Media Filters and the V-Chip– Part III

Edited Version Originally Published in 45 Duke L. J. 1133 (1996). 
Copyright 1998 Jack M. Balkin. All Rights Reserved.

VIII. The V-Chip and the Delegation of Informational Filtering


So far, I have spoken only about the constitutional issues raised by the V-chip. Yet the deeper problems that the V-chip raises lie elsewhere, and it is likely that these problems are not constitutionally cognizable ones. They concern the power over individual thought and national culture that arises with increasingly powerful forms of delegation of informational access. This problem is by no means new. Delegation of informational access has always existed in one form or another. But my concern is that, in the Information Age, the shape of culture will increasingly be determined by those persons and organizations who organize, filter, and present information for others and to others. I fear that neither the proponents nor opponents of the V-chip fully grasp this fact. Although these features already exist in the world we now inhabit, they will surely be magnified in the world we now enter.

The regulatory apparatus surrounding the V-chip will work an enormous new delegation of informational filtering to bureaucractic institutions, whether operated by the federal government or by private industry.(54) This new bureaucracy will be entrusted with the task of devising and implementing filters for virtually all of the television programs available in the United States. It will have to determine both the salient characteristics of all programming and evaluate which programs fit within the boundaries defined by these characteristics. These characteristics and these evaluations will in turn be employed by viewers and, more importantly, by advertisers, cable providers, video rental stores, public libraries, television production companies, writers, composers, and directors. As these evaluations become commonly employed, further choices and social arrangements will then be organized around them. In this way, the divisions of the cultural and informational world created by the custodians of the V-chip, however innocent, will be amplified throughout our culture, shaping and skewing the social world in unforeseen ways. It is possible that we shall have nothing to fear from these effects. But it is equally likely that there is much to fear. It is probable that some version of these effects is inevitable. But it is certain that no particular version is inevitable.

Filtering mechanisms are not neutral means of organization, blocking and selection. They have important effects on what kinds of materials are subsequently produced and how social arrangements are subsequently organized. People who produce and receive information respond to and organize their lives around the existing forms of filtering. I do not yet know the many ways that the filtering mechanisms devised for the V-chip will affect our culture. Indeed, I am quite sure that we will not be able to recognize them for many years after they have already taken hold. All I can do here is offer the most minor examples of mechanisms that may have major consequences.

I want to focus on three basic kinds of effects. The first has to do with what characteristics are salient in forming categories–for example, bad language or nudity. The second has to do with coarseness–how fine-grained the filtering categories are. The third concerns equivalency–what kinds of things are seen as parts of equivalent categories. These factors overlap, but they are also distinct. Two ratings systems can be equally coarse and yet view different characteristics as salient. Moreover, two ratings systems can be equally coarse and view different sorts of things as equivalent in each category. Consider two ratings systems that each have only two categories. The first system holds that any profane language or any mention of contraception places a program in the adult category, while the second includes only profanity. The two systems are equally coarse, but they have different senses of equivalency. In the first system, profanity and discussions of contraception are treated alike as inappropriate for children; in the second system they are treated differently.

The first problem of any ratings system is what characteristics count in making programming unsuitable for children. The industry television ratings system in effect since January 1997 focuses on the categories of sexual content, nudity, violence, and profane language; these factors basically track the considerations currently employed by the Motion Picture Association of America (MPAA) ratings system.(55)

Racist, sexist, and homophobic depicitions are not specifically included as salient categories. Yet if parents are concerned with what their children pick up from television, they might be particularly concerned whether their children are picking up habits of intolerance. The harm to our children from these influences, one might think, would be equally as great as the harm from exposure to sex, violence, and bad language. And both sets of criteria involve content-based distinctions.

It is even less likely that an FCC-appointed television advisory committee would code for racist, sexist, or homophobic expression. The Telecommunications Act of 1996 expressly states that ratings systems are to avoid political and ideological categorizations.(56) Such a commission would probably argue that coding or blocking programming as racist, sexist, or homophobic would give the unmistakable appearance of political favoritism.

Nevertheless, this objection reveals the problems that already exist with a system of ratings organized around depictions of sexual conduct, violence, and profanity. The choice to protect our children from these things rather than others cannot be said to be truly apolitical, even if it can be assured to be mainstream. While overt expressions of homophobia are likely to remain uncoded, overt homosexual expressions of affection will probably be among the first to be coded as inappropriate for children. The social equality of homosexuals is currently a political hot potato, and one is quite sure in which direction this particular potato will get dropped.

Coding for violence and for homophobia both involve content-based distinctions of subject matter. One might object that coding for violence, unlike homophobia, is viewpoint neutral, and therefore less controversial. But the example of homosexuality shows how tenuous the distinction between subject matter and viewpoint neturality can be in practice. Simple, ordinary demonstrations of affection between gays–the kind that would pass unnoticed between heterosexuals–are important means of showing the normalcy of gay lives and the commonality of their basic concerns with those of straight audiences. Yet these displays are more likely to be judged unsuitable for children while negative portrayals of gays will pass unfiltered by the system.

In any case, the very assumption that exposure to racist messages is less harmful to our children and our community than exposure to violence already carries considerable political freight. Although coding for violence but not for racism seems to exclude political and ideological controversy, it does not avoid politics or ideology. Rather it installs them in the very process of coding. The actual practice of political and religious “neutrality” will be achieved by the selective avoidance of topics; it will produce the appearance but hardly the reality of apolitical judgment.

My point in raising these difficulties is not to call for the coding of racist expressions. It is rather to note the politics implicit in a coding system that focuses on violence and indecency to the exclusion of other factors. Coding for racist messages, whatever its constitutional problems, would prove very difficult in practice. Often racial stereotypes are used in ironic ways, in which it is difficult to tell their actual meaning, much less their long term effects. Black-oriented comedy shows like “In Living Color” and “Martin” routinely employ exaggerated racial stereotypes of minorities. It is difficult to know where one would begin in classifying this material.

What advocates of rating systems may not realize, however, is that similar problems apply to depictions of violence. Violence is often used to show a character in a bad light, or to punish the wicked and the violent. Much violence is portrayed in a cartoon-like fashion. The many ways in which violence can be depicted, and the many social meanings it can convey, underscore that, like racist expression, there will be no easy way to code it.

Many people would probably be content with a ratings system that, even if not guaranteed to be nonideological, would at least be doggedly centrist. That is probably the best reason to have an industry-sponsored ratings system, which will cater to the tastes (or, more appropriately, the fears) of advertisers. Of course, it is hard to know whether this is cause for rejoicing. In any case, industry-developed ratings will not be unaffected by politics. Any industry-developed ratings system must still be approved by the FCC. Moreover, it is likely that future politicians will attempt to make political hay by bashing any industry ratings system and threatening a government takeover. To be sure, such a threat would be of dubious constituitonality, but constitutional proprieties about the first amendment have rarely dettered the pontifications of American politicians. Just as Senator Dole attempted to boost his 1996 campaign for the presidency by denouncing the wickedness of Hollywood, pseudopopulists of the future will discover an irresistible temptation to denounce whatever ratings system emerges as toothless and sinful, endangering the lives of our children and the future of America. Thus, even though the industry has adopted its own system, the eventual result may still be heavily politicized. The use of industry-developed ratings is only the lesser of two considerable evils.

Coarseness of the ratings system is a second major concern. The new industry sponsored ratings are based on the MPAA motion picture ratings system, yet that ratings system is perhaps the best example of how coarseness operates in practice. The MPAA currently offers a rating system featuring six categories–Unrated, G, PG, PG-13, R, and NC-17.(57) Ratings are determined by a panel of full-time employees using a combination of factors, including theme, violence, sexual content, and language.(58) Because these factors are taken together rather than differentiated, motion picture producers face a relatively coarse filtering mechanism. In fact, the PG-13 category was added later on because the previous system included too much in the PG category.(59)

Some effects of the system occur at the far end of the spectrum. Producers know that an NC-17 rating will significantly cut into movie sales. Many movie theaters will not show NC-17 movies,(60) many newspapers and television stations will not advertise them,(61) and they are not carried by major video chains like Blockbuster.(62) Hence producers take great pains to gain an R rating from the MPAA board, often offering to cut out offending materials.(63) Although the desire to obtain an R rating may produce self-censorship, movies with an NC-17 rating can still be shown to consenting adults.

A more curious and perverse effect happens on the other side of the ratings spectrum. Although a G rating signifies that a movie is suitable for all audiences, it also tends to drive away teenagers and young adults, who are among the most avid consumers of movies. As a result, the ratings system produces a perverse incentive to “dirty up” pictures to make them attractive to a wider audience.(64) Apparently many Americans demand genuine family entertainment; they just don’t want to have to see it themselves.

Any system of ratings will produce self-censorship because movie makers fear losing a desired audience. A movie producer has to balance the potential gains that might come from a change in content with the loss resulting from a corresponding change in movie rating. But the more important coarseness effects occur in the middle of the ratings spectrum. A ratings system that does not differentiate between sex, violence, and profanity will actually encourage the use of all three. For example, suppose that as a result of using several four-letter words a movie gains an R rating. At that point the movie director has every reason to put in additional sexual content and violence if she believes this will increase audience attention, as long as she doesn’t cross the line into NC-17. She is guaranteed not to lose audience share because of a change in rating but she can hope to gain audience share by strategically increasing sexual or violent content. As a result, movies in the middle range of ratings may tend to get progressively more violent and more sexually explicit at the same time.

If the current V-chip system uses ratings as coarse as the MPAA, we can expect that the broadcast world will display similar effects. The MPAA ratings resemble the anthropologist’s two basic categories of the sacred and the profane. There is a category of that which is suitable for children (taking the role of the sacred) and a category in which everything else–violence, bad language, nudity, homosexuality–gets thrown in indiscriminately (the profane).(65) What is profane is then subdivided not by kind of expression but by degree of profaneness, resulting in a world consisting of what is sacred, a bit profane, a lot profane, and seriously profane.

By contrast, if substantive categories are increasingly differentiated– for example into three separate categories of language, sexual content and violence, with ratings from 1 to 5 in each category– the ratings system produces a different set of incentives. It may pay for the director to produce a film with increased violence but not sexual content, and vice versa, because a change at the margins is better reflected in the ratings system. Of course, the more categories are added, the more difficult it becomes for parents to operate the system. As noted earlier, one of the most important constraints on the V-chip system will be ensuring ease of use to technologically-challenged adults. So the result is likely to be a compromise between coarseness and adequacy of ratings.

The July 10th, 1997 compromise (entered into by all of the networks except NBC) looks at first glance like a move away from the MPAA model to something more like the separate category-based system described above. In fact, it is not substantially less coarse than the original MPAA-inspired ratings system. There are still only four categories for non-children’s programming–TV-G, TV-PG, TV-14, and TV-M. The compromise plan merely offers the viewer the rationale for the choice of category. For example, “moderate” violence or “some” sexual situations are sufficient to garner a TV-PG rating, “intense” violence or “intense” sexual situations gain a TV-14 rating, and “graphic” violence or “explicit” sexual situations produce the dreaded TV-M rating.(66) But under the current plan, a parent cannot set the V-Chip to let in programs with “graphic violence” but not more than “some” sexual situations. She must choose either to blank out all programs rated higher than TV-PG (for whatever reason) or only programs rated TV-M (for whatever reason). To move to a truly more fine grained system, the industry would have to consent to make the levels of violence, sexual content, coarse language, or sexually suggestive dialogue fully independent ratings categories.

A third and final set of problems with any ratings system concerns equivalency. Even after the basic categories are determined, any ratings system will have to decide what gets coded within each category. More important for present purposes, it will have to decide what gets coded as possessing equal levels of inappropriateness. Like decisions about the categories themselves, these decisions cannot avoid political controversy; they are likely to have wide-ranging effects.

Take, for example, discussions of homosexuality or of safe sex as a means of preventing AIDS. How should these be coded in a ratings system? And what should they be coded as equivalent to? Some parents would see a big difference between such discussions and a sexually titillating love scene, while other parents would find both categories equally unsuitable for people under the age of eighteen. Now imagine a made-for-television movie that depicts a fictional cover-up by the church hierarchy of child abuse allegations made against Catholic priests, and a movie in which Freddy Krueger murders a hapless teenage couple having sex in the woods at midnight. It is not difficult to imagine different groups of parents disagreeing heatedly about the relative inappropriateness for children of these two examples.

Questions of equivalency severely test any facade of political neutrality. Does the ratings system regard two men kissing as equivalent to a woman being raped or another being slashed with a knife? Does the system regard a discussion of contraception as more or less inappropriate than a discussion of drug use? Whether or not we regard these events as really being different in kind is irrelevant. What is important is whether the ratings system makes them equivalent, by coding them as equally appropriate or inappropriate for children. Once materials are coded as equivalent, they become equivalent for all purposes for which the ratings system is used. And, make no mistake, the ratings system will be used for purposes other than its designers intended.

Advertisers deciding where to invest their dollars, video rental stores purchasing and organizing inventory, parental groups demanding tighter controls on undesirable programming, and consumers searching for suitable entertainment will not easily be able to differentiate within categories created by a ratings system. They will not have to. Rather, they will use the ratings system to avoid having to engage in such differentiations. They will rely on the categories already provided to choose what to purchase, what to watch, what to protest, and what to invest in. The ratings system will come ready-made as a division of the programming universe, and the efficiency and ubiquity of the system will make its distinctions real in practice.

Just as parental groups today do not watch NC-17 movies before protesting their inclusion in suburban movie complexes or local video stores,(67) people will use the television ratings system as a guide to the content of rated programming. The categories produced by a ratings board, whether public or private, will be the key informational filters that others will use to organize their decisions, whether monetary, political, or aesthetic.

Nevertheless, it is possible that events will play out quite differently. If cable bandwidth is expanded–for example, through digital delivery systems–there may be room for several different ratings systems. Groups like the Christian Coalition may offer their own ratings system using V-chip technology, employing their own conception of what is family-friendly and what is not. Consumers can then subscribe to the ratings system of their choice, much as they now subscribe to magazines like TV Guide. Moreover, an explosion of space on cable systems promises the possibility of filtering systems based on any number of programming criteria. The only limitation upon would-be filterers is their ability to catalogue and categorize the millions of hours of materials that will eventually exist for television, and their ability to gain sufficient market share to underwrite the costs of rating this material.

We do not yet know whether the economy will produce and support a wide variety of V-Chip ratings systems. There may be economies of scale in producing a commercially viable ratings system. If so, then the number of ratings systems that can survive will be quite small, and the results will not be too dissimilar from what I have described above. But the more interesting possibility is that ratings systems and related forms of media filters can and will proliferate. Consumers will be able to insulate themselves in increasingly specialized programming universes. By delegating their choices to specialized media filtering companies, they can filter out the great mass of programming to focus narrowly on their own special interests. Some, I suspect, will see this as the ultimate vindication of autonomy. Others will mourn the loss of a common televisual culture. In any case, this scenario produces effects completely opposite of the first. Instead of a single filtering system (or a handful of systems) uncannily structuring and skewing thought and culture, the alternative scenario imagines an increasingly fractured community of individuals fixated on their personal programming universe and increasingly oblivious to everything else.

Standing as we are, still in the infancy of the Information Age, it is impossible to tell how events will play out. But we can already appreciate the deep irony of our situation. The call for the V-chip, like the call for censorship of the Internet, stems from a sincerely felt anxiety that our culture is spinning out of control and an earnest desire to strike back at those new technologies thought to form part of the cause. The promotion of the V-chip as the solution to this cultural anxiety is at once appropriate and perverse. It is appropriate because it uses technology to fight the perceived effects of technology. It is perverse in that, like all other technologies before it, our submission to it is destined to have immeasurable and unexpected consequences.

The inevitable emergence of filtering organizations, whether public or private, underscores the importance of distinguishing between delegation and choice–the distance between the informational future that awaits us and the attractive homilies of autonomy and personal empowerment now used to describe it. We are on the verge of installing a series of new filtering mechanisms that will transform the most important systems of mass communication available to us. We do this to satisfy the concerns of parents and the ambitions of politicians. But as we do this, we might be well advised to stop for a moment, and try to imagine what is as yet unimaginable–the profound though unintended effects of this potent combination of bureaucracy and technology on the health of our democracy and the evolution of our culture.


54. The Telecommunications Act of 1996, for example, provides that the advisory committee shall be assigned “such staff and resources as may be necessary to permit it to perform its functions efficiently and promptly.” section 551(b)(2)(B), 110 Stat. at 141.

55. One major problem with the current MPAA system is that the ratings are awarded by a committee of laypersons who have no particular expertise in what kinds of violence are actually the most harmful for children to watch. So it is likely that some movies rated R for their violence may actually be less harmful than some rated PG-13. Of course, there is no guarantee that an industry-produced ratings system will match the results of psychological studies any better.

56. The Act insists that “nothing in [the requirement of ratings provisions] shall be construed to authorize any rating of video programming on the basis of its political or religious content.” section 551(b)(2), 110 Stat. at 140. Furthermore, the advisory committee that informs the television commission is to be “composed of parents, television broadcasters, television programming producers, cable operators, appropriate public interest groups, and other interested individuals from the private sector,” and is to be “fairly balanced in terms of political affiliation, the points of view represented, and the functions to be per- formed by the committee.” Id.

One assumes that the same caveats will apply to any industry-created ratings system, since under section 551(e)(1)(A) of the Act the FCC will not issue guidelines for ratings if private industry has established rules “acceptable to the commission.” section 551(e)(1)(A), 110 Stat. at 142.

By contrast, under the experimental three category ratings system tested in Canada, “language offensive to minorities or ethnic groups” can be blocked out by a V-chip. It gains a rating of 3 out of a possible 5 on the language scale, with 5 being the most offensive. Verne Gay, Ratings Soon: TV Industry to Code Shows by Next Year, Newsday, Mar. 1, 1996, at A3. In effect, this system equates the “F-word” with the “N-word;” hence it raises many of the problems of equivalent degrees of offensiveness discussed infra. Moreover, under this system, even though racial epithets are coded, they are judged less offensive than many other possible expressions. Id.

57. The X rating originally devised by the motion picture industry has been abandoned, and is now used primarily by adult video producers as a way of emphasizing the salacious nature of their product. See Leonard Klady et al., Sticks Can’t Nix Naughty Pix: ‘Showgirls’ Wide Release Pushes NC-17 Envelope, Variety, July 24-30, 1995, at 1.

58. Jack Valenti, The Voluntary Movie Rating System, in The Movie Business Book 396, 401-02 (Jason E. Squire ed., 1992); Richard P. Salgado, Regulating a Video Revolution, 7 Yale L. & Pol’y Rev. 516, 519-20 (1989).

59. Richard Zoglin, Gremlins in the Rating System: Two Hit Films Raise New Concerns About Protecting Children, Time, June 25, 1984, at 78.

60. Rachel Eisendrath, Film Industry Rates NC-17, Montgomery Advertiser, Oct. 29, 1995, at 1H (reporting decision by Carmike Cinemas, a chain of 2,104 screens scattered through the South, not to show “Showgirls” because of its NC-17 rating).

61. See Richard Corliss, What Ever Became of NC-17?, Time, Jan. 27, 1992, at 64.

62. See John Greenwald, Wayne’s World, the Sequel, Time, Oct. 11, 1993, at 64, 67.

63. Although the director does not directly negotiate with the film board, strategic behavior is apparently not uncommon. Martin Scorcese was reported to have deliberately added ultraviolent material to his film “Casino” so that later cuts would seem tame by comparison, thus enabling him to keep in material he thought essential. Steve Daly, In the Ratings Game, Ultraviolence is the Ace in the Hole, Ent. Wkly., Aug. 18, 1995, at 8.

Another possible alternative is to refuse the MPAA rating and release the film as unrated. See Edward Guthmann, Director Finds Gender Does Matter, S.F. Chron., Nov. 25, 1995, at C1 (describing the decision to release “When Night is Falling,” a drama which included two lesbian love scenes, as unrated). But this does not avoid stigma, and indeed, may even invite it.

64. Film Censors; Child-Minders, The Economist, Aug. 13, 1994, at 78.

65. Richard Corliss, It’s Great! Don’t Show It! A Misguided Rating System Slaps an X on a Discreetly Erotic Film, Time, Sept. 17, 1990, at 70 (discussing NC-17 ratings given to Director Philip Kaufman’s film, “Henry and June,” and controversies over other films involving lesbian scenes).

66. Kim Asch, Rating system to help parents pick programs, Washington Times, August 19, 1997, at E1.

67. See, eg., Michael Granberry, ‘Temptation’: Some Resist, Others Yawn, L.A. Times, Aug. 20, 1988, section 2, at 2; Julia McCord, Film on Fallen Priests Sparks Pastor Protest, Omaha World-Herald, Apr. 7, 1995, at 13SF; Sean P. Means, Pickets Organize to Protest ‘Showgirls’ in Utah: Kids’ Movies and Skin Flicks a Bad Mix, Protesters Say, Salt Lake Trib., Sept. 28, 1995, at B1; Victor Volland, Video Stores Have ‘Priest’; Groups Protest, St. Louis Post-Dispatch, Jan. 27, 1996, at 5D.