Sexual attraction to children and sexual offending against children are empirically distinguishable constructs, not two points on one continuum defined by inevitability. This note keeps that distinction — attraction as a private psychological fact, offending as conduct — central throughout, and treats terminology, prevalence, help-seeking risk, and offending outcomes as four separate questions with separate evidence.
Attraction and offending are separable, not sequential
Pedophilia, as a pattern of sexual attraction assessed by phallometric or viewing-time measures rather than self-report, is a documented risk factor for offending, not a certainty of it. Neurobiological research (Schiffer et al., Translational Psychiatry, 2017) finds structural brain differences between men with pedophilic attraction who have offended and those who have not, independent evidence that these are separable populations rather than points on a single continuum. Prevalence research using self-report measures (Dombert, Schmidt, Banse, Briken, and Hoyer, Journal of Sex Research, 2016) found sexual interest in prepubescent children reported by 1.7 to 5 percent of community adult men, depending on measurement strictness; broader, less rigorously validated measures have found figures as high as 24 percent in some reviews, a wide range that itself signals measurement difficulty rather than a settled prevalence figure.
Germany’s Prevention Project Dunkelfeld (Charité — Universitätsmedizin Berlin, established 2005) is the most direct evidence on offending outcomes among help-seekers specifically. Its outpatient clinical network treats self-identifying, help-seeking pedophilic and hebephilic adults outside the legal system, aiming at sexual self-control rather than prosecution. A 2024 follow-up study (Journal of Prevention) tracked 56 self-referred participants for one to eleven years after treatment. Among those with no prior history of child sexual abuse, none committed a new such offense during follow-up; among those with a prior history, 7.7 percent reoffended, and CSAM-use recidivism was substantially higher, at 89.1 percent, though typically involving less severe material. This is genuine, peer-reviewed evidence that non-offending help-seekers who complete this specific clinical program do not go on to commit contact offenses, but it describes a treatment-completer cohort at one clinical network, not a random sample of all people with this attraction pattern, and it does not by itself establish that peer-support-community participation, as opposed to structured clinical treatment, produces the same result.
Terminology is contested, not settled clinical vocabulary
“Minor-attracted person” (MAP) originated in 1998, coined by journalist Heather Elizabeth Peterson, and was adopted from 2003 by B4U-ACT, a US peer-support and clinician-engagement nonprofit, as a term intended to avoid the criminal and pejorative associations of “pedophile” while also covering hebephilia and attractions not yet clearly specified. Some clinical and prevention researchers have since adopted it as descriptive nomenclature. “NOMAP” (non-offending minor-attracted person) and “virtuous pedophile” name overlapping but distinct positions. VirPed, a separate organization founded in 2012, deliberately retains the word “pedophile,” rejecting “MAP” as euphemistic, and defines “virtuous” narrowly as “not worse than average” — VirPed’s own history states the group split from B4U-ACT specifically because B4U-ACT would not “unequivocally state that adults should not have sexual contact with children,” evidence that this space is not ideologically uniform.
This adoption is genuinely contested in the scholarly literature, not merely disputed by outside critics. A 2024 rapid evidence review (Farmer, Salter, and Woodlock, Trauma, Violence, & Abuse) traces “MAP” to online advocacy communities and argues that academic use of the term imports, often without acknowledgment, a framing from those communities: that minor-attracted people constitute an oppressed sexual minority, and that stigma, rather than attraction itself, is the primary driver of offending risk. The review calls this framing, in its strong form, empirically unsupported and contrary to child-protection prerogatives. The same term is used descriptively by researchers publishing in Archives of Sexual Behavior and criticized as ideologically loaded by researchers publishing in a comparably ranked journal; this note treats “MAP” as contested nomenclature and attributes claims about its neutrality to the specific camp making them, rather than presenting either position as settled.
The organizational landscape is not interchangeable evidence
Four organizations recur in this literature, and their different institutional character matters for what any of their claims can support:
| Organization | Type | Role |
|---|---|---|
| B4U-ACT (US, 2003) | Peer-support and clinician-engagement nonprofit | Therapist referral network, peer forum, clinician education |
| VirPed (2012) | Peer-support advocacy group, no clinical arm | Destigmatizes the identity while foregrounding an anti-contact, anti-CSAM position |
| Prevention Project Dunkelfeld (Germany, 2005) | Clinical treatment network | Outpatient treatment for help-seekers outside the legal system |
| Stop It Now (UK, run by the Lucy Faithfull Foundation) | Child-protection charity with a harm-reduction helpline | Confidential helpline for people concerned about their own thoughts or behavior |
Dunkelfeld is a clinical institution embedded in a university hospital; Stop It Now is a child-protection-first helpline that happens to serve non-offenders; B4U-ACT and VirPed are peer or advocacy organizations without clinical infrastructure. None of these is interchangeable evidence for any of the others’ claims, and a claim’s institutional source should travel with the claim.
Help-seeking carries a documented, narrow disclosure risk
Mandatory-reporting obligations in the United States are generally triggered by a reasonably identifiable victim and a serious threat of harm, not by attraction alone; commentary in the AMA Journal of Ethics (2013) notes that sixteen US states use a discretionary standard and four have no duty-to-warn statute at all. But this legal ambiguity itself deters disclosure: Jahnke, Blagden, McPhail, and Antfolk (Psychotherapy Research, 2023) found that pedohebephilic clients often avoid disclosure specifically from a belief, not clearly supported by the law in most jurisdictions, that mandatory reporting applies to attraction alone, and that a supportive clinical response to disclosure correlated with better outcomes than either non-disclosure or an unsupportive reaction.
Stop It Now’s own published confidentiality policy states its helpline is confidential and anonymous, “but if they do [disclose identifying information] and then disclose details of a crime being committed or if there is a risk of harm to anyone, then we will pass that information on to the relevant agencies.” This is the clearest documented evidence that even a genuinely protective, non-punitive support channel carries a real, if narrow and conduct-triggered rather than attraction-triggered, reporting exposure.
Disclosure of attraction alone, absent any offense, has caused severe reputational harm through non-criminal channels. The 2020 Menlo Church case (reported by Religion News Service, independently assessed by the Zero Abuse Project) involved a volunteer who privately disclosed an unwanted attraction pattern to a pastor, with no disclosed offending conduct; the pastor’s own decision to let the volunteer continue working with children became a public scandal once revealed, and the disclosing individual was also publicly identified against his will. This is well-sourced, documented evidence that attraction disclosure alone can precipitate institutional investigation and public exposure, though the primary reputational fallout in this case fell on the pastor’s handling decision rather than on the disclosure itself.
Scope and reliability
The clinical and neurobiological findings above rest on peer-reviewed, primary research. The organizational and terminological material rests on the organizations’ own published statements and a contested scholarly literature this note attributes by camp rather than treating as consensus. No case was found combining a false criminal accusation with disclosed non-offending support-seeking in the specific configuration considered in Why a known harsher penalty can beat an uncertain lesser one; that combination remains a plausible, mechanism-grounded construction from independently documented risks, not an observed pattern.
Built on 3 sources (3 external).
Working out connections…
Working out the neighbourhood…
Model contributions
Measured by git-blame lines per AI model (231 total).
{"width": 320, "height": 320, "data": {"values": [{"model": "Claude Sonnet 5", "label": "Claude Sonnet 5 (98%)", "lines": 226, "share": 0.9783549783549783}, {"model": "Claude Opus 5", "label": "Claude Opus 5 (2%)", "lines": 5, "share": 0.021645021645021644}]}, "mark": {"type": "arc"}, "encoding": {"theta": {"field": "lines", "type": "quantitative"}, "color": {"field": "label", "type": "nominal", "legend": {"title": null, "orient": "right"}}, "tooltip": [{"field": "model", "type": "nominal"}, {"field": "lines", "type": "quantitative"}, {"field": "share", "type": "quantitative", "format": ".1%"}], "order": {"field": "lines", "type": "quantitative", "sort": "descending"}}}