The ethics of pedophilia

Nordic Journal of Applied Ethics

Moen, Ole Martin
Type of WorkEssay
Original Publication


The References of this article are here:


Pedophilia is bad. But how bad is it? And in what ways, and for what reasons, is it bad? This is a thorny issue, and sadly, one seldom discussed by ethicists. I argue in this article that pedophilia is bad only because, and only to the extent that, it causes harm to children, and that pedophilia itself, as well as pedophilic expressions and practices that do not cause harm to children, are morally all right. I further argue that the aim of our social and legal treatment of pedophilia should be to minimize harm to children, and that current practices are often counterproductive in this respect.


Few things upset us more than pedophilia, and with good reason. Pedophilic sex harms children, and roughly one in five girls, and one in twelve boys, are victims of sexual abuse. [*1]

  • [*1] Pereda, N., Guilera, G., Forns, M., & Gómez-Benito, J., 2009. This meta-analysis sampled 65 original studies from 22 countries and concluded that the best estimate of the global prevalence of child sexual abuse is
    • 19.7% for girls and
    • 7.9% for boys.

What is pedophilia?

According to the World Health Organization, pedophilia is

  • “sexual preference for children, boys or girls or both, usually of prepubertal or early pubertal age” (WHO 2010).

The discussion in this paper stays within the boundaries of WHO’s definition, but focuses on sexual preference for prepubertals. Sexual preference for prepubertals provides the clearest example of pedophilia, and by focusing on clear cases we can set aside, for the moment, the added complexities that arise in the borderline case of sexual preference for those who have entered puberty. [*2]

  • [*2] Attraction towards those in early puberty is called hebephilia. Attraction towards those in late puberty is called ephebophilia.

Pedophilia is prevalent.

In an influential 1989 study, John Briere and Marsha Runtz sampled 193 male college students, and found that

  • 9% reported that they had had at least some (more than zero) sexual fantasies involving prepubertal children,
  • 5% admitted to having at least once masturbated to such fantasies, and
  • 7% indicated at least some likelihood of seeking sexual contact with a child if they were certain that they would avoid detection and punishment
  • (Briere & Runtz 1989).

In 1996, Kathy Smiljanich and John Briere sampled 279 students, and arrived at a similar result. They also found that there is a significant gender discrepancy among pedophiles: roughly seven out of eight pedophiles are men (Briere & Smiljanich 1996).

In the most recent and most comprehensive study to date, Kathryn Becker-Blease, Daniel Friend, and Jennifer J. Freyd sampled 531 male students, and found that

  • 7% admitted to having some sexual attraction to children and that
  • 3% would consider seeking sexual contact with a child if they were certain that no one would find out (Becker-Blease, Friend & Freyd 2006).

Nathaniel McCognathy (1998) and Michael C. Seto (2009), in their respective meta-analyses, estimate that around 5% of men, or slightly less than that, are to some extent sexually attracted to prepubertal children.


Pedophilia concerns many different disciplines, most obviously psychology, psychiatry, and criminology. It is also, however, a subarea of sexual ethics, and as such, a concern for ethicists. Disappointingly, however, ethicists have given little attention to pedophilia. [*3]

  • [*3] There are some exceptions:
    • Ehman (1984) defends adult-child sex.
    • Primoratz (1999) and Kershnar (2001) criticize Ehman.
    • Spiecker & Steutel (1997) present and criticize a number of arguments defending adult-child sex.
    • Benatar (2002) argues that if we accept casual sex, it is hard to reject pedophilia.
    • Luck (1999) argues that if we accept murder in virtual reality, it is hard to reject adult-child sex in virtual reality.

This is unfortunate, because ethicists can make valuable contributions to how pedophilia should be understood, assessed, and handled.

My goal with this paper is to discuss the ethics of pedophilia somewhat comprehensively,
[1] beginning
with a consideration of the ethical status of being a pedophile.
[2] I then proceed
to discuss adult-child sex: What makes it wrong for an adult to seek sexual contact with a child, and under what conditions are the adults who do so blameworthy?
[3] Finally
, I turn to the ethical status of producing, distributing, and enjoying fictional stories and computer-generated graphics with pedophilic content.

The overall approach I will defend is that pedophilia is bad only because, and only to the extent that, it causes harm to children, and that pedophilic expressions and practices that do not cause harm to children are morally all right. I explore a number of practical implications of this view in [4] the conclusion.

[1] Being a Pedophile

According to WHO’s definition (quoted above) a pedophile is someone who has a sexual preference for children. Note that a pedophile need not ever have acted on his preference in order to qualify as a pedophile, according to this definition. Is it immoral to have a sexual preference for children?

Before answering this, it is worth noting that a moral assessment of pedophilia is different from both an evaluative and a psychiatric assessment. Pedophilia might well be evaluatively bad in the sense that it would be better if a given person were not a pedophile. Being a pedophile is unfortunate for the pedophile himself, who will most likely not have a good sexual and romantic life, and for a number of children, who might be seriously harmed by his actions. Pedophilia might also be psychiatrically bad in the sense that it might be a mental disease. But is being a pedophile, in the sense of having a sexual preference for children, immoral?

For something to be subject to moral assessment, we usually assume that it must be something that, in some sense, we control or choose. If something is entirely outside of our control, such as our genetic makeup, it seems that though it might be more or less fortunate, and perhaps more or less pathological, it is not properly subject to moral assessment.

Do pedophiles choose or control their sexual preferences?

Nothing indicates that they do. The competing scientific explanations of pedophilia are

  • conditioning,
  • childhood sexual abuse, and
  • neurodevelopmental perturbations.

In terms of phenomenology, pedophiles report that they discover their sexual preference (usually in their teens), not that they choose it (see Seto 2007).

Once someone is a pedophile, moreover, there is little evidence to suggest that it is any easier for them to choose to be sexually attracted to adults than it is for those of us who are not pedophiles to choose to be sexually attracted to children. Being a pedophile is also highly disadvantageous, so it is also puzzling, on a more general level, why anyone would choose a pedophilic preference in the first place. Though this issue cannot be settled here, it seems most plausible that pedophiles do not choose their preferences, and that though their preferences might well be both unfortunate and pathological, the mere fact of having such preferences is neither moral nor immoral.

[2] Adult-Child Sex

Even if we grant that it is not immoral to be a pedophile, it might still be immoral for pedophiles to seek sexual contact with children. To do so is an action, or a series of actions, and actions are subject to moral assessment. Though there are several possible arguments as to why it is immoral for adults to seek sexual contact with children, let me here examine what I take to be the two central arguments:

  • the harm argument and
  • the consent argument.

The Harm Argument

The harm argument is a simple argument with two premises:

  • An empirical premise, that adult-child sex seriously harms children, and
  • a normative premise, that it is immoral to seriously harm children (at least as long as the only justification is that it gives an adult sexual pleasure).

In what ways does adult-child sex harm children?

Presumably, we can distinguish between two main forms of harm.

  • On the one hand, there is bodily harm. If small children are penetrated, or otherwise raped, they have a significant chance of being physically harmed. So much is indeed uncontroversial, and I shall take this for granted in the rest of the paper.
  • What I shall discuss in more detail is psychological harm.

Psychological harm is more complex and more controversial than bodily harm, and it is also the kind of harm to which we must appeal if we seek to explain why most adult-child sex is harmful. The reason why is that most adult-child sex does not involve violence or penetration. Though there are many tragic counterexamples, the most characteristic pedophile activities are cuddling, caressing, and genital fondling, and when full intercourse takes place, it occurs most commonly when the child is well into adolescence (Howitt 1995). It seems hard to argue that cuddling, caressing, and fondling causes physical harm to children’s bodies, so if we think that these cases are problematic as well, then the harms to which we appeal must, at least in part, be psychological.

There are several studies on the psychological effects of adult-child sex on children. One of the largest studies, funded by the US National Institute of Drug Abuse, found that in a sample of 1,400 adult women, childhood sexual abuse was significantly correlated with increased likelihood of drug dependence, alcohol dependence, major depression, and general anxiety disorder (Zickler 2002).

Other studies identify a strong correlation between sexual abuse and various psychological disorders such as

  • dissociative identity disorder,
  • post-traumatic stress disorder,
  • borderline personality disorder,
  • and various eating disorders.

Meta-analyses estimate that between 51% and 79% of sexually abused children display symptoms of psychological disorders (see Hornor 2010). Presumably, psychological disorders, or symptoms thereof, constitute harm, and arguably, the threshold for being harmed is even lower than the threshold for having identifiable psychological disorders or symptoms of such disorders. As such, the evidence for the harmful psychological effects of adult-child sex seems solid.

There are, however, arguments to the effect that these findings fail to show that adult-child sex is harmful. Robert Ehman, a philosopher who has defended adult-child sex, has advanced three such arguments. Let us consider these.

Ehman’s first argument against the harmfulness of adult-child sex is that the available research was conducted using a non-representative sample of children who have had sexual contact with adults.

  • “Although there is a good deal of clinical data on the effects of adult-child sex on the children,” Ehman argues, “there is little in the way of controlled scientific research.”

The children who are studied are those that have come to clinical and legal attention, which is problematic, because

  • “[t]he people who come to clinical and legal attention are not necessarily a typical cross section of the relevant population. … [T]hey are apt to be more negatively affected than those who do not come to clinical attention.” Indeed, he argues, it is often “the negative effects [that] bring them to clinical attention in the first place.”

Instances of adult-child sex that are voluntary and considered good by both parties, and thus result in neither criminal charges nor psychiatric treatment, “remain invisible to clinical observation” (Ehman 1984: 433).

Ehman’s second argument is that adult-child sex is harmful only, or predominantly, because of society’s way of viewing and handling sexual contact between adults and children. The harm, he suggests, is the result of a self-fulfilling prophecy. In Ehman’s view, adult-child sex is made traumatizing by a culture that labels it with strongly evaluative terms such as “abuse,” “assault,” and “molestation,” which tells the child it has been scarred for life. The harms are culturally contingent, and to support this view, Ehman points to places where adult-child sex were viewed and handled very differently, such as Ancient Greece (Ehman 1984: 435-436).

Ehman’s final argument is that children are sexual beings: Children of both sexes are capable of and interested in various types of sexual experiences, and most children have at some point engaged in sexual activities with other children. Because of this, Ehman takes it that the burden of proof rests on those who claim that a child’s sexual gratification may not, in any circumstances, come from sexual contact with an adult. If the burden of proof cannot be met, the argument goes, we should be open to the view that sexual contact between an adult and a child will sometimes be mutually desirable.

What should we make of Ehman’s arguments?

Regarding his first argument, the appeal to an unrepresentative sample, we must concede that it is problematic that many studies are concerned only with cases that are brought to clinical or legal attention. It is worth noting, however, that not all studies suffer from this problem; the US National Institute of Drug Abuse study, for example, relies on a non-forensic and presumably representative sample, and still finds a significant correlation between adult-child sex and psychological problems.

Even if we focus exclusively on the studies that were available when Ehman published his paper (1984), however, it is still unclear how much Ehman’s argument really shows. Though he might have been right that a group of unharmed children escaped the researchers’ attention, nothing in Ehman’s argument tells us how large that group is. His argument is compatible with the unharmed group being very large, but it is just as compatible with it being very small.

We must also take into account the potential underreporting of harmful sexual abuse, so Ehman cannot assume that all, or even most, unreported cases did not result in harm. Children might have failed to tell anyone due to fear or shame, due to few or no prospects of being believed, or, in the worst cases, due to not living to tell. As such, the unrepresentative sample argument, though it does point to a weakness in some of the research, does not show that Ehman is right; it merely allows for the possibility that he is, and since more representative studies also find similar correlations, his first argument is very much weakened.

What concerns Ehman’s second argument, the appeal to a self-fulfilling prophecy, it seems hard to deny that cultural attitudes influence children’s psychological reactions to adult-child sex. In fact, this view has gained some recent support by the findings of Harvard psychologist Susan Clancy (2011), which indicate that victims of child sexual abuse typically do not suffer psychological problems because the abuse was traumatizing when it happened, but rather, that the abuse tends to become a problem later on when the memories are processed and examined, and the actions more fully understood. If much of the harm occurs only after conceptual evaluation, this might suggest an important role for cultural attitudes in determining children’s negative responses.

This is not, however, enough to save Ehman’s argument.

First, even if Ehman is right that the harms are culturally contingent, this does not make the harms any less real, for presumably, a child’s suffering is just as bad when it is contingent on culture as when it is not. A child, moreover, cannot be held responsible for having internalized the norms of his or her society. Though Ehman is right that if the harms are culturally contingent, this raises the question of whether our norms can and should be changed, it still holds true that until or unless they are changed, adult-child sex causes harm to children.

Second, the fact that we use strongly value-laden terms such as “molestation,” “abuse,” and “assault” does not show that our cultural attitudes cause the harms of adult-child sex through a self-fulfilling prophecy, for we might use strongly value-laden terms for the very reason that adult-child sex is harmful. “Rape” is not a neutral concept either, but that does not show that the harms of rape are the result of a self-fulfilling prophecy.

Third, regarding the appeal to Ancient Greece, it should be pointed out that what was condoned in Ancient Greek culture was sexual relationships between men and teenagers, not between men and prepubertal children (see Nussbaum 1994). Moreover, the fact that a certain practice was accepted in Ancient Greece does not show that the practice was harmless. Consider, for example, slavery or the oppression of women.

Our response to the self-fulfilling prophecy argument is given further strength by how we respond to Ehman’s last argument, the appeal to children’s sexuality. Here I think we must concede that children are, in some sense, sexual beings, and for the sake of the argument, let us also concede that this places the burden of proof on those who claim that there should not be any sexual contact between adults and children. Placing a burden, however, is not the same as settling a case, for the question then becomes whether the burden can be satisfied. To satisfy the burden, we would need to point to a difference, or a range of differences, between adults and children that explains why sexual contact between them harms children.

There are, I think, several differences that can play this explanatory role.

One difference, pointed out by Ben Piecker and Jan Steutel, is that although children are sexual beings in one respect, child and adult sexualities are importantly different, and children and adults have different aims in sexual engagement. For one, children might often seek nothing sexual at all in physical intimacy with an adult, but rather, seek confirmation and affection.

Moreover, even in cases where the child’s motivation for physical intimacy is in some sense sexual (presumably, a child can find genital fondling pleasant), the aims in sexual involvement are still different, for though children can experience sexual pleasure, and can be curious about others’ bodies, they do not feel sexually attracted to others and desire others’ bodies the way adults do (Piecker & Steutel 1997: 332-338).

In addition to having different aims, there is also an asymmetry of power and knowledge between children and adults.

  • On the one hand, David Finkelhor observes, “adults control all kinds of resources that are essential to [children] – food, money, freedom, etc.” (Finkelhor 1979: 693).
  • On the other hand, adults are more knowledgeable than children and children tend to trust adults.

For these reasons, it might be very hard for a child to say no to an approaching adult, especially if the adult insists and has already gained the child’s trust. Presumably, the pressure can be overwhelming even if the adult does not intend it this way.

Finally, children are vulnerable, so a bodily invasion and a breach of trust might cause significant difficulties for how children perceive their own bodies and how they perceive adults. Differences in aims, power, knowledge, and vulnerability seem to come far in explaining why, in spite of the fact that children are sexual beings, adult-child sex harms children.

Importantly, however, by pointing to differences in aims, power, knowledge, and vulnerability, we are also pointing to often present but still contingent features of relationships between children and adults. As long as our argument rests on an appeal to such differences, we must concede that it possible that in some cases children are not harmed by adult-child sex. There might be special cases where differences in aims, power, knowledge, and vulnerability do not come into play, or do not do so in a harmful manner.

Indeed, we must concede that, in exceptional cases, the child might not only fail to be harmed but might also, retrospectively, view the incident positively. This conclusion is backed up by empirical findings. In an influential meta-analysis of 59 studies, Rind, Tromovitch & Bauserman (1995) looked not merely for harm or lack of harm, but also for reports of positive experiences, and found that a minority of college students who had had sexual contact with an adult when they were children, and who retrospectively described this experience as positive.

Rind also found that the extent of psychological damage depended heavily on whether or not the incident(s) were retrospectively described as “consensual.” Rind’s study was condemned by the US Congress for providing material that could be used by pedophile organizations to justify their activities. It is doubtful, however, if Rind’s findings give pedophile organizations what they want.

After all, granted the significant harms that often occur, the fact that some children are not harmed — and that a minority reports that their sexual experiences were positive — fails to establish that harm is not expected to result from adult-child sex. A useful parallel to engaging in adult-child sex might be that of giving children hard drugs. Even if we concede that giving children hard drugs might sometimes end well, and even if we discovered that a minority later on reported that being given hard drugs was in sum a positive experience, it will still be true that harm should be expected and that, in some cases, devastating harm will be the result.

It therefore seems that we should reformulate the harm argument in terms of risk of harm. The empirical premise, on this variant of the argument, is that adult-child sex exposes children to a high risk of being seriously harmed; the normative premise is that it is morally wrong to expose children to a high risk of being seriously harmed (absent very strong overriding reasons to the contrary). If we accept both of these premises, as it seems that we should, we have a sound argument as to why it is wrong to engage in adult-child sex.

[2] The Consent Argument

Another common argument as to why it is wrong to engage in adult-child sex is the consent argument. This argument can also be formulated in terms of two premises:

  • first, that it is wrong to engage in sex without consent;
  • second, that children cannot consent to sex, such that all sex involving children becomes non-consensual.

The consent argument has strong intuitive appeal, and David Finkelhor — one of the world’s leading pedophilia researchers — makes the case that the consent argument is even stronger than the harm argument. He explains that the harm argument is empirically vulnerable: It depends on the empirical fact of harm, which is sometimes questionable. The consent argument, by contrast, does not depend on uncertain empirical facts to the same degree. As such, the consent argument is more robust (Finkelhor 1979).

Though Finkelhor might be right that the harm argument is empirically vulnerable, he is wrong in claiming that the consent argument is any less vulnerable. The reason why is that the consent argument is dependent on the harm argument.

To see why, consider first the mundane fact that there are many things to which children may rightfully consent. If I ask my 10-year-old son if we should go play basketball, and he says yes — and we then go on to play basketball — nothing wrong has happened. The same would be true for going on a ski trip, watching a children’s movie, or baking a cake. On the other hand, there are things to which children may not consent. If I suggested that my son and I go play with guns, get drunk, or have sex, it would not be permissible to follow through on these suggestions, irrespective of his consent.

Why may children consent to some things but not to other things? The central explanation seems to be that while some things are harmful, other things are not, and while adults have the privilege to consent to harmful things (within certain limits), children do not have the same privilege, or do not have it to the same extent.

The reason why the consent argument depends on the harm argument, therefore, is that only if the harm argument is sound do we have a good explanation of why children cannot rightfully consent to sex. Had adult-child sex posed no risk of harm at all, it is unclear why children could not consent to it.

Perhaps it could be suggested that children cannot consent to sex because they are not sufficiently physically and psychologically developed to know what they are consenting to. True as this might be, in the absence of any risk of harm, this does not seem to be problematic either. If my son ventured to read Hegel, it is evident that he would not know what he was doing. Still, since this would presumably not expose him to any significant risk of harm, there would be nothing wrong in letting him do so.

Even if we accept that the consent argument depends on the harm argument, however, it does not follow that the consent argument is eliminated. Arguably, the harm argument vindicates the consent argument, and does so by virtue of providing an explanation as to why sex is something to which children, given their level of autonomy and understanding, cannot consent.

Still, within the scope of the present discussion, the consent argument does not add much either. While the harm argument states that it is wrong to expose a child to a significant risk of serious harm, the consent argument states that this is wrong even if the child consents. Since this is already implicit in the harm argument, I shall proceed on the assumption that it is the harm argument that provides the central explanation of why it is morally wrong for adults to engage in adult-child sex.


Even if we agree that it is morally wrong for adults to engage in adult-child sex (because doing so exposes the child to a significant risk of serious harm), the question of moral wrongness is not the only relevant moral question that we might ask. A further question is to what extent, and under what conditions, those who engage in adult-child sex are not only wrongdoers, but blameworthy wrongdoers. Though blameworthiness is a complex issue that cannot be explored in depth in this article, let me suggest two reasons why some pedophiles who engage in adult-child sex might be less blameworthy than we intuitively assume, even though what they do is wrong.

One factor [a]  that might mitigate blame on the part of a wrongdoing agent is that the agent is ignorant of the fact that what he is doing is wrong. Let us call this the appeal to ignorance.

For the appeal to ignorance to detract from the blameworthiness of pedophiles who engage in adult-child sex, then (as long as we relate our discussion to the harm argument) the pedophiles in question would need to be ignorant of the truth of one of the two premises of the argument:

  • either the empirical premise that adult-child sex exposes children to a significant risk of serious harm 
  • or the normative premise that it is immoral to expose children to a significant risk of serious harm (absent very strong reasons to the contrary).

Let us take for granted that pedophiles cannot be ignorant of the truth of the normative premise that it is immoral to expose children to a significant risk of serious harm. Could it be, however, that many pedophiles are ignorant of the truth of the empirical premise that adult-child sex exposes children to such harms?

I think some reasons support this view. To see why, we must first appreciate that sexual attraction towards children is often a deep and integral part of pedophiles’ personalities, and that from the pedophiles’ perspective, adult-child sex is very appealing. Presumably, adult-child sex is just as appealing to pedophiles as sex between adults is to non-pedophiles.

Due to self-serving biases, therefore, it will take very serious reflection on the part of pedophiles to realize that the actions that they like so much systematically expose the other party to a serious risk of harm.

It is also worth noting that the case against the harmfulness of adult-child sex is not obviously mistaken (cf. Ehman’s arguments), and in fact, pedophiles have at least some justification for rejecting the mainstream view. To the extent that the mainstream view is that adult-child sex is categorically very harmful, it is an overly strong view, and being overly strong, it is easily falsified.

Accordingly, when pedophiles discover, for instance, that NAMBLA, the North American Man-Boy Love Association, has had a number of teenage members, some pedophiles might think, not entirely without justification, that the mainstream view must be mistaken. It is paradoxical, therefore, that those who promote the view that adult-child sex is categorically very harmful, by virtue of promoting too strong a view, contribute to justifying, to some extent, a pedophile’s thinking that society at large must be mistaken about adult-child sex.

This, in turn, is a pragmatic reason for promoting the more accurate view: that adult-child sex is not categorically very harmful, but that it exposes children to a significant risk of being seriously harmed. While the categorical view is falsified by a single instance of adult-child sex that does not harm the child, the more accurate view, though it is in principle falsifiable, is not so easily falsified.

Perhaps we can demand both of pedophiles and of others that they be epistemically responsible: that they counteract their biases, even when doing so is emotionally difficult, and that they charitably modify opposing hypotheses as these get falsified. To the extent that this is a tall order for the average pedophile, however, we have a reason that might detract from the blameworthiness of the average pedophile who seeks out (at least certain forms of) adult-child sex.

Another factor [b] that might mitigate blame is bad moral luck on the part of pedophiles. Pedophiles are unlucky to be pedophiles, and we should concede that for many of them, it must be very difficult to go through life without ever seeking sexual contact with a child.

Why, we might ask, should we expect it to be easier for pedophiles to abstain completely from adult-child sex than it is for others to abstain completely from adult-adult sex? Approaching this from the perspective of moral luck, we might ask how many of those who currently blame pedophiles for engaging in adult-child sex would have managed to abstain completely from adult-child sex in case they themselves were pedophiles. Of course, it is possible that they would all have been able to abstain, but it is far from obvious, and we should be careful before expecting a much higher level of willpower from pedophiles than we know to be the norm.

In spite of the appeal to ignorance and the appeal to moral luck, it might well be that some forms of adult-child sex, such as rape, are so clearly harmful that our intuitive assumptions about blameworthiness are fully justified. Moreover, observe that what counts against blameworthiness in these cases is exclusively limitations on the part of pedophiles qua agents, so if a given pedophile is sufficiently enlightened and autonomous, neither the appeal to ignorance nor the appeal to moral luck will detract from his (or her) blameworthiness.

Finally, even if we acknowledge that the above reasons have the potential to count against blameworthiness in individual cases, they need not count against criminal prosecution. We might still have reasons for prosecution grounded in the protection of victims and the prevention of future harms.

[3] Fictional stories and computer-generated graphics

So far I have considered the ethical status of being a pedophile and of engaging in adult-child sex. How, however, should we assess ways to satisfy pedophilic preferences that do not involve any actual children, such as the enjoyment of fictional stories and computer-generated graphics with pedophilic content? Even though most of us might think of such stories and graphics as less bad than adult-child sex, there still seems to be something troubling about them, and in most countries they are banned. But what makes such texts and graphics bad? It seems clear that we cannot appeal to harm to virtual children since, presumably, virtual children cannot be harmed.

One explanation [a] might be that even though virtual children cannot be harmed, real children can be harmed as a result of what goes on in virtual reality. It might be argued, for example, that exposure to texts and graphics that sexualize children makes pedophiles more prone to engage in adult-child sex in the real world.

This argument has some intuitive appeal, as it seems likely that repeated virtual engagement in an activity lowers one’s barriers to engaging in that activity outside of the virtual world. In spite of the intuitive appeal, however, the claim is empirically questionable.

Dennis Howitt, in a study of pornography usage among pedophiles, concludes that

  • “no clear-cut causal link has been demonstrated between … exposure to pornography and sex crime.” (Howitt 1995b: 17)

A similar conclusion has been drawn by Jerome Endrass, who recently found that

  • “[c]onsuming child pornography alone is not a risk factor for committing hands-on sex offenses – at least not for those subjects who had never committed a hands-on sex offense.” (Endrass 2009: 1)

David Riegel has gone one step further and suggested that for many pedophiles, pornography is a tool that helps them redirect their urges and drives, and gives them an outlet for their sexual desires in a way that does not involve having sex with children (Riegel 2004).

Howitt, Endrass, and Riegel’s conclusions are also given tentative support by broader sociological findings. Milton Diamond has found that when the Czech Republic lifted its ban on pornography (including child pornography) in 1989, there was a drop in rape and child sexual abuse (Diamond, Jozifkova & Weiss 2011). Diamond also found similar negative correlations between the availability of child pornography and adult-child sex in Japan (Diamond & Uchiyama 1999).

Granted our current knowledge, it therefore seems that texts and computer-generated graphics with pedophilic content may result in less adult-child sex.

Another possible argument  [b] as to why it is wrong to enjoy fictional stories and computer-generated graphics with pedophilic content is that all attempts at satisfying pedophilic sexual preferences are in some sense abusive or disrespectful, or expressive of ill will, which makes them morally bad.

Let us grant, for the sake of the argument, that a practice can be made morally bad by virtue of being abusive or disrespectful, or expressive of ill will. The question then becomes whether a pedophile’s enjoyment of texts and computer-generated graphics with pedophilic content involves these features.

Though it is tempting to assume that it does, I do not think we are justified in making that assumption. Note that the assumption concerns the mental states of pedophiles, and though it is true that the sexual actions that pedophiles desire tend to harm children when carried out in the real world, we have no reason to believe that what pedophiles desire is to harm children.

Pedophiles desire to have sex with children, and as long as it is only a contingent fact about the world that children are harmed by adult sex, there is no reason to posit intentions to harm on the part of pedophiles. When pedophiles themselves are asked about their intentions and motivations, they commonly report that they fall in love with children, long for physical intimacy with children, and want their feelings to be reciprocated (Green 2002).

Unless we find good reasons for disbelieving them, we should be careful before assuming that abusive or disrespectful intentions, or expressions of ill will, are perennial features of the satisfaction of pedophilic preferences.

This is not to deny that some pedophiles’ intentions are in fact abusive or disrespectful, or expressive of ill will. Very likely, some pedophiles prefer outright violence against children. Even if we grant that it is always wrong to seek to satisfy such preferences, however, we still do not have an argument against the satisfaction of all pedophilic preferences. A preference for violence can, after all, also be a feature of adult-oriented homo- and heterosexuality, but we do not think that that is sufficient to make the satisfaction of all adult-oriented homo- and heterosexual preferences wrong.

Perhaps it can be argued that violent preferences are more closely tied to pedophilia than to other sexual preferences, but this is an empirical claim for which we need evidence, and even if we had this evidence, we would still only have an argument against the satisfaction of a subset of pedophilic preferences, not against the satisfaction of pedophilic preferences as such.

Accordingly, the argument that it is wrong to enjoy fictional stories and computer-generated graphics with pedophilic content because doing so is abusive or disrespectful, or expressive of ill will, is weak.

A final argument [c] as to why it is wrong to enjoy stories and graphics with pedophilic content is that pedophilic preferences are sick, which makes it wrong to satisfy them.

Let us start by considering a number of arguments in favor of the view that pedophilia is a sickness or disease.

One argument appeals to correlations between pedophilia on the one hand and psychological traits such as poor social skills, introversion, and low IQ, on the other (Seto 2009). Though several such findings have been made, and though they might perhaps provide reasons for believing that pedophilia is a mental disease, it is doubtful how much the findings really tell us about pedophiles. The reason why is that the pedophiles who have been studied are almost exclusively pedophiles whose sexual preference was revealed because they got caught for engaging in adult-child sex.

This, moreover, skews the sample in a very problematic manner.

First, those pedophiles in the subset of individuals who engage in adult-child sex are presumably, on average, less able to curb their desires, less concerned with adhering to social norms, and more willing to take risks than pedophiles who never engage in adult-child sex – and only the former are included in the sample.

Second, pedophiles who are studied are not a representative sample of the pedophiles who have engaged in adult-child sex; the sample is restricted to those who got caught. It should not be surprising, however, that those who first engaged in adult-child sex and then were unable to hide their acts have a lower IQ and poorer social skills on average.

In studies that have examined a more representative sample, the only significant psychological correlate to pedophilia is introversion (Howitt 1995a). Glenn Wilson and David Cox conclude that

  • “…the most striking thing about these results is how normal the paedophiles appear to be according to their scores on these major personality dimensions — particularly the two that are clinically relevant [neuroticism and psychoticism]. . . . introversion . . . in itself is not usually thought of as pathological.” (Wilson & Cox 1983: 57)

Though it might be interesting to explore the explanation for the correlation with introversion, the correlation argument for why pedophilia is a mental disease is not, by itself, convincing.

There are, however, arguments as to why pedophilia is a mental disease that do not rely on correlations between pedophilia and other psychological problems, but rather, on pedophilia as such and the ways in which pedophilia diverges from normal sexuality. This view assumes there is something pathological about the very state of being sexually attracted to children. This suggestion cannot be rejected out of hand, for presumably, not all arguments for why something is a mental disease can rest on correlations with further mental diseases.

It is doubtful, however, if an appeal to the nature of pedophilia makes us justified in concluding that it is a disease. A central reason for this is that pedophilia diverges less from normal sexuality than what we often assume.

One reason is that, at least in one sense of the term “pedophile,” many of us have been pedophiles at one point. When you were 11, it is not unlikely that you were sexually attracted to prepubertal children. Admittedly, when we discuss pedophilia in psychiatric and legal contexts, we are concerned with adult sexual attraction toward children. That, however, does not change the fact that the mental state of finding children sexually attractive is very common, and is indeed part of many fully healthy, albeit younger, psychologies.

A second reason is that children look much like adults. Though the difference between an 11-year-old and an 18-year-old is significant, the development from childhood to adulthood in humans is hardly equivalent to that from, say, a larva to a butterfly. The skin, the skeleton, the bodily contours, the way the body moves — even the genitals — are fairly similar in children and adults.

A third reason is that youngness, the property which children have in excess, is a property that, when present to a lesser extent, is widely considered sexually attractive. As such, what makes pedophiles diverge from others is that they prefer more youngness than does the average person. This emphasizes that a main way in which pedophiles differ from non-pedophiles is quantitative.

A fourth reason is that virtually all adults are drawn to be physically intimate with children: adults kiss them, cuddle them, find them cute, and develop fond emotions for them. Sexual attraction, moreover, is one of many forms of positive affection, and since human psychologies are associative, it is not surprising that some adults are also sexually attracted to children.

A fifth reason is that the ways in which pedophilia does in fact diverge from normal sexuality must be seen against the backdrop of human sexuality in general, which includes many non-reproductive sexual practices and expressions (such as homosexuality, masturbation, oral sex, attraction toward postmenopausal women, etc.). Within the wide spectrum of human sexualities, it is not clear that pedophilia stands out as exceptionally divergent.

For pedophilia to be a mental disease, it seems that it must be so in

  • either a pragmatic 
  • or an evolutionary sense.

In a pragmatic sense, the view that pedophilia is a mental disease is true if and only if good consequences follow if pedophiles are given a diagnosis and subsequent psychiatric treatment. Although pedophilia might qualify as a disease in this sense, this tells us nothing about pedophilia as such or about the moral status of any pedophilic expressions.

In an evolutionary sense, the view that pedophilia is a mental disease is true if and only if pedophilia is reproductively disadvantageous. Reproductively disadvantageous as pedophilia might be, this is a problematic criterion for disease, since from an evolutionary perspective, pedophilia is no worse than the examples listed above: homosexuality, masturbation, oral sex, and attraction toward postmenopausal women.

Even if we were to grant that pedophilia really is a disease, however, we would still not have a complete argument against the enjoyment of texts and computer-generated graphics with pedophilic content, for it would also need to be true that it is wrong to satisfy preferences that result from a disease.

This is a doubtful premise, however, for even if we grant that it is bad to have a disease, once we have it and cannot get rid of it, it seems to be an open question whether or not we should give in to the disease in ways that do not cause harm.

Admittedly, doing so might be wrong if it intensifies the disease, but we lack evidence that pedophiles become more pedophilic by virtue of enjoying texts or computer-generated graphics with pedophilic content. No research indicates this, and when it comes to other sexual preferences, such as homo- and heterosexuality, we seem to be accustomed to the idea that they are largely stable and that exposure to homo- or heterosexual pornography cannot be expected to change anyone’s orientation.

As such, we lack good reasons for why it is bad that pedophiles satisfy their preferences by means of texts and computer-generated graphics.

In the absence of convincing arguments to the contrary, we should conclude that it can be morally permissible for pedophiles to enjoy fictional stories and computer-generated graphics with pedophilic content.

It is important to note, however, that even if we accept this conclusion, it does not follow that we must be comfortable with such practices. Given that pedophilia predisposes people to seek adult-child sex, and that adult-child sex exposes children to significant risks of serious harm, it is understandable that most of us strongly dislike it.

Moreover, it does not follow from the above that it is irrational to be worried if one discovers that someone enjoys fictional stories or computer-generated graphics with pedophilic content. One might have very good reason to worry since this is a strong indication that the person is a pedophile, and pedophilia predisposes people to seek adult-child sex.

Observe, however, that what is troubling here is the discovery of the person’s sexual preferences via his enjoyment of virtual pedophilia, not his enjoyment of virtual pedophilia as such. Once he has those preferences and cannot change them, he might do nothing wrong in enjoying such fictional stories and computer-generated graphics. Indeed, doing so might be one of the best things he can do granted the unfortunate circumstances in which he finds himself.


If my arguments in this article are sound, then being a pedophile — in the sense of having a sexual preference for children — is neither moral nor immoral. Engagement in adult-child sex is immoral, but perhaps not always blameworthy to the extent that we intuitively assume.

Finally, the enjoyment of fictional stories and computer-generated graphics with pedophilic content is, in and of itself, morally acceptable. If these conclusions are correct, what practical implications follow?

A central implication is that in dealing with pedophilia, our aim should not be to find outlets for our disgust and outrage, but rather, to minimize what is the real problem: harm to children. On the least revisionist side, the aim of reducing harm provides us with a good justification for upholding current bans on adult-child sex and child pornography.

There are, however, also a number of more revisionist implications.

One revisionist implication is that we should stop the outright condemnation of pedophiles. Condemning pedophiles for being pedophiles is unjust, and non-offending pedophiles, rather than deserving condemnation for their pedophilia, deserve praise for their admirable willpower.

Possibly, today’s condemnation also prevents pedophiles from telling health professionals about their attraction to children, and insofar as detection and counseling can help prevent abuse, this is very unfortunate.

[Second,] To prevent harm to future children, we would also be well advised to start teaching high school students not just what to do in case they are victims of sexual abuse (which, thankfully, we have started telling them over the last few decades), but also what to do in case they themselves are pedophiles. A certain percentage of high school students either are or will become pedophiles, and currently they are not given any advice on how to handle their sexuality.

[Third,] The production, distribution, and enjoyment of texts and computer-generated graphics with pedophilic content should almost certainly be made legal. Until or unless it can be shown that such texts and graphics lead to more adult-child sex, the justification for today’s widespread ban is weak.

[Fourth,] Of course, a policy plan for dealing with pedophilia requires interdisciplinary work far beyond the scope of this article. Nevertheless, I have sketched how pedophilia should be approached under such a plan. The aim of our social and legal treatment of pedophilia should be to minimize harm to children, and to the extent that current practices are counterproductive in this respect, we should change our ways.

  • [* End note:] There is now a network of non-offending pedophiles called Virtuous Pedophiles. None of its members dare be open about their identities.