Analysts for a Department of Homeland Security program that monitors social networks like Twitter and Facebook have been instructed to produce reports on policy debates related to the department, a newly disclosed manual shows.
The manual, a 2011 reference guide for analysts working with the department’s Media Monitoring Capability program, raises questions about recent claims by Homeland Security officials who portrayed the program as limited to gathering information that would help gain operational awareness about attacks, disasters or other emerging problems.
Last month, a previous disclosure of documents related to the program showed that in 2009, when it was being designed, officials contemplated having reports produced about “public reaction to major governmental proposals with homeland security implications.”
But the department said it never put that category into practice when the program began in 2010. Officials repeated that portrayal in testimony last week before an oversight hearing by a House Homeland Security subcommittee.
“I am not aware of any information we have gathered on government proposals,” testified Richard Chavez, the director of the office that oversees the National Operations Center, which runs the program.
Still, the 2011 manual, which was disclosed this week as part of a Freedom of Information Act lawsuit, lists a series of categories that constitute an “item of interest” warranting a report. One category is discussion on social media networks of “policy directives, debates and implementations related to DHS.”
It is not clear whether the department has produced such reports. Matthew Chandler, a department spokesman, said Wednesday that in practice the program had been limited to “social media monitoring for situational awareness only.”
He also said the department would review the reference guide and related materials to make sure they “clearly and accurately convey the parameters and intention of the program.”
Ginger McCall of the Electronic Privacy Information Center, an advocacy group that filed the lawsuit and obtained the document, argued that the manual shows that the monitoring may have gone beyond its limited portrayal by department officials.
“The D.H.S. continues to monitor the Internet for criticism of the government,” she said. “This suspicionless, overbroad monitoring quells legitimate First Amendment activity and exceeds the agency’s legal authority.”
A federal statute cited by officials last week as the legal basis for the program gives the National Operations Center the authority “to provide situational awareness” for officials “in the event of a natural disaster, act of terrorism or other man-made disaster” and to “ensure that critical terrorism and disaster-related information reaches government decision makers.”
Officials have stressed that the program does not collect personally identifying information, like the names or Twitter account handles of the people making comments, and that it does not monitor, review or collect First Amendment-protected speech.
Still, the program also monitors articles and broadcasts by traditional media outlets. The 2011 manual says that analysts, in addition to flagging information related to matters like terrorism and natural disasters, should also identify “media reports that reflect adversely on D.H.S. and response activities” and collect “both positive and negative reports” on department components as well organizations outside of the department.
The manual includes keywords that analysts should search for. A list of agencies in the keyword section includes not only those in the department dealing with matters like immigration and emergency management, but also the Central Intelligence Agency, several law enforcement agencies in the Justice Department, the Red Cross and the United Nations.
At the hearing last week, lawmakers of both parties said it made sense for the department to use the Internet to gather information about emerging events, but they voiced concerns that if it went further than that, the program might chill people’s freedom of speech and willingness to express dissent online.
“Other private individuals reading your Facebook status updates is different than the Department of Homeland Security reading them, analyzing them and possibly disseminating and collecting them for future purposes,” said the chairman of the subcommittee, Representative Patrick Meehan of Pennsylvania.
Mary Ellen Callahan, the department’s chief privacy director, testified that the program was interested only in events within the department’s mission — like disasters, attacks or continuing operational problems. As an example, she cited a situation in which people post to Twitter about an unusually long line at a particular airport checkpoint.
She also played down the use of keyword searches the program uses for articles and postings on social networks, portraying them as simply related to disasters — “you know, flood, tornado and things like that.”
The 2011 manual contains a fuller list. Many keywords are closely related to various disasters. But a handful are potentially more sweeping, like China, cops, hacking, illegal immigrants, Iran, Iraq, marijuana, organized crime, police, pork and radicals.