Those of us who value free speech as a paramount value that is essential to liberty tend to be skeptical of privacy laws. Privacy is often in tension with speech rights because protecting one person’s private information can mean preventing another from speaking the true things they know. In many cases, and especially when it comes to commercial speech online, the harm that privacy laws aim to mitigate is an amorphous feeling of “creepiness.”

When it comes to revenge porn, however, the harm is very real.

As Mitchell Matorin, an attorney who has represented involuntary porn victims, wrote recently in Talking Points Memo, free speech advocates can give short shrift to the plight of people who find their lives turned upside down when their most intimate photos are posted online along with their names, phone number, email address, and social media profiles. It’s not merely an inconvenience for the victims–mostly women–who took erotic photos (as adults are allowed to do) only to find themselves hacked or betrayed by an ex-lover.

Because photos are posted along with the victim’s name and contact details, they may soon find their way into her father’s email inbox. Potential employers or dates who search for her name online find the photos among the top results. Strangers approach her on the street to make lascivious remarks. Imagine if this happened to your sister or daughter.

Free speech absolutism can seem naive in the face of revenge porn’s very real consequences. As a result, anyone who dares question the wisdom of criminalizing speech is quickly accused of “blaming the victim.

“If you consider this ‘speech’ and you prize ‘free speech’ above all else, be honest and admit that your only concern is an abstract ideal and the victims are just collateral damage.” wrote Matorin. “Don’t pretend that they already have sufficient remedies.”

The sad fact, however, is that no reasonable remedy may ever be sufficient. Pointing out that reality is not blaming the victim.

Through civil lawsuits for invasion of privacy and infliction of emotional distress, among other torts, victims do indeed have recourse against those who posted their images online. These causes of action may not seem sufficient, however, because it may be impossible to identify the person who posted the images, or the poster may be judgment proof. More importantly, even if a victim obtains a judgment against her tormentor, the photos will likely remain on the Internet, copied by countless others to dozens of sites.

This month California enacted a law making it a crime to post intimate photos of another with an intent to harass. Both campaigners for stricter privacy protections and free speech advocates criticized it as ineffectual. The former on the grounds that the law only applies if the poster was also the photographer, the latter arguing that it unnecessarily encroaches on the First Amendment without making victims any better off than they would be employing civil suits. As satisfying as revenge against the perpetrators of revenge porn may be, incarcerating the perpetrator is not going to erase the victim’s photos from the Internet.

So how does a victim go about having her photos removed from the Internet? The answer is that it’s almost impossible.

Section 230 of the Communications Decency Act grants immunity to websites for content posted by their users. If user generated content posted to a site does not violate copyright or a federal criminal laws, then the site is not liable as the speaker might be, nor does it have an obligation to take the content down. Believe it or not, this simple law is the font of our amazing modern Internet. Without it, it’s not likely we would have seen the explosion in content and technical innovation the last decade has brought.

Imagine if YouTube were responsible for the content of the videos that its users uploaded? Or if Reddit was responsible for user’s posts and photos? Or if Reason was responsible for the speech in the comments section of articles? It’s likely that without Section 230, these sites would either not exist, or would not take the risk of carrying user generated content, and we’d be much the poorer for it.

In Europe, where intermediary protection is not as strong, company executives have been found liable for the libelous or harassing content that users have posted. This may have something to do with the fact that almost all user generated content innovation–from Facebook to Tumblr to Vine–happens in the U.S.

Unfortunately, Section 230’s intermediary immunity also applies to sites that do nothing but solicit and host revenge porn. As a result, some argue for an exemption to deal with revenge porn. We have to ask ourselves, though, whether such an exemption could ever provide the sufficient remedy victims seek, and what we would be giving up by tinkering with Section 230.

The good news is that even without an exemption, revenge porn sites do not tend to fare well. Infamous sites like IsAnyoneUp, Texxxan, and PinkMeth were all driven out of business shortly after they launched under pressure from victims, media, the public, and vendors that refuse to service the sites. The bad news is that the photos they hosted might now be found elsewhere on the web, there’s no exemption that will eradicate them all, and it is not treating victims as “collateral damage” to face that fact.

To the extent Section 230 should be changed to deal with revenge porn, an appropriate model might be the law’s existing exemption for copyrighted content. Known as “notice and takedown,” it allows copyright holders to give notice to websites that a user has posted copyrighted content without authorization and the website must take it down immediately in order to retain its immunity. The site must also notify the poster and allow them to appeal the decision. Such a system would help balance victim’s interests with important free speech rights and the intermediary immunity that makes today’s Internet possible.

Controlling the spread of information once it’s released on the Internet is practically impossible. This doesn’t mean we shouldn’t try when the information is harming real people, but it does mean that we should factor in the potential futility of our efforts as we consider how much we want to alter our simple, and indeed sweeping, free speech and intermediary liability rules that have served us so well.