10. Intervention Decisionmaking in the Bush Administration
by Arnold Kanter*

The Context: Characteristics and Implications of Post-Cold War Conflicts

Another consequence of the end of the Cold War is that classic cases of inter-state aggression are being supplanted by ethnic, nationalist, and separatist conflicts.[4] As noted above, U.S. interests at stake in such conflicts are more ambiguous. In addition, the political and legal grounds for intervention in such conflicts are unfamiliar and controversial. Traditionally, there has been a strong presumption in U.S. foreign policy against "interference in the internal affairs of others." This presumption is reinforced by objections from governments whose political support we increasingly require, but who themselves may be worried about establishing precedents which might affect them in the future. Finally, the basis in international law for intervening in EN&SC is often problematic. That, in turn, is important because countries whose political support we seek will often insist that there be a solid international legal rationale for the proposed course of action.

Such political and legal considerations loom large in U.S. decisionmaking about whether and how to intervene in EN&SCs precisely because the end of the Cold War both increased the political need for support from other countries and eliminated our strongest claim on that support. One need look no farther than negotiations with our allies to establish "no-fly zones" over northern and southern Iraq and respond militarily to Iraqi defiance of U.N. inspection teams for evidence both of the importance the United States attaches to multinational support for intervention and of the price paid in the coin of constraints on U.S. action.

Plowing the unfamiliar ground of EN&SC in the post-Cold War world also gives rise to what might be called the "paradox of policy principles." There is considerable value--if not a clear necessity--to articulate some principles to guide intervention decisions. These principles are needed to provide both internal guidance to the bureaucracy and public rationales to explain the intervention decisions taken--and not taken. At least at this early stage of the post-Cold War world, however, it has proven to be very difficult, if not impossible, to articulate a set of principles which the United States is prepared to apply and invoke consistently. Nevertheless, decisions somehow get taken, explained, and defended.

For example, if we are prepared to consider using force to uphold (or restore) democracy in Haiti, why will we not do so in Angola, much less in Georgia? If we are prepared to intervene to end the starvation borne of anarchy in Somalia, why are we not willing to do the same in Sudan or Burundi? If we really do support the principle of self-determination, why do we oppose the secession of Nagorno-Karabakh, Trans-Dniester, etc., from the entities which emerged from the breakup of the Soviet Union? Indeed, how do we reconcile our support for the principle of self-determination with our own Civil War (a question Russian officials take regular delight in asking)?

This "paradox of policy principles" can probably be explained by reference both to the novelty of dealing with EN&SC in the post-Cold War world and the ambiguity of U.S. interests at stake in any such conflict. Whatever the reasons, however, this paradox imposes both internal and external costs. Internally, the absence of a relatively consistent set of decision guidelines adds to the confusion and controversy which surround any intervention issue. Externally, the absence of a stable set of principles which can be invoked contributes to accusations of inconsistency and hypocrisy which, in turn, reinforce internal confusion and controversy and add to the time and other costs required to reach, implement, and defend intervention decisions.

In brief, the central features of the post-Cold War world have made the decision-making process related to intervention issues both more complicated and less orderly. While there has always been some finite probability that a decision about whether or not to intervene could have gone the other way, in the post-Cold War world that uncertainty has substantially increased. This increased unpredictability also has potentially significant implications for our ability to deter EN&SC.

Characteristics of the Decision-making Process

1. The manner in which "candidates" for intervention get on the decision agenda is relatively idiosyncratic and unpredictable. As the contrasting responses to the seemingly similar Somalia and Sudan cases suggest, media coverage can have a significant impact. Media coverage helps to put the issue on the agenda of senior decisionmakers who, like the rest of us, watch television, read newspapers, and are prey to human emotion. Moreover, intensive media coverage--reinforced by questions at regularly scheduled press briefings and Congressional inquiries, hearings, and floor speeches--all but compels senior level decisionmakers to address the issue, whatever their personal predilections. As the Bosnia case illustrates, however, media coverage can substantially increase the chances that an issue will get onto the decision agenda, but it does not necessarily lead to a decision to intervene.

   Ad hoc, self-generated assessments by officials about "what is at stake" is another way in which candidates for intervention get on and rise to the top of the decision agenda. For example, the ebb and flow in U.S. policy toward the conflict in ex-Yugoslavia can be explained in part by "the" issue being variously defined as a humanitarian tragedy, a test of the North Atlantic Treaty Organization (NATO) in the post-Cold War world, a test of the European Community's (EC) ability to act, a test of U.S. leadership, a risk of spillover and escalation, an important precedent for dealing with EN&SC in the post-Cold War world, and a clear signal to Russia about the constraints on its actions toward breakaway republics. The ways in which issues such as ex-Yugoslavia get defined, however, are highly idiosyncratic depending on factors such as who has access to senior level decisionmakers, who reads what, who talks with whom, and even the sequence in which seemingly diverse issues get addressed.

2. Although the identity of the participants is fairly stable and predictable, the decision-making process itself is fairly ad hoc. To some extent, this feature is simply a reflection of the decision-making process with respect to any "important" issue, particularly issues such as intervention decisions which are non-routine and often take place in a "crisis" atmosphere. It is also a reflection of the still inchoate "intervention policy." We are still very much feeling our way toward a "new world order" in the post-Cold War world, including the central questions about how various U.S. interests are engaged and how we should try to protect or advance them. We are in an inductive, "learning from experience" mode at this point, with little precedent, much less policy, to serve as a guide to action.

   That said, it is important to add that "learning" almost certainly is taking place. Not only is there a growing body of precedent as we face more and more intervention decisions, but the criteria and checklists--the questions that get asked and answered with respect to potential interventions--are becoming increasingly standardized. Perhaps because of the unfamiliarity of both the issues and the context, the ideas and views of pundits, essayists, and scholars have played a significant role in this learning process.

3. The decision-making process with respect to intervention issues frequently begins at relatively senior levels and works up from there. During the Bush administration, issues related to basic policy decisions about whether or not to intervene with U.S. forces were often initially addressed by the Deputies Committee, an interagency group composed of second- and third-ranking officials from the concerned departments and agencies. Frequently, these senior-level interagency deliberations were not supported by substantial staffing or written analyses. This was due to several factors, including time constraints and a fear of leaks.

   The issues were then referred to Cabinet-level officials for consideration before being presented to the president for decision. The Cabinet-level meetings tended to be relatively informal, unlike the formal, structured meetings of the National Security Council (NSC) or the Principals Committee (a group whose membership closely paralleled the NSC minus the president). No doubt in large part due to the level at which the process operates and paucity of staffing, the decisions which emerge often seem to be surprising, abrupt changes in policy. The decision to send a 30,000 man force to Somalia is a vivid case in point.

4. The process by which the military frames an issue and formulates its assessment and advice is a "black box" to most participants outside the Pentagon. The civilian leadership in the Office of the Secretary of Defense (OSD) may not be much better informed. While the Joint Staff routinely tasks the relevant command(s) for assessments, recommendations, and plans, it is not clear how much--or what kind of--guidance and context is provided with that tasking. It, likewise, is unclear how much and what kinds of informal communications between the field and Washington take place in service and other military channels. Finally, the joint impact on the process of the Goldwater-Nichols reforms and Colin Powell's tenure as Chairman of the Joint Chiefs of Staff (CJCS) is uncertain but appears to have been substantial. It often seemed to outsiders (i.e., civilian policymakers) that General Powell had about as much latitude vis-à-vis the other members of the Joint Chiefs of Staff (JCS) as the Secretary of State or Secretary of Defense ordinarily exercise vis-à-vis their respective departments.

5. There are relatively stable and distinguishable "agency perspectives" on issues related to intervention. Cognizant of the pitfalls of generalizations, it nevertheless is possible to broadly characterize the positions which the key organizations tend to adopt. The State Department tends to be more willing than the other participants to threaten, deploy, and employ military forces. This approach is probably due in part to a perspective which views the threat and use of force as a key instrument of U.S. foreign policy and American leadership. It may also be attributed in part to the fact that the State Department is substantially insulated from the price--in blood as well as treasure--that military interventions exact.

   In addition, the State Department tends to be more willing than the other participants to take risks. In part, this is a corollary to the observation that the State Department does not pay the price of military interventions, be they successful or unsuccessful. But it is also probably a reflection of a bureaucratic tactic to make intervention options more attractive by reducing their apparent force requirements.[5]

   The Pentagon--both civilian and military--tends to be much more conservative on "use of force" issues.[6] This is particularly true of the JCS representatives.[7] They are typically insistent that military forces should be deployed only if the president is ready to employ force. In contrast to the State Department, which views the threat and use of force as a complement to other force policy instruments, they are also more likely to view the use of force as a last resort, i.e., when diplomacy and other policy instruments have been tried and failed.

   In addition, JCS representatives--beginning with General Powell himself--typically argue that if the United States is going to commit military forces, we should be prepared to bring overwhelming force to bear.[8] In part, this position reflects a thoroughly understandable desire to ensure victory. In part, it probably also reflects an effort to minimize casualties by intimidating would-be opponents and discouraging engagements. Finally, they clearly understand that if intervention options entail very large force requirements, it often has the practical political effect of virtually ruling out military intervention.[9] Consideration during 1992 of options to provide convoy security in northern Iraq or to secure the route from Split to Sarajevo in ex-Yugoslavia are good illustrations of this process at work.

6. Personalities matter. If the foregoing reflects the truism of bureaucratic behavior that "where you stand depends on where you sit," there is a related truism that the higher one goes in an organization, the less one can predict positions on issues from a knowledge of organizational membership. Given the high level, ad hoc, informal nature of the process described above, however, the personal beliefs, experiences, and perceptions of the participants played an unusually large role in the decisions which emerged.

   In understanding U.S. policy toward Bosnia, for example, one needs to give full weight to the "lessons" learned by Brent Scowcroft and Lawrence Eagleburger from their respective tours of duty in Yugoslavia,[10] to General Powell's views on the use of force and the commitment of forces, and to President Bush's determination not to put U.S. ground forces into Bosnia under virtually any circumstances. More generally, President Bush was strongly inclined both to accept military recommendations regarding the forces required to accomplish the mission he had given them, and to grant broad operational latitude in the conduct of that mission. For his part, General Powell brought not only his forceful personality to the process, but the healthy discipline of regularly insisting on an answer to the question: "What is the military objective we are trying to accomplish?"

7. "Slippery slopes" are a frequent source of concern, in part because they really exist. One of the arguments regularly invoked in interagency debates to oppose U.S. military involvement in an EN&SC is that our role, however modest initially, will start us down an uncontrollable "slippery slope" to growing political responsibility for the outcomes which emerge and increasing military intervention to ensure that the outcomes are satisfactory. Put differently, there is concern that we will lose control over our stakes and therefore over our involvement, specifically, that we will be unable to keep either one limited. An important corollary is that it becomes increasingly difficult to terminate U.S. military involvement short of having achieved "success" or setting an arbitrary deadline for withdrawal.

   Although these "slippery slope" arguments may be invoked by bureaucratic participants who are opposed to U.S. intervention for other reasons, there is enough real world experience to preclude simply dismissing such objections out of hand. Somalia provides a good illustration. The U.S. airlift of relief supplies to a handful of Somali airfields seemed to be a relatively modest operation with clear limits, but it paved the way for the United Nations Task Force Somalia (UNITAF). Given the overwhelming U.S. role in and responsibility for UNITAF, it proved to be virtually impossible politically to do a clean hand-off to the United Nations Operation in Somalia (UNISOM).

   It is easy to imagine how the United States could have found itself on an analogous slippery slope in Bosnia. In early 1992, the United States reached a consensus with several of its NATO allies to initiate an airlift of relief supplies to Sarajevo. Such an airlift required that the airport at Sarajevo be reopened, which in turn necessitated the deployment of air controllers, other technicians and experts, and a security force to protect them, i.e., a contingent totaling several hundred military personnel. The United States was prepared to provide this on-the-ground capability, but the French pre-empted its deployment with a contingent of its own.[11] One is left to wonder whether subsequent U.S. policy toward Bosnia, particularly with respect to the use of military force, might have been different if U.S. rather than French ground personnel were sent to reopen and secure the Sarajevo airport.

8. Concerns about slippery slopes notwithstanding, planning paradoxically is often shortsighted. This may be due, in part, to a tendency to discount "slippery slope" objections to intervention as being little more than stalking horses for other, often unstated, objections. Thus, if a decision is made to intervene, the worries, concerns, and risks expressed by those who have been overruled tend to be forgotten or ignored. This tendency may be reinforced when the origins of an intervention decision stem from a seemingly irresistible pressure to "do something." Whatever the reasons, decisionmakers often fail to look more than one or two moves ahead, and to take into account responses available to the other side.

9. Decisionmaking is probabilistic and context-dependent. It was argued above that the combination of the lower risks and reduced stakes which characterize U.S. intervention options in the post-Cold War world make it increasingly difficult to predict what the United States will decide and how it will respond in any given instance. Most of the characteristics of the decision-making process described here add to that uncertainty. The decision to intervene in Somalia in late 1992 with nearly 30,000 American troops caught most people, inside as well as outside the government, by surprise. A review of the record will likely show that "even" the decision to take military action to drive the Iraqis out of Kuwait was neither certain nor obvious in advance. The net result of this unpredictability--reinforced by a long list of cases in which we did not intervene, e.g., Sudan, Georgia, Tadjikstan, Nagorno-Karabakh, Bosnia--is to undermine the deterrent effect of threats to intervene.

Guidelines for Decision

1. Do not intervene--especially on the ground--absent high confidence the intervention will be relatively brief, inexpensive, and cause minimal casualties and collateral damage. This guideline reflects the intersection of ambiguous (but clearly less-than-vital) stakes and uncertain domestic political support for intervention. Quick, clean, and cheap interventions are more likely to be commensurate with the stakes, and are unlikely to generate significant domestic political problems. Conversely, the prospects for sustaining domestic political support over an extended period of time, especially as things go wrong, setbacks occur, and costs mount, usually appear foreboding.

2. Do not intervene unless there is a high probability of success. This guideline adds another condition to the ones listed above, viz., if we intervene, we need to "win" or otherwise "succeed." The implication of this added condition is that we should not intervene simply to "raise the price of aggression" or send a message to would-be aggressors. An important corollary is that the perceived costs of a "failed" intervention are always seen to exceed the costs of inaction, both at home and with respect to deterrence of future EN&SC. This perspective creates a strong presumption that it is better to take no action than to launch an intervention which risks failure. Finally, the application of this guideline tends to equate the conditions for terminating the intervention with "success," i.e., the intervention typically must continue until we "win." A distant second choice is to set a fixed, but essentially arbitrary, deadline for withdrawal.

3. Avoid Congressional involvement in the decision process. The reluctance to inform and involve Congress stems in part from a view about the respective roles of the executive and legislative branches in foreign policy.[13] It is also the result of a pragmatic calculation that to involve Congress is to impose the requirement that Congress support--rather than merely acquiesce in--the proposed intervention.[14] At best, the price of Congressional support will be a loss of presidential flexibility. Depending on the specifics, of course, Congress could well impose other conditions and costs. Taken together, these prospects tend to discourage U.S. involvement in contingencies that carry a significant probability of Congressional involvement.[15]

4. Minimize the need for political support and the risk of negative political consequences. This guideline is an extension of the one above and is motivated by many of the same considerations. Paradoxically, the desire to avoid an intervention issue becoming fodder for political pundits is probably strengthened as a consequence of the growing importance of domestic political considerations in foreign policy.[16] Put simply, the greater the chances of controversy, the lower the probability that the United States will intervene, and the higher the premium on keeping the intervention quick, clean, and cheap.

5. Insist that U.S. involvement is qualitatively different in political terms. The public justification for this position is that because the United States is the world's only remaining superpower, its participation--and especially any casualties it might suffer or inflict--has distinctive implications and consequences. Its involvement, therefore, should be the exception rather than the rule and it is perfectly appropriate for the United States to urge and expect others to go where we will not. This rationale was invoked to help explain why the United States would not contribute observers on the ground in Bosnia to help monitor the "no-fly zone" established by the U.N. Security Council.[17] Whatever its legitimacy, the result of this perspective is to create yet another presumption against U.S. involvement.

6. Avoid committing U.S. ground forces. This guideline is usually applied by insisting that U.S. contributions be limited to when and where they would be "unique." Typically, this has been interpreted to refer to unique or distinctive American military capabilities, e.g., strategic lift, intelligence, communications.[18] The clear implication is that the United States has no comparative advantage in the ground forces which might be required as part of an intervention force and, therefore, would look to others to contribute these capabilities. The effect of this guideline is to increase the chances that the U.S. role in a given intervention will meet the standards of "quick, cheap, and clean."

7. Retain operational control over U.S. combat forces, particularly ground combat forces. In part, this guideline reflects the increased stakes which result from the commitment of U.S. forces. It also reflects greater confidence in the competence of U.S. military leadership, certainly compared to third world commanders. Finally, and perhaps most important, it reflects an appreciation of the fact that the president will not only be held responsible for American casualties, no matter under whose command they occurred, but also that he is likely to be more severely criticized if those casualties occurred while American forces were carrying out the orders of a foreign (and, by implication, less capable or less caring) commander.

   Whatever its origins and rationale, this guideline has the potentially perverse consequence of making U.S. military participation tantamount to an "all or nothing" proposition. If the United States insists that our forces may only serve under U.S. commanders, then other potential contributors may expect, and in any event will argue, that the United States should contribute the bulk of the forces. This means that if, for whatever reason, the United States determines that it is in its interest to have U.S. forces play some role in a given intervention, it will be under considerable pressure to play a substantial role.

8. Secure authorization by the U.N. or other international organizations. In the post-Cold War world in which only less-than-vital interests are militarily threatened, unilateral interventions such as Operation Just Cause may increasingly become rare exceptions to the rule of operations authorized by U.N. Security Council resolutions. In part, this requirement is imposed by allies as a condition of their participation. (See below.) In part, it is a condition imposed by the political rhetoric and realities at home. Whatever its origins, however, it is a source of additional legal and political constraints on U.S. action.

9. Obtain multilateral participation. This dictum might be considered to be a straightforward corollary to the guideline above, but that should not obscure the fact that it imposes an additional set of constraints. It is one thing for an ally to offer political support, such as voting in favor of a U.N. Security Council resolution, but it is quite another thing to agree to join with the United States in a military operation. The price of that participation is likely to be additional conditions and constraints on the operation (as well as perhaps "side payments" on other, unrelated issues). Operation Provide Comfort and Operation Southern Watch--the no-fly zones over northern and southern Iraq which are enforced by British and French as well as American aircraft-are good illustrations.

Conclusions

The picture of the decision-making process which emerges from this examination of the early post-Cold War period is one which is, paradoxically, both very insular and highly constrained by external factors. It is also one which has strong, systemic biases against intervention. Indeed, the question is less why the United States decides to intervene in any particular instance than how it ever manages to overcome these strong, pervasive presumptions against intervention.

The most difficult and controversial issues typically concern the commitment of U.S. ground forces. Those decisions are qualitatively different than decisions about the commitment of "unique" U.S. capabilities such as strategic lift and intelligence. To overstate only slightly, U.S. decisionmaking about intervention is about decisions concerning the deployment of U.S. ground combat forces. Finally, "success" is the primary--if not sole--criterion for deciding when to terminate an intervention. On the one hand, this is a high standard which constitutes yet another obstacle to an intervention decision. On the other hand, it has profound implications for "termination decisionmaking" by making it very difficult to extricate ourselves from an intervention short of "victory."

[1] Given the limited objectives of this essay, no effort was made to conduct interviews or to research primary and secondary source material. That said, there is every reason to believe that the observations it records are probably general, rather than peculiar, to the Bush administration. Both for that reason, and as a stylistic convenience, it is written in the present tense.

[2] On the one hand, Operation Desert Shield/Desert Storm (ODS) falls outside the boundaries of EN&SCs. On the other hand, the ODS experience is relevant to an analysis of EN&SC decisionmaking if only because it is regularly invoked as a successful model and precedent when EN&SC interventions are being considered.

[3] An important residual of Cold War thinking remains. There has been great sensitivity about getting involved in any conflict which is located on the Russian periphery or presents the likelihood of confronting Russian soldiers. This sensitivity all but ruled out serious consideration of intervening in Moldova, Georgia, or Tadjikstan. Russian officials did not hesitate to reinforce that sensitivity by such tactics as drawing analogies between the Monroe Doctrine and Russian "responsibilities" in countries on their periphery, i.e., in the "near-abroad."

[4] The actual number of EN&SCs has probably increased with the end of the Cold War and the demise of the Soviet Union, but it is also the case that these same factors permit, or in some cases require, the United States to pay more attention to EN&SCs.

[5] As will be seen below, military participants employ the opposite bureaucratic tactic of overstating the apparent force requirements. The result is mutual suspicion and reciprocal incentives to engage in bureaucratic gaming.

[6] During the Bush administration, OSD and JCS made it a regular practice to arrive at a common position which they would then jointly present in the interagency process. This practice may have had the collateral consequence of muting any "pro-intervention" proclivities among senior OSD civilians.

[7] In fact, it is very difficult to distinguish between JCS Chairman Powell's approach to "use of force" issues and the positions taken by JCS representatives who participated at lower levels in the decision-making process.

[8] At the same time, JCS representatives often worry aloud about the scarcity of critical force elements and the implications of a proposed military operation for our ability to deal with other contingencies. Likewise, the projected dollar cost of a proposed operation can be a substantial factor in deciding whether or not to proceed. The combination of increasingly tight defense budgets and negligible allocations for "peacekeeping" requires either that other military capabilities--typically readiness--will be shortchanged to cover the costs of an unbudgeted intervention, or that the administration pay the political price of seeking a supplemental appropriation from Congress.

[9] Needless to say, JCS recommendations about force requirements are rarely subject to second-guessing and carry enormous weight in the decision-making process, whatever other participants might suspect about what lay behind those stated requirements.

[10] Eagleburger served as American ambassador in Yugoslavia, and Scowcroft as defense attachŽ in Belgrade.

[11] The tacit competition between the United States and France over which would provide the personnel to reopen the Sarajevo airport was a reflection of intra-Alliance politics, intramural disputes about the "European security and defense identity," and the respective roles for NATO and the European Community (EC) in expressing this identity.

[12] As it enters its second year in office, the Clinton administration, likewise, has yet to issue a "Presidential Decision Directive" with its policy guidelines.

[13] The recent debates about the "Byrd amendment" on U.S. involvement in Somalia, and the "Dole amendment" on U.S. involvement in Haiti are good contemporary examples of this issue.

[14] Its public grumbling notwithstanding, Congress often welcomes the opportunity not to have to take a formal position for or against controversial foreign policy decisions, including decisions related to intervention.

[15] Indeed, presidents sometimes establish Congressional approval as a condition for U.S. intervention in order to avoid pressure from allies to become involved. Many analysts believe that Eisenhower employed this tactic to deflect a French request for assistance in Indochina. Some observers see the same tactic at play in the administration's insistence on Congressional approval of any deployment of U.S. ground forces in Bosnia.

[16] As the Bosnia case demonstrates, inaction by the United States can also become a source of political controversy and grist for editorial writers.

[17] Undoubtedly, there were several other reservations about U.S. participation, including "slippery slope" concerns.

[18] There is another sense in which U.S. capabilities can be unique, viz., when American participation is politically indispensable.