Bloggers are up in arms over this weekend’s revelation that the NYPD conducted extensive surveillance of mostly-peaceful protest groups in the run-up to the 2004 Republican National Convention. (I was up there for Reason at the time and ended up covering a lot of the protests myself.)

It would be disturbing is if the city’s keeping dossiers on peaceful people’s political activities, and Lindsay Beyerstein is certainly right that the city shouldn’t be able to block the release of records from that surveillance on the grounds that some of us might find it newsworthy. But I also don’t think it’s hugely problematic in principle for police to be attending public meetings or following public message boards to get some sense of how a major event like this is likely to unfold.

I can, however, think of at least one way in which this sort of practice might be counterproductive from the police’s own perspective. Even completely peaceful protest groups will sometimes be planning acts of civil disobedience that are essentially harmless but technically illegal. If you create the expectation that these groups are invariably going to be monitored or infiltrated, their natural response is going to be to adopt a more closed structure, limit membership to known, committed activists, cultivate a stronger us-versus-them mentality, and so on. And it’s a well established phenomenon that if you gather a group of like-minded people and close them off from moderating external feedback, they will over time become more extreme than the median group member was at the outset. Encouraging that tendency seems likely to promote lawless or violent forms of protest, not preempt them.