The Supreme Court’s Approach to Restitution For Victims of Child Pornography Possession


In April of this year, the U.S. Supreme Court decided the case of Paroline v. United States. In Paroline, the Court confronted a question about restitution for victims of the sexual abuse of children in the child pornography industry. In the particular case, the victim (whose pseudonym for the litigation is “Amy”) endured sexual abuse by her uncle when she was eight and nine years old. Her uncle perpetrated these violations against his niece in order to produce child pornography. Amy experienced tremendous psychological distress and other complications because of the abuse, and her suffering was severely compounded when she learned that her images were circulating on the Internet among consumers of child pornography.

In the case before the Supreme Court, Amy sought restitution from Doyle Randall Paroline, a man who pleaded guilty in federal court to possessing images of child pornography, including two of Amy. Federal law permits victims of child pornography to seek mandatory restitution from their perpetrators, covering the full amount of the victim’s losses. The argument for restitution for Amy under this federal law was that Paroline, by possessing Amy’s images, was part of the infliction of tremendous distress on her and should therefore be required to pay for the total amount of her resulting lost income, future treatment, and counseling costs, amounting to almost $3.4 million. Paroline, in response, claimed that he could not properly be expected to pay restitution in such amounts, given that a causal link could not be traced between his particular crime and any (let alone all) of the harm experienced by Amy, as required by the causation provision of the restitution statute.

In this column, I will discuss the issue that came before the Court, the resolution of that issue by a majority, and the alternatives offered by two diametrically opposed dissenting opinions. I will then discuss an important narrative missing from the Court’s restitution story.


The primary issue that presented itself to the Supreme Court under the statute was that of factual causation. By its terms, the statute provides restitution for the full amount of a victim’s losses that resulted from the perpetrator’s offense. Therefore, if a causal connection between the offense and any harm or loss experienced by the victim could not be established, it would appear to follow that the court could award no restitution.

In figuring out whether and how much a court might be equipped to award to Amy, a majority of the U.S. Supreme Court wrestled with two competing difficulties that result from application of the causation requirement in a statute plainly intended to provide compensation to victims of child pornography crimes. Let us consider each difficulty in turn.

The first difficulty is that if the ordinary test for factual causation—“but for” causation—applied, then almost no one victimized by child pornography would ever be able to recover restitution under the federal statute. This is because, as all parties agree, it is not possible to say accurately that “but for” the actions of the particular offender here, the victim would not have suffered the harm she suffered. Even if Doyle Randall Paroline had never possessed Amy’s images, the collective thousands of people who have possessed her images would likely have generated more or less the same traumatic consequences for Amy that she in fact experienced. She did not know Paroline individually, after all, and she did not have any direct connection to him beyond her knowing of his possession (along with that of so many others) of her images.

At the same time, if Paroline had been the only person to possess Amy’s images, it is quite likely that she would not have suffered the degree of injury that she actually suffered. In other words, not only was Paroline not a “but for” (i.e., necessary) cause of Amy’s suffering; he was not even a sufficient cause of it. A sufficient cause, sometimes applied to multiple offenders who each do more than enough to inflict all of the injuries suffered, but none of whom are “but for” causes, requires that the particular wrongdoer’s actions alone would have resulted in the same injury. Under this test too, we come up short in assessing Paroline’s conduct as a causal agent relative to Amy’s injuries.

Justice Kennedy, for a majority of the Court, accordingly explains that if we were to apply the usual tests of causation to Amy’s case, then Amy would recover nothing. This, suggests Justice Kennedy, is reason enough to reject the ordinary tests of causation in this case. The Court thus deals with the first (causation) difficulty by construing the statute to better line up with the purposes of the legislation.

Federal law, the Court reasons, provided that people in Amy’s situation would be entitled to “mandatory” restitution from those who had victimized her, and those perpetrators plainly include people who possess child pornography. To apply a test of causation (“but for” or even “sufficient” causation) that eliminates restitution for the vast majority of cases—in which many, many perpetrators have participated in harming the victim of child pornography through possession of her images—would be contrary to the plain intent of the statute. By interpreting causation more flexibly in contemplating the causal relationship between the activity to which this petitioner contributed and the harm that the victim suffered, a flexible interpretation that is found in the tort law concept of joint and several liability, it becomes possible to identify causation and thus to effectuate the statute’s plain purpose to provide restitution to victims like Amy.

There is, however, a second difficulty that confronts the Court, one that emerges more fully once the “causation” problem is managed through creative statutory construction. This problem is proportionality. It strikes the Court as grossly disproportionate to the particular defendant’s share in the misconduct victimizing Amy to require him to compensate her for all of the harm that she has suffered (and will suffer) from the thousands of people who have created, distributed, and consumed her images, given Paroline’s relatively minor causal role in her suffering.

Furthermore, the statute does not expressly provide for seeking contribution from other perpetrators (to help reduce the disproportionate burden on the particular perpetrators who are forced to pay restitution first), and such contribution might—even if theoretically available—do little to address the inequity, because most people who possess images of Amy will never be caught with her image and therefore will never have to pay her (or indemnify Paroline, if he were held fully liable) even a penny. The Court explains that “[t]he reality is that the victim’s suggested approach would amount to holding each possessor of her images liable for the conduct of thousands of other independently acting possessors and distributors, with no legal or practical avenue for seeking contribution. That approach is so severe it might raise questions under the Excessive Fines Clause of the Eighth Amendment.”

A majority of the Court thus feels torn between the counterintuitive outcome of applying a strict “causation” analysis, on the one hand, and, on the other, the apparently disproportionate outcome of holding one convict in possession of child pornography responsible for all of the (arguably indivisible) harm suffered by a victim of an extremely large (and not internally coordinated) number of people trafficking in and consuming her image. The solution at which the majority arrives is to “split the baby.” The Court directs judges to “order restitution in an amount that comports with the defendant’s relative role in the causal process that underlies the victim’s general losses,” arriving at “a reasonable and circumscribed award imposed in recognition of the indisputable role of the offender in the causal process underlying the victim’s losses and suited to the relative size of that causal role.”

The first dissent, by Chief Justice Roberts (joined by Justices Scalia and Thomas) agrees with the majority that Congress seemingly meant to provide compensation to victims. Contrary to the majority, however, the Chief Justice’s dissent says that the plain text of the statute requires actual causation and that the majority’s attempt to invent a fair (but fictional) causal standard that balances the text and purpose of the statute is both inappropriate and vague. One could, under this standard, award either $500,000 or $100, consistent with the majority’s opinion, and the statute does not appear to leave things open to the boundless discretion that the majority thereby invites. In the dissent’s words, “[w]hen it comes to Paroline’s crime—possession of two of Amy’s images—it is not possible to do anything more than pick an arbitrary number for that ‘amount.’ And arbitrary is not good enough for the criminal law.” This dissent thus concludes that under the statute (including its causation requirement), as presently written, Amy must, regrettably, recover nothing.

A second dissent, by Justice Sotomayor, shares the majority’s commitment to providing compensation for the victim but, contrary to the majority opinion, rejects the notion that full restitution for all of the victim’s injuries would be disproportionate. Justice Sotomayor agrees that the plain statutory purpose of the legislation invites a flexible understanding of factual causation—one that encompasses all of the injury caused by the indivisible misconduct in which the petitioner and the many other perpetrators who possessed her image engaged. Justice Sotomayor, however, is less concerned about any disproportionality between the petitioner’s actions and the restitution that she would order (and that he might not successfully recover from other perpetrators) than she is about the possibility of leaving a victim with only a small portion of the restitution to which she is entitled, driven to bring many different actions against various perpetrators for potentially only nominal recoveries. As between the perpetrator and the victim, Justice Sotomayor says, let the perpetrator bear the risk of being unable to recover the inequitably distributed cost of the harm that he participated in inflicting.

The Complexity of Causation for Abstract Harms

In pleading guilty, Doyle Randall Paroline, though admitting to possession of only two images of Amy, also admitted to possessing between 150 and 300 child pornography images in total. Yet the other victims of his misconduct will likely never have the opportunity to seek restitution, because it may be impractical or impossible to identify who they are from the pictures.

A focus on this practical (but inevitable) failure of the restitution scheme overlooks an even more glaring gap: Paroline’s conduct inflicted harm not only upon those individuals whose violated childhood images he possessed. His behavior also harmed the many other victims whose later sexual violation was driven in part by the demand that he expressed (through his consumption choices) for the products of recorded child molestation. To make this point differently, the harm committed by those who purchase child pornography does not consist only of the shame, anxiety, and despair that the pictured children feel as a result of that possession. The harm consists as well of fueling a market in recorded sexual violence against children. And because such violence occurs after the earlier perpetrator’s purchases and likely affects children who do not appear in the particular child pornography that this individual possessed, it will be literally impossible to draw a causal link between his actions and the specific sexual violation that other children experience as a result of his conduct.

Notwithstanding the impossibility of providing proof of causation, the act of feeding the market in the creation of child pornography—in the sexual violence that is recorded for later viewing—is one of the reasons that the Supreme Court ruled in New York v. Ferber that distribution of child pornography may be criminalized, consistent with the First Amendment. The Court there noted, among other things, that “[t]he advertising and selling of child pornography provide an economic motive for, and are thus an integral part of, the production of such materials, an activity illegal throughout the Nation.”

That is, the First Amendment does not protect child pornography, given the economic relationship between the money to be earned in the child pornography business and the violence against children required to produce the product that earns that money. It follows that the possession of child pornography by the person who demands the product also drives its actual creation through child molestation.

The specific woman who was previously molested—like Amy—obviously suffers tremendous trauma from knowing that unwanted parties have previously viewed and will continue to view her image, thereby aggravating the initial violation that she suffered. But an additional grievous harm that proceeds from possession is the demand that possession manifests for more molestation of children in the future. Indeed, it was likely Amy’s uncle’s knowledge that he could make money by filming child molestation—a knowledge derived from his observation of the great demand for images of child molestation manifested through other offenders’ possession of such images—that motivated him to rape his young niece in front of a camera in the first place.

The causal role of possession offenses in generating demand for future victimizations is plain, but it is not the sort of causation that lends itself well to conventional restitution. To oversimplify, consider the following hypothetical fact pattern. Producer1 molests Victim1 in front of a camera and then sells the images to Trafficker1, a child-pornography trafficker. Trafficker1 then sells the images to Consumer1, a pedophile who derives sexual satisfaction from viewing such images.

After making money in this way, Producer1 comes to realize that this behavior is very profitable and is motivated to continue to molest more children, such as Victim2 and Victim3, in front of a camera, in the hopes of making even more money. Producer1 then sells the resulting images to Trafficker1 again and perhaps to Trafficker2 as well, who in turn sell those images to another pedophile, Consumer2. At least two significant victims of Consumer1’s behavior in this scenario are now Victim2 and Victim3, even though Consumer1 may never view images of Victim2’s or Victim3’s violation. It was, in this scenario, Consumer1’s purchase that communicated to actual and potential producers the fact that “you have customers out here who want this product, so please make more,” and that message resulted in Victim2 and Victim3 being attacked.

In the real world, it will be essentially impossible to know which victims are causally tied to which possessors, because it is the aggregate demand expressed by the possessors that together generate the aggregate molestation in front of a camera. It is difficult, as the Paroline Court recognizes, to identify precisely who saw images of Amy and of other specific victims, but that difficulty pales in comparison to the difficulty of tracing definite causal links between the possessor who motivates further exploitation and the specific victims of that further exploitation. This is an argument for fines rather than restitution or, alternately, for a non-specific restitution that is due to any victim of child pornography from any participant in the child pornography business, regardless of whose specific images turn up in an offender’s possession.

By analogy, readers may know that I am an ethical vegan. One reason for my decision to avoid consuming animal-derived products is the fact that when a consumer expresses demand for flesh, dairy, eggs, and other such products, all created by hurting and slaughtering animals, the consumer thereby solicits more violence against animals, just as purchasers of child pornography thereby solicit more violence against children. A consumer, however, could not hope to provide restitution to the animal whose flesh or hormonal secretions she has eaten, because by the time she ate the violence-derived food, the victim of that violence was likely dead already (and that includes the victims of dairy and egg production, who are allowed to live only a fraction of their lifespans before being slaughtered). Yet the victims of a consumer’s animal consumption consist not only of the animals she has actually consumed but include as well the animals who will be bred into existence for later violence and slaughter, as a foreseeable and solicited consequence of her consumption.

When I consumed animal products, many of the animals I harmed did not yet exist at the time that I manifested my own (ill-informed and ill-considered) demand for violence against animals. And by the same token, many of the children who will suffer sexual violence in front of a camera because people purchase images of that kind of violence have not yet been violated at the time that an offender demands an image that fuels demand for more such images. Requiring precision in drawing causal lines thus frustrates the rights of victims to obtain compensation for their suffering and the need to hold an offender accountable for the tremendous cost to countless future victims of his seemingly “victimless” act of possession.

If Justice Sotomayor had taken account of the contributing role of possession in manifesting consumer demand for future child molestation, she might have had an easier time defending the imposition of seemingly disproportionate restitution awards on a criminal defendant responsible for only a small fraction of a specific victim’s injuries. Paroline unquestionably participated in hurting Amy and in similarly hurting the other 149-299 people whose childhood images he admitted to possessing. But equally important, Paroline also participated in hurting other child victims of recorded molestation whose molesters knew—from people like Paroline—that there was money to be made in the production of more such images. The apparent gravity of Paroline’s offense thus becomes more significant when we take into account the role of demand in a market economy. With a properly drafted restitution (or fine) statute that takes all victims, past and future, into account, the fairness of holding people like Paroline accountable for their behavior through large restitution awards will be far clearer and less in doubt.

  • Emma Sweet

    You missed a significant matter in this article. Paroline did not purchase any of the materials found in his possession. He merely used a file sharing application, which he used to download the images/videos free or charge. There was no money and no transaction.

    It is true that commercial child pornography fuels demand for other children to be abused, but the material that is widely available (non-commercial) does not apply. In addition to this it seems that you have emphasized a very commonly touted idea that viewing is the equivalent of victimizing, which is merely a metaphor used in courts all over the world to convict the defendant on the child pornography charges. However the harm does not arrive to the victim until he/she is notified of the trial proceedings.

    On the other hand the victim is left with the general idea that images of abuse portraying him/her in sexually abusive situations is available for anyone to see, but not someone they know.

    A majority, probably around 90%+ are convicted on child pornography charges relating to possession and distribution for materials that have circulated for years, however imprisoning the suspect has had no effect the availability of the material. In order to end the production of child pornography, everyone must be educated on the issue, such as signs to look for; being proactive. Prison does nothing to protect the next child from being abused and plastered on the Internet.

    • In this piece above, I can see your point and can agree with it. To say that viewing is the equivalent of victimizing, opens up many can’s of worms. Such as someone researching the physiology of this kind of crime, etc… The people viewing these types of films are not the ones creating these films or the one abusing these victims.

      We would need to find a way to track down the people who are on the films themselves in order to bring down the number of victims of sexual abuse. Is this already being done? If not, why aren’t they?

    • Maurice

      Emma, you have nailed it, clarifying for the general public what I have been saying, in vain, for years. Causation just isn’t there. The truth is, child pornography is NOT a normal supply & demand market. Unlike the illegal drug market, child pornography is never consumed away: downloading an image is only copying an image and if no more new material were produced, there would still be an internet full of it for consumers forever.

      Furthermore, I think Colb, the author of this article, falls into another trap common in sex offense discussions, that of presuming she understands the production side of the issue. She believes that it necessarily runs like a business, more purchases fuel more abuses. I, for one, find it difficult to believe that someone would not be abusing children were it not for the money. For example, I happen to be straight; I also know there is good money to be made in making gay pornographic movies. However, I am not about to start making gay films because of the financial incentive. Not only do I have no sexual interest in doing so, I find the idea somewhat repulsive and it would not work, physicially speaking. In this case, what is required is a perpetrator that is sexually attracted to, and physically capable of, abusing children. It is far more likely that the production happens as an afterthought, a by-product of the abuse that would be happening anyway.

      If there is any mandatory restitution or fines, I think the best use for those funds would be to abuse prevention programs. That is money that, without question, would be money well-spent.