Censoring US imagery: Is there any point?

On May 8, AP’s Katherine Shrader had an interesting interview with Vice Admiral Robert Murrett, director of the US National Geospatial-Intelligence Agency (NGA) that got picked up by a number of newspapers. Murrett seems to be indicating he’d like a shift in US policy when it comes to the availability of satellite imagery in the public domain. It is well worth reading in its entirety, if you haven’t already.

That article, in turn, inspired a piece in the San Francisco Chronicle on May 18 (“Top secret, in plain view / Google Earth may blur the image, but others don’t“) that looks at the feasibility of efforts to censor imagery, and the patchwork results that localized censorship attempts produce. Nut graf:

This patchwork censorship raises doubts about efforts to protect some of the nation’s key facilities, a matter of high urgency after the terrorist attacks of Sept. 11, 2001. It also raises the question of whether obscuring aerial views serves any purpose, given that overhead photographs of many important installations are already widely available elsewhere online. And it shows the futility of attempts to control Internet information in the digital, online age.

And to underline that point:

Surprisingly, some of California’s most sensitive locations, including the Diablo Canyon nuclear power plant or Lawrence Livermore National Laboratory in the Bay Area, are not blacked out or blurred on any of the mapping sites.

What neither article really manages to drive home, however, is that the reason so many sites in the US are uncensored is because most security experts agree that the publicly available high resolution imagery is not a threat. They understand that such imagery is not current, not a replacement for an actual scoping of a place, and far more likely to do good than harm. And let’s not forget that there are always other countries’ satellite services available if US sources are censored.

There actually exists a very liberal US policy on disseminating such imagery (direct link to PDF), prepared by the USGS, though neither article mentions it. The policy paper implicitly berates self-styled “experts” at individual government organizations for taking ad hoc and paranoid positions on censorship.

The risk now, of course, is that Murrett intends to replace an enlightened policy with something more paranoid, in keeping with the current US administration’s wider obsession with secrecy. But that just leaves me wondering: It’s Murrett’s job to run an agency that analyses satellite and aerial imagery and hands the results to the Pentagon. Where in that job description is there a formal policy-making role?

I bet he doesn’t have one, though he certainly has access to a tool for making a de facto censorship policy, and it has even a name, according to Shrader: “Checkbook shutter control”:

During the 2001 invasion to overthrow Afghanistan’s Taliban regime, the geospatial intelligence agency bought up all the imagery over that country for several months, creating a blackout for private groups at the height of the fighting. The agency was criticized for embarking on “checkbook shutter control” and hampering relief work and public understanding of the fight.

In the past, the agency has said publicly that it doesn’t plan to take such steps again. But Murrett, who took over last summer, clearly sees moments where such information may have to be restricted, especially to protect U.S. forces.

“I think we may need to have some control over things that are disseminated. I don’t know if that means buying up all the imagery or not. I think there are probably some other ways you could do it,” he said, leaving the specifics to legal and policy experts.

Yuck. That’s a good example of how you can circumvent the letter of laws that protect the public interest with tactics that completely demolish the spirit of those laws.

5 thoughts on “Censoring US imagery: Is there any point?”

  1. I’m still trying to think back to when it was – that in order to fight or locate terrorists, abroad or domestic, that it also meant we should give up our freedoms in this society. Funny how that’s all worked out.

    Then again, if I were in the business of selling to a government imagery redaction under contract – or could benefit greatly from such a contract, financially – then perhaps I wouldn’t complain about that loss of freedom. I suppose I’d just keep painting targets on imagery maps for terrorists to know exactly what’s sensitive, so they can go scope them out, for that all mighty dollar.

  2. I couldn’t help but reflect back on one of the comments made by Murret, as I was waking up this morning. Of course, one of those ‘things’ that just kind of pops into your head when you’re pouring a first cup of coffee.

    Murret states, “I could certainly foresee circumstances in which we would not want imagery to be openly disseminated of a sensitive site of any type, whether it is here or overseas,” he said. “This would include imagery on Web sites such as Google Earth, because the companies that supply the photos get help from the NGIA with launches.”

    This comment haunts me, because it sounds more like a veiled or underhanded threat more than anything. The NGIA may have some influence on launch authority, but ultimately, the NGIA cannot stop any commercial company from seeking out and gaining launch services in other parts of the world. Or can they?

    I’m curious what the point to making such a comment was attempting to acheive. It’s interesting to me that something of this nature could ever be said – unless of course I’m misconstruing what’s been said. It certainly raises an eyebrow, however.

  3. The article points out, correctly, the futility of censorship of images acquired, whether by aerial or spaceborne sensors. The Rand Corporation concluded in a 2004 study (http://www.rand.org/pubs/monographs/2004/RAND_MG142.pdf) that “although publicly accessible geospatial information has the potential to be generally helpful in selecting and locating a target, potential attackers, such as terrorists, are likely to need more reliable, more detailed, and more up-to-date information to plan and carry out a strike than is typically publicly accessible.” and less than 6% of Federal geodatasets contained any useful information to a terrorist group.

Comments are closed.