Hacker News Reader

There is a categorical difference with your examples for "phantom increase due to reporting". On one side it's something done to you/someone, in particular a serious crime, on the other side it's something about you/someone, in particular not a crime.

This is important, because the risk of getting caught, not necessarily severity of punishment, is the main factor in crime deterrence, as far as I know. I presume, increased vigilance and ease/encouragement of reporting has a significant deterring effect on potential perpetrators. As with all sexual abuse, the vast majority of cases happens in settings where the victim knows/trusts the abuser beforehand, therefore conviction entirely depends on the victim's willingness, prospect and ability to name/accuse the abuser. Coming with the reporting, there is probably also greater awareness and knowledge about the manipulation tactics and situations enabling abuse, helping guardians, friends and authorities to identify concerning dynamics and encounters.

However, I presume, the increased ease of reporting and awareness also indirectly prevents further abuse by subsequently encouraging victims to seek help and work with their trauma. I've heard a significant share of perpetrators have been victims of abuse themselves in the past, so seeking therapy after being victimized may also break the cycle of violence.

Indeed, people like to label anyone who is attracted to people under the age of consent as "pedophiles" but the definition is as you point out.

https://en.m.wikipedia.org/wiki/Chronophilia

I would guess that pedos today are far less likely to commit sexual abuse given that the chances of being caught are high and the repercussions are extreme. In the past this was much less the case.

Perhaps? With no evidence, it strikes me as a crime that is not weighed in ROI terms; it’s people with a sick compulsion, and I have always believed that kind of thing isn’t affected by the severity of possible punishment. I’d be interested in data for sure.

With no evidence the proper assumption would be no assumption.

But evidence was presented child sexual abuse declined since 1990. The question was how to explain it.

Punishment severity doesn't much reduce child sexual abuse probably. But not because it's a sick compulsion. Punishment severity doesn't much reduce most crimes. Financial crimes included.

They mentioned chances of being caught also. Punishment likelihood is an effective deterrent generally.

Pornography could be a factor. Many studies found rape decreased in countries when pornography became more available.

You're way off but I don't feel like taking the opposite side of this argument. If you want to research this there's plenty of data and information available.

I'd certainly also like to see a source for the basic claim. It would be wonderful news if true.

Why don't you Google it? This is fundamental basic knowledge about CSA incidence.

If it's so trivial, you should find the claim no harder to substantiate than to insist upon as here.

I encourage the attempt. Seriously. Go show me how easy reliable data is to find on this. Prove me wrong and you'll have earned a friend for life, or at least a dinner at the nice place down the street the next time you're in town.

OK... but there really are so many results upon literally just searching it. You immediately get multiple prominent results. You'd think if people actually cared about substantiating this, they could do one (1) Google search to find out. But here you go, courtesy of lazyweb.

https://files.calio.org/BIBS/declining-rates-bib4.pdf

https://www.stopitnow.org/sites/default/files//prevalence-ra...

https://www.unh.edu/ccrc/resource/decline-sexual-abuse-cases

(Of course the answer is that people don't care. They'd be happy to ignore the lack of sources if the conclusion being drawn agreed to them, but can't do a single fucking Google search if it doesn't.)

I'd be interested to know what search query you actually used. It's also been a while since I compared DDG and Google result quality. Maybe I'm getting poorer results than in the past. Queries on the theme of "child sexual abuse statistics US by year" don't get me anything very useful from DDG, much more advocacy around headline prevalence figures than anything relevant to incidence. (Statista does reflect a drop, but I didn't want to make an account to see their sources. Fair, my fault, I can't evaluate that well.)

The bibliography is interesting and useful but not immediately evaluable, and seems mainly Canada-focused. The Stop It Now/D2L data is very clear up front that it relates to prevalence and not incidence, a distinction that matters for reasons to which we'll return in a moment. The UNH data substantiates some level of decrease since the early 90s, as does every other source, but disagrees with itself on magnitude in different places and comes out of one researcher's shop, which does not invalidate it but does mean I'd like to see someone else produce something similar before I weight their novel results too heavily.

I guess maybe we have differing definitions of "reliable," but I think you and I have clashed on this subject before. I don't blame anyone for assuming that a victim of such abuse would tend to overestimate its prevalence, and I try to watch out for the tendency in myself, but what's all the anger about?

The difference between prevalence and incidence matters because the former tells us how many people have been abused as children, and the latter tells us how frequently children are abused over time, which is the thing we're actually talking about. How we count multiple instances of abuse against the same person confounds everything, and intuitively I would expect prevalence by victim (ie for a given victim how frequently they themselves were abused) to follow something like a power-law distribution, but in general if we just take the headline number of "X total kids" it's hard to see how that gets us to a nuanced understanding of incidence rates that lets us talk knowledgeably about how and to what extent the shape of this problem is changing over time.

So when you said:

> I'd certainly also like to see a source for the basic claim. It would be wonderful news if true.

You actually already knew about all of these sources and have debated this before, and decided now to start talking about this bit of nuance?

I'm not going to play. Sorry.

I'm familiar with the subject through long study. The sources you identified were not previously known to me but I feel confident evaluating them as I have based on prior knowledge and review. Do you mean to argue whose web search skills are the best - a point I already spent a paragraph conceding - or do you mean to argue the subject actually under discussion?

I still don't understand where the anger is coming from. Why does it piss you off that I want to argue child sexual abuse is more common and more of a problem than it's broadly acknowledged to be? Are you assuming I mean to use that claim as an argument against porn or something? I really can't figure out what's gotten so far up your nose on this, and I wish I could because it's really making it impossible for us to have a conversation.

Last I recall, we spoke cordially enough last year, if not generally in agreement, on the subject of AI capable of CSAM generation, and how it might be regulated. That was coincidentally a very strange day for me, granted, but again, I struggle to see what's changed since to earn this sort of acrimony.

With whom do you understand yourself to be needing to have this fight right now?

> I still don't understand where the anger is coming from. Why does it piss you off that I want to argue child sexual abuse is more common and more of a problem than it's broadly acknowledged to be? Are you assuming I mean to use that claim as an argument against porn or something? I really can't figure out what's gotten so far up your nose on this, and I wish I could because it's really making it impossible for us to have a conversation.

It's pretty simple honestly, any online discussion about pornography has turned into a complete and total clown show and it has become increasingly obvious that people who are arguing are doing so in bad faith.

I don't know what previous conversations we've had, as I spend far too much time on this damned website and have probably made tens of thousands of comments, but I kind of had a bad feeling the moment I saw your first comment because I had a pretty good feeling you knew what sources people were referring to and decided to play it as if you didn't, for one reason or another. I can see strategically why one would do this when the data seems to contradict the conclusion you have drawn so unilaterally, but at the same time, it's an annoying strategy. It's like laying a trap.

I really don't like this sort of discussion thread:

> Oh, it would be great news if incidence of CSA has gone down. Care to link to a source?

> Sure, here: [d2l link]

> Ah, but, you see, that talks about prevalence, not incidence. [...]

Because you knew what direction it was going to go all along. Playing coy here feels like theatrics.

If I took your position, I'd probably go more along the lines of:

> I am aware of the abundance of studies that discuss the prevelance of CSA, but [...]

(And I'm sure people would probably downvote that, but also, I get downvoted any time I respond in one of these types of threads, so it wouldn't make any difference to me.)

Anyway, the truth is that when I saw your first post, I didn't know what type of trap was being set, but my expectations were pretty low. The actual nuance you brought up is more interesting than that, but still not something I really care to debate.

I'm completely lost.

You raise sources I've never seen, I evaluate them as fairly as I can, you don't like the result so I'm arguing in bad faith.

I mentioned incidence, one source raised to try to counter not only relates to prevalence rather than incidence but explains why this matters, but when I point out that a source you chose itself disclaims support for how you want to use it, I'm arguing in bad faith.

I still don't even know what kind of pernicious dishonesty I'm meant to be advancing here, but sure, if you like then apparently I'm setting some kind of trap, in service of good God, I still don't know what.

I don't know who you imagine yourself to be at war with here, but whoever you're talking to clearly isn't me, so...I don't know, good luck with the rest of your day, I guess. You're not wrong we both have better ways to spend our time, at least.

I can explain a lot of the disconnect: I simply don't believe you.

For one thing, because the decrease in CSA over time is so well-known that I am pretty sure one of the PDFs I have saved refers to it as having been "common knowledge" in the 2000s. I have a hard time understanding how someone could apparently have studied this topic in detail and not come across this.

For another, I posted 3 sources and you 'evaluated' them within 15 minutes of me posting them, and of course found issues with all of them.

Yes, I can see that you are replying cordially, but that doesn't necessarily mean you're being forthcoming, either. I've grown extremely tired of this type of discourse.

Currently it's 2025 - in the early 2000s when most of the sources you linked come from there wasn't even Facebook let alone pornhub. The ones reporting on more recent times seen to have consensus that factors in reporting, especially covid, make it difficult to come to any conclusion. So a fair point of discussion that seems to be a far cry from "common knowledge". I don't know what you're aiming for but it's hard to take it in good faith really.

My basic search turned up:

https://www.unh.edu/ccrc/trends-child-victimization

Scroll down to the "Updated Trends" link and click it for more up-to-date info.

In the latest one (looks like it covers 2023, released last month), child sexual abuse is down 67% since 1990. (Also, physical abuse is down 63% and neglect down 27%.) Child sex abuse down 63% from the '90s.

Since you turned up one of the same sources as my other interlocutor, I'll refer you to my reply there at https://news.ycombinator.com/item?id=43593671