The NYT liberated the specific answer to a question that Russ Feingold asked in March 2009, but which DOJ didn’t respond to until November 2010, when Feingold was a lame duck Senator. At issue were new investigative guidelines Attorney General Michael Mukasey issued in late 2008, on his way out the door, which allowed the FBI to investigate Americans for First Amendment reasons so long as that First Amendment reason was not the only reason they were being investigated.
Here’s how the ACLU described the new guidelines:
Under the new “assessment” authority, FBI agents can investigate anyone they choose, so long as they claim they are acting to prevent crime, protect national security, or collect foreign intelligence, with absolutely no requirement of a factual connection between their authorizing purpose and the conduct of the individuals they are investigating. FBI agents can start “assessments” without any supervisory approval, and without reporting to FBI headquarters or the Department of Justice. The Guidelines do not require the FBI to keep records regarding when “assessments” are opened or closed and “assessments” have no time limitation. The FBI can even start an “assessment” of you simply to determine if you would make a good FBI informant. Innocence no longer protects ordinary Americans from being subjected to a wide range of intrusive investigative techniques. The techniques include:
- collecting information from online sources, including commercial databases.
- recruiting and tasking informants to gather information about you.
- using FBI agents to surreptitiously gather information from you or your friends and neighbors without revealing their true identity or true purpose for asking questions.
- having FBI agents follow you day and night for as long as they want.
So in response to Feingold’s questions about how many assessments had been initiated and closed, FBI responded:
The FBI has initiated 11,667 Type 1 and Type 2 assessments, 3,062 of which are ongoing. 427 preliminary and full investigations have been opened based upon information developed in these Type 1 and Type 2 assessments. 480 Type 3, 4, 5, and 6 assessments have been initiated, of which 422 remain open.
To do the math, 95% of the Type 1 and 2 assessments that have been closed have resulted in no further investigation, suggesting the FBI was on a wild goose hunt.
But here’s the tricky thing: the FBI records on those people can be entered into FBI’s investigative databases!
Even if information obtained during an assessment does not warrant opening a predicated investigation, the FBI may retain personally identifying information for criminal and national security purposes. In this context, the information may eventually serve a variety of valid analytic purposes as pieces of the overall criminal or intelligence picture are developed to detect and disrupt criminal and terrorist activities. In addition, such information may assist FBI personnel in responding to questions that may subsequently arise as to the nature and extent of the assessment and its results, whether positive or negative. Furthermore, retention of such information about an individual collected in the course of an assessment will alert other Divisions or Field Offices considering conducting an assessment on the same individual that the particular individual is not a criminal or national security threat. As such, retaining personally identifying information collected in the course of an assessment will also serve to conserve resources and prevent the initiation of unnecessary assessments and other investigative activities.
So that says the FBI may be entering those 95% innocent people into a database with personally identifiable information.
Now, to be fair, FBI also mandates that these personally identifying information contain a warning that the person “does not warrant further FBI investigation at this time.”
As a result: (i) when records retained in an assessment specifically identify an individual or group whose possible involvement in criminal or national security threatening activity was checked out through the assessment; and (ii) the assessment turns up no sufficient basis to justify further investigation of the individual or group, then the records must be clearly annotated as follows: “It is noted that the individual or group identified during the assessment does not warrant further FBI investigation at this time. It is recommended that this assessment be closed.”
And, as Charlie Savage notes, the numbers FBI gave Feingold may not be all that accurate.
Some aspects of the statistics are hazy, officials cautioned.
[snip]
F.B.I. officials also said about 30 percent of the 11,667 assessments were just vague tips — like a report of a suspicious car that included no license plate number. Such tips are entered into its computer system even if there is no way to follow up on them.
Finally, they said, it is impossible to know precisely how many assessments turned up suspicious facts. A single assessment may have spun off more than one higher investigation, and some agents may have neglected to record when such an investigation started as an assessment.
Still, if what the FBI had wanted was just a database of information on all the young African American and Muslim men out there, maybe they should have just been straight up about it and simply retained the census workers to put together their database?