Defense Media Network

Government Secrecy and “Mosaic Theory”

DMN Q&A: University of Missouri Professor Charles Davis

Mosaic theory doesn’t ask government to prove very much to keep information secret, Charles Davis observes. Rather he says, “It asks government to tell a good story.”

Ours is largely an open society, but when it comes to secrecy Americans generally defer to their government. Information deemed potentially harmful to national security is, more often than not, just that. But a new study published by University of Missouri associate professor of Journalism Studies Charles Davis, finds that government secrecy claims based on “mosaic theory” are often too broad.

Mosaic theory is a rationale used to uphold the classification of information. The basic idea is individuals or states may be able to piece together bits of information from multiple unclassified but relevant documents to form an intelligence picture or “mosaic.” It is unknown when the mosaic idea was first employed but Davis says that a range of government agencies are increasingly using it to keep select information from public view and from Freedom of Information Act (FOIA) requests, a trend he finds disturbing.

A former director of the National Freedom of Information Coalition (a coalition of state Freedom of Information groups) and director of the Freedom of Information Center at the University of Missouri, Davis has studied secrecy and public information issues for 20 years.

Defense Media Network spoke with Davis shortly after his work was published in Continuum: Journal of Media & Cultural Studies in November 2011.


Your basic contention is that mosaic theory is so ambiguous that it can be used by government to justify non-disclosure of material deemed dangerous to national security even if the risk may be questionable?

Yes. Federal fears have to be tempered by the public’s right to know. If FOIA’s provisions for exemption are narrowed, then the public will retain its right to know, and the government will still be able to act in accordance with national security interests.

The public’s interest in knowing should be given a degree of importance in clashes between government secrecy and calls for transparency. Right now the government says “just trust us.” That doesn’t cut it.


When did mosaic theory emerge as a secrecy rationale?

It started even before Sept. 11, 2001. There were arguments being made by various agencies that one could parcel together seemingly disparate piles of information from systematic releases of information in a shrewd way and cause harm. But that basic argument, that disparate information can be [fused] to create harm, has been around for decades. It was after Sept. 11 that it began to take shape in the imagination. The idea that the mosaic theory could truly come to life took wing.


Though you’ve directed a Freedom of Information NGO you say you’re not an absolutist regarding information secrecy?

Yes, I know that there is information at the federal, state and local level that is dangerous, frankly, and would do our adversaries good. I come at this from the perspective of someone who believes that there is justification for secrecy.

That having been said, I have real problems with this doctrine. It’s a doctrine without limits, a moat without end. You can put just about anything around the rubric of mosaic theory by cooking up a worst-case scenario. If terrorist X gets hold of records A, C, and G something terrible is going to happen.


Do you worry that mosaic theory is being over-used for defense-related information?

From a defense perspective, I don’t think that’s an argument. It worries me more in other contexts. You can make a very plausible argument in the military industrial complex or national intelligence gathering areas that the mosaic theory is a real worry.

Once you do that and have given birth to this doctrine, however, what you see – and this is what our paper set out to establish – is [mosaic theory] beginning to leap over to agencies that have nothing to do with either defending the borders of the United States or gathering intelligence to keep us safe. The theory gets picked up by the EPA and Interior, places where I think it has absolutely no business.


So your concern with mosaic theory is the potential to use it to control information flow?

My unease with mosaic theory is not categorically the notion that there may be information that can be pieced together in a dangerous way. I do reject the notion that [the risk] is so commonplace that we need a categorical approach to it. When you accept mosaic theory doctrine you accept an unbelievable amount of deference to administrative agency positions for secrecy. That’s my fundamental problem with it. It doesn’t ask government to prove very much to keep information secret. It asks government to tell a good story. If government can produce a scary enough specter, then courts are swayed and the information is closed.


Could the defense establishment use mosaic theory as a sort of public relations management tool? Often the military has been accused of shaping the message.

That’s right. You’re already in a context – military and national intelligence gathering – where information access is incredibly deferential. When you’re talking about Exemption 1 under FOIA, information falls into an entirely different category of treatment. You have FOIA requesters arguing uphill against impressive guys in uniform who say “That information is dangerous and presents a risk to the United States.” That shuts down critical inquiry about whether or not there is a rationale for secrecy.


And yet a significant percentage of the people who make FOIA requests are military or former military people who are writing about or highlighting some aspect of national security.

That’s a very good point. There are also a lot of citizen-requesters. Some are working on personal histories. Some are working on books, some are simply curious. Under the mosaic rationale everyone is treated the same way. Any release [of information] is potentially disastrous regardless of the source requesting. That’s not so much a criticism as it is an observation.


You say there hasn’t been any formal challenge to mosaic theory in court. I suppose it would have to be carefully undertaken because the courts might uphold it, thereby establishing precedent.

You have to be really careful about the case you present if you’re the requester community. If the facts lined up well, say there are four documents that some guy in a cave in a country that wishes us harm could patch together information from, you would lose that case and there would be a rock-solid precedent in favor of the mosaic theory forever.


Bottom line, you believe that mosaic theory is too vague a standard to justify sequestration of information?

Yes, and I don’t think this is a dominant rationale for [information] denial, but I think it’s an increasingly popular one. It’s a rationale without limits.

I’m not saying there aren’t occasions where it’s perfectly justified. There are, but as a standard it’s too loose. It doesn’t require enough scrutiny of secrecy claims.



Eric Tegler is a writer/broadcaster from Severna Park, Md. His work appears in a variety...