A Conundrum

Earlier today, someone directed me to this article on the online harassment of women. I've been pondering the problem today, and it's a very difficult one.

I'm going to take the author at her word about the scale of the problem. I don't actually know that she's right about it, and as she apparently writes a column about sex, it may be that there is a lightning-rod effect in terms of drawing sexually-aggressive responses. On the other hand, she cites some evidence that backs up her position that this is a problem on the kind of very large scale she's describing. So, for the purpose of this discussion, I'm just going to assume she's completely right about the facts, and consider what might be done about it.

She has several implied measures that she thinks would improve things, all of which prove to be problematic on even a moment's consideration:

1) Police action. The problems here are twofold:

(a) The very scale of the problem defies policing as a workable response. Even with the right tools (see the next point), it would take hours to chase down a positive identification on an anonymous comment left on a blog, or one of these "tweets." You'd have to contact the ISP or online service, get the data, and then do the work of tracking it back to the specific IP address. Then, you'd have to do the work necessary to prove that the individual you're planning to arrest is the one guy who was using that anonymous account at that specific IP address at that moment.

This is all very workable if the problem we're talking about is, say, terrorism. The incidents of terrorism are rare enough that you can run each one to ground. But she's talking about something that, according to her report, happens millions of times a day. If every cop in America did nothing else, the very scale of the problem puts it outside their power to solve. You couldn't even prosecute enough of a percentage to make an impact, so the prosecutions would have to serve as meaningful symbols. But given the difficulty of proving that the IP address ties to a specific person beyond a reasonable doubt, as well as given the possibility of 1st Amendment defenses ("She misunderstood: that was intended as parody, which is protected free speech")... well, you could easily end up losing your meaningful symbolic prosecutions, sending exactly the opposite message intended.

There's a problem with symbolic prosecutions anyway, but if you're going to make an example, it has to work.

(b) As she is herself aware, many of the tools that the police would need to address these issues effectively are the very tools that people are objecting to the NSA leveraging. Now presumably there would be less problem with police doing it, in an open environment of due process and subpoenas. Still, there is a legitimate counterbalancing interest in limiting the government's ability to do what the police would have to be able to do to be as effective as they could be. It may not be the case that there is the political will, or trust in the state, to hand over the powers they would require.

This compounds the difficulty of bringing off effective symbolic prosecutions. You can't afford to lose, because the symbol is all you've got, but you may be denied some of the evidence you require by privacy advocates (and may encounter a jury hostile to police snooping on internet activity, who could therefore find the 1A defenses more palatable than one moved chiefly by outrage at the things said to the women).

2) More female police. There are two problems with this, too:

(a) There's no draft for police. Women aren't choosing to be police officers in greater numbers because that's not what they want to do with their lives. You could make a case that women have a duty to do this, but unless women are persuaded by that case, you certainly can't make them.

(b) Even in the case of the FBI, which works extremely hard to recruit as many women agents as possible, her own evidence suggests that the very high percentage of women (19%) has not led to institutional changes making prosecutions more likely. This may be because the ratio of hot air to serious threats has proven to be so low. Again, we're talking about apparently millions of offenses a day; the actual number of these that turn into physical stalkers or attackers is so much lower that the FBI may be acting rationally in focusing its efforts elsewhere. Compared to their counterintelligence mission, for example, time invested here is much less likely to uncover and stop a serious threat; and if it does, it's a threat to one person, whereas a counterintelligence risk could threaten very many.

3) More female game designers and software engineers. The problem here is the same as 2a: "While the number of women working across the sciences is generally increasing, the percentage of women working in computer sciences peaked in 2000 and is now on the decline."

4) Get offline. This is a solution for the individual, if they're willing to pay the price she talks about in great detail. It's not a solution for the society, unless we want the internet to be a public space like a Saudi shopping mall.

5) Enable software to block hateful messages. As she points out, this will greatly improve the experience for the woman, though it takes a constant effort. However, it doesn't actually do anything at all to deal with the one person who is really a danger.

6) Treat the whole internet as a Title VII area, which is subject to intense Federal scrutiny aimed at preventing harassment of women. This has all the problems of (1), especially because (as the author of the suggestion admits) the real intent is to pressure police into working harder on this problem. It's also not going to improve the underlying tension between the sexes to extend all the pleasures of the office or campus Equal Opportunity Department to all our private internet activity. If anything, I'd think this would increase the number of men inclined to hate and lash out at women.

7) Protect yourself. The author tried to use a protective order, and describes how difficult it was to obtain (and how overwhelmed the courts are anyway). She lives in California, so she can't carry a gun (and perhaps wouldn't anyway); but even if she had one, she would have to wait until she found herself in reasonable fear of death or serious bodily injury to use it. This could solve the most serious aspect of the problem -- an actual assault or rape attempt -- but only for those women who are willing to kill another person. That's not every woman, and that's not their fault.

I would have to say that the best workable solution is some combination of 5 and 7, combined with some efforts by everyone to make clear that this kind of behavior is not acceptable. Of course, aside from deleting comments here at the Hall, there's nothing I can do to actually enforce that. And, also of course, the whole pleasure of doing it is that it is offensive to people. Making clear that it's offensive and inappropriate isn't going to stop them, or even slow them down.

It's a difficult problem, at least for the non-DL-Sly's of the world. I suspect she's got this. But not everyone is like her!

UPDATE: Cathy Young attacks the premise I was granting in paragraph two. She's got a good argument in parts, though in the end I think she is monumentally unfair to the NYT's Douthat. I don't think I agree with his conclusion either -- we need not a new vision of masculinity, but a restoration of the old one -- but it ought to be clear that his position is far better than the kind of attacks that we're talking about here. His final lines are remarkable for their respect for the quality of women's influence, while asserting men's responsibility: "Forging this vision is a project for both sexes. Living up to it, and cleansing the Internet of the worst misogyny, is ultimately a task for men."

He may be wrong about that one-sided responsibility, too. But he's not the enemy of women: if anything, he's erring in the other direction.