Joint Intelligence in Support of Peace Operations

by Colonel H. Allen Boyd (USA, Retired)

In recent years, the U.S. joint military intelligence (MI) community has made substantial advances toward assembling a more responsive and cohesive intelligence support system for the operational commander. Common systems and increasingly joint tactics, techniques, and procedures (TTP) continue to mitigate individual Service parochialism and its “stovepipes.” Spearheaded by a technology revolution, these efforts are applying the lessons of Operations DESERT SHIELD and DESERT STORM to an objective joint MI system that can quickly focus on a threat situation and provide the joint force commander with sufficient battlespace awareness to win a decisive military victory. However, it remains a system primarily focused on conventional scenarios where the military tasks are clear, the threat is homogenous, and technology is the pre- dominant means of resolving ambiguity. It is a system designed for use against an opponent whose intent we know or can readily presume by virtue of its formations and patterns that the high- technology sensors detect.1

Since Operation DESERT STORM, those situations have been few, as peace operations have dominated the use of U.S. forces. Among the many lessons learned from recent peace operations are that—

In all peace operations from Somalia to the Balkans, joint intelligence efforts have required major modifications to the offensive- and target-oriented model vali- dated in part by Operation DESERT STORM.

There are more clearly defined and practical intelligence planning principles and realities for peace operations than those currently found in operational doctrine; future peace-operations commanders will not be able ignore them

While peace operations conduc- ted to date have been very different from one another in purpose and scope, supporting joint intelligence efforts contain a number of commonalities (oper- ational threads), which appear to be lessons that future peace opera- tions commanders should recog- nize and heed. Joint and individual Service doctrine consists of generic guidance and planning considerations that do not adequately emphasize the differences and complexities of intelligence support to peace operations. Nor does it sufficiently reflect the essential lessons learned over the course of recent operations. There are more clearly defined and practical intelligence planning principles and realities for peace operations than those currently found in operational doctrine; future peace-operations commanders will not be able ignore them

Operational Intelligence Planning Imperatives

From the lessons of Somalia, Haiti, and Bosnia, one can distill four future operations planning imperatives. They are:

Intelligence Support to Force Protection

Given the caution and even reluctance to commit U.S. forces to peace operations and the fragile nature of the peace, operational commanders enter them with a heightened mandate to prevent casualties.

 Force protection in peace operations receives more emphasis than combat operations because it directly and immediately affects the force’s political and popular support.

 As a case in point, the J2 of Atlantic Command’s Joint Task Force- (JTF) 180 stated that his most critical task during Operation UPHOLD DEMOCRACY was to—

develop a level of situational awareness that would prevent friendly casualties from any threats short of accidents…a zero-casualty mission in an environment where the potential for casualties lay inside many Port-au-Prince doorways and alleys.2

What are the implications for the operational commander and intelligence officer given this mandate? First, they must directly allocate intelligence resources to the force protection mission and aggressively employ them. Force protection intelligence must not be a secondary effort or the by-product of intelligence geared to traditional fire- and maneuver-type concerns nor prioritized along close, deep, and rear operational lines.3 Force protection requirements resemble those associated with rear operations in that they are the operations which customarily receive the fewest intelligence resources and less focus. The commander and J2 must immediately reverse that convention in a peace operations mission.

The intelligence collection effort must begin with the headquarters and staging bases of the peace forces and then extend outward along their patrol routes and bases into the zones of separation (ZOS) between belligerents. This requires an increasing commitment of intelligence resources (sensors as well as analysis and dissemination means) until the ZOS area achieves a steady state.

At that point, however, the commander and J2 must devote that level of resource commitment to the force protection mission. Success at this point is a tempting, but dangerous, opportunity to redirect limited, high-demand intelligence resources from secured areas deeper beyond the ZOS.

The joint intelligence system’s coverage must form an expanding threat indications and warning “bubble” around the elements of the peace force as they deploy throughout their assigned operating areas and maintain that bubble with vigilance. This grows harder to do as massed brigades and battalions disperse to companies, platoons, and squads at isolated checkpoints and small observation bases where the potential for casualties is highest and where any small incident may easily have significant political consequences.

Successive peace operations commanders and J2s have learned that, while they fully recognized the importance of force protection, they often underestimated the enormous demand it placed on the intelligence systems and support architecture as well as the peace forces themselves. A notable example is the level of effort required to locate and to characterize the millions of antivehicle and antipersonnel mines in Bosnia. As threats, mines are a concern for force protection intelligence, and as such, will occupy the time of hundreds to thousands of information collectors and dozens of intelligence analysts and reporters.

This requires an enormous investment in resources that must share capabilities and time with others concerned with the belligerents’ more visible military and less visible political capabilities and intentions. These resources will very likely need to support an expanded, combined joint MI architecture as well. Another equally important example is the extraordinary time and effort required to develop what then Major Ralph Peters coined “urban intelligence”—information ferreted from the domain of Mogadishu, Port-au- Prince, Tuzla, and Sarajevo.4

HUMINT is Paramount

Existing and emerging peace operations doctrine notes the importance of human intelligence (HUMINT) to varying degrees, but does not assign to it the overwhelming importance that commanders and J2s continue to learn from one operation to another.5 This is perhaps the easiest imperative to understand but possibly the most difficult to implement because it counters technology-based, DESERT STORM-influenced joint operations. Successive peace operations clearly reflect the necessity to establish a full-spectrum HUMINT network throughout the operational area and the reality that poor HUMINT simultaneously risks overall mission failure and protection of the peace force itself. Repeatedly peace operations commanders and J2s have reiterated that HUMINT was their most important intelligence resource.

The purpose here is not to recommend any specific application of a complex intelligence-col- lection effort but rather to point out directly that several common shortfalls occur in HUMINT support to peace operations. The first is that MI planners, especially during the early planning stages, can easily equate HUMINT, to the products of the MI organizations capable of producing it (interrogators, counterintelligence (CI) teams, and long-range surveillance units). More than one peace-operations J2 has at least initially planned and organized so-called HUMINT operations in those terms alone. In an era where we commonly equate sensors to systems organic in MI organizations, remember that every individual soldier, sailor, airman, and civilian in the mission area is an information provider and potential HUMINT collector.6

CI teams are essentially force protection assets, highly skilled individuals trained to look inward through the lens of an adversary. We train interrogator teams to elicit HUMINT from persons and documents that have come under friendly control but who also introduce some question in terms of credibility and reliability.

Together, CI and interrogator teams do provide critical HUMINT functions, but we can and must complement their limited capabilities with the eyes and ears of non-MI personnel throughout the peace force. Both Somalia and Haiti taught us that lesson, and Bosnia is reiterating it with even greater effect as MI and non-MI personnel alike rediscover the fundamentals of human reconnaissance, surveillance, and liaison.7

Peace operations both facilitate and require the broad application and management of human collectors to collect the many fragments of localized, low-level, but often factual information that technology-based sensors can neither find nor process in such highly charged situational dynamics. Infantry soldiers, engineers, civil affairs teams, drivers, aircrew members, special operations teams, medical personnel, and staff members conducting liaison are all valid information collectors. MI must help them see themselves as such, trained to observe and report, and constantly educated about what information is important within the context of the military and political environment.

The commander must instill this sense of purpose, and the J2 must establish a network in which it can operate. That network would provide a mechanism for disseminating collection requirements to the lowest observer levels and reporting observations upward to the level where a single report may serve as the critically needed information that guides an important military or political decision. It must contain appropriate filters to place informational needs in the proper local context and direct them to the right collectors. It must then similarly ensure the validation of the collected information and its placement in the correct context as it moves upward. The network must have the ability to function in the native language, a reality that we consistently relearn in the face of severely underestimated interpreter requirements.8

Technology remains a force multiplier in peace operations, but is more limited than in conventional combat operations

Commanders and J2s must quickly make critical decisions (particularly in the early operational stages) on the placement of reliable, mission-educated interpreters and constantly review these decisions as language capabilities develop and the number of interpreters expands.

This is an enormous effort for which modern commanders and intelligence officers are largely unprepared. It is an effort that requires the re-development of World War II-vintage reconnaissance, surveillance, and liaison skills; the assembly and understanding of thousands of small pieces of diverse information; and the personal manipulation of a broad, human-based network. This is a network where automation can help organize and collate information—but can rarely interpret its meaning.

Finally, commanders and J2s must recognize that beyond the roles of MI-trained teams, there are HUMINT assets trained to perform the classic role of developing and managing intelligence sources within the belligerent parties. Whether from the Central Intelligence Agency (CIA) or the Defense HUMINT Service (DHS), these individuals have critical roles to play in operations where ambiguity abounds. These assets are unlikely to be directly available to the military. Commanders and J2s should, however, make their operational-level HUMINT requirements and the scope and products of their internal HUMINT efforts known through their supporting National Intelligence Support Team (NIST). This coordination enables the tactical, operational, and strategic HUMINT actions to support each other.9 This is particularly critical in an environment where HUMINT developed at a low level can have very pronounced operational and strategic impacts.

Judicious Use of  Technology

Our high-technology systems include national and theater optical and synthetic aperture radar imagery systems, unmanned aerial vehicles, multimode signals intelli- gence platforms, and the sophisticated processors and information fusion systems that they feed. Technology remains a force multiplier in peace operations, but is more limited than in conventional combat operations.

The design of technology-based intelligence, surveillance, and reconnaissance (ISR) systems permits us to map out quickly a conventional, standardized threat for which the visible nature of its formations and orientations leads us to presume its capabilities and intentions. Their use against a small number of mortars or artillery pieces hidden in the debris of Bosnia’s urban areas or its rugged terrain is certainly limited.

This is not to say that we should not plan for or use high-technology systems in peace operations. We should use them. They provide a significant advantage when we can closely match their capabilities to specific information requirements.

Imagery can be similarly effective to update or refine information...required for political negotiation and military enforcement.

Commanders and J2s must understand how the environment governs their capabilities and limitations. The Joint Surveillance Target Attack Radar System (Joint STARS), for example, proved useful in Bosnia under conditions that limited its application more than we originally expected. Its introduction to the JOINT ENDEAVOR area of operations was accompanied by DESERT STORM-induced pros- pects for success that were immediately limited by the severity of Bosnian terrain and the radar shadowing it introduces. The Bosnia environment does not have the long convoy lines and massed formations of the Iraqi Republican Guard.

Joint STARS’ successes have been in monitoring the evacuation of belligerent forces from the ZOS where formations of equipment were in assembly areas at known points and time and departed in an organized fashion.10 Elsewhere the system has been more difficult to apply because of typically small force movements, the inevitable mixing of belligerent military equipment with civilian traffic, the concentration of belligerent’s equipment in and around the urban areas, or the simple lack of movement at all.

Imagery systems have proven similarly useful and limited. Imagery can be used with great effect to verify compliance of troop and equipment withdrawals around the ZOS, the integrity of equipment holding areas, and general conditions in areas of interest.11 Imagery in support of peace operations is best applied to surveil areas where some known condition has been established and from which changes can be measured. Where it has proven least successful is in the general search mode where the same limits described above for Joint STARS commonly prevail. Imagery can be similarly effective to update or refine mapping and geodesy information in areas where precision geography is required for political negotiation and military enforcement but is lacking.

High-capacity information pro- cessors and analysis tools are limited. This is principally because the bulk of information that MI personnel must analyze (predominantly HUMINT) needs to be put in a non-technical context (characterization of intent) that does not easily lend itself to machine-formatted reports or their terse data fields. By design, the sensor-processor combinations upon which we build joint intelligence architectures and the associated tactics, techniques, and procedures (TTP) will quickly and accurately find visible crucial threat entities and, from those, develop targets.

The intelligence architecture must be robust, flexible, and governed as much as possible by the principles of interoperability and commonality

Those capabilities to acquire, collate, and correlate information should be applicable in peace operations but with the clear under- standing that their inherently target-focused processes and products need tempering in a human judgment framework that mechanical systems cannot provide. J2s have consistently discovered that intelligence architectures must be modified (and soldiers trained) to incorporate assessment of both a political and military context with every analysis, adding a very important human dimension to powerful, but limited, technology-based systems.12

Simplify the Architecture

The intelligence architecture must be robust, flexible, and governed as much as possible by the principles of interoperability and commonality. However, it is important we understand and plan for several realities. The architecture of the U.S. peace forces will usually dominate those of our allies in terms of capabilities (particularly sophisticated sensors, processors, automated analysis tools, and supporting dissemination systems) and we must share the resulting data with them.13 It should be able to accommodate inputs from them, understanding that they can bring intelligence talents, perspectives, and capabilities to complement our own capabilities. This has proved to be particularly true in the HUMINT arena where allies have, in fact, been stronger, better prepared, and more experienced because their more limited high- technology means often force them to develop compensating HUMINT expertise.

The intelligence architecture must also interact with the information-gathering capabilities of non- governmental organizations (NGO) and private volunteer organizations (PVO). Most are extremely sensitive to the negative connotations of “military intelligence” and are habitually (frequently by charter or politics) reluctant (if not overtly opposed) to being involved with it.14 They are nevertheless superb sources of facts, impressions, and context if we can respect, accommodate, and protect their politics and policies. Until substantial changes occur, however, commanders must plan early and with great (political) care to interact with allied MI organizations as well as to develop information- sharing relationships with NGOs and PVOs.

The lesson for the future is clear: the more complex and dynamic the peace situation is, the simpler and clearer the supporting intelligence architecture must be.

Similarly, its capacity to accept, process, and effectively deal with HUMINT in general must, in large part, shape the intelligence architecture. The design of the complex information-handling subsystems permits them to deal with facts and things, not the context in which they exist. They conventionally support a “sensor-to-shooter” process that we should avoid altogether or apply very surgically under firm operational and possibly even strategic-level control.

 Within such architecture, it is difficult to inject the context of the information in an increasingly automated analytical process that accepts information, applies logical algorithms to it, and then produces an artificial intelligence-based result. We must retain sensor-to- shooter mechanisms, particularly for force protection, but we must retrain them to accept a constant inflow of context-based HUMINT and to operate under careful control.

All of this points to a central need to keep the intelligence architecture as simple and understandable as possible. Our allies cannot trust the U.S. architecture if they cannot understand it. U.S. intelligence liaison teams to allied commands cannot deliver products or submit requirements if our allies cannot understand the mechanisms or if we confuse them with a constantly changing architecture. Most importantly, the system must be clear and simple enabling its U.S. users to know precisely where and how the thousands of informational bits and pieces are processed.

The application of high technology means we must measure the architecture not in quantity but rather in terms of overall simplicity. This was a hard-learned lesson in Somalia where numerous analysis and dissemination problems resulted from an ineffective kluge of intelligence systems and processes.15 The application of technology had a more positive effect in Haiti, where the architecture centered on a HUMINT effort and technology supported it in a simple but effective manner.16 We are relearning the lesson now in Bosnia, where the joint MI architecture has a broader array of military and non-military players, and where the complexity of the mission environment is much greater.


The lesson for ther future is clear: the more complex and dynamic the peace situation is, the simpler and clearer the supporting intelligence architecture must be. Complicated architectures may be impressive and appear to promise much. They appear to serve as convenient proving grounds for even more complex improvements, and the less threatening environment of peace operations tends to encourage that experimentation. However, there is greater promise of mission effectiveness in the elegance of simplicity. Technology application must simplify and streamline joint MI processes. Commanders must constantly ask if we are applying it to that end. J2s must extend that outlook further and constantly measure the operational value of technologies against the complexity those technologies inherently add to the architecture.



1. Clapper, James R. Jr., “Joint Military Intelligence,” Joint Forces Quarterly, Spring 1994, page 94.

2. U.S. Army, Headquarters, XVIII Airborne Corps Deputy Chief of Staff for Intelligence Briefing, “Joint Intelligence Operations: Operation RESTORE DEMOCRACY,” given at the U.S. Army Worldwide Intelligence Conference, Fort Huachuca, Arizona, January 1995.

3. Atkinson, Rick, “GI’s Signal Bosnians: Yes, We’re Listening,” The Washington Post, 18 March 1996, page A-14.

4. Peters, Ralph, “Our Soldiers, Their Cities,” Parameters, Spring 1996, pages 48-49.

5. HUMINT is a planning consideration in all current Joint Chiefs of Staff (JCS) and Army doctrinal publications that address peace operations; however, most fail to give it prominence among collection means. None address the enormous workload that widespread HUMINT operations (usually manifested in hundreds of observer reports per day) place on standard joint MI collection management and analysis systems.

6. Rababy, David A., “Intelligence Support During a Humanitarian Mission,” Marine Corps Gazette, February 1995, page 41.

7. Duncan, Alastair, “Operating in Bosnia,” RUSI Journal, June 1994, page 12-15.

8. Rababy, page 41.

9. Wilson, Thomas R., “Joint Intelligence and UPHOLD DEMOCRACY,” Joint Forces Quarterly, Spring 1996, pages 57-58.

10. “Bosnia Patrol is Big Test for Joint STARS,” Jane’s Defence Weekly, 17 January 1996, pages 25-27.

11. Rababy, page 42.

12. Shelton, page 40.

13. Department of the Army, FM 100-23, Peace Operations, December 1994, page 47.

14. Leach, Raymond J., “Information Support to UN Forces,” Marine Corps Gazette, September 1994, page 49.

15. Shelton, page 37.

16. U.S. Army, Headquarters, XVIII Airborne Corps Deputy Chief of Staff for Intelligence Briefing, “Joint Intelligence Operations: Operation RESTORE DEMO- CRACY,” given at the U.S. Army Worldwide Intelligence Conference, Fort Huachuca, Arizona, Januay 14, 1995.

Colonel Allen Boyd (USA, Retired) was the Director of Futures at the U.S. Army Intelligence Center at Fort Huachuca before his retirement. He commanded the 302d MI Battalion in Weisbaden, Germany. He graduated a Distinguished Military Graduate from the Georgia Institute of Technology Army Reserve Officers Training Corps (ROTC) program with a Bachelor of Science degree in Aerospace Engineering. He also holds a Master of Science in Electrical Engineering from Georgia Institute of Technology and he earned a Master of Arts in National Security and Strategic Studies from the Naval War College. Readers can contact Colonel Boyd at