13 February 2017

Do child sex dolls count as child porn? Should they?

Canada and Australia now have both reported charging men with breaking child pornography laws for possession of child sex dolls (CBC News, The Independent). This has led to several discussions about whether such dolls constitute child pornography and what the relevant science says. To understand properly whether child sex dolls constitute child pornography depends on whether we mean within the worlds of science and mental health or under the law.

Scientifically, my colleagues and I have published research demonstrating that people who possess erotic or nude photographs of children are usually pedophilic. (At least, this is the case for people who have been apprehended and convicted of possession.)
Seto, M. C., Cantor, J. M., & Blanchard, R. (2006). Child pornography offenses are a valid diagnostic indicator of pedophilia. Journal of Abnormal Psychology, 115, 610–615.
doi: 10.1037/0021-843X.115.3.610
Because sex doll cases are rare, the equivalent research about that kind of possession has not been done. I would predict (and I think most sex researchers would predict) that possessing child sex dolls would be an effective indicator of pedophilia. So, within the world of psychological and psychiatric diagnosis, I would agree with Dr. Peter Collins, the prosecution expert in this case whom the CBC quoted as saying “In my professional opinion, the possession of a sex doll is just another form of depicting a child for a sexual purpose and therefore would meet the criteria for child pornography.”

Whether child sex dolls constitute child porn under the law, however, is the pertinent issue for the courts. Although I am a scientist rather than a policy expert, I believe I can point out that the purpose of child porn laws is to protect children from being harmed in the production of child porn (where the images are themselves evidence of abuse) and to prevent there being a market for child porn. With child sex dolls, however, there is no actual victim being protected, and it does not matter whether there is a market for latex dolls. Although it is very reasonable to conclude that the consumers purchasing sex dolls are genuinely attracted to children, there is no evidence to indicate that sex doll owners are at an elevated level of risk to go on to commit crimes that do involve a victim.

Different jurisdictions use different language in their anti-child porn laws, including different definitions of child porn. The language of such laws can usually be interpreted either way. So, in practice, whether dolls are covered by the laws generally meant for images of real people is up to the courts.

This invites, of course, the question of what one thinks the courts should find. This is now a pure value judgment rather than a scientific question at all. For what it is worth, my personal value judgment is on the side of free speech (however distasteful one might find that speech to be) until there is a compelling societal need for otherwise. Although many child molesters also possess child pornography, it does not appear to work the other way around: Of people who have been apprehended for child porn, very few are found to have molested any children. (In such cases of child porn, investigators attempt to find any children in the porn-possessor’s environment who might have been targeted.) That is, as best as can be found with the current science, there is no compelling need here.

Men who are attracted to children have no healthy outlet for their sex drives, and it will be that way for their entire lives. For someone who did not ask to be pedophilic and who cannot have romantic and sexual relationships like the rest of us, the only outlet they have is masturbation. For society to hamper even that, to me, is cruel when it is in the absence of a demonstrated societal benefit.


  1. I honor this courageous post, James. The only thing I would change is the comment; "Of people who have been apprehended for child porn, very few are found to have molested any children." My understanding from Michael's work is that 50/50 is more like it, which to me, is more than "a few". I would point out that a toss of the coin would still be WAY lower than evidence beyond a reasonable doubt, SO, your points herein would be valid, and you'd avoid providing food to those that want to avoid your core message - which, IMO, is spot on!

  2. Thanks! I'm not sure which work you're referring to, however. Do you have a specific article in mind?

    [For other readers: The above refers to Dr. Michael Seto. He is a top notch researcher, and I strongly recommend following him! https://www.researchgate.net/profile/Michael_Seto]

    1. I meant various articles that Michael has published, wherein he sites (and in personal communications w him, has said) data that supports a conclusion that there is about a 50/50 chance that child porn offenders have contact offenses in their history, thus, supporting your statement that "Although many child molesters also possess child pornography, it does not appear to work the other way around".

  3. The number of identified incidents of child sexual abuse decreased at least 47% from 1993 to 2005-2006. This coincides exactly with an even larger increase in the general availability of child pornography over that same time period.

    This isn’t an argument that child pornography is good, however it perhaps supports the evidence (linked much earlier in this blog) that it at least diverts potential hands-on sex offenders into what can be less harmful ways of managing their sexual orientation.

    And as you can make harm reduction arguments for hard drug abusers, so you can make similar arguments for those who are on the borderline of becoming child sex abusers (though there are many pedophiles or hebephiles, myself included, who manage fine without child porn).

    If there is anything other than real child porn that can divert potential offenders from offending (those for whom “adult women” doesn’t cut it), ought we not to do everything we could to make these means available to those who need then?

    Yet people still want to make sex dolls illegal.

    Is keeping these people in the shadows, assuming that their ultimate offending is inevitable anyway, and routinely threatening them with death a better option? You tell me, but I have more faith in human nature than that. Even the human nature of pedophiles.

  4. I cannot find any data on non-offending pedophiles. Has no one done such research? Seems to me that if heart disease is being studied persons with structural heart irregularities who *don't* have myocardial infarctions are as important to be studied as those who do - in order to determine differences, if any. And regarding data on offending pedophiles, which study on utility of various therapies in preventing repeat offences is the best in reliability and validity? I am skeptical of any studies that do not have large samples, are not long-term, etc.

  5. Such a very useful article. Very interesting to read this article.I would like to thank you for the efforts you had made for writing this awesome article. flypoint glattbrugg

  6. The only thing I disagree with in this article is the assertion that the possession of Child Porn as an indicator of pedophilia. To me this is more correlation than anything else. yes, one can be a pedophile and possess child porn, but one can also possess child porn and not be a pedophile. There are 2 key points here...

    1. Those that will fap to anything
    2. Those that are actual children

    particularly those that are literal children being found to possess "child pornography" often of themselves or of their significant other or ex. and those that will literally fap to anything and have a vast variety of stuff in their possession, these 2 categories, might have "child porn" in their possession but that doesn't make them pedophiles. The same can be said for those that almost exclusively have an intrest in lolicon, but have an active disdain for any real children. In this case, they could be attracted to the style not the age or the "perceived" age of the drawing.

    though i'm also not a psychologist, or any type of professional in the field, and would assume that an "indicator" doesn't necessarily mean that someone is. I would assume that more evidence would be needed to make that determination. but based off what was said, in this article, I do feel that this point needs more information as the above examples would state.

  7. Correlates ARE indicators. When given characteristic A, the size of the correlation indicates the probability of also having characteristic B. Indicators are not guarantees, however. It is indeed true that there exist each of the combinations of pedophilie/nonpedophile and porn possesser/nonpossessor. What I am point out (and what the correlations are pointing out) is that these four squares do not have equal 1/4 shares of the population. The proportion of child porn possessors who are nonpedophiles is relatively few---few enough that it child porn possession is a meaningful indicator. It is true also that kids sending each other pics of themselves do not constitute child porn in the way that indicates pedophilia (or a crime).

  8. New Zealand has similar laws now (no doubt influenced by Australia) and the first conviction for importing a sex doll happened not that long ago [https://www.stuff.co.nz/timaru-herald/news/110451089/timaru-man-admits-importing-child-sex-doll-a-first-in-new-zealand]