect of exposure to pornography have not yet been substantiated. Currently we have bits of knowledge about the effects of pornography; future research is required to unite these bits into a more comprehensive statement. While convincing evidence exists about the effects of pornography on perceptions, attitudes, and behavior, it is important to remember that the rela- tionships between measured perceptions, attitudes, and behaviors themselves are un- clear. They are not necessarily linked in a straight causal pathway. For instance, while it is a common belief that attitude changes lead to behavioral changes, research has consistently shown otherwise. Behaviors are as likely to influence attitudes as attitudes are to influence behavior. As a consequence, conclusive statements about the extent to which attitude or perceptual changes brought about by pornography are ever reflected in changed behavior are not currently possible. This absence of clear information does not argue for dismissa of the hypotheses that perceptions about the frequency of uncommon acts influences attitudes or that attitudes toward sexual coercion promote sexually coercive behavior; it only argues that there is not a clear causal link yet demonstrated. Indeed, the evidence of the coexistence of the very attitudes that are shown to be affected by sexually violent pornography and self- reported or verified sexual aggression is an association that warrants concern and continued investigation. There is substantiation for the basic concern that sexually violent material has more consistent and marked effects than nonviolent erotic pornography. Attitudes condoning sexual coercion have been fostered by materials combining sex and vio- lence, particularly those in which sexual assault eventually produces arousal in the victim. Increases in punitive behavior toward women in a laboratory setting have also 36 been observed after exposure to sexually violent materiaL Although the findings re- garding the unique effects contributed by the violent and the erotic material have not been fully elaborated, their combination has repeatedly brought about attitudes and behaviors which are considered to be antithetical to the aspirations of our society. There is clear evidence that youth involved in the production of pornography are adversely affected by their participation. It is unreasonable and contrary to reported experience to think that such sexual victimization will not leave its mark on these children as they attempt to forge a social image of themselves. Programs to stop the production of child pornography and to intervene with the children involved in its production are clearly needed to forestall these consequences. Summary of Effects on Children As has been noted throughout this report, because of ethical concerns it is impossible to directly test for effects of pornography on children and adolescents. As a result, research on effects of exposure has been done almost exclusively on college students in late adolescence. Estimating the extent to which these effects can be generalized to children and young adolescents is a complex process, requiring interpretation and integration of knowledge regarding child development. This task is beyond the scope of this section, but a few caveats could be helpfuL As children mature they develop new cognitive and emotional skills, and their interests shift. As a result of these changes in basic understanding and orientations, the message that an 8- or 12- or 16year+ld would get from a certain pornographic movie may be quite different from that of an 18year+ld. In addition, relationships to family and peers undergo considerable change in the ten years between 8 and 18, meaning that the mediating factors on any effects will also shift during this develop- mental period. Finally, there is the likelihood of a "cohort effect" interacting with normal patterns of development. Growing up in the 80'9 is different than growing up in the 60'3, and so&historical changes can affect the rates of many things from juvenile crime to views of interpersonal relationships. These simultaneously changing factors make it extremely difficult to predict the exact effects of particular influences on children. Reasonable prediction can really only be done regarding influences that would be so powerful as to exert marked negative or positive effects in spite of shifts in other influences. Repeated physical 37 abuse of a child, for example, could clearly be stated to produce significant negative effects regardless of shifts in other factors in the child's life. It is likely, however, that exposure to pornography is in that vast grey area of influences that will probably not be shown to act so strongly as to be consistently harmful independent of changes in the child and the social situation. There are certain changes in the child, however, that are rather predictable and probably important in mediating the effects of exposure to pornography. Many developmental psychologists believe that around the age of 12 or 13, children aquire many of the basic thinking processes that they will use, with refinements brought about through experience, throughout the rest of their lives (e.g., Colby, Kohlberg, Gibbs & Leiberman, 1983). Usually by this age, children are able to understand social causation and the i&a of reciprocity in interpersonal and social relationships. If this is the case, one might expect that the effects of pornography 6n those 12 years or older would be generally similar in quality to those seen in 18yearolds. A host of other factors, however, could magnify or reduce these effects. For example, it is possible that the effects could be magnified in those who are 12 years old since they do not have the number of life experiences against which they can compare the contents of pornography in order to determine its accuracy. On the other hand, young adolescents may not be as interested in the sexual activity of the por- nography as older adolescents, and their resulting lowered attention may reduce the effects of viewing the pornography. Unfortunately, the exact direction or magnitude of differences resulting from factors such as these between young and old adolescents is largely a matter of speculation. What does seem worthwhile speculating further upon and investigating, however, is the general question of the effects of cognitive level and an adolescent's attributional framework regarding social relationships as mediators that increase or decrease susceptibility to pornography exposure. Speculating about the effects on children less than 12 years of age is even more of a problem. Younger children think in a qualitatively different manner from those on whom research regarding the effects of pornography are done. In addition, the focus of interest may be quite different in those who have not completed the emotion- al and physical changes that accompany puberty. The fear of some ls that the sexual and emotional patterns to be followed by these children when they are grown will be "imprinted" on them by seeing pornography at a younger age. Others believe that young children are less affected since they do not have the cognitive or emotional capacities needed to comprehend the messages of much pornographic materiaL Again, though, we really do not know which of these statements is more accurate, and each of them may be accurate for specific individual children. 38 A final factor that makes extrapolation of these results to children troublesome is that the susceptibility of children to a variety of influences has been shown to vary widely. Children bring individual temperments and adaptive skills to situations, and the predictability of how particular influences will affect a child is lower than we might expect. Children are amazingly resilient to a number of influences, and the exact causes of this resilience are not clear (Kagan, 1985). In the end, then, it ia really rather difficult to say much definitive about the possible effects of exposure to pornography on children. The direct research is not present, and probably never will be. `There are reasons to believe that the effects seen in older adolescents would probably generalize to younger adolescents as well, but we know little about the possible mediators of these effects that may be different in younger adolescents. Finally; effects on younger children are very difficult to determine, given their malleable and adaptive natures. 39 FUTURE RESEARCH AGENDA As indicated by the previous statements, social science has provided conclusive evidence about only a few of the possible effects of pornography on children, adolescents, and adults. Much remains unknown. If researchers are to make a substantial contribution to future policy decisions in this area, exploratory applied studies about the effects of pornography in everyday life and laboratory studies clarifying the theoretical mechanisms related to exposure must be done. The fact that much still needs to be done in this area is not too suprising. For one thing, research into the effects of pornography has occurred primarily over only the past 15 years. Several questions, such as the effects of prolonged exposure or the effects of sexually violent materials, have only very recently been investigated. In addition, numerous related studies, rather than any one diapositive study, must be pursued. Many questions in this area cannot be addressed directly because of ethical constraints, and investigations of a somewhat tangential nature must be done instead. Converging evidence from many tangential studies using different methodologies must then be related before conclusive statements can be made confidently. Delineating which studies should be undertaken was the task of a second con- sensus session of the Workshop, and a far-reaching agenda for future research was devised. Time constraints prohibited the development of specific research designs, but selected areas that require more investigation were isolated. It is hoped that researchers will use this formulation of needed research as a set of general guideposts in designing specific research questions and approaches. The agenda laid out here requires both field and laboratory studies, and there is considerable advantage to pursuing questions in this area using both strategies simul- taneously. Certain leads that have come from laboratory investigations (e.g., the dis- tinct effects of violent sexual material) now need to be examined in more realistic settings. Conversely, certain real world phenomenon of interest (e.g., habituation to exposure) must be fleshed out theoretically by systematic laboratory investigations. Overall, this area of inquiry is in an ideal position to advance if methods are tailored to existing questions rather than vice versa and an active exchange is fostered be- tween laboratory and field researchers. The extent to which any of this research agenda can be accomplished will depend on the willingness of private foundations and various government agencies to provide funding. A number of the proposed investigations (particularly longitudinal studies and those that require large numbers of paid subjects) could be quite expensive. Given the clear need for more information on this topic, however, the amount of funding provided for research will reflect the priority that society places on addressing the effects of pornography. Definitional Issues The term pornography has been used in so many ways that its meaning has become increasingly ambiguous. Researchers and policy makers alike may be thinking of a wide range of materials when trying to speak in common terms regarding pornog- raphy. The inevitable confusion resulting from these differing interpretations hampers the design and interpretation of social science research as well as the application of knowledge to policy. More descriptive studies regarding the existing types and prevalance of certain materials and s&vey research regarding people's implicit categorizations.of pomog- raphy are needed. This information is central to the development of a usable typology of pornographic materials to guide research and discussion. Standard descriptive terms for the various types of pornography that take into account the levels of sexual explicitness, violence, and types of participants would be a starting place in this effort. A subsequent, more involved approach would be to develop dimensions or axes along which different types of materials could be characterized in a multivariate fashion. Regardless of the exact form such a typology takes, however, it is essential that it both reflects the range of materials found in the marketplace and has theo- retical relevance to social scientists and policy specialists. Clear definitional boundaries are particularly needed across the range of sexual activities between completely mutually consenting sexual activity and rape. `Ihe fact that depictions of rape seem to have particularly powerful effects in controlled research studies and the problems with estimating the incidence of rape both highlight the need to focus on this content area in efforts at definition. Questions such as the extent to which certain activities (e.g., expecting sex as part of an unequal status relationship) fit the defmition of coercion need to be discussed and debated. 41 Production, Distribution, and Consumption of Pornography More accurate information regarding the production, distribution, and consumption of pornography is needed to focus research onto topics of real world relevance and to target prevention and intervention programs. Currently, information regarding the amount of various types of pornography being produced and used is sparse and often contradictory. Widely varying estimates can be found about the amount of production, the numbers of children and adults involved in production, and the number of consum- ers. Patterns of use are not any clearer, and accurate information about the extent to which certain types of pornography are used by different groups of children, adoles- cents, and adults is needed. Information regarding typical consumption patterns of children, adolescents, and adulta is particularly critical to hrture theoretical and applied research. Of most concern are the effects resulting from long term, repeated exposure occurring in contexts that promote unfavorable reactions in children, adolescents, or adults. Knowing how much is too much or what conditions are critical to changes in attitudes or behavior are questions that cannot be posed meaningfully until the parameters of current patterns of use are clarified. Descriptive studies in this area should provide information about the most popular type of pornography among different age groups, the reasons for its popularity, the amount that it and other forms of pornography are viewed, and the environment and context in which it is viewed. Developmental Patterns Research has indicated that individuals have different patterns of reaction to sexual and aggressive material. Abel and his colleagues, for instance, have shown that rapists are aroused by aggressive material more often than nonrapists, and that pedophiles are aroused by sexual material involving children. Exactly how these arousal patterns develop, however, is still rather unclear. We are unaware of, and need to know more about, the timing and mechanism of this area of development. 42 There needs to be careful longitudinal research on the role of early develop- ment and of later experience in the shaping and changing of these patterns. There is little doubt that pornography has different effects on different people and that a majority of sex offenders make regular use of these materials. The next problem to address is the one regarding how these patterns of use and sexual arousal are inter- twined in the developmental process. Related to these developmental questions are other important questions about propensity to commit sexual aggression and the role of pornography in activating that propensity. Malamuth, Check, and Briere (1986), for example, found that some college students' arousal to scenes of aggression was similar to the arousal observed in rapists, but these students did not report engaging in any sexually coercive behavior. If arousal patterns develop early and if certain patterns would seem to place an individual at risk for being sexually coercive, it is important to also explore the reasons why some individuals who would seem to be at risk do not exhibit sexually coercive behavior. The way that sex education can affect development and sexual expression should be investigated as part of this effort. This information would have direct applicability to the prevention and treatment of sexually coercive behavior as well as theoretical relevance to the understanding of the mitigation of arousal. Increased general theoretical work on the link betwen sexual and aggressive behavior would also be of value in providing a larger framework for these focused studies. Several theories have been proposed that link sexual and aggressive behavior (e.g., Barclay, 1971). Questions remain, however, about whether aggression enhances sexual desire or sexual arousal in anyone, everyone, or some paticular set of people. A clearer understanding of any Iink between these two general energies or behaviors would provide a structure that could facilitate the interpretation of specific findings about effects of exposure to sexually aggressive materials and arousal patterns. Stability of Observed Changes Further laboratory research is warranted to replicate and extend the existing lines of investigation on the effects of pornography on perceptions, attitudes, and behaviors. Replication of previous studies would add credence to (or call into question) earlier findings and clarify the potential for generalizing from these results. Most directly, inclusion of subjects from segments of the population that have not been closely ex- amined (e.g., women, nonstudents of college age, those older and younger than college 43 age) in replication studies would be important steps in determining how far we can extend the findings of earlier research. Also, the duration of behavioral and attitudinal changes seen in the laboratory needs to be more clearly determined. Almost all of the investigations done to date have been concerned with immediate changes, and the susceptibility of any of these changes to mitigating life influences over time is still basically unknown. Although presenting many ethically thorny issues, this question is central to knowing how much significance should be given to the series of present findings. If these effects appear to be rather stable, the issue of the effects of pornography takes on increased social importance; the effects become more that just the short-term results of a clever research design manipulation. The effects of nonviolent sexually explicit pornography is another question that laboratory studies should continue to investigate. While the behavioral effects immediately following sexually violent pornography have been consistent, those following nonviolent sexual films have not been consistent. Further investigation in this area could contribute to the development of theory about the aspects of the pornographic presentation or viewing context that interact to produce harmful effects. Habituation Research by Zillmann and others has indicated that those who watch pornography on a regular basis in a laboratory setting eventually become less aroused when they see it; they become "habituated" to the pornography. This habituation, however, appears to be limited to only the type of pornography that they have been watching. Subjects' arousal returns to earlier levels when they watch a new type of pornography. This phenomenon has been investigated in only a few studies, but it should be analyzed further because of its potential importance for our understanding of long term use of pornography. It is possible that this mechanism is an important aspect of a process by which certain individuals become interested in more deviant forms of pornography. If regular users do tire of particular types of pornography, they may be more inclined to seek out more uncommon pornography in order to maintain enjoyable levels of arousaL One of the implications of this hypothesized process could therefore be that the ready availability of pornography could habituate users on a grand scale, contributing to an increasing spiral in the deviance portrayed in new pornographic ma- 44 terials. On the other hand, habitutation may make many individuals use pornography selectively or not use pornography extensively because of the awareness that its effects are short-lived. More deviant forms of pornography may just not be that in- trinsically arousing and patterns of use may be altered to accomodate to the phenom- enon of habituation. Regular users may seek intermittent exposure to the same type of material in order to achieve acceptable arousal. Presently, we do not know how habituation works to promote different patterns of preference or use. In order to gain a fuller picture, more laboratory studies of the limits and mechanisms of habituation in different populations will have to be conducted. At the same time, it will be important to conduct field-based research of extensive users, focusing on how habituation affects selection of different types of pornography and the schedule of exposure. Documentation of Natural Changes Researchers should also attempt to capitalize on naturally occurring changes in viewing patterns or availability of pornography to assess simultaneous changes in potentially linked phenomena such as sexual assault or attitudes. Retrospective cor- relational studies of social indicators that have been done in this area are generally marred by their inability to rule out the possibility of mediating factors being responsible for observed associations. Examination of changes in the frequency of events in a small region over a time period during which the availability of por- nography clearly increased at aparticular point would make a stronger test of any association. For example, examination of behavior or attitude changes corresponding to the availability of cable television carrying pornographic movies would give more exact information than is presently available. These studies could possibly use a variety of quasi-experimental methodologies, such as interrupted time series analyses or repeated observations of selected groups. What must be done in every case, however, is to anticipate a change in order to gain adequate baseline data before the level of avaiIability increases or decreases. Such foresight and planning is necessary to add to the presently limited set of correlational studies on social indicators. 4s Attitude Research The present body of studies on the effects of pornography on perceptions and attitudes warrant more elaborate investigation. We presently know that certain perceptions and attitudes toward sexual coercion can be altered in a laboratory setting by exposure to specific types of materials. It would be worthwhile to determine what other perceptions and attitudes might be affected by exposure and how the development of affected attitudes is linked to behavior. The concern with attitudes toward sexual coercion has been primed by the relevance of this attitude formation process to our understanding of the subjugation of women and the etiology of sexual aggression. There is the possibility, however, that exposure to pornography also has an effect on the formation of other attitudes of interest. Attitudes toward such things as traditional family values or the attractiveness of one's partner (see, e.g., Zillmann, in press) may also be affected by repeated exposure, and changes in attitudes like these would be important to know about in order to assess the Iong term effects of viewing pornography on the quality of interpersonal relationships. Of equal or even more importance, however, ia the need to conduct longitudinal investigations regarding the development and influence of attitudes on sexual behavior. Longitudinal studies, similar to Ageton's (19831, that chart attitudin