Instead of Ramping Up, Declassification Activity Slowed in 2011
The total number of pages of government records that were reviewed for declassification last year, as well as the number that were actually declassified, declined slightly from the year before, according to the 2011 annual report from the Information Security Oversight Office (ISOO) that was published today.
Not only is this trend line unfavorable in itself, it also means that the goal set by President Obama of reviewing the entire backlog of 25 year old historical records awaiting declassification by December 2013 is out of reach and will not be achieved. [Update: This is not quite correct, since most of the backlog that needs to be processed for release does not require “review.” Only a subset of the backlog is being referred to agencies for review.]
The latest ISOO annual report, like its predecessors, is a sometimes bewildering of collection of raw statistics about government classification and declassification activity, some of which have little or no meaning or are actually misleading.
So, for example, ISOO reports that there were precisely 127,072 original classification decisions to create new secrets throughout the government in 2011. But upon close inspection this combined total of all agency classification actions conveys no meaningful information since the individual agencies exercise their classification authority in different and incommensurate ways. Thus, CIA and ODNI each generated only four original classification decisions — 0.003% of the total — though they are among the most secrecy-intensive agencies in government. Meanwhile, the much less secretive Department of State supposedly accounted for 48,968 original classification decisions last year, or 38% of all new secrets. These figures are simply not an accurate representation of executive branch classification activity as it exists in practice, and adding them together does not improve their quality.
The ISOO report also indicates that derivative classification activity — that is, the restatement in new form of information that was previously classified — increased sharply by 20% over the previous year. But the report warns that the new data reflects revised reporting requirements, so that it cannot be properly compared to previous years’ numbers. In other words, it has no particular significance or utility.
To its credit, ISOO seems cognizant that the current reporting format is not very useful or informative. The report states that ISOO “has begun to re-evaluate the elements of information that the executive branch agencies are asked to provide for this annual report” and that the “re-evaluation covers most aspects of the reporting process.”
Still, some of the data presented by ISOO are striking, though their actual meaning needs to be teased out by the reader.
So, for example, a total of 52,760,524 pages were reviewed for declassification in 2011, and 26,720,121 of those pages were declassified. These are not trivial numbers, but they are a reduction from the 2010 total of 53,087,345 pages reviewed and 29,050,290 pages declassified. More significantly, the reported level of activity means that the President’s 2009 goal of reviewing 400 million pages of classified records of historical importance by December 2013 cannot and will not be achieved. Instead of ramping up to meet the presidentially-mandated requirement — to review an average of 100 million pages per year for four years — declassification activity last year actually leveled off and declined. Curiously, the new ISOO report to the President made no mention of this disappointing fact. [Update 2: As noted above, this is not correct, since most of the 400 million pages awaiting processing for public release do not require additional declassification review. Even so, the December 2013 deadline is not likely to be met.]
The ISOO report does make the important observation that, as in past years, the majority of agency classification determinations that were appealed by requesters to the Interagency Security Classification Appeals Panel were overturned by the Panel in whole or in part, resulting in the declassification and release of records that agencies had wanted to withhold as classified.
Because this pattern has persisted for 15 years (since the Panel was established), it represents empirical proof that overclassification has been and still remains pervasive, even by internal executive branch standards. In fact, there are indications that the Panel itself is too conservative in its handling of classification disputes. Recently, even the hyper-retentive National Security Agency decided to fully release a document despite a Panel finding that it should remain partly classified.
The radical implications of ISCAP’s unbroken record of overturning a majority of the agency classification positions it reviews — which suggest that agencies are consistently misclassifying and failing to properly declassify information — are not examined in the ISOO report.
However, ISOO Director John P. Fitzpatrick noted in his transmittal letter to the President that an initiative known as the Fundamental Classification Guidance Review holds promise for improving the integrity of classification practice. The Review, which is now drawing to a conclusion, is an effort to update agency classification guidance and to identify currently classified information that no longer should be classified. “We believe that significant results will be obtained from this program,” Mr. Fitzpatrick wrote.
The research community lacks strategies to incentivize collaboration on high-quality data acquisition and sharing. The government should fund collaborative roadmapping, certification, collection, and sharing of large, high-quality datasets in life science.
The potential of new nuclear power plants to meet energy demand, increase energy security, and revitalize local economies depends on new regulatory and operational approaches at the NRC.
In anticipation of future known and unknown health security threats, including new pandemics, biothreats, and climate-related health emergencies, our answers need to be much faster, cheaper, and less disruptive to other operations.
To unlock the full potential of artificial intelligence within the Department of Health and Human Services, an AI Corps should be established, embedding specialized AI experts within each of the department’s 10 agencies.