Robert David Steele
Marine Corps University
Cleared for Publication 16 March 1992
NOTE: For reasons unknown, this version did not include endnotes. It draws heavily on the work of Jack Davis and Greg Treverton. Proper credit to them can be found in ON INTELLIGENCE: Spies and Secrecy in an Open World, Fairfax, VA: Armed Forces Communications and Electronics Association, 2000.
1992 was a good year. Everyone tried to do the right thing, but the forces of passive aggressive opposition were over-whelming. Within the military, only the U.S. Marine Corps took this seriiously, and within the U.S. Intelligence Community (more like an archipelago) only the Defense Intelligence Agency–and within that agency only one man, Paul Wallner, took this seriously. Everywhere we went, “nice to have, not invented here, certainly not interested in redirecting funds” was the refrain. Below is a decent effort by decent people.
After four years of advocacy by the US Marine Corps, in 1992 the Director of Central Intelligence (DCI) agreed to take action on Open Source Intelligence (OSINT). As would be natural, the Foreign Broadcast Information Service (FBIS) was asked to provide a report on behalf of a Joint Open Source Task Force. As is the custom at CIA, everything having to do with OSINT is classified. Below is the UNCLASSIFIED Marine Corps evaluation of the Final Report. So very little has changed within CIA and FBIS in 20 years that most of this evaluation–and especially those aspects of this evaluation pertaining to the CIA/FBIS (or Open Source Center/OSC) ability to a) understand the military and b) meet military needs, remain valid as a starting point for discussing how DoD and other elements of the US Government not now satisfied by CIA or OSC, should address their needs for OSINT in support of policy, acquisition, and operations.
How to Avoid Future Strategic Intelligence Failures
This article focuses on three concerns of mine that are central to ensuring that the restructuring effort is meaningful.
First, what “sins” of strategic intelligence persist in the face of restructuring?
Second, how must the nature of the individual intelligence analyst, their working conditions, and their relationship to policy-makers change if we are to avoid strategic intelligence failures in the future?
And third, how must we relate defense intelligence restructuring to a broader national effort to establish a truly national knowledge management and information technology strategy, a strategy to empower our enterprises and schools while enabling our government to make information policy decision in all areas?
Six Sins of Strategic Intelligence Today
Here are the major sins we are committing today:
(1) Excessive collection of technical intelligence (including much too much emphasis on repetitive collection against higher priorities instead of baseline collection against lower, e.g. Third World, priorities).
(2) Cursory attention to both open source collection, and the need for a modest and redirected expansion of our clandestine human intelligence collection capability.
(3) Severe shortcomings in control over intelligence resources – those responsible for billions of dollars in each year’s budget have no capability to evaluate relative returns on investment across programs or elements of the intelligence cycle, and no adequate mechanism for ensuring government-owned capabilities are shared and not duplicated.
(4) Mindset inertia. We still have very senior bureaucrats and appointees insisting that we maintain our traditional priorities against the Soviet Union and major economic powers. To be clear on this problem:
(a) It will continue to be difficult to be difficult for our policy-makers and senior intelligence managers to focus on the need for changed priorities because our intelligence and foreign affairs communities are at least two generations away from fully understanding the Third World and dimensions of change outside the political-military and transnational economic environment. We do not have an adequate methodology for studying the preconditions and precipitants of revolutionary change (including ideo-cultural, techno-demographic, and natural-geographic change), and no indications and warning (I&W) suited to this challenge.
(b) Our entire intelligence structure, our designs and methods, do not lend themselves to being restructured and reconstituted. It is as if, after decades of learning to build Cadillacs, our very fine Cadillac, accustomed to traveling the same super-highway back and forth, must suddenly be taken apart and put back together as an off-road vehicle able to deal with the treacherous terrain and back roads of the Third World. It is obvious we not only need to pay much more attention to different “designs and methods,” but that the fastest way to create our off-road vehicle, given our lack of resources, is by melting down and recasting some portions of the community (Cadillac) in their entirety.
(5) Lack of accountability among acquisition managers and the intelligence professionals who support them. We spend billions on complex weapons systems which cannot be supported by existing or planned communications, computer, or intelligence capabilities. This sin also merits elaboration.
(a) Many of our acquisition managers and action officers want nothing to do with classified information – their offices are not cleared to hold what they would want to hold; they tend to assume that once the Required Operational Capability (ROC) is approved that the “threat” ticket has been punched; they don’t understand the intelligence community or how to make it work; no one has sponsored many of them for appropriate clearances; and they have no process for prioritizing their needs for ongoing threat support to their respective life cycles.
(b) Our concept for providing intelligence support to acquisition is flawed. We tend to focus on the technical lethality aspect of the threat, while ignoring the equally if not more important aspects of tactical reliability, operational availability (and mobility), and strategic sustainability.
(6) Finally, our worst sin, a lack of commitment to people. Our grade structure, working conditions, and turnover rates (both job reassignments and resignations) leave us with a largely “un-expert” analysis population whose historical memory is both conventional (what is in the files) and of short duration. We are not growing the kind of analyst so immersed in their topic that they can sense change and underlying analytical trends and anomalies. When someone says “protect the people in the budget” what they really mean is “keep as many serfs on board as possible.” They do not mean “nurture our best, give them time and money for travel, training, and reflection, protect them from day-to-day ‘must have update’ calls.” Our personnel strategies, some of which seek to keep personnel costs down by having a “bulge” in the most junior analytical ranks, do not provide the career opportunities needed to keep the “best and the brightest” focused on analysis for an entire career, and literally drive people away from analysis and toward “management” or administrative positions, if not out the door entirely. We compound this sin by failing to provide the analysts we do have with the tools they need to manage raw multi-media data and carry out higher-level analysis tasks including pattern analysis and modeling. In combination, our existing tools, training policies, and production requirements perpetuate the “cut and paste” syndrome. This is all part of a broader national failure, my final concern.
The Larger Failure: No Strategy
The six since discussed above come together in our failure to develop a national knowledge management strategy and a related national information strategy.
We spend too much on classified collection which we cannot process in time (or ever), and no enough on open source information including foreign scientific and technical literature vital to our national competitiveness.
We have done well as linking a vast array of different computer databases and capabilities but at a huge cost in terms of people and maintenance dollars, and without significantly improving the individual analyst’s access to data.
We have failed completely at developing a standard advanced analysts’ toolkit (workstation with integrated applications), and we are therefore wasting millions building hundreds of different workstations and application packages that provide slightly different implementation of the same generic functionality at thousands of sites throughout the world.
In short, we have done nothing to improve the quality of life for our individual analysts and little to improve their intellectual reach. In a broader context, outside the intelligence arena, we have failed to use federal funds in the knowledge management arena to support, direct, and synergize private outlays in the commercial and academic sectors.
Our nation is significantly behind its potential in exploiting the available knowledge in the world, and the available information technologies, and this is a “grand strategy” failure of enormous proportions.
Within intelligence, we will continue to have strategic failures so long as we continue to intellectually shackle and starve our diminishing population of analysts by failing to act in the two areas offering very significant returns on investment: the integration of now-operational advanced information processing technologies into a single standard analysis “toolkit” exportable to any enterprise; and the development of a multi-level and multi-media database architecture that seamlessly merges classified and unclassified data, and extends the analyst’s reach to every corner of the globe.
As an aside, let me note my support for those initiatives sponsored by the Federal Coordinating Council for Science, Engineering, and Technology (“Grand Challenges: High Performance Computing and Communications”), and the related “computer superhighway” concepts coming off the Hill. Both reflect our national tendency to focus on “big problems” and “technical solutions.” Where my emphasis differs from these two major initiatives, in a complementary way, is through my focus on “enabling tools” that give large numbers of people greater access to data, rather than great computing power to a few select scientists and their acolytes.
What Is To Be Done?
(1) Adopt David Abshire’s idea of an Advisor to the President for Long-Term Planning and make that individual the Presidential champion of a national knowledge management strategy, working in concert with the Office of Science and Technology Policy and other interested parties.
(2) Establish a Senior Inter-Agency Group (SIG/C4I) tasked with directing resources toward a global C4I system that provides multi-level security access (to include foreign nationals with no clearances), integrates multi-media databases, and establishes a standard advanced analysis “toolkit.” The Information Handling Committee (IHC) and the Advanced Processing and Analysis Steering Group (AIPASG) should serve as focal points for inter-agency coordination while the SIG/C4I provides a decision-making forum and ensures that the external investments in communications and unclassified computing are part of an integrated continuum of government-private sector spending. We should use the Defense Information Systems Agency (DISA), the Intelligence Communications Architecture (INCA) Project, and the Joint National Intelligence Development Staff (JNIDS) as executive agents for implementing a national knowledge management campaign plan. The new Advisor to the President for Long-Term Planning should chair this group, with an assistant to serve as Executive Secretary.
(3) Use the Corporate Information Management (CIM) initiative to begin exploring inter-agency solutions and mechanism for fully integrating open source and unclassified databases into a global C4I architecture. Provide a mechanism for conveying to Comptrollers the evaluations and recommendations of the IHC and AIPASG as a means of accelerating the retirement of inefficient installed bases while consolidating resources to attack generic problem sets. In particular, end the isolation of intelligence systems from all other C4 systems – C4 must improve its personnel security levels and adjust its approach to accommodate intelligence, but intelligence managers must understand that their days of pipeline management and compartmented resource allocation are over.
(4) Establish a new National Information Agency (NAI) that folds in the National Technical Information Service (NTIS) of the Department of Commerce, the Foreign Broadcast Information Service (FBIS), the Joint Publications Research Service (JPRS), the Defense Gateway Information System (DGIS), and the Defense Technical Information Center (DTIC), while also folding in and revitalizing the Federal Research Division (FRD) of the Library of Congress and creating a new consolidated joint government-business Center for the Exploitation of Open Sources (CEOS). Such a national investment could be fruitfully directed to:
(a) Engage in “competitive analysis,” using only open sources, as a means of challenging the assumptions of the remainder of the intelligence community regarding the value of extremely expensive and fragmentary classified sources; and
(b) Emphasize direct support to national and private research endeavors, with a view to stimulating and reinforcing business and academic research and development in all domains.
(5) Establish an Open Source Committee under the Director of Central Intelligence, to serve as a focal point for intelligence community collection and processing of open source information (which would include multi-spectral imagery as well as public signals, unclassified documentation, and open debriefings and interviews). Utilize military intelligence personnel and capabilities in peacetime to “jump start” the open source collection and exploitation process – this will help the military because many of the Third World intelligence gaps stemming from our obsession with the Soviet Union can be filled relatively quickly through systematic, legal, and overt access to unrestricted foreign information.
(6) Consider reorganizing the Central Intelligence Agency to provide for four distinct capabilities:
(a) a national intelligence analysis capability with numerous inter-agency collection management and analysis centers along the lines of the existing centers focused on special topics;
(b) a consolidated clandestine operations agency with its own communications and computing capabilities but integrating tactical SIGINT, necessary technical support and a new separate Office for Military Contingencies manned jointly by military and civilian personnel;
(c) a national technical intelligence agency to manage overhead technical collection systems; and finally
(d) a national intelligence research and development (R&D) agency under a new deputy director responsible for consolidating and managing the now fragmented intelligence R&D efforts scattered among different services and agencies.
Knowledge is power. Technology has broken down the walls that previously required vast technical and human endeavors to isolate nationally vital information about plans, intentions, and capabilities. At the same time, the vast outpouring of multi-media multi-lingual knowledge has presented us with an enormous technical and intellectual challenge, one worthy of the same kind of national attention occasioned by past energy crises.
There is still a role for clandestine human collection and covert technical collection, but it must be more tightly focused.
Our emphasis must shift from collection to analysis, from indiscriminate collection to integrated processing, from analysts as assembly-line producers chained to their desks to analysts as observers and partners in the national decision-making process – not making policy, but informing policy.
Finally, we must shift away from a strategy of producing highly classified compendiums of information for a few select customers, and toward maximizing public access to basic knowledge in all areas of endeavor.
The sins of intelligence will always be with us in one form or another; restructuring will cure some ills and bring on others; our greatest challenge continues to be one of strategic vision – if we can change the way we view analysts and their role in the daily decision-making process; if we can adopt a national knowledge management strategy, accelerate our integration of national C4I systems, and address the open source challenge; then we will have accomplished a far more fundamental and constructive “restructuring” – applied a “new paradigm.” This strategic interpretation is consistent with the present restructuring plans, but of far great import to how our Nation “does business” in the future.