Animal-cruelty videos & free speech: some observations from data

Thursday, July 9, 2009

First Amendment principles are majestic until, it often seems, they’re applied in controversial contexts.

Take racial hate speech, or the commercial-speech rights of tobacco advertisers, or the free-speech rights of abortion-clinic protesters. In each instance, it is far easier to support the general principle than its application in practice. Hard cases test the mettle of those who defend the principle of free expression in America. And now those who hold to that creed are being put to a new and formidable test — to what extent, if any, should the First Amendment protect the right to make and distribute animal-cruelty videos? That is the basic question before the Supreme Court in United States v. Stevens, a case to be argued this fall.

It is a troubling case. Then again, some of the most important cases in the history of the First Amendment were cut from just such a cloth. While reasonable minds may disagree on what result the Court should reach, what is problematic from a First Amendment vantage point are some of the bold arguments advanced by the U.S. solicitor general in defense of a federal law outlawing the commercial creation or distribution of animal-cruelty videos. There is reason for serious concern when those arguments urge the Court to create an exception to the First Amendment — something that has not happened since New York v. Ferber (1982) — and do so by relying heavily on an old and First Amendment-hostile precedent (Chaplinsky v. New Hampshire, 1942). Equally alarming is the fact that the government urges such a course of action based, in important part, on a questionable proposition. This analysis will examine that proposition.

The governing law
In 1999 Congress passed a law (18 USC § 48) concerning depictions of animal cruelty. The law prohibits the “knowing” creation, sale, or possession of “depictions of animal cruelty with the intention of placing” such depictions in “commerce for commercial gain.” The penalty is a fine, or five years' imprisonment, or both. The law, however, does “not apply to any depiction that has serious religious, political, scientific, educational, journalistic, historical, or artistic value.”

This statute defines depictions of animal cruelty as “any visual or auditory depiction, including any photograph, motion-picture film, video recording, electronic image, or sound recording of conduct in which a living animal is intentionally maimed, mutilated, tortured, wounded, or killed, if such conduct is illegal under Federal law or the law of the State.”

The knowledge requirement and the commercial-purpose condition, combined with the exceptions clause, were intended to cure any possible due-process or First Amendment problems, including overbreadth. Whether those problems were indeed cured is now one of the questions before the Supreme Court.

The case
In March 2004, a Pennsylvania federal grand jury returned a three-count indictment against Robert J. Stevens, a resident of Virginia. The counts charged the defendant with knowingly selling depictions of animal cruelty with the intention of placing those depictions in interstate commerce for commercial gain in contravention of 18 USC § 48. The charges stemmed from two videotapes portraying pit bulls fighting and a third, titled “Catch Dogs,” offering instructional footage of dogs hunting prey (wild boars).

The U.S. District Court for the Western District of Pennsylvania denied Stevens’ motion to dismiss the indictment based on his claim that Section 48 violated his First Amendment free-speech rights. When the matter went to trial, a jury found Stevens guilty on all three counts. He was sentenced to 37 months in prison on each count, to be followed by three years of supervised release.

When the 3rd U.S. Circuit Court of Appeals heard the matter, it sustained Stevens’ First Amendment claims, reversed the district court’s decision, and vacated the conviction. The majority applied the five-factor test found in Ferber and concluded that, unlike child pornography, depictions of animal cruelty are safeguarded by the First Amendment. It then held 18 USC § 48 to be facially unconstitutional under the First Amendment. Three judges dissented.

The government’s arguments
The government appealed the 3rd Circuit ruling to the Supreme Court and on April 20, 2009, the justices agreed to hear the case. Oral arguments are scheduled for the fall.

The solicitor general’s merit brief sets forth two basic arguments for reversing the judgment. First, that Section 48 does not regulate protected speech; and second, that Section 48 is not substantially overbroad. As a part of her overall arguments for sustaining the law, Solicitor General Elena Kagan put forth the justification that “animal fighting ventures” are “nearly impossible to prosecute directly” and that  criminalizing the depiction thus constitutes a “critical tool to reach acts of animal cruelty that long have been shielded from prosecution.” That argument speaks directly to the government’s alleged interest in this case and to its purported need to create an exception to the First Amendment.

Though the government’s brief does not mention the fact, the legislative history of Section 48 reveals that its chief impetus were so-called “crush videos” in which women squash small rodents with their high heels. The three witnesses who testified before the House in support of the legislation confined their testimony to “crush videos” without mention of any difficulties in prosecuting animal-fighting or any other type of animal abuse. Of course, Stevens is not a “crush video” case. Moreover, the New York Times article (“Dog-fighting subculture,” Dec. 7, 2008) that Kagan cited for the proposition that animal-fighting is almost impossible to prosecute illustrated the Texas subculture of dog-fighting and related matters, not the general ease or difficulty of prosecuting animal-abuse cases nationwide.

If the government’s assertion is, in part or whole, a sufficient justification for abridging rights otherwise assured by the First Amendment, then the empirical record must be examined to determine whether animal-cruelty cases generally (as distinguished from “crush video” abuses) are in fact quite difficult to prosecute, so much so that depictions of animal cruelty must also be prosecuted.

Prosecuting animal cruelty cases: Pre- and post-arrest data
In Stevens, the underlying offense of engaging in dogfighting, by causing, provoking or encouraging dogs to fight, is a felony in all 50 states and the District of Columbia. One step removed from this unlawful activity is the possession of the implements of the crime (including the animals). Possession of dogs for fighting is likewise illegal, either a felony or a misdemeanor, in all states and D.C. Two steps removed is attendance at the event. Spectators at dogfights are criminally liable in 48 states and D.C. Only Illinois has taken a third step and outlawed, albeit as a misdemeanor, depictions of animal cruelty (510 Ill. Comp. Stat § 70/3.03-1).

As noted, the government’s brief contends that the clandestine nature of dogfights and the tight-knit communities of those who stage and watch them make prosecution of these crimes unacceptably difficult. The 3rd Circuit, however, was unconvinced that “banning depictions of animal cruelty is a necessary or even particularly effective means of prosecuting the underlying acts of animal cruelty” (533 F.3d 218, 234). For purposes of an empirical investigation, the process of prosecution can be divided into pre-arrest and post-arrest periods.

The methodology employed in this investigation consisted of consulting online databases (Westlaw, LexisNexis, and as well as canvassing animal-rights organizations and animal-cruelty enforcement task forces. As a member of the Animal Legal Defense Fund’s Criminal Justice Program noted in a June 11 e-mail to the author, the lack of tracking mechanisms at the national level makes the collection of such data difficult. In virtue of this fact, as well as the inevitability of unreported crimes, any survey of the data must strive to be a representative sample and not the entire universe of facts.

  • Pre-arrest data
    The most significant impediment to the collection of pre-arrest data is the lack of a systematic cataloging of complaints at any level higher than individual municipalities or local animal-protective organizations. A few studies in the mid-to-late 1990s (one from Ohio, one from Massachusetts, one from New Jersey) found that the ratio of animal-cruelty prosecutions to complaints was low — there are many more complaints than prosecutions. Nonetheless, a good deal of evidence suggests that this rate, especially with regard to dogfighting, has increased in recent years with the visibility of the 2007 Michael Vick dogfighting incident, prosecution and conviction.

    Indicative of this heightened awareness, several new animal-protection/dogfighting task forces have been established in places such as Los Angeles, Indianapolis and North Carolina. The Los Angeles Times reported early this year that the district attorney had filed charges in 50% more cases for the year ending in August 2008 than for the one ending in 2007 (Jack Leonard, “LA Cracks Down on Animal Cruelty,” Feb. 15, 2009). Alone, the increase in such filings does not show that the number of prosecutions per complaint has increased. But when combined with data (the database of animal cruelty) that shows reported incidents of animal cruelty in California and around the country to be fewer in 2008 than in 2007 (and fewer in 2007 than in 2006), it is a fair inference that the ratio of prosecutions to complaints has increased. And it has increased without the putatively necessary assistance of Section 48 enforcement, which has been quite rare at best.

  • Post-arrest data
    Post-arrest data, by contrast, is more readily accessible. The site offers a detailed repository of some 15,000 incidents of animal abuse and more than 1,300 incidents of criminally-induced animal fighting, with most of the data from within the past 10 years. According to Alison L. Gianotto, the president/CEO of, its information is compiled from news outlets, animal-control officers and prosecuting attorneys.

    The compilation cannot hope to capture every incident, but the details of each incident emerge from reliable sources. Of the 1,252 animal-fighting cases listed in the United States, 820 involve dogs, the vast majority being dog-on-dog fighting. Criminal charges were brought in more than 84% of those dog cases. The data, however, may make it appear that animal-fighting incidents were more likely to be prosecuted than is actually the case — because most incidents reported by prosecutors or news media would come only after arrests were made. data may be more useful for what it can show about what happened in cases after arrests were made and whether it's “nearly impossible” to prosecute such cases after they get through the courthouse door.

    Concerning dogfights, lists 246 convictions as contrasted with a small number of acquittals (four), and a similarly small number of dismissals (six) and conditional dismissals (seven), the latter usually contingent upon pleading to other charges or donating to an animal-protection agency. Such figures reveal a conviction-success rate of over 93%. (More than 400 of the other dogfighting cases are listed as “alleged,” which denotes that the case is still in progress.) According to John Goodwin, who manages the Humane Society's campaign against animal-fighting, the success rate of federal prosecutions of dogfighting exceeds 98%. (See Joe Biddle, “Vick Raises Bar on Cruelty,” July 21, 2007, The Tennessean, Nashville).

    Similarly, in the case of cockfighting, lists 65 convictions as contrasted with no acquittals and a single dismissal. The conviction success rate in these cases tops 98%. data shows cases of dog- and cockfighting have a combined conviction success rate of over 94%. For all types of animal cruelty, the total prosecutorin-success rate is 90%.

    Appellate data: At the appellate level, the defendants in animal-fighting cases fare poorly when challenging their convictions on “sufficiency of the evidence” grounds. A search of state and federal courts via the legal-research engines Westlaw and LexisNexis reveals that such challenges fail more than 88% of the time. (The following search terms were used on June 17: “animal fighting,” “dogfighting,” “dog fighting,” “cockfighting,” “animal cruelty.”)

    A counterproductive practice? Furthermore, data from also reveals that deterring the production of videos and DVDs may be counterproductive, thus making it more difficult to convict for the underlying offense of cruelty to animals.

    A keyword search for “video” on June 18 revealed 107 cases in which the individuals who committed a crime of animal cruelty also videotaped their own criminality. And in many other cases, neighbors or other good Samaritans documented the events. In these cases, the videotape often proves to be a pivotal piece of evidence for use at trial. Because good Samaritans are immune from the law’s reach, surveyed below are only those animal abusers who videotaped their own acts, presumably for entertainment and often for commercial gain.

    The overall 90% success rate of prosecutions for all types of animal cruelty noted above is irrespective of whether the defendants had videotaped their acts. In contrast, in prosecutions with such tapes the prosecution succeeded in 43 of 44 completed cases for a success rate of almost 98%. The one acquittal involved the hunting of a wild boar.

    In prosecutions irrespective of such videotape evidence, the prosecution success rate specifically for animal-fighting incidents was 94.5%. By contrast, in prosecutions with such tapes the success rate was 100% (23 convictions in 23 completed cases). Furthermore, prosecutors pressed charges in 54 of 55 (more than 98%) reported animal-fighting incidents involving videotaping. For all animal-fighting reports, regardless of the presence or absence of videotape, charges were filed in 88.3% of cases.

    This data supports the intuition that videotape has evidentiary value in prosecuting the underlying offense. Prosecutors seem more willing to go forward when there is videotape evidence, and juries seem more willing to convict in such cases. Hence, by outlawing and thereby deterring videotaping of cruelty, Section 48 may not only fail to promote the justice of animal cruelty prosecutions, it may also actively impede such justice.

    The above data suggests the following: (1) that once an arrest has been made, it is not difficult to prosecute animal-cruelty cases generally; (2) that it is even easier to prosecute dogfighting and cockfighting than average animal-abuse cases; (3) that once a conviction is secured, the reversal in animal-fighting cases is very low; and (4) that the absence of videotape evidence decreases the chances for a successful prosecution.

    While parties may argue in the abstract whether animal-cruelty videos ought to be entitled to First Amendment protection, it is an entirely different proposition whether an exception to the First Amendment ought to be allowed on the basis of questionable factual assertions at odds with the empirical evidence as we know it.

    If the constitutionality of 18 USC § 48 were to be sustained in United States v. Stevens — the first prosecution under the statute — it would have to be on grounds other than the allegation that such laws are “critical” to the successful prosecution of animal cruelty involving dogfighting.

    Adam Ezra Schulman is a law student at Georgetown University Law Center and an intern at the First Amendment Center in Washington, D.C.

  • Tags: ,