r18 livingston

CLARIFYING THE CNN EFFECT: An Examination of Media EffectsAccording to Type of MilitaryIntervention

by

Steven Livingston

Research Paper R-18June 1997

PRESS POLITICS

PUBLIC POLICY

The Joan Shorenstein Center

Harvard UniversityJohn F. Kennedy School of Government

CLARIFYING THE CNN EFFECT: An Examination of Media EffectsAccording to Type of MilitaryIntervention

by

Steven Livingston

Research Paper R-18June 1997

Copyright 1997, President and Fellows of Harvard College

All rights reserved

Steven Livingston 1

IntroductionIn recent years, observers of international

affairs have raised the concern that media haveexpanded their ability to affect the conduct ofU.S. diplomacy and foreign policy. Dubbed theCNN effect (or CNN curve or CNN fac-tor), the impact of these new global, real-timemedia is typically regarded as substantial, if notprofound.

Two key factors have joined to bring thisabout. One is the end of the Cold War. With itspassing the United States lacks an evident ratio-nale in fashioning its foreign policy.1 The otherfactor is technological. Advances in communica-tion technology have created a capacity to broad-cast live from anywhere on Earth. As a result, thevacuum left by the end of the Cold War has beenfilled by a foreign policy of media-specified crisismanagement.

While William Randolph Hearsts New YorkJournal and Joseph Pulitzers World may havecreated the climate for war with Spain in 1898,the extent, depth, and speed of the new globalmedia have created a new species of effects. It isthis global, real-time quality to contemporarymedia that separates the CNN effect from ear-lier media effects on foreign policy. Yet exactlywhat those effects are, when they are likely to beseen, and even whether they exist at all is thesubject of intense debate.

Despite numerous symposia, books, articles,and research fellowships devoted to unravelingthe CNN effect, success at clarifying itthispaper will arguehas been minimal. In part, thismay be due to the imprecise use of the termCNN effect. Writers too often and too easilyslip back and forth between related but otherwiseconceptually distinct understandings of the effector effects in question. The first objective of thispaper is to clarify exactly what is meant by theCNN effect. The second objective concerns

policy. Just as we must speak more preciselyregarding the type of effect we might expect tofind as a result of media coverage, so too must wespeak more precisely about foreign policy. Ratherthan treat foreign policy as an undifferentiatedmonolith, we need to discriminate between dif-ferent foreign policies, each with its own objec-tives, means, potential and actual costs (mea-sured in dollars, lives, and political prestige) andsensitivities to media and public pressures. Wemust develop, in other words, a greater apprecia-tion for the possibility that different foreign poli-cy objectives will present different types and lev-els of sensitivity to different types of media. Atypology of policy-media effects will be developedin the last half of this paper that demonstratesseveral different potential consequences for poli-cy, some harmful, some salubrious, dependingupon the nature of the policy objectives andmedia content. A matrix of media effects, policytypes, and objectives is offered last.

Differentiating Several CNN EffectsFor many journalists, policy-makers, and

scholars, there really is little doubt that mediaprofoundly affect the foreign policy process.Former Defense Secretary James Schlesinger, forexample, has argued that in the post-Cold Warera the United States has come to make foreignpolicy in response to impulse and image. Inthis age image means television, and policiesseem increasingly subject, especially in democra-cies, to the images flickering across the televisionscreen.2 A commonly-cited example is theClinton administrations response to the mortarattack on a Sarajevo market in February 1994that killed sixty-eight people.3

Despite the frequency, volume, and intuitiveappeal of this argument, a growing number ofscholars and commentators have begun to ques-tion whether media actually do have the abilityto affect the foreign policy process as presumed.4The key variable to medias effect on foreign poli-cy is not the presence or absence of cameras, butrather the presence or absence of political leader-ship. James Hoge, Jr., editor of Foreign Affairs, forexample, argues that while a CNN effect of somesort may have once existed immediately follow-ing the end of the Cold War, it no longer does, orat least not to the same extent.

CLARIFYING THE CNN EFFECT: AN EXAMINATION OF MEDIA EFFECTSACCORDING TO TYPE OF MILITARY INTERVENTION

by Steven Livingston

Steven Livingston was a Fellow at the Shorenstein Center inthe Spring of 1996. He is Associate Professor of PoliticalCommunication and International Affairs and can bereached at: The School of Media and Public Affairs,The George Washington University,801 22nd Street, NW, Washington, D.C. 20052.Telephone: (202)[email protected]

2 Clarifying the CNN Effect

It seems to me that about two years ago we reachedthe high-water mark on standing in awe over thepotential CNN effect on things. Since then, therehave been a lot of conferences and things writtenthat have essentially gotten us to where we aretoday, which is that television news has a tacticaleffect from time to time, but not a strategic one;that it operates more when humanitarian issues areat hand than when actual security issues are.5

Hoges point is important. Whether his specif-ic assertions are correct is less important thanthe approach he takes to the question. He is sug-gesting that effects on policy are conditional andspecific to policy types and objectives.

A reading of the growing literature suggests atleast three conceptually distinct and analyticallyuseful understandings of medias effect on theforeign policy process.

We may speak of the CNN effect as 1) a poli-cy agenda-setting agent, 2) an impediment tothe achievement of desired policy goals, and 3)an accelerant to policy decisionmaking. Eachroughly corresponds to various stages of the use-ful, though slightly contrived, notion of a linearpolicy process. The initial formulation of policycorresponds with concerns that media are policyagenda-setting agents. Secondly, policy imple-mentation corresponds with concerns thatmedia may serve as accelerants of the process orimpediments to the achievement of policyobjectives. (Figure One provides an outline ofthese effects.)

It is important to keep in mind that each ofthese possible effects may be evident over

timesometimes a very short timeon a singlepolicy issue. It is possible, for example, thatmedia as policy agenda-setters may raise theprominence of an issue, placing it before higher-level policymakers. It may then shorten thetime those policymakers have to deal with orresolve the issue (accelerant). Finally, it maythenwith coverage of some traumatic event ordisclosure of tactically important information,impede the development or implementation ofpolicy meant to address the problem. U.S. policyin Somalia, in some measure, fits this mold.These are, nevertheless, analytically distincteffects, and as I will argue later, each is likely tobe associated with different types of policy. Eachwill be taken up in turn.6

Media as AccelerantOne of the potential effects of global, real-time

media is the shortening of response time for deci-sionmaking. Decisions are made in haste, some-times dangerously so. Policymakers decry theabsence of quiet time to deliberate choices, reachprivate agreements, and mold the publics under-standing.7 Instantaneous reporting of events,remarks State Department SpokespersonNicholas Burns, often demands instant analysisby governments . . . In our day, as events unfoldhalf a world away, it is not unusual for CNNState Department correspondent Steve Hurst toask me for a reaction before weve had a chanceto receive a more detailed report from ourembassy and consider carefully our options.8

Former Secretary of State James A. Baker, IIIhighlights this understanding of the CNN effect.

Accelerant Media shortens decision-making response time.Televisiondiplomacy evident. During time of war, live, global televisionoffer potential security-intelligence risks. But media may alsobe a force multiplier, method of sending signals. Evident inmost foreign policy issues to receive media attention.

Impediment Two types: 1. Emotional, grisly coverage may underminemorale. Government attempts to sanitize war (emphasis onvideo game war), limit access to the battlefield.2. Global, real-time media constitute a threat to operationalsecurity.

Agenda Setting Emotional, compelling coverage of atrocities or humanitarian Agency crises reorder foreign policy priorities. Somalia, Bosnia and

Haiti said to be examples.

Figure 1. Conceptual Variations of CNN Effect

Steven Livingston 3

The one thing it does, he says, is to drivepolicymakers to have a policy position. I wouldhave to articulate it very quickly. You are in areal-time mode. You dont have time toreflect.9 His adviser and former press secretary,Margaret Tutwiler, echoes his assessment:Time for reaction is compressed. Analysis andintelligence-gathering is out in the new worldof global media.10

Richard Haass, former member of theNational Security Council and one of PresidentBushs closest advisers during the Persian Gulfconflict, also notes this effect, saying thatCNN has changed the concept of a daily newscycle. We no longer have the old rhythms,everything is telescoped. So, if he (SaddamHussein) was going to get out there at 4:00 or5:00 in the afternoon, we had to get out by 4:30or 5:00 in order to make sure that the eveningnews was not a disaster or that people in theMiddle East some seven or eight hours aheaddidnt go to sleep thinking that somehowSaddam had made some great new offer, whenin fact he really hadnt.11

Understood as an accelerant to the policyprocess, global, real-time media have also had aneffect on the operation of the foreign policybureaucracy, particularly intelligence agenciesand desk officers in the State Department.Former presidential press secretary MarlinFitzwater remarked, In most of these kinds ofinternational crises now, we virtually cut outthe State Department and the desk officers . . .Their reports are still important, but they dontget here in time for the basic decisions to bemade.12 Intelligence agencies now must com-pete with news organizations, thus speeding uptheir assessments, and be prepared to defendtheir assessments against the evidence presentedon television or other real-time media, such asthe Internet and telephone.

While often treated as a detriment to goodpolicy, Haass has argued that the availability ofglobal, real-time television can just as well beconsidered an asset.

People are looking at the medias impact as adowner, . . . a problem for policymakers to copewith. That is true. But it was also an opportunity.One of the things about the CNN effect for peo-ple like me at the time (of the Persian Gulf war)was it gave you some real opportunities. One waspenetration. CNN gave you tremendous access tomarkets that normally you couldnt get to.13

Besides the Middle East, it was useful, saidHaass, in sending signals into Europe. And itgave us a real capacity to reach people at home.The media which brought information ininstantaneously also gave us the chance torespond and to get our message out instanta-neously. This had consequences, in Haasssview, but not of the sort supposed by thoselamenting the CNN effect. We felt we couldmanage public opinion in this country and thatwe could manage the alliance, or the coalitiondimensions of the war, as well as get to the Iraqipeople and the Arab world. Much of the time,global, real-time media offered opportunities fora policymaker, rather than only presenting prob-lems.14 This more inclusive understanding ofthe CNN effect was evident even at the dawn ofthe global reach of media over thirty years ago.

During the Cuban missile crisis in 1962, theKennedy administration had several days duringwhich the public knew nothing of the threatlooming over the horizon. According to historianMichael Beschloss, Kennedys successors mightwell look back longingly at the episode, forKennedy had the luxury of operating in whatthey would probably consider to be the halcyonage before modern television news coverage.15Kennedy used the first six days of the crisis toconvene his advisers and rationally consider theoptions in quiet, without public hysteria.16

What is often overlooked, however, is theconstructive role played by the real-time, glob-al media, such as they were in 1962, in endingthe crisis. At the time, government-to-govern-ment communication between Moscow andWashington was so primitive, according toBeschloss, it took six to eight hours to send andtranslate messages. In an attempt to overcomethis barrier, and to side-step the KGB andSoviet military, Khrushchev began sending mes-sages to the Americans via Radio Moscow,which he knew was constantly monitored bythe United States. Robert McNamara recalledthat on Sunday, October 28, the day the crisiswas finally defused,

He (Khrushchev) instructed that the public radiotransmitter in Moscow be held open for his mes-sage. And his message was sent over that so that itwould avoid the long interval of coding and decod-ing . . . It was to eliminate that time gap of six oreight hours that Khrushchev insisted that the finalmessage be transmitted immediately, because hefeared that we were engaged at that moment intime in initiating military action.17


Meanwhile, CBS News Moscow correspon-dent Marvin Kalb, having anticipatedKhrushchevs announcement, secured a line toNew York to coincide with the key RadioMoscow broadcast that morning. As Kalb simul-taneously translated Radio Moscows announce-ment, President Kennedy and his advisers lis-tened in the White House.18

Ironically, it seems, the pre-global televisionhalcyon age included a scramble to find ameans to achieve what is today one of the chiefcharacteristics of the CNN effect: accelerated,real-time diplomacy.

Today, the instantaneous transmission ofdiplomatic signals via global media is routine.Tutwiler points out that other governmentswatched her briefings with great care, lookingfor nuanced policy shifts. Consequently, theState Department would use this to theiradvantage to inform their counterparts overseasof U.S. reactions or intentions.19 StateDepartment spokesperson Nicholas Burns doesthe same thing.

I sometimes read carefully calibrated statements tocommunicate with those governments with whichwe have no diplomatic relationsIraq, Iran, Libyaand North Korea . . . . Given the concentration ofjournalists in Washington and our position in theworld, the U.S. is uniquely situated to use televi-sion to our best advantage, with our friends as wellas our adversaries.20

While the new environment constitutes a sig-nificant change to the slower, more deliberateprocesses of yesteryear, it is less clear whetherthis is necessarily injurious to sound policymak-ing. Rather than a liability, a resourceful diplo-mat may just as well find global, real-timemedia an asset.

Media as ImpedimentThere are at least two types of media-related

policy impediments. One is rooted in theinhibiting effects of emotional coverage andoperates through the agency of public opinion,both actual and latent. The other is rooted inthe potential for global, real-time media to com-promise operational security, the veil of secrecyespecially needed with some types of militaryoperations. We will take up each of these typesin order.a) As an Emotional Inhibitor

Following the decisive American military victory in the Persian Gulf, President Bushenthusiastically remarked, By God, weve

kicked the Vietnam syndrome once and for all.21At the heart of the Vietnam syndrome was theconcern that media coverage had the potentialto undermine public support for an operationand erode troop morale on the ground. As such,perceived American credibility and resolve inthe world was undermined.

Yet two years later, in October 1993, picturesof a dead American soldier being draggedthrough the streets of Mogadishu revived someof the same fears and concerns evoked byVietnam. The Clinton administrations decisionto withdraw U.S. troops from Somalia as soon aspossible was the more immediate result.22 AsThe New York Times put it, the recent fightingcrystallized American public opinion on anissue that previously was not particularly press-ing to the average citizen. And the pictures of adead American soldier being dragged throughthe streets of Mogadishu seem to have made itall but impossible for Mr. Clinton to changemany minds. Indeed, public opinion pollsfound that more than half the respondents didnot approve of President Clintons handling ofthe situation in Somalia.23

During the Gulf war, fear of an unsanitizedpresentation of the carnage of battle was per-haps central to the militarys efforts to controlthe media through the use of press pools andmilitary escorts. John J. Fialka, a Wall StreetJournal correspondent, remarked, We wereescorted away from most of the violencebecause the bodies of the dead chopped up byartillery, pulverized by B-52 raids, or laceratedby friendly fire dont play well, politically.24Military planners insisted, on the other hand,that they were motivated by a legitimate con-cern for operational security, as well as a con-cern for the well-being of the journalists. Theyfurther pointed to the logistical difficultiesencountered in accommodating the large num-ber of journalists who wanted to cover the war.

But for many the impression remained that atthe heart of the militarys concern was thecapacity of media to undermine public andpolitical support for an operation involvingcasualties. Ted Koppel, speaking of the PersianGulf war, remarked, Im not sure the publicsinterest is served by seeing what seems to havebeen such a painless war, when 50,000 to100,000 people may have died on the other side.Obviously this was done so they could maintainthe closest possible control over public opinion,to increase support for the war.25

Control of the reporter was a central compo-nent of the militarys effort to limit the potential

Steven Livingston 5

for public relations damage. Just before theground war there were 25 to 30 pool reporters tocover six Army and two Marine divisions nearthe Kuwaiti border. No reporter from The NewYork Times bureau in Saudi Arabia was givenofficial access to a pool slot before February 10.26

Use of officially sanctioned pools had a par-ticularly pronounced effect on the availability ofpictures during combat. One editor at the timewas quoted as remarking, The pictures comingout of pool arrangements are quite ordinary.There are no negative aspects to the war.27

Tomorrow's wars will most likely look morelike the conflicts in Somalia, Bosnia, and Haiti,and less like the Gulf war. In such circum-stances, journalists will already be in the zoneof conflict, making their control far more diffi-cult for military planners. Yet, in the long-run,pictures may not matter as much as context andleadership. The key variable may be the pres-ence of a clearly articulated policy and a publicsense that the policy is worth it. Colin Powellexpressed this point:

Theyre (the American people) prepared to takecasualties. And even if they see them on live tele-vision it will make them madder. Even if they seethem on live television, as long as they believe itsfor a solid purpose and for a cause that is under-standable and for a cause that has something to dowith an interest of ours. They will not understandit, if it cant be explained, which is the point I havemade consistently over the years. If you cantexplain it to the parents who are sending theirkids, youd better think twice about it.28

Media scholars suggest that government offi-cials and agencies are becoming more sophisti-cated in their effort at offering the sort of credi-ble explanations referred to by Powell andthat in most circumstances they are assisted intheir efforts by the American media.29 Forexample, political scientist W. Lance Bennetthas found that the media closely index theircoverage to the contours of official debate andcontroversy.30 That is, the levels of criticismdirected at government policy rises and falls inaccordance with the intensity of criticisms ema-nating out of other institutionally-based officialsources. As Bennett and political scientist JarolManheim put it, As a practical matter, newsorganizations routinely leave policy framing andissue emphasis to political elites (generally, gov-ernment officials).31

b) As a Threat to Operational SecurityWhile it may still be an open question

whether media content, live or otherwise, hasthe ability to hinder the pursuit of desired policygoals because of their emotional freight, the factremains that some operations are extremely sen-sitive to media exposure. Maintaining opera-tional security during conventional war and tac-tical operations, such as anti-terrorism opera-tions, is essential. In these circumstances, mediahave the technological capacity to hinder sometypes of operations simply by exposing them.

This is true, for example, in conventionalwarfare. As communication equipment becomesmore mobile and global in its reach, and real-time reporting of all types becomes more perva-sive, the danger to operational security willbecome more pronounced. It isnt like WorldWar II, when George Patton would sit around inhis tent with six or seven reporters and muse,with the results transcribed and reviewedbefore being released, remarks Powell. If a com-mander in Desert Shield sat around in his tentand mused with a few CNN guys and pool guysand other guys, its in 105 capitals a minutelater.32 In the process of covering an operation,news organizations may reveal information thatleads to unnecessary casualties and even thepossible failure of a mission.

This is not to say that journalists will seek (orhave sought) to deliberately expose operations.The disclosures are inadvertent. Retired generalNorman Schwarzkopf has told of such a case.

It was reported (by an American television net-work) that at this time, right now, we are witness-ing an artillery duel between the 82nd AirborneDivision and the Iraqis. If they (the Iraqis) had anykind of halfway decent intelligence, they wouldhave made note of the time . . . and through theirintelligence network they would have pinpointedthe location of the 82nd Airborne. Until that timeeverything they ever saw of the 82nd was on theeast coast. All of a sudden they would have foundthe 82nd way to the west and it would certainlyhave telegraphed something to them.33

Shortcomings in Iraqi military intelligencemeant they were not able to take advantage of akey piece of information that would haveinformed them the American military wasmassing 200 miles west of Kuwait City.

In summary, there are two understandings ofmedia effect-as-policy-impediment. One is psy-chological and concerns the corrosive effect


some types of media content may have on pub-lic opinion, particularly public support for war.The seriousness of this potential effect is opento question. As Colin Powell remarked, picturesof dead American soldiers, as one example, mayjust make the American public madder. Theother, and potentially far more profound effectinvolves violations of operational security, as wehave just reviewed.

Media as Agenda Setting AgentsOf the presumed media effects on foreign pol-

icy, perhaps the most disturbing is the sugges-tion that the U.S. foreign policy agenda itself isat times merely a reflection of news content.This is not to say that issues are necessarily cre-ated ex nihilo by media content, but rather thatpriorities are reordered by coverage.34 Whatwould have been handled by mid-level officialsin a routine fashion instead become the focus ofhigh-level decisionmaking. Former Secretary ofState James Baker makes the point this way,

All too often, television is what determines whatis a crisis. Television concluded the break-up ofthe former Yugoslavia and the fighting in theBalkans was a crisis, and they began to cover itand cover it. And so the Clinton administration(was left) to find a way to do something. (Yet) theydidnt do that in Rwanda where the excesses wereevery bit as bad, if not worse. And so, you have toask yourself, does that mean you should do foreignpolicy by television? Are we going to define crisesaccording to what is covered, by what the editorsdecide to cover? I dont think we should do that.35

This has been a constant theme of criticismsince the end of the Cold War. JamesSchlesinger argues this when he remarks,National policy is determined by the plight ofthe Kurds or starvation in Somalia, as it appearson the screen. Jessica Mathews makes thesame point this way, The process by which aparticular human tragedy becomes a crisisdemanding a response is less the result of arational weighing of need or of what is remedia-ble than it is of what gets on nightly newsshows.36

Most of the post-Cold War interventionsSomalia, Haiti, and Bosniahave been done inthe name of humanitarianism, what MichaelMandelbaum has referred to as the foreign poli-cy of Mother Teresa, or foreign policy as abranch of social work.37 In his view, foreignpolicy as social work, particularly during the

Clinton administration, has tended to be aboutperipheral issues. It intends

to relieve the suffering caused by ethnic cleansingin Bosnia, starvation in Somalia, and oppression inHaiti. Historically the foreign policy of the UnitedStates has centered on American interests, definedas developments that could affect the lives ofAmerican citizens. Nothing that occurred in thesethree countries fit that criterion.38

It is debatable whether this is a fair criticismof the Clinton administrations foreign policy. Inparticular, Mandelbaums reliance on a tradition-al values/national interests dichotomy is prob-lematic, as Stanley Hoffmann pointed out in hisresponse to the argument.39 National interestsare not self-evident, but are instead constructsthe choices and preferences made by nationalleaders. Because the selections are often contro-versial, those who support them cover themwith the mantle of the national interest, andthose who do not back them argue, likeMandelbaum, that they deal with developmentsthat could (not) affect the lives of . . . citizensand thus are not in the national interest.40

Whatever its shortcomings, Mandelbaumsargument does point us to the fact that of all thehumanitarian crises found at any given point intime, the inclination will be to address thosewhich happen to be featured on television,rather than those which are the more severe, orthose with the greatest likelihood of successfulredress by outside intervention. To put it anoth-er way, media-as-agenda-setting-agent arguesthat the choices and selections of national inter-ests are too heavily weighted in favor of whathappens to get covered by CNN or other media.

These are important considerations, for clear-ly the conditions that have given rise to human-itarian interventions by the United States in thepast are only likely to grow more severe in thefuture. According to a 1996 study by the UnitedStates Mission to the United Nations, regionalconflicts in the mid-1990s have put 42 millionpeople around the world at risk of disease andstarvation.41 If media coverage of crises has hadan effect on U.S. foreign policy in the past, assome argue, then the potential for similar effectsin the future are great.

Television, for a variety of commercial andprofessional reasons, is drawn to the dramaticvisuals found in mostbut not allhumanitari-an emergencies. The pitched battles betweengun-totting teenagers in the streets of somehitherto unheard of place, massive flows of

Steven Livingston 7

refugees, the pathos of a starving child, all makefor compelling television. Once engaged, oncethe U.S. foreign policy priorities align them-selves with media coverage, the other two mani-festations of the CNN effect may come intoplay. Decisionmaking may be accelerated andrash. Events may cascade out of control, leadingto confrontations for which the public and poli-cymakers themselves are psychologically unpre-pared. Vivid imagery, in such a scenario, drivesboth ends of policy, to reiterate an earlier quote.First there is the politics of humanitarian inter-vention and then of disillusioned withdrawal.42

While this version of the CNN effect has themost profound potential consequences, it is alsothe most problematic for several reasons.

First, what few empirical investigations existhave not borne out the contention. AndrewNatsios, the Bush administration official whoheaded-up the relief effort in Somalia, hasargued that if one examines the record ofAmerican policy involvement in overseas

humanitarian crises, one comes away with theconclusion that the so-called CNN effect hastaken on more importance than it deserves as anexplanation for responses emanating from thepolicymaking process in Washington.43

The majority of humanitarian operations areconducted without media attention. In 1991, forexample, the United States Agency forInternational Developments Office of ForeignDisaster Assistance and Food for Peace Programhad shipped some 12,000 tons of food toSomalia. This was well before the news mediadiscovered the crisis there in August of the fol-lowing year.44 Furthermore, the eventual mediacoverage itself was the consequence of officialactions. Specifically, it resulted from the effortsof one part of the foreign policy community(Office of Foreign Disaster Assistance, and itsallies in Congress and portions of the StateDepartment) to pursuade other elements of theforeign policy community (primarily senior deci-sion makers in the White House) to sign on to

25

20

15

10

5

0

A

B

C

D

E

F

G

H

I

J

K

12/3012/1011/2012/1010/109/208/308/107/20

A

250

200

150

100

50

0

B

C

D

E

F

G

H

I

J

K

7/20 8/10 8/30 9/20 10/10 10/30 11/20 12/10 12/30

Figure 2.

Aggregated Broadcast

Network News Coverage

of Somalia:

July-December 1992

(a) Daily story count; (b) cumulativestory count. A, U.N. secretary gener-als report to Security Council; B,U.N. Security Councils report onSomalia; C, OFDA press confer-ence: Washington, DC; D, WhiteHouse announces relief effort; E,U.S. airlift commences; F, first flightsinto Somalia; G, President Bushadvocates use of security forces; H,Bush approves 28,000 U.S. troops;I, secretary of state tells U.N. thatUnited States will send troops; J,U.N. Security Council authorizestroops; K, first U.S. troops go ashore in Somalia.

2a

2b


desired policy goals. The media were used bysome officials to get the attention of other offi-cials, a tried and true practice of bureaucraticpolitics that predates CNN by many years.45The great majority of Somalia coverage followedrather than preceded official action.

Of the events noted on the timeline presentedin Figures two (a) and two (b), the White HousesAugust 14 announcement (point designationD) that it would use U.S. aircraft to sendrelief supplies precipitated the first wave ofAmerican news media attention to Somalia,which jumped fivefold almost overnight. As anexecutive at NBC said at the time, With theinternational relief effort growing, the Somaliasituation is likely to be examined more often bythe network news shows in the coming weeks.Were going to cover it more.46 The announce-ment of the planned deployment of troops two

months later (point designation H) caused asecond expansion of coverage. In both instances,media attention followed official actions.

This trend was even more pronounced withCNN coverage, evident in Figures three (a) andthree (b).

CNN coverage of Somalia prior to theannouncement of the airlift of emergency foodand medicine in August (notation E) was spo-radic to non-existent. By logical necessity, it isdifficult to conceive how media could have beenthe cause of policy developments in Somalia in1992.47 Media were doing as they have for gener-ations: they followed the troops.

While disease and starvation are commonlyseen by American journalists in Africa and else-where in the developing world,48 they are notcommon news stories. In fact, it may be thatjournalists tend to dismiss humanitarian crises

Figure 3.

CNN Daily Coverage

of Somalia:

July-December 1992

(a) Daily count by minutes andstory; (b) cumulative count by min-utes and story. A, U.N. secretarygenerals report to Security Council;B, U.N. Security Councils report onSomalia; C, OFDA press confer-ence: Washington, DC; D, WhiteHouse announces relief effort; E,U.S. airlift commences; F, firstflights into Somalia; G, PresidentBush advocates use of securityforces; H, Bush approves 28,000U.S. troups; I, secretary of statetells U.N. that United States willsend troops; J, U.N. SecurityCouncil authorizes troops; K, firstU.S. troops go ashore in Somalia.

Figures 2a, 2b, 3a and 3b from Steven Livingston and Todd Eachus, Humanitarian Crises and U.S. Foreign Policy:Somalia and the CNN Effect Reconsidered, Political Comunication, Volume 12, pp. 413-429.

STORIES MINUTES200

150

100

50

0

AB

C

DE

FG

H

I

J

K

12/3012/1011/2010/3010/109/208/308/107/20

STORIES MINUTES2,500

2,000

1,500

1,000

500

0

AB

C

DE

FG

H

I

J

K

7/20 8/10 8/30 9/20 10/10 10/30 11/20 12/10 12/30

3a

3b

Steven Livingston 9

because they are so much a part of the land-scape in some regions of the world. In 1993,when approximately 50,000 people were killedin political fighting between Hutus and Tutsisin Burundi, American broadcast television net-works ignored the story. When regional leadersmet in Dar es Salam in April 1994 in an attemptto reach a regional peace accord, only CNNmentioned the meeting. Yet a more sophisticat-ed analysis of events in Rwanda that samemonth required understanding that fear of suc-cessful peace talks encouraged Hutu extremistsin Rwanda to launch their campaign of carnageagainst Tutsis and Hutu moderates who favoredaccommodation.49

Overall, the lack of media coverage of human-itarian emergencies is most striking.50 Figurefour presents the percentage of mentions of all13 of the most severe humanitarian emergenciescombined and total individual mentions in TheNew York Times, Washington Post, ABC Newsprogramming, CNN news programming, andNPRs All Things Considered and MorningEdition from January 1995 to mid-May 1996.

This figure, of course, does not capturechanges in news coverage over time, as FiguresTwo and Three did regarding Somalia, but it doesclearly indicate the tendency of the Americannews media to give uneven attention to humani-tarian crises. This is seen in the cases of Liberiaand Sierra Leone. Next to Tajikistan, they each

received the least attention by the news organi-zations represented here.51 Afghanistan and theSudan have more people at risk than Bosnia, buttogether they received only 12 percent of thetotal media coverage devoted to Bosnia alone.Tajikistan, with one million people at risk, has alittle over one percent of the media coveragedevoted to Bosnia alone. Put another way, of allnews stories between January 1995 and May1996 concerning the thirteen worst humanitari-an crises in the worldaffecting nearly 30 mil-lion people, nearly half were devoted to theplight of the 3.7 million people of Bosnia.52 Thisis not to dismiss their condition, but only tomake the point that media coverage of humani-tarian crises is not uniform, and, more impor-tantly is typically triggered by official actionsand associated with the presence of U.S. troops.

The second reason the CNN effect may havetaken on more importance than it deserves, touse Natsiosis's characterization, is that if onelooks more closely at some of the more promi-nent post-Cold War U.S. humanitarian inter-ventions, one is likely to find equally com-pelling geostrategic reasons for the intervention.This was certainly true of the response to theKurdish refugee crises along the Iraqi borderwith Turkey following the Persian Gulf war inthe spring of 1991.

President Bushs National Security Adviser,Brent Scowcroft emphatically makes this point:

Figure Four

Percentage and Frequency of Total at Risk Per Country&

Percentage and Frequency of Total Coverage by News Organizationof Each Listed Country

Country Percent atRisk(millions)

Percent ofMentionsin Times

Percent ofMentionsin Post

Percent ofMentionsby ABC

Percent ofMentionsby CNN

Percent ofMentionsBy NPR

Afghanistan 14 (4) 4.7 (274) 4.8 (225) 1.5 (19) 1.2 (57) 2.9 (57)Sudan 14 (4) 3.3 (190) 3.5 (166) 0.6 (8) 1.1 (54) 1.5 (31)Bosnia 13 (3.7) 45.8

(2,633)43.7(2,046)

66 (833) 66.7(3,062)

61.3(1,204)

Ethiopia 11 (3 - 4) 0.2 (15) 0.2 (10) 0 (0) 0 (3) 0.3 (6)Angola 9 (2.5) 2.0 (120) 3.0 (144) 0.7 (9) 0.4 (22) 1.7 (34)Rwanda 9 (2.5) 6.9 (401) 5.9 (277) 3.9 (49) 9.8 (150) 6.0 (118)Sierra Leone 6 (1.8) 1.0 (63) 1.6 (78) 0.3 (4) 0.5 (26) 1.0 (20)Liberia 5 (1.5) 2.8 (164) 3.2 (150) 2.5 (32) 1.0 (49) 2.3 (46)Iraq 5 (1.3 -


We were actually quite cynical about mediasimpact. Media are too fickle (to have animpact). Media attention to any given crisiscould therefore be ridden-out by policy makers.Geopolitical factors, in Scowcrofts view, weremore important concerning the implementationof Operation Provide Comfort. Without Turkeyfactored in, with just television pictures, I dontknow what our response would have been. Wewere very sensitive to Turkeys anxiety aboutallowing the Kurds to stay. That was fundamen-tally what motivated us.53 Turkey, a staunchAmerican ally and a member of NATO, hasbeen in a long and bloody guerrilla war with ele-ments of its own Kurdish population in easternTurkey. The idea that thousands of Kurdishrefugees from Iraq might become permanentlylocated in or near the border was anathema tothe Turkish government.

James Baker made the same point, Oncethey (the Iraqi Kurdish refugees) all went intoTurkey, it was important to get them back toIraq.54 When Saddam Husseins survivingforces ruthlessly crushed first the Shiite rebel-lion in the south and then the Kurdish rebellionin the north in March 1991, the policy of theUnited States was to let the rebellions fail,despite the gruesome pictures coming out ofnorthern Iraq at the time. When asked if it wasaccurate to suggest that U.S. policy at the timewas to not get involved, regardless of what thepictures showed, Baker responded, Thatsright. I think that is an accurate description. Itwould have been a mistake to be involved.55For geopolitical reasons, the United Statesallowed the rebellions to fail and implementeda policy designed to resettle the Kurdishrefugees back in their towns and villages innorthern Iraq. This was not done because ofpictures. It was done because Turkey, a staunchally during the Persian Gulf war and NATOpartner, needed it done for its stability and forthe stability of the entire region.

The third reason the CNN effect as agendasetter may be overstated is the formal policyrequirements put into place following events inSomalia in 1993. Under the provisions ofPresidential Decision Directive 25 (PDD 25),issued in May 1994, approval of the use of U.S.forces for humanitarian relief missions becamemore difficult. Before forces may be used, aseries of strict conditions must be met, includ-ing: a clear statement of American interests atstake in the operation, the approval of Congress,the availability of funding for the operation, afixed date of withdrawal of U.S. forces, and an

agreed upon command and control structure.56The first application of PDD 25 was in

Rwanda. The U.S. Representative to the UN,Madeleine Albright, insisted it would havebeen folly for a UN force to venture into themaelstrom of killing.57 Despite the biblicalproportions of the bloodshed, the United Statesdid not intervene until later, when Huturefugees in camps in Zairesome of whomwere the perpetrators of the massacre in thefirst placebegan dying from the effects ofdehydration, malnutrition, and disease. Butfeeding and watering, as it is sometimescalled in the Pentagon, and what we will call aconsensual humanitarian operation below, isa considerably different policy objective thanshooting and pacifying, as would have beennecessary to stop the bloodshed and possiblyavoid the exodus of refugees. It would haveinvolved far higher risks and potential costs,measured on several scales.

There are at least three understandings ofthe CNN effect: media as an accelerant to theprocess, as an inhibitor, and as an agenda set-ter. The next step is to clarify further the rela-tionship between these various possible effectsand different policy types.

Types of Intervention

Eight types of military interventions andtheir possible relation with one or another ofthe media effects outlined above will bereviewed next. The objective simply is to illus-trate the thesis, not to offer a comprehensivereview of potential media effects on all policytypes. It is intended to stimulate thoughtanchored in a more refined understanding ofmedia and policy interaction. The policy typesreviewed are conventional warfare, special oper-ations and low-intensity conflict (SOLIC),strategic deterrence, tactical deterrence, peace-making, peacekeeping, imposed humanitarianoperations, and consensual humanitarian operations.58

As one moves from top to bottom of FigureFour, one sees reduced potential costs resulting from failure, measured in money andlives, andless preciselyin political prestige,international standing, and confidence felt byalliance partners. Beginning with peace making,as one moves toward the bottom of the figure,one finds policies designed to respond to crisesstemming from war or other human-createdconditions. Failed states or civil wars such as inSomalia, the Sudan, and Zaire offer examples.

Steven Livingston 11

1) Conventional WarfareThe stakes are highest in conventional war-

fare, which also generates the greatest mediaand public interest. Reprising the difficultiesfound between the military and the media inwartime is not necessary here.59 Experience inrecent wars indicates that when and where pos-sible, the military will attempt to control themovements of journalists and the content oftheir reports, behavior rooted in the two con-cerns outlined above: fear that the wrong pic-tures will undermine public or congressionalsupport for the effort and, second, that journal-ists will inadvertently disclose tactical or strate-gic information to the enemy.

Pool systems, prior clearance of dispatches,and other forms of censorship will continue tobe a part of military planning for conventional

warfare. At the same time, high public interestand the journalists ambition and sense of inde-pendent professionalism will lead to efforts toavoid and undermine the militarys attempts tocontrol them. The media will be assisted inthese efforts by the greater mobility provided tothem by smaller, light-weight equipment capa-ble of point-to-point transmissions from any-where to anywhere on Earth.

In conventional warfare, media are most like-ly to serve as accelerants and impediments inthe policy process. It is highly unlikely thatmedia content alone might in some fashion leadthe United States into a conventional war. Thedegree to which media serve as accelerants todecisionmaking in war, and the degree to whichthis is necessarily injurious cannot be answeredin the abstract. The answer depends on the cir-

Figure 5.Intervention Types and Accompanying Media Considerations

Policy Goals &Objectives

Likely Media Interest Government Policy Likely Media Effects Public Opinion

Use of Force

Conventional Warfare Destruction of enemyand his war-fightingcapability.

Extremely high. High degree ofattempted mediacontrol.Indexed news

Accelerant,impediment (bothtypes).

High public interestand attentiveness to it

Strategic Deterrence Maintain status quo.Ex.: Cold war.Korean Peninsula.

Moderate to highinterest. Routinizedcoverage.

Routine newsinteraction. WhiteHouse, DoD, Statebriefings, etc.

During stability, littleeffect. Accelerateduring periods ofinstability.

Scrutiny only byattentive public.Expansion of baseduring instability.

Tactical Deterrence Meet challenge tostatus quo.Example: DesertShield. PRC-Taiwan,March 1996.

Moderate to highinterest but episodic.

Controlled butcooperative. Forcemultiplier.

All three effects, butnot necessarilyinjurious.

Attentive publicscrutiny. Latent publicopinion a concern topolicy makers

SOLIC Counterterrorism,hostage rescue,specialized operations.

High interest,particularly hostagesituations, someterrorism.

Secrecy. Barring of allaccess.

Impediment(Operational securityrisk).

Little to no publicawareness in mostcases.

Peace Making Third party impositionof political solution byforce of arms.Example: LateSomalia, Yugoslavia.

High interest at initialstages of operation.Variation afterwarddependent on level ofstability.

Volatile conditions.Danger in reporting.Access with risk.Impediment (bothtypes).


Peace Keeping Bolster an acceptedpolitical solution bypresence of third party.

Moderate interestunless accord isdestabilized.

Generally unrestrictedaccess to theater ofoperation.

Impediment(emotional impedimentmost likely).


Imposed HumanitarianOperations

Forceful, apolitical aidpolicy

Low /moderate interestunless violence ensues.

Volatile conditions.Reporting risky.

Impediment (emotionalimpediment mostlikely). Attentivepublic scrutiny. Latentpublic opinion aconcern to policymakers

ConsensualHumanitarianOperations

Agreed humanitarianassistance.

Initial operation metby moderate to lowinterest.

Unrestricted, evenencouraged mediacoverage.

Media effect unlikely. Attentive publicscrutiny.

Operations Other Than War


cumstances and the resourcefulness of the poli-cymakers. The same is true of the possibilitythat pictures may undermine public support forthe war. As Colin Powell remarked, picturesmay just as well make people madder.

The media effect of greatest concern to themilitary in conventional warfare is their abilityto provide adversaries sensitive information. Inan era of highly mobile, decentralized, global,real-time media, the risks to operational securityare considerable.

2) Strategic DeterrenceDeterrence may be defined as the persuasion

of ones opponent that the costs and/or risks of agiven course of action he might take outweighits benefits.60 Persuasion, of course, involvescommunication. The idea is to communicate awillingness to use force with the hope that,ironically, it will negate the necessity to do so.The movement of force and a projected willing-ness to use it is an important component ofdeterrence. It can take the form of a long-termdeployment, such as was the case in WesternEurope during the Cold War, the continued posi-tioning of forces on the Korean Peninsula, themore recent stationing of U.S. troops inMacedonia, and the stationing of mostly AirForce units in southern Turkey. The objective isto maintain the status quo.61

Typically, media coverage of strategic deterrentoperations during times of relative stability willbe highly routinized. Regular correspondentsassigned to institutional settings such as theState Department, the Defense Department,White House, and foreign correspondents in theregions of the operation, will monitor usuallyincremental developments over an extended timeperiod. The level of media and public interestwill vary according to the perceived stability ofthe situation, that is, according to the perceivedeffectiveness of deterrence. Signs of instability,such as the rash of North Korean incursions intothe demilitarized zone between the two Koreas in1996, will spark an increase in attention. Thelikely media effect is subtle. Elite debate and dia-log in columns and opinion journals regarding thebest way to maintain (or perhaps dissolve) thestatus quo is most common. Except during timesof crisis, foreign policy debate of this sort is com-monly left to the attentive public and policyelite. Television, with a few exceptions, does notdwell on such matters, again except during timesof instability, such as when the Eastern Bloc andSoviet Union collapsed.

3) Tactical DeterrenceDeterrence may also come in the form of a

rapid response to tactical developments, such asthe deployment of a Navy carrier group or rapidreaction force to some trouble spot in the world.In these circumstances media interest is likelyto be extremely high, as is almost always thecase of troop deployments in circumstances ofpotential conflict.

With tactical deterrence, global media areoften important and valuable assets to the mili-tary, particularly when time is short and condi-tions are critical. Admiral Kendell Pease, Chiefof Information for the United States Navy, hascalled global media in such circumstances aforce multiplier. After showing a CNN videoclip of carrier-based U.S. fighter-bombers takingoff on a practice bombing run against an impliedIraqi target during Desert Shield, Peaseexplained that the Navy had arranged for aCNN crew to be aboard the carrier to film thehardware in use and to send a message toSaddam Hussein. The Iraqis, the Navy realizedand counted on, monitored CNN.62 The samething is going on now, said Pease inTaiwan.63 Prior to Taiwans March 1996 elec-tions, which China opposed and threatened tostop with military force if necessary, theClinton administration sent two aircraft carriergroups to the seas off Taiwan. Television crewsaccompanying the Navy ships sent pictures ofthe American defenders to the Chinese and therest of the world.

By using media as a force multiplier in con-junction with deterrent force, U.S. policy mak-ers are, in effect, attempting to create a CNNeffect in the policymaking of a potential oractual adversary. As Richard Haass noted at thebeginning of this paper, global, real-time mediashould not be regarded solely as an impedimentor obstacle to policy makers. It may just as wellbe an asset.

4) SOLICSpecial operations and low-intensity conflict

(SOLIC) is a general term applied to an array ofmilitary missions employing highly trained andspecialized commando forces. Navy Seals, ArmyRangers, special operations wings of the AirForce, and the Delta force offer examples ofunits typically involved in SOLIC. SOLIC mis-sions include counter-terrorism operations,hostage rescue, and during conventional war-fare, infiltration into enemy territory. Duringthe Persian Gulf war, many of the Scud missile


batteries in Iraq were destroyed by special opera-tions units, rather than by airstrikes.

SOLIC operations take place in hostile envi-ronments, are usually limited in scope, and areconducted in an envelope of extreme secrecy. Infact, for years the most elite of the Americanmilitary units involved in counter-terrorism/hostage rescue operations, Delta Force, was notpublically acknowledged. The necessity of secre-cy makes SOLIC operations the most sensitiveto media coverage. The disclosure of an opera-tion would, in all likelihood, lead to its termina-tion. It is unlikely other forms of media effectswould be associated with SOLIC, though it isconceivable that special operations units may beactivated, for example, in response to somehighly publicized situation involving hostages.

5) PeacemakingWith peacemaking operations we begin a dis-

cussion of several non-traditional militarymissions, each differentiated by subtle butimportant features. Peacemaking, also some-times referred to as robust peacekeeping andaggravated peacekeeping, presupposes thatone or more parties to a conflict are not inter-ested in peace and do not agree to the presenceof outsider peacekeepers. Ergo, such an outsideforce is a heavily armed. On the other hand,while there is little or no peace in a peacemak-ing environment, there is considerable restraintin the use of force. Peacekeeping is distin-guished from conventional war by its objectivesin using violence, if necessary. In war, the objec-tive is to inflict massive destruction on anenemy. In peacemaking, the goal is to create theconditions necessary for the implementation ofan accord. In a sense, the goal of peacemaking isto create the environment required for peace-keeping operations discussed below.

The hostile, unstable nature of the peacemak-ing environment means media and public inter-est is likely to be extremely high, at least ini-tially. As with peacekeeping, if and when asense of stability is established, media interestwill diminish accordingly. Also as with peace-keeping, the most likely potential media effectwith peacemaking is as an emotional impedi-ment. Casualties may undermine public andelite support for the operation. Knowing this,opponents of peace may deliberately targetpeacemaking forces. Due to the hostile nature ofthe environment and the potential for open con-flict between the peacemakers and one or moreof the hostile forces in the war, operational

security is also a concern. Descriptions of peace-maker capabilities may serve as a force multi-plier, but if it is too detailed it may just as wellsuggest vulnerabilities of the peacemaking force.Furthermore, it is feasible that in highly unsta-ble, fluid situations, media content may acceler-ate decisionmaking. Finally, it seems unlikely,particularly after implementation of PDD 25,that media content alone will lead to U.S.involvement in risky peacemaking operations.

6) Peacekeeping In peacekeeping missions, lightly-armed forces

are deployed in a permissive environment tobolster a fragile peace. A permissive environ-ment is one in which the outside military forceis welcome in sufficient measure by all combat-ants to allow for relative safe operation. Haassdescribes peacekeeping as the deployment offorce in a largely consensual framework inwhich there are at most only periodic, relativelyisolated, and small-scale breakdowns of thepeace.64 Their role is relatively passive.

As with all U.S. troop deployments, the newsmedia will show considerable interest in peace-keeping operations, though after a period ofapparent stability, media interest is likely toflag. U.S. peacekeeping forces in Macedoniaoffer an example. Coverage, to the degree thereis any, will be limited to the elite press, such asThe New York Times. Television, at best, willpay only passing interest. Media interest willrise in direct proportion to the sense of potentialinstability. Put another way, the more fragile thepeace the peacekeepers are there to protect, thegreater will be media and public interest. If thesituation appears unstable, and political leadershave not made the case that American nationalinterests are involved in preserving the peace,media coverage of casualties may quickly under-mine support for the mission. Here the mediaeffect is an emotional impediment. Operationalsecurity, though a concern, is not central.Peacekeeping missions are often transparentin any case, an openness in operations that ismeant to suggest evenhandedness.

7) Imposed Humanitarian InterventionsAs the title implies, imposed humanitarian

interventions differ from peacekeeping opera-tions in that the scope of their objectives ismore limited. Examples of imposed humanitari-an interventions are Somalia after December1992 but before the summer of 1993, Iraq sinceApril 1991, and in and around Sarajevo in 1994.The objectives are limited to providing food,


medicine, clean, safe water, and a secure butlimited geographical location.65 Whereas themission of a peacekeeping operation is to main-tain a fragile status quo (peace), and the missionof a peacemaking operation is to impose a polit-ical solution, by force of arms if necessary, onone or more sides to a conflict, the mission ofan imposed humanitarian intervention is to feedand care for a population in need.

In large measure, in these circumstances themilitary is used for their technical capabilities,such as water purification, field medicine, and,most importantly, logistical capabilities. Withstrategic airlifters, the American military isunmatched in its ability to move massiveamounts of cargo great distances to almost anysort of terrain.

Media interest is likely to be quite high, particularly at the beginning with the introduc-tion of U.S. troops, as discussed earlier. Thiswill be particularly true if correspondents canoperate safely in the secure zone established bythe military. One of the very legitimate andunderstandable reasons so little coverage wasgiven to the massacres in Rwanda prior to theGoma refugee coverage was the inability of jour-nalists to move about safely in Rwanda.66

Though media content alone is not likely tolead to an imposed humanitarian intervention,it cannot be ruled out. The media effect of great-est potential in imposed humanitarian missionsis as an impediment. The argument for continu-ation of a policy with possibly little or no directAmerican interest would be difficult to sustainin the face of pictures of the injured and dead.

8) Consensual Humanitarian InterventionsConsensual humanitarian intervention, as the

name implies, involve the use of the military inaddressing the urgent needs of a distressed popu-lation. As with imposed humanitarian interven-tions, consensual humanitarian interventionsare intended to save lives, not alter political cir-cumstances on the ground through the use offorce.67 The U.S. response to a devastatingcyclone that hit Bangladesh on May 3, 1991offers an example. In the midst of OperationProvide Comfort in northern Iraq, the U.S. mili-tary sent military teams to assist the survivors.U.S. assistance to refugees in Goma, Zaire in1994 is another example of a consensualhumanitarian intervention.

Consensual humanitarian interventions arerelatively low-cost, not only in materialresources but also in terms of the potentialpolitical capital at stake. Because American sol-diers are working in a relatively permissiveenvironment, political leaders face relatively lit-tle risk in deploying them on such a mission.What risk there is may be found in a generalsense of unease that pervades such missions, atleast with some, over the possibility that mis-sion creep will lead to a deeper involvement,as happened in Somalia. If truly consensual, andif it remains so, there will probably be little sus-tained media interest in the story.

Shifts in Media-Policy EffectsEach operation outlined above tends to offer

different sensitivities to media content. Further,the potential effects in question are interactive.Shifts in policy will produce changes in mediacoverage, just as media coverage may changepolicy. U.S. policy in Somalia offers an exampleof the dynamic interactive nature of foreign pol-icy making and media coverage.

From late 1991 to about July 1992, the U.S.policy response to the worsening conditions inSomalia was non-military in nature, with reliefoperations working through the auspices of non-governmental organizations, the Office ofForeign Disaster Assistance, the InternationalCommittee of the Red Cross, and UN agencies.As noted above, during this time there was prac-tically no media attention to Somalia. In August1992, the Bush administration undertook a con-sensual humanitarian intervention, transportingrelief supplies to Somalia using military cargoplanes. A small contingent of security and othersupport personnel were also involved. With ToddEachus, I have argued elsewhere that media con-tent had no effect on this decision, contrary topopular belief. Instead, it resulted from a numberof bureaucratic and domestic political (presiden-tial campaign) considerations.68 That is not tosay that media did not play a role in the unfold-ing policy developments. With the introductionof American military personnel in August,media coverage of Somalia skyrocketed, notbecause conditions had worsened, but becauseAmericans were there.

The continued fighting and banditry inSomalia made the environment there somethingless than consensual; not all of the players on


the ground agreed that giving food and otherforms of aid to those in need was necessarily agood idea. As a result, in December the Bushadministration sent in Marines to provide secu-rity. At that point the policy changed, for thethird time, and became an imposed humanitari-an intervention. Media attention, as Figure Twoindicates, rose accordingly.

Then, by the summer of 1993, the Clintonadministration and its counterparts in the UNallowed the mission in Somalia to becomesomething else again. It drifted into becoming apeacemaking operation. The problem theadministration created for itself in the processwas found in the fact that it had not put anappropriate force structure on the groundthetroops and equipment necessary to achieve thenew political mission. Whereas humanitarianmissions, strictly speaking, do not pursue politi-cal objectives, at least not in theory, peacemak-ing missions do. More importantly, the Clintonadministration failed to build the political sup-port with Congress, opinion leaders, and theAmerican public necessary for sustaining amore demanding political mission in Somalia.As a result, the policy was derailed in October1993, as is often said, with the pictures of adead American body on macabre display inMogadishu. Different policies with differenttypes and levels of media scrutiny produced dif-ferent results.

ConclusionEach policy outlined above obviously has dif-

ferent objectives, actual and potential costs, andoperational requirements. As a result, the levelof interest media have and the potential conse-quence of that interest vary substantially. Beforewe can make theoretical and empirical progressin understanding the effects of media on foreignpolicy we must refine the debate to meaningfulterms. The grand, interesting, and often heateddebate about the CNN effect will continue tofail us. unless we distill it into its constituentparts. That means speaking more precisely aboutthe likely effects relative to specific policies.


1. James Schlesinger, Quest for a Post-Cold WarForeign Policy, Foreign Affairs, Winter 1992. On-linewithout pagination. See also Richard N. Haass,Paradigm Lost, Foreign Affairs, Winter 1995. On-linewithout pagination.

2. James Schlesinger, Quest for a Post-Cold WarForeign Policy. See also Richard N. Haass, ParadigmLost, Foreign Affairs, Winter 1995, on-line withoutpagination, and James F. Hoge, Jr., MediaPervasiveness, Foreign Affairs, Summer 1994, on-linewithout pagination.

3. Charles Krauthammer, Intervention Lite: ForeignPolicy by CNN, The Washington Post, February 18,1994, p. 25.

4. Nik Gowing, Real-time Television Coverage ofArmed Conflicts and Diplomatic Crises: Does itPressure or Distort Foreign Policy Decision Making?,Working Paper 94-1, The Joan Shorenstein Center onthe Press, Politics and Public Policy, John F. KennedySchool of Government, Harvard University, p. 9. Seealso Johanna Neuman, Lights, Camera, War: IsTechnology Driving International Politics? (St. MartinsPress, 1995) and Warren Strobel, The CNN Effect,American Journalism Review, May 1996, p. 32;Hearing Before the Committee on Foreign Affairs, U.S.House of Representatives, 103rd Congress, second ses-sion, April 26, 1994, p. 5. See also For the Record,The Washington Post, April 27, 1994, p. A22.

5. The Robert R. McCormick Tribune Foundation-George Washington University conference on Military,Media, and Humanitarian Crises, held on the campusof The George Washington University, May 5, 1995.

6. This does not exhaust the possibilities, of course.Murray Edelman pointed out in several importantworks (Constructing the Political Spectacle, PoliticalLanguage) that the policy process often acts in reverseof what has just been implied. At times, particular pol-icy goals are first identified by officials. Second, a prob-lem of sufficient emotional or symbolic importance isthen offered as a justification for adopting a set offavored solutions.

For example, reducing Iraqs ability to conduct hos-tile actions against its neighbors was a policy goal ofthe Bush administration in the fall and winter of 1990.The administration needed to find ways to frameAmerican interests there as something more com-pelling than protecting oil. Secretary of State James A.Bakers Its jobs, jobs, jobs just didnt do the trick.Instead, the administration and its allies at Hill &Knowlton, a Washington public relations firm, wereable to cast Saddam Hussein as Adolph Hitler and cre-ate emotionally compelling but factually unsound sto-ries about Iraqi soldiers removing Kuwaiti babies fromincubators (See William Dorman and StevenLivingston in Taken By Storm, W. Lance Bennett and

David Paletz (eds.) (Chicago: University of ChicagoPress, 1994)). Desired policy was achieved by creatingcompelling news stories and frames.

I do not mean to suggest that the analytical frame-work offered here explains more than a portion of theforeign policy process. Not all policy is the result ofmedia content. There are many sources of policy. Andjust as often as media influences policy, media them-selves are the instruments of policymakers. If weassume some portion of policy and policy outcomes arethe result of media content (knowing also that mediacontent is itself often a reflection of policy), whatmight those effects be?

7. James F. Hoge, Jr., Media Pervasiveness, ForeignAffairs, Summer 1994. On-line without pagination.

8. Nicholas Burns, Talking to the World AboutAmerican Foreign Policy, The Harvard InternationalJournal of Press/Politics, Fall 1996. Vol. 1, No. 4. pp.10-14.

9. Author interview with James A. Baker III,Houston, Texas, May 13, 1996. See also James A. Baker,III, The Politics of Diplomacy: Revolution, War &Peace, 1989-92 (G.P. Putnams Sons, 1995) p. 103.

10. Author interview with Margaret Tutwiler,Alexandria, Virginia, March 29, 1996.

11. The Robert R. McCormick Tribune Foundation-George Washington University conference on Military,Media, and Humanitarian Crises. May 5, 1995.

12. Frank J. Stech, Winning CNN Wars, Parameters,Autumn 1994, p. 38.



15. Michael R. Beschloss, Presidents, Television, andForeign Crises, (Washington, D.C.: The AnnenbergWashington Program, 1993) p. 9.

16. Beschloss, Presidents, Television, and ForeignCrises, p. 10.

17. Ibid.

18. Author interview with Marvin Kalb, ShorensteinCenter on the Press, Politics and Public Policy, HarvardUniversity, May 29, 1996.

19. Author interview with Tutwiler. See also Baker,Politics of Diplomacy, p. 34.

20. Nicholas Burns, Talking to the World AboutAmerican Foreign Policy, The Harvard InternationalJournal of Press/Politics, Fall 1996. Vol. 1, No. 4. pp.10-14.

Endnotes


21. Maureen Dowd, War Introduces a Tougher Bushto the Nation, The New York Times, March 2, 1991.How much this has proven to be the case is an openquestion.

More recently, Thomas Friedman wryly remarkedthat if the Vietnam syndrome was eliminated in 1991,then it has been replaced by the Gulf War syn-drome. The Gulf War syndrome says that the U.S.will engage in military operations abroad only if theytake place in a desert with nowhere for the enemy tohide, if the fighting can be guaranteed to last no morethan five days, if casualties can be counted on onehand, if both oil and nuclear weapons are at stake, ifthe enemy is a madman who will not accept any com-promise, and if the whole operation will be paid for byGermany and Japan. Thomas Friedman, GlobalMandate, The New York Times, March 5, 1995, p. 15.

On a more general level, the role of public opinionin U.S. foreign policy has been the subject of consider-able debate. See Donald L. Jordan and Benjamin I.Page, Shaping Foreign Policy Opinion: The Role ofTelevision," Journal of Conflict Resolution, June 1992,pp. 227-241; Philip J. Powlick, The Sources of PublicOpinion for American Foreign Policy Officials,International Studies Quarterly, Vol. 39, 1995, pp.427-451.

22. See B. Drummond Ayers, Jr., A Common CryAcross the U.S.: Its Time to Exit, The New YorkTimes, October 9, 1993, p. 1.

23. Ayers, A Common Cry p. 1. It may well havebeen the case that once the October battle occurredthe Clinton Administration seized the opportunity toextricate itself from a situation not of its choosing.The pictures were less an impediment and more anopportunity, unfortunate as it otherwise was.

24. John J. Fialka, Hotel Warriors: Covering the GulfWar (Washington, D.C.: The Woodrow Wilson CenterPress, 1991), p. 2.

25. Fialka, Hotel Warriors, p. 1.

26. RW. Apple, Correspondents Protest PoolSystem, The New York Times, February 12, 1991, p.14.

27. Editors Criticize Picture Limits, The New YorkTimes, February 21, 1991, p. 14.

28. Barrie Dunsmore, The Next War: Live?, TheJoan Shorenstein Center on Press, Politics andPublic Policy, Harvard University, March 1996, p. 9.Edward N. Luttwak has questioned whether televi-sion plays a role at all in creating an intolerance ofcasualties, suggesting it is a superficial explana-tion. He noted that the Soviets demonstrated suchan intolerance, despite the fact that Soviet televisionwas not allowed to show the sort of gruesome pic-tures that are of concern to some policy-makers inthe U.S. He instead suggested that the smaller aver-age size of families in contemporary post-industrialsociety has concentrated emotional familial capitalinto fewer children, therefore raising the level ofsacrifice in war to intolerable levels. Edward N.

Luttwak, Where are the Great Powers?; At HomeWith the Kids, Foreign Affairs, (Summer 1994), on-line without pagination.

29. Jarol B. Manheim, Managing Kuwaits ImageDuring the Gulf Conflict, and William A. Dorman andSteven Livingston, News and Historical Content: TheEstablishing Phase of the Persian Gulf Policy Debate,in W. Lance Bennett and David L. Paletz (eds.) Taken ByStorm: The Media, Public Opinion, and U.S. ForeignPolicy in the Gulf War (Chicago: University of ChicagoPress, 1994). See also Jarol B. Manheim, StrategicPublic Diplomacy & American Foreign Policy: TheEvolution of Influence (Oxford: Oxford UniversityPress, 1994).

30. W. Lance Bennett, Marginalizing the Majority:Conditioning Public Opinion to Accept ManagerialDemocracy in Michael Margolis and Gary Mauser(eds.), Manipulating Public Opinion (Pacific Grove,CA.:Brooks/Cole, 1989), pp. 320-361.

31. W. Lance Bennett and Jarol B. Manheim, Takingthe Media by Storm: Information, Cuing, and theDemocratic Process in the Gulf Conflict, PoliticalCommunication, 10, 331-351.

32. Jason DeParle, Long Series of Military DecisionsLed to Gulf War News Censorship, The New YorkTimes, May 5, 1991, p. 1.

33. Barrie Dunsmore, The Next War: Live?, p. 9.

34. Thanks to Jonathan Moore of the Kennedy Schoolat Harvard for helping me appreciate this importantdistinction.

35. Author interview with James A. Baker, III.,Houston, Texas, May 14, 1996.

36. Jessica Mathews, Policy Vs TV, TheWashington Post, March 8, 1994, p. A19.

37. Michael Mandelbaum, Foreign Policy as SocialWork, Foreign Affairs, January 1996, on-line withoutpagination.

38. Mandelbaum, Foreign Policy as Social Work,unpaginated.

39. Stanley Hoffmann, In Defense of Mother Teresa:Morality in Foreign Policy, Foreign Affairs, March1996, on-line without pagination.

40. Hoffman, In Defense of Mother Teresa.

41. United States Mission to the United Nations,Global Humanitarian Emergencies, 1996, p. 1.

42. David C. Unger, Taking Haiti, The New YorkTimes Magazine, October 23, 1994, p. 50. JimHoagland has also noted this whipsaw effect ofmedia coverage. A wave of emotion roiled up by hor-rific images that demand immediate action ispushed back by new sets of heart-rending images. Agovernment initially lambasted for callousness is sud-denly vulnerable to accusations of being foolhardy.This whipsaw effect is the politicians nightmare.Hoaglands larger point, however, is that this is a con-


dition created by the lack of a clearly articulated for-eign policy. (Jim Hoagland, Dont Blame CNN, TheWashington Post, March 3, 1994, p. 29.)

43. Andrew Natsios, Illusions of Influence: TheCNN Effect in Complex Emergencies, in FromMassacres to Genocide: The Media, Public Policy, andHumanitarian Crises, Robert I. Rotberg and ThomasG. Weiss (eds.), (The Brookings Institution, 1996), p.150.

44. Steven Livingston and Todd Eachus,Humanitarian Crises and U.S. Foreign Policy: Somaliaand the CNN Effect Reconsidered, PoliticalCommunication, Vol. 12, 1995, pp. 417.

45. Livingston and Eachus, Humanitarian Crises andU.S. Foreign Policy, p. 417.


47. Events in 1993 presented another story. CNN-as-impediment almost certainly explains the withdrawalof U.S. troops from Somalia. My only point here is thatCNN-as-agenda-setter does not serve as an acceptableexplanation for the intervention in Somalia in 1992.

48. A point made clear to me in interviews of foreigncorrespondents in Nairobi, Kenya in May 1994.

49. Steven Livingston and Todd Eachus, Too LittleToo Late: American Television Coverage of theRwanda Crisis of 1994, in Howard Adelman and AstriSuhrke (eds) Untitled (Boulder, CO.: Westview Press,forthcoming).

50. Steven Livingston, Suffering in Silence: MediaCoverage of War and Famine in the Sudan, in FromMassacres to Genocide: The Media, Public Policy, andHumanitarian Crises, Robert I. Rotberg and ThomasG. Weiss (eds.), (Washington, D.C.: The BrookingsInstitution, 1996), pp. 68-89. See also Howard Adelmanand Astri Suhrke, The International Response toConflict and Genocide: Lessons from the RwandaExperience, Early Warning and ConflictManagement, (York, Ontario: Centre for RefugeeStudies, York University), pp. 46-48.

51. The figures for Liberia are even more pronouncedwhen one takes into consideration that most of theprint news items, and nearly all of the broadcast newsitems, occurred in April and May 1996, at a time ofheavy combat and American evacuation missions tothe stricken capital of Monrovia. For The New YorkTimes, 43 percent (71 of 164) of the news items tomention Liberia between January 1995 and mid-May1996 were found in April and half of May 1996. ForABC News, nearly 88 percent (28 of 32) of its reportsfrom Liberia came in April and half of May 1996.

52. Why Bosnia in 1995 and 96? To address this ques-tion well would require another study. But certainlyhigh on the list of reasons for the coverage would beNATOs involvement and the deployment of U.S.troops there (as in Somalia).

53. Author interview with Brent Scowcroft,

Washington, D.C., March 27, 1996.

54. Author interview with James A. Baker, III,Houston, Texas, May 14, 1996.

55. Author interview with James A. Baker, III,Houston, Texas, May 14, 1996.

56. Larry Minear and Thomas G. Weiss,Humanitarian Policies (Headline Series, Foreign PolicyAssociation, 1995), p. 36.

57. Douglas Jehl, US is Showing a New Caution onUN Peacekeeping Mission, The New York Times,May 18, 1994, p. 1.

58. Each intervention type has been adapted fromRichard Haasss Intervention: The Use of AmericanMilitary Force in the Post-Cold War World (CarnegieEndowment Book, 1994)

59. See Nancy Ethiel (series editor) and StevenLivingston (ed), Reporting the Next War (Chicago: TheRobert R. McMormick Tribune Foundation, 1992), TheMilitary and the Media: the Continuing Dialogue(Chicago: The Robert R. McMormick TribuneFoundation, 1993).

60. Alexander L. George and Richard Smoke,Deterrence in American Foreign Policy: Theory andPractice (New York: Columbia University Press, 1974),p. 11.

61. Even the Reagan Doctrines claims to rolling backCommunist control was careful not to risk the totaldestabilization of Europe All-out war on the Europeancontinent was not an objective.

62. Author interview with Adm. Kendal Pease,Pentagon, September 1995.

63. Author telephone interview with Adm. KendalPease, March 12, 1996.

64. Richard N. Haass, Intervention: The Use ofAmerican Military Force in the Post-Cold War World(Washington, D.C.: Carnegie Endowment, 1994), p. 57.

65. Haass, Intervention, p. 62.

66. I was in Nairobi during this time and saw first-hand the attempts made by correspondents such asTerry Leonard of the Associated Press to get intoRwanda. It was simply too dangerous.

67. Of course, this formulation might be criticized asmissing the point that saving lives in political conflictis not apolitical in nature. Certainly in the case of theefforts to save the lives of largely Hutu refugees inGoma and elsewhere in Zaire, the results were notpolitically neutral. Some of these very refugees wereresponsible for the massacre of Tutsi and Hutu moder-ates in the weeks before.


The Joan Shorenstein Centeron the Press, Politics and Public PolicyJohn F. Kennedy School of Government

Harvard University79 John F. Kennedy Street

Cambridge, MA 02138Telephone (617) 495-8269 Fax: (617) 495-8696

Web Site Address: http://ksgwww.harvard.edu/~presspol/home.htm

r18 livingston

Documents