Violence in the media essay

Controlling Communications That Teach or Demonstrate Violence: “The Movie Made Them Do It”.

by Lillian R. BeVier

Violence sells. Americans have what sometimes seems to be an insatiable appetite for it. Depictions and descriptions of violence saturate our culture. Songs urge us to rape women, kill police officers, and commit suicide. Movies portray–indeed they glorify–violence as an intrinsic element of every imaginable plot line.

Despite substantial evidence that an individual’s repeated exposure to portrayals of violence is associated with significantly increased likelihood that the individual will commit aggressive acts against others, (1) no legal regime currently regulates such portrayals either on television, in music, in movies, or in video games. Neither Congress, the Federal Communications Commission, the Federal Trade Commission, nor any state legislature has yet mustered the political will to impose substantial or systematic legal constraints upon producers or purveyors of violent images. Official censorship is rightly to be feared, but unreflective invocations of our commitment to freedom of speech provide incomplete justification for our legal regime’s apparent indifference to the possibility that media-induced violence may impose substantial costs on innocent victims. Industries other than entertainment are subject to substantial regulation; and in other industries profit-making corporations are held by tort law to far-reaching duties to protect not only their customers but also innocent bystanders from harm that their products cause. (2) What reasons of legal policy counsel against imposing similar regulations and duties on the entertainment industry? Ought we to reconsider them?

This paper offers a modest beginning to the exploration of the reasons that support the legal and political decision to refrain from regulating media portrayals of violence. It will summarize the basic outlines of several common-law doctrines that could conceivably be deployed to impose liability on publishers or purveyors of communicative material–books, movies, video games, and music–that might be thought to have induced individuals to engage in violent behavior that harmed innocent victims. The paper will not consider legislation or regulations that propose restricting or punishing such material. Since the beginning of 2003, for example, legislators in various states and cities have introduced 16 anti-video game bills alone, and Congress and the Federal Trade Commission also periodically consider similar restrictions. (3) The policy and constitutional issues that legislative or administrative agency regulations would present are in many respects similar to the kinds of issues that the prospect of judge-imposed liability raises, but the legal and institutional contexts of legislation and regulation are sufficiently different from the context of litigation to warrant excluding them from the present discussion. The paper will briefly describe the common law doctrines of negligence, product liability, and aiding and abetting as potential sources of media accountability. It will identify the principal reasons why courts have generally been reluctant to use them to impose liability on media defendants. In addition, on the assumption that the tort policy roadblocks to liability could–or perhaps should–be overcome, it will discuss the First Amendment roadblocks. Readers should understand that the paper does not aim, nor does it purport, to provide a complete or nuanced analysis of legal doctrine. Rather, it seeks to use legal doctrine as a framework for structuring a discussion of why liability for media-induced violence is the exception rather than the role. It keeps legal jargon to a minimum while seeking to supply a transparent, though necessarily truncated, summary of the legal basics.


The relevant legal questions have arisen in three related situations. First is the publication of manuals containing specific and explicit instructions about how to commit a crime-how to build a bomb or how to become a hired killer and commit a successful murder. An example of the latter is the publication Hit Man: A Technical Manual for Independent Contractors, which contained 130 pages of detailed factual instructions on how to become a professional killer and, once hired to kill, how to carry out the murder. In Rice v. Paladin Enterprises, Inc., (4) for example, family members of the hired killer’s victims sued the publisher of the manual. Materials that explicitly teach audiences how to engage in violent acts pursue an agenda and raise legal issues that are different from those that merely portray or glamorize violence and thus “teach” it only implicitly or by example. Both kinds of materials, however, have been alleged to have caused members of the audience to commit violent acts for which the victims or their families have sought recompense not merely from the perpetrators but from the media corporation that produced them. Thus both raise the kinds of issues with which this paper is concerned

The second example is the so-called “copy-cat” situation, in which particular violent or harmful acts portrayed in film or on television become the inspiration for audience members’ violent acts, and the victims’ families sue the producers. In Olivia N. v. NBC, (5) for example, a gang of teenagers allegedly inspired by a showing of “Born Innocent” on television savagely assaulted a 9-year old girl and raped her with a bottle. And in Yakubowicz v. Paramount Pictures Corporation, (6) a teenager was knifed to death by a gang who had just come from watching “The Warriors,” a film that included numerous scenes of juvenile gang-related violence. Sometimes the victim of the copy-cat crime is the copy-cat himself. In Herceg v. Hustler Magazine, (7) for example, an adolescent who read an article called “Orgasm of Death” in the magazine attempted “auto-erotic asphyxia” and killed himself doing so.

The third paradigmatic situation is closely related to the copy-cat case. It gives rise to the claim that “the movie–or the television program, or the music, or the video game-made me do it.” In Zamora v. State, (8) for example, a teenager claiming to have become a sociopath by reason of having been “involuntarily subliminally intoxicated” by watching thousands of murders on television brutally murdered his 82-year old neighbor. And in another case, McCollum v. CBS, (9) a depressed teenager committed suicide after listening for hours and hours to Ozzy Ozborne recordings containing the message “that life is filled with nothing but despair and hopelessness and suicide is not only acceptable, but desirable.” (10)


When victims of allegedly media-induced crimes sue the producers, they allege one or more of three theories of liability: negligence, products liability, or aiding and abetting the commission of a crime.


Negligence is the cause of action most frequently invoked by victims of copy-cat and “movie-made-me-do-it” crimes. Though the law varies in its particulars from state to state, in general it is true that a plaintiff-victim of media-induced violence could recover on a negligence theory if she could persuade a court “that the media entity owed to the public in general or the plaintiff in particular a duty to exercise reasonable care with respect to the risks of harmful audience responses; that the defendant breached that duty by disseminating a depiction that unreasonably risked harmful audience response; that the depiction caused such a response; and that the response resulted in foreseeable harm to the plaintiff.” (11)

Courts have been singularly unsympathetic to alleged media-induced violence victims’ negligence claims. Plaintiffs have foundered principally on the requirements that they prove that the media defendants owed them a duty to protect them from unreasonable risks of harm and that the media portrayals were both the cause in fact and the proximate cause of the harm that in fact ensued. The most significant stumbling block in all the cases has been the fact that the actual perpetrator of the crime or injury has been an independent third party who has committed an intentional criminal or harmful act. It is easy to intuit why courts would be reluctant to impose liability on media defendants who did not themselves commit the acts that directly injured the victims. This intuitive reluctance is embodied in the way courts apply the law of negligence in media-induced violence cases. In particular it finds expression in courts’ unwillingness to find that media defendants owed plaintiff victims a duty to protect the victims from harmful audience responses and in their refusal to conclude that the requisite causal link exists between defendants’ portrayal of violence and plaintiffs’ injuries,

Consider the question of duty. Whether media defendants owe a duty to protect victims from the risk of harm caused by the violent acts of third parties is treated by courts as a question of law. Judges treat it forthrightly as a question that they are free–indeed that they have an obligation to–decide based on their assessment of its policy implications. Judges have been almost uniformly reluctant to decide the question in favor of plaintiffs, in large part because they have concluded that media defendants cannot reasonably be held to have foreseen that their products would stimulate criminal acts by members of the audience. No one has a duty to protect others from risks of harm that cannot reasonably be foreseen. Whether the crimes that harmed the alleged victims of media-induced violence were foreseeable is a question that must be answered in hindsight, after the crimes have been committed. The difficulty is that the perspective from which the determination is made is supposed to be before the fact: the question is whether the defendant should have foreseen them. It is an uncertain proposition at best to say with certainty that a harm that did materialize but was not in fact foreseen was so likely to happen that the defendant should have foreseen it–or was so unlikely that it need not have done so.

No clear standard has emerged for gauging how probable–and thus how foreseeable–an actually unforeseen harm was, but a couple of examples suggest the way courts approach the inquiry and reveal why judges have such a strong inclination to conclude that the harms were not foreseeable. In one case, the parents of the victims of the Columbine, Colorado, school shootings sought relief from several media defendants. The court acknowledged that the defendants “might have speculated that their motion picture or video games had the potential to stimulate an idiosyncratic reaction in the mind of some disturbed individuals. (12) But, said the court, the defendants had no reason to suppose that the two boys who did the shooting “would decide to murder or injure their fellow classmates and teachers.” (13) Similarly, in a case brought by victims of the Paducah, Kentucky, school shootings, the court acknowledged that the shooter’s reaction to depictions of violence “was not a normal reaction. Indeed, [the shooter] is not a normal person[.] But it is not utter craziness to predict that someone like [him] is out there.” (14) Nonetheless, the court concluded that defendants could not in fact have foreseen the particular crime, and thus should not be found to have been under a duty to prevent it: it was simply

too far a leap from shooting characters on a video
screen (an activity undertaken by millions) to
shooting people in a classroom (an activity undertaken
by a handful, at most) for [the shooter’s]
actions to have been reasonably foreseeable to
the manufacturers of the media that [he] played
and viewed. (15)
The second issue on which victims of media-induced violence have foundered in negligence cases is the issue of causation. That media portrayals of violence were a cause-in-fact of–or even a substantial factor in producing-violent behavior turns out to be a highly problematic conclusion, given the range of influences that bear on those who commit crimes. True, a substantial body of social science research suggests a correlation between viewing violent images and a viewer’s subsequent violent acts, but it is also true that correlation is not cause and standing alone it does not support an inference of causality. Though

[v]iolence on television has been shown in
hundreds of studies to have an influence on
aggressive behavior, … the causes of violence are
manifold, and include biological and psychological
factors as well as broader social and cultural
ones…. Peer influences, family role models,
social and economic status, educational level,
and the availability of weapons can each significantly
alter the likelihood of a particular reaction
to viewing violence on television. (16)

This point bears emphasis. In a recent study that it undertook following the Columbine shootings, the Federal Trade Commission found that researchers generally agree that media violence explains only a very small amount of youthful violent behavior. (17) The effects of images “come through a process of socialization … [and] depend on mental intermediation” (18) and it is extremely difficult to measure “the actual probability that any single image will result in imitative behavior.” (19) It follows that the conclusion that any single image did in fact result in imitative behavior cannot confidently be reached. Even if a causal link could confidently be established between a particular media portrayal and violence inflicted on a particular victim, courts would be inclined to find that the intervening criminal act of the third party perpetrator breaks the causal chain between the media defendant and the victim plaintiff. “Proximate cause” is the doctrinal rubric for considering whether a criminal act that supercedes between the media defendant’s portrayal of violence and the harm that befalls the plaintiff breaks the causal chain, and the foreseeability of the crime is again relevant to the inquiry. The general rule is that “a third party’s criminal action that directly causes all of the damages will break the chain of causation.” (20) Two reasons provide the rule’s rationale. First, the commission of intentional criminal acts is relatively rare, a fact which has considerable bearing on their foreseeability. Despite the fact that our culture is saturated with media depictions of violence, most people obey the law most of the time. Media defendants have been held to be entitled to assume that members of their audiences will do likewise. Second, the criminal law and the sanctions it imposes assume that those who commit crimes are responsible moral agents. Imposing liability on defendants who do not actually participate in the crimes would tend unjustifiably to diminish the criminal actor’s own moral accountability. As one commentator argued,

[i]f we decide speech is responsible for causing
harmful acts, we are giving potential criminals an
excuse for their behavior. One of the goals of our
criminal justice system is to insure responsibility
for our actions. We should not diminish that goal
by suggesting that the criminal would not have
acted violently had he or she not read that book
or seen that movie. By suggesting that media
causes violence, we are encouraging criminals
to blame something besides themselves for their
antisocial behavior. (21)

Thus the tort of negligence has been a dead end for victims claiming to have been harmed by media-induced violence. The most salient reasons are that the particular acts of violence are not foreseeable, and that they are in any event committed not by the media defendants but by third parties deemed to be independent moral agents. Liability for Defective Products

Consider next the product liability cause of action. Though as in negligence the law varies in detail from state to state, again a generally accurate statement is possible: manufacturers, distributors, and retailers are liable in tort for injuries caused in fact by defects in the manufacture or design of their products or in their failure to warn consumers of a danger of which they should have been made aware. (22) Liability for defective products does not require a finding that defendant breached a duty owed plaintiff to exercise “reasonable care” to prevent the harm. Product liability is strict liability. In other words, if a product is determined to be defective, no amount of caution–no amount of care however reasonable–on the part of the manufacturer can forestall liability once the defect has caused an injury.

Plaintiff-victims pursuing product liability claims in media violence cases have argued that video games, songs, movies and books are “products” and that their violent content constitutes a “defect” that was a cause-in-fact of the harm they suffered, but courts have nearly always rejected such claims. (23) They have regarded as decisive the fact that the plaintiffs suffered their injuries or deaths on account of consumers’ reactions to and not on account of defects in the products themselves, as would be the case for example if a video game had exploded and injured them. In addition, plaintiffs’ injuries were caused by the effect on the audience of the ideas and images conveyed by the products. But when dealing with claims of liability for injuries allegedly caused by ideas and images, courts have been almost uniformly inclined to separate the tangible containers of the ideas and images, which they treat as products for the purposes of products liability, from their communicative elements-the ideas and images they convey-which they do not so treat.

Aiding and Abetting

Consider finally the possibility of imposing civil liability on media defendants pursuant to a theory that their publications aided and abetted the commission of a crime. This was the theory successfully invoked in the exceptional Paladin case, a case that merits significant discussion not because it is typical or because there are likely to be more cases closely analogous to it in the future but rather because it is so unique. It is unique on account of its gruesome facts, on account of the procedural posture in which it reached the Fourth Circuit Court of Appeals, and on account of the nature of the publication for which liability was sought.

In the book he wrote about the case, Rod Smolla, the attorney and law professor who represented the victims in Paladin, described the effect the book had on him:

I was depressed at the absolute incarnate evil of
the thing, the brazen, cold-blooded, calculating,
meticulous instruction, and repeated encouragement
in the black arts of assassination … I didn’t
even want to touch the damn book. I couldn’t
leave it on the night table–I had to take it back
to my office in the house and lock it in my briefcase.
It didn’t even seem like it was a book at all,
really. It was more like someone had sent me a
loaded pistol, or a vial of poison. The physical
thing had a stench of evil to it. (24)

The opening of the powerful and passionate opinion that Judge Michael Luttig wrote for the Fourth Circuit makes Professor Smolla’s reaction completely comprehensible, for it consists of several pages of hair-raising, blood-curdling, even sickening quotes from “Hit Man” itself. The extensive quotations from the book begin with the author’s bold assertion that “the professional hit man fills a need in society,” (25) and go on to summarize the book’s detailed instructions on how to equip oneself (with flesh-toned surgical gloves, for example, and handcuffs); what to wear (dark clothes); what kind of knife to use and how to deploy it (6-inch double edged serrated blade, thrust into a vital organ and twisted); how to build a silencer if the plan is to kill the victim with a gun; how to kill with a gun at short range, with a rifle, with explosives, with tire; and, finally, how to dispose of the corpse (the first thing to do is chop off its head). After this eye-opening introduction, the opinion proceeds to recount how, on March 3, 1993, “readied by these instructions and steeled by these seductive adjurations,” (26) one James Perry murdered three people: an 8-year old quadriplegic boy, whom he strangled, and the boy’s mother and nurse, whom he shot point blank through the eyes. Perry committed the murders pursuant to a contract with the boy’s father, who stood to receive his son’s $2 million insurance settlement. Moreover, “[i]n soliciting, preparing for, and committing [the] murders, Perry meticulously followed countless of Hit Man’s 130 pages of derailed factual instructions on how to murder and how to become a professional killer,” (27) and the opinion offers a lengthy summary of the instructions and what Perry did to follow them. Relatives of the victims sued Paladin Enterprises, Hit Marts publisher for the wrongful death of their loved ones, their theory being that through its publication of the manual Paladin had aided and abetted the crimes. The district court granted summary judgment to the defendant Publisher. (28) The Fourth Circuit Court of Appeals reversed, holding that sufficient facts had been stipulated or pleaded under Maryland law to render Paladin civilly liable for aiding and abetting murder and that, if liability were otherwise appropriate under state law, the First Amendment did not preclude imposing liability on the publisher (the issue discussed in the next section).

As to civil liability for aiding and abetting, the Fourth Circuit held that Maryland law recognizes such a cause of action, and that its elements are very much like those that define the crime of aiding and abetting:

The state defines ‘aider’ as one who ‘assist[s],
support[s] or supplement[s] the efforts of another,’
and defines ‘abettor’ as ‘one who instigates, advises
or encourages the commission of a crime.’ (29)

The one difference between civil and criminal liability is in the intent standard, which in the context of civil torts only requires that the criminal conduct be “the natural consequence of the defendant’s original act,” but in the criminal context requires the defendant to have had a “purposive attitude” toward the commission of the offense. (30) This difference was not relevant in Paladin, however, because Paladin had stipulated to facts that would have justified imposing liability even under the more rigorous standard for establishing criminal intent(30) Paladin had stipulated that Perry followed the instructions enumerated in Hit Man for planning, executing, and attempting to cover up his triple murder; that in marketing Hit Man, Paladin intended to attract and assist would-be criminals desiring information and instructions on how to commit crimes; that Paladin intended and had knowledge that Hit Man actually would be used by criminals and would-be criminals to plan and execute the crime of murder for hire; and that Hit Man assisted Perry in the perpetration of his triple murder. (31) These stipulations alone were sufficient to establish as a matter of law that Paladin was civilly liable for aiding and abetting Perry’s crime, the court held. But the court pointed to four additional bases upon which, collectively, “a reasonable jury could find” that Paladin possessed the intent required under Maryland law: the book’s declared purpose was to facilitate murder; its pointed promotion of murder is probative of the publisher’s intent; the way it was marketed would support an inference of the requisite intent; and it would be reasonable for a jury to conclude that the only genuine use for the book is that of facilitating murder. (32) These additional bases for liability turned out to be crucial to a finding of liability in the later case of Wilson v. Paladin Enterprises, (33) a case that was very similar to Paladin except for the fact that the defendant did not stipulate to having marketed Hit Man with the intention of attracting and assisting would-be criminals. As bases for concluding nonetheless that Paladin indeed had published with the requisite intent, the Wilson court relied on Hit Man’s declaration that it proposes to facilitate murder; the book’s actual and extensive promotion; Paladin’s marketing strategy targeting would-be criminals; and the impossibility of conjuring a purpose for the book other than an unlawful one. (34)


Assume for the moment that courts put aside their doubts about the wisdom–purely as a matter of tort policy–of imposing liability for negligence or defective products or civil liability for aiding and abetting on media defendants whose products seduce, glorify or, implicitly or explicitly, teach violence. Courts might become persuaded that profit-making media corporations ought to be held to a duty to protect innocent victims from violent acts that, at least in hindsight, and on account of the fact that such acts have in fact been perpetrated in the past, appear to have been reasonably foreseeable effects of their unrestrained, exploitive depictions of violence. And they might be persuaded that particular crimes have in fact been caused by the depictions, notwithstanding that the crimes themselves are actually committed by third parties who cannot by any stretch of the imagination have been said to have been “incited” to commit them. Other courts than the Fourth Circuit–like the district court in Wilson–similarly confronted with a plaintiff victimized by a criminal who learned his methods from a publication that meticulously and in great detail described how to commit a criminal act, might become convinced that the defendant publisher ought to be held to be an aider and abettor of the crime.

This brings us to the question whether the First Amendment stands–or ought to stand–as a barrier to liability in any of these situations. To what extent does our commitment to free expression, embodied in doctrines protecting advocacy of ideas (even of the idea that laws exist to be violated) and in the principle that “debate on public issues should be uninhibited, robust and wide-open,” (35) trump whatever substantive law policies might counsel in favor of holding media defendants accountable, such as prevention of violent crimes against innocent victims and perhaps even forestalling the debasement of our culture.

The First Amendment inquiry begins with the Paladin case, for it presents a situation about which it seems fair to say the policies in favor of accountability are at their most powerful while the free speech values that would be sacrificed by finding liability are at their most tenuous. If free speech values trump accountability in a case such as Paladin, they would presumably trump it in every case and we could simply dispense with any notion of holding the media liable for events that take place on account of its depictions or teachings of violence. Indeed, it would perhaps not be much of an overstatement to say that publishers, simply “because they are publishers … [would then] have a unique constitutional right to aid and abet murder.” (36) On the other hand, as Judge Luttig pointed in his Paladin opinion, to impose liability for publishing Hit Man cannot realistically be said to imperil First Amendment values or significantly increase the risk that other media defendants will routinely be held liable in copy-cat or movie-made-me-do-it cases. Hit Man was simply unique on account of Paladin’s stipulations regarding its intent in publishing the book; the book’s comprehensive, detailed and clear instructions on how to commit murder; its exhortation to murder; the effectiveness of its instructions; its total lack of ideas or of any “even arguably legitimate purpose beyond the promotion and teaching of murder.” (37)

The most obviously relevant First Amendment doctrine in all the cases in which victims sought to impose liability on media defendants for allegedly media-induced crimes was the Supreme Court’s holding in the seminal case of Brandenburg v. Ohio (38) that “the constitutional guarantees of free speech and free press do not permit a state to forbid or proscribe advocacy of the use of force or of law violation except where such advocacy is directed to inciting or producing imminent lawless action and is likely to incite or produce such action.” (39) In other words, only advocacy amounting to incitement of imminent lawlessness may be punished consistent with the First Amendment. Judge Luttig offered several reasons why this doctrine was not a barrier to media liability in Paladin. First, invoking the so-called “speech-act” doctrine, he pointed out that Brandenburg does not immunize speech that is an integral part of conduct that violates a valid criminal statute, which the court held Hit Man was. Second, Brandenburg’s requirement that, to be punishable, advocacy of law violation must be directed to inciting or producing imminent lawless action, was inapplicable because Hit Man could not really be said to advocate law violation. Instead, it was a successful effort to assist would-be criminals by telling them precisely how to commit their crimes. Finally, neither of two possible qualifications that might have brought the First Amendment into play applied. Thus, a heightened intent requirement might be thought appropriate where the speech act doctrine was invoked in the context of civil liability for aiding and abetting. First Amendment doctrine has been crafted to prevent the punishment, and even the “chilling” of innocent lawful speech, said Judge Luttig, and thus sometimes it prohibits liability based merely on the fact that the impermissible misuse of information one imparts was foreseeable. This limitation would allay the concern, about exposure to suit under lesser standards, expressed by publishers, broadcasters or distributors with large undifferentiated audiences) But, as we have seen, the facts of Paladin satisfied even such a heightened intent requirement, since the defendant had stipulated that it both knew and intended that Hit Man would be used by criminals to solicit, plan, and commit murders. Similarly, the First Amendment bar to both criminal and civil liability for speech that amounts to abstract advocacy of–as opposed to incitement of imminent–violence did not apply. The court could find no hint of abstract advocacy in Hit Man, nor instructions of any value–only the “detailed, focused instructional assistance to those contemplating” murder, which is the opposite of the kind of speech that Brandenburg was intended to protect. (41) Ideas were simply not the focus of the book. (42)

A noteworthy feature of the Paladin litigation was the number and stature of organizations that filed or joined friend-of-the-court (amici curiae) briefs on defendant’s behalf. They comprise a long and impressive list that included, among others, The Thomas Jefferson Center for the Protection of Free Expression, the American Civil Liberties Union, ABC, America Online, Association of American Publishers, The Baltimore Sun, E.W. Scripps Company, Freedom to Read Foundation, Magazine Publishers of America, McClatchy Newspapers, Media General Corp., National Association of Broadcasters, Newspapers Association of America, The New York Times, The Reporters Committee for Freedom of the Press, the Society of Professional Journalists, and The Washington Post. Though Hit Man itself might not be of genuine value, these amici argued, it ought to be protected in order to fortify the painstakingly constructed First Amendment barrier to liability for speech that did have merit. They claimed that decades of First Amendment jurisprudence would be put at risk by a finding of liability, and asserted that civil liability would hang over the head of and exert a chilling effect on all expression–“music, video, books, even newspaper articles.” (43) For his part, Judge Luttig found it “breathtaking” that national media organizations of such stature would defend Paladin so vigorously, since what Paladin was asserting was

a constitutional fight to intentionally and knowingly
assist murderers with technical information
which Paladin admits it intended and knew
would be used immediately in the commission
of murder. (44)

And in answer to the amici’s claim that liability in this case would have far-reaching chilling effects, Judge Luttig stressed Paladin’s unique facts and strongly emphasized how different they were from the facts in copy-cat and movie-made-me-do-it cases. He pointed out that broadcasters and publishers in such cases are completely unlikely to intend to assist in commission of murder, nor, he emphasized, would the speech itself, as a matter of law, permit any inference that the speaker intended to facilitate the criminal conduct that it depicted. (44) Along with the fact that plaintiffs in copy-cat and “movie-made-me-do-it” cases are unable to prove that defendants’ products constituted “incitement” to “imminent” lawless action, the chilling effect argument provides the principal rationale behind courts’ general unwillingness to impose liability in such cases. For this reason, Judge Luttig’s opinion took great pains both to explain his conclusion that it did not apply in Paladin and narrowly to circumscribe Paladin’s potential effect as precedent. For this reason, also, it is important to understand the “chilling effect” argument that the Paladin amici were making.

The chilling effect argument builds on a fundamental insight about the way liability law works as a mechanism of social control, which is that any decision imposing–or refusing to impose liability always has two components, one backward–looking, the other looking forward. The backward-looking component settles a dispute between a plaintiff and a defendant and, if liability is found, it entails an order that plaintiff compensate defendant for the wrong done. The forward-looking component–we might call it the regulatory component–affects the way potential defendants will behave in the future. A decision imposing liability for negligence or a product defect on a media defendant in a case brought by the victim of a copy-cat crime, for example, would both compensate the victim and signal to producers or publishers of violent content that they too risk liability in the future should someone in the audience commit a copy-cat crime. The signal would have a substantial impact on their future decisions about what to produce or publish. Intuition tells us that, if nothing else, media defendants would take care to produce or publish less violent content–both content that is less violent, and less content that is violent–even if they were otherwise convinced of the violent portrayal’s political, informational, educational, entertainment, or artistic value. It is this predictable reaction to which the “chilling effect” refers. A commonplace interpretation of First Amendment jurisprudence is that rules (or findings of liability in particular cases) that have this effect of chilling future speech are likely to be unconstitutional.

If it were possible for producers and publishers to know ex ante what works were particularly (although still unreasonably) likely to induce copy-cat crimes to be committed, and if we could confidently conclude that such works had no redeeming social value, so that producers’ decisions to suppress them would produce only benefits (no more copy-cat crimes) and no costs (no works of value go unproduced), then we might conclude that the chilling effect is a good thing and that we ought to adopt liability roles precisely so as to bring it into play–and stop wringing our hands about it. But legal rules do not work so exactly. They simply do not, nor can they be crafted to, have such precise effects. They always hit more than they aim at and in the process often deter conduct that is merely only innocuous and often quite valuable. It might be possible for producers to predict in a general way that there are crazy people out there who will sometimes–though very occasionally–react to some depictions of violence by committing violent acts themselves. It is not, however, possible for them to know before the fact that in the audience for a particular work there will be unstable people who will be induced by the work to commit a copy-cat crime. (This fact is what courts are getting at when they say that such crimes are not “foreseeable.”) In addition, recall that it is not even possible after the fact to be sure that a particular work was indeed the legal cause of the harm done by a crime committed in its wake. This is not only because a criminal act committed by a third person will have been the direct cause of the victim’s harm, but also because it is not possible to determine the extent to which other factors than the particular depiction of violence contributed to the perpetrator’s behavior–for example, the environmental characteristics of the audience member, such as living in an unsafe neighborhood and being raised by neglectful parents. Remember, evidence that violent images actually cause crime is scant. Finally, and perhaps most importantly, we can not be confident of a conclusion that even works that induce audience members to commit violent acts have no redeeming political, informational, educational, entertainment, or artistic value. Audiences do in fact plunk down good money to see and to listen to and to play games that depict and describe and glorify violence. And it is certainly conceivable that such works are as likely to portray ideas–albeit ideas that many find loathsome–or to provide satisfying relief from the cares of the day as works that generate feelings of loving kindness.

Courts’ concern with the “chilling effects” of rules that would impose liability on media defendants has its source in these realities, for they create the probability that the prospect of liability would dissuade–“chill”–media defendants from producing many works that either would not turn out to induce violence or would have redeeming social value, or both. It is worth emphasis, too, that it would not be merely works of marginal value that the prospect of liability might chill. Consider, for example, the possibility that the producers of the movie The Godfather might have been persuaded not to release the film had they thought that they might be held liable were someone in the audience induced to commit a copy cat crime. Or that the producers of the recent enactment of Medea in New York might have similarly refrained had the prospect of liability for a copy-cat crime loomed. It is the cost of foregone works which would turn out to have genuine merit that our traditions of flee expression seem to have made us unwilling to bear, or so our courts and judges seem to think.

I do not mean to suggest that our freedom of expression is itself “free,” in the sense that its exercise does not impose real costs both on society as a whole, in that our popular culture is degraded, and on the particular victims of media-induced crimes. We might wish that we used our freedom in more enlightened or uplifting ways, but of course the whole point of freedom is that it eschews coercion even in a good cause. Freedom is and always has been a risky proposition. And, with the exception of the peculiar and so-far unique circumstances of the Paladin case, we seem to have decided that if the cost of freedom of expression is a degraded culture and the occasional violent crime perpetrated on an innocent victim, they are costs we are willing to bear ourselves–or to let fall on the victims. This conclusion is not belied by the fact that the Supreme Court has interpreted the First Amendment to permit the prohibition of obscenity and child pornography, for the reason the Court has permitted the prohibition of such speech is not that it causes members of their audience to commit violent acts. Obscene speech has been held to be completely outside the realm of First Amendment protection because the Court thinks it has no social value other than to appeal to the prurient interest in sex, not because of its potential to lead to criminal acts. (46) Child pornography may be prohibited, but only when real children are portrayed, and then only because of the interest in protecting children from being forced to perform in such portrayals. (47) Child pornography that only appears to depict minors, but that was in fact produced without using real children, may not be prohibited. (47)

What my analysis boils down to is this. Despite the distaste–even perhaps the revulsion–we might experience at much of the violent content that the media produces today, the plain fact is that we do not know with the requisite degree of certainty that it actually causes people to commit crimes. We do believe, though, that if we imposed liability on the media for media-induced crimes, publishers and producers would censor themselves and decide not to produce a great deal of material that would have proved valuable. Society would be culturally the poorer for such decisions. In the end, the conclusion seems warranted that while freedom of expression for producers of violent content undeniably has its costs, the costs of suppression are probably greater–and its benefits are surely more uncertain.


(1.) J.G. Johnson, P. Cohen, E. M. Smailes, S. Kasen and J. S. Brook, “Television Viewing and Aggressive Behavior During Adolescence and Adulthood,” Science 295 (2002): 2468-2471, at 2470.

(2.) See, e.g., Elmore v. American Motors Corp., 451 P.2d 84 (Cal. 1969); Codling v. Paglia, 298 N.E.2d 622 (N.Y. 1973).

(3.) Ronald D. Rotunda, “Current Proposals for Media Accountability in Light of the First Amendment,” forthcoming, Journal of Social Policy, Philosophy, and Law (2004): 1-74, at 6-7.

(4.) 128 F.3d 283 (4th Cir. 1997).

(5.) 178 Cal. Rptr. 888 (Cal. App. 1981).

(6.) 536 N.E. 2d 1067 (Mass. 1989).

(7.) 814 F.2d 1017 (5th Cir. 1987).

(8.) 361 So. 2d 776 (Dist. Ct. Fla. 1978).

(9.) 249 Cal. Rptr. 187 (Cal. App. 1988).

(10.) Id. at 190.

(11.) L.W. Brill, “The First Amendment and the Power of Suggestion: Protecting ‘Negligent’ Speakers in Cases of Imitative Harm,” Columbia Law Review 94 (1994): 984-1044, at 987 (citing Restatement (Second) of Torts, [subsection] 281, 302-302B (1965)). For a thorough and engaging account, see Robert M. O’Neil, The First Amendment and Civil Liability (Bloomington: Indiana University Press, 2001): 137-62.

(12.) Sanders v. Acclaim Entertainment, Inc., 188 F. Supp. 2d 1264 at 1272 (D. Colo. 2002).

(13.) Id.

(14.) James v. Meow Media, 300 F. 3d 683, 693 (6th Cir. 2002).

(15.) Id. at 693.

(16.) J. Federman, ed. National Television Violence Study, Executive Summary, Volume 3 (Santa Barbara: Center for Communication and Social Policy, University of California, Santa Barbara, 1997); at 5.

(17.) Report of the Federal Trade Commission, Marketing Violent Entertainment to Children: A Review of Self-regulation and Industry Practices in the Motion Picture, Music Recordings & Electronic Game Industries(FTC, Sept. 2000), Appendix A, “A Review of Research on the Impact of Violence in Entertainment Media,” at 9.

(18.) American Booksellers v. Hudnut, 771 F. 2d 323, 329 n. 2 (1985).

(19.) Brill, supra note 11, at 1033.

(20.) James v. Meow Media, 300 F. 3d at 699.

(21.) Christopher Campbell, “Murder Media–Does Media Incite Violence and Lose First Amendment Protection?” Chicago-Kent Law Review 76 (2002): 637-69, at 664. In Zamora v. State, 361 So. 2d 776 (Dist. Ct. App. Fla. 1978), the defendant attempted to defend a murder conviction based on the allegation that he was insane on account of “involuntary subliminal television intoxication.” The Florida courts rejected the defense.

(22.) Restatement (Second) Torts, [section] 402A.

(23.) See, e.g., James v. Meow Media, Inc., 300 F. 3d 683 (6th Cir. 2002)(victims of Paducah school shooting unsuccessfully claimed that the violent content of defendants’ video games, movie, and internet transmissions constituted product defects that caused the shooter to act violently); Wilson v. Midway Games, Inc., 198 F. Supp. 2d 167 (D. Conn. 2002)(boy stabbed with kitchen knife by a boy who was allegedly obsessed with the video game Mortal Combat unsuccessfully sued for product liability; court finds a video game is not a product for purposes of the relevant state statute); Davidson v. Time Warner, Inc., 1997 U.S. Dist LEXIS 21559 (S.D. Texas 1997)(victim of shooting by young man who had been listening to a tape of rap songs titled 2PocalypseNow, which contained lyrics that expressed extreme hostility to law enforcement, unsuccessfully claimed that the tape was an inherently dangerous product). See generally O’Neil, supra note 10, at 107-19.

(24.) Rod Smolla, Deliberate Intent (New York: Crown Publishers, 1999): at 38-39.

(25.) Rice v. Palladin, 128 F.3d at 236.

(26.) Id. at 239.

(27.) Id.

(28.) Rice v. Paladin Enterprises, Inc., 940 F. Supp. 836 (D. Md. 1997).

(29.) Rice v. Paladin, 138 F.3d at 233.

(30.) Id.

(31.) Id. at 241.

(32.) Id. at 253-55.

(33.) 186 F. Supp. 2d 1140 (D. Ore. 2001).

(34.) Id. at 1144.

(35.) New York Times Co. v. Sullivan, 376 U.S. 254, 270 (1964).

(36.) 128 F.3d at 264-65.

(37.) 128 F.3d at 267.

(38.) 395 U.S. 444 (1969).

(39.) Id. at 447.

(40.) Id. at 247.

(41.) Id. at 249.

(42.) Id. at 262.

(43.) Id. at 265, quoting Brief of Amici at 3, 22.

(44.) Id.

(45.) Id. at 265-66.

(46.) Miller v. California, 413 U.S. 15 (1973).

(47.) Ferber v. New York, 458 U.S. 747 (1982).

(48.) Ashcroft v. Free Speech Coalition, 535 U.S. 234 (2002).

Leave a Reply