Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Stop the press : are the news media taking over foreign policy? Tettensor, Erin. 2000

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Item Metadata


831-ubc_2000-0283.pdf [ 4.31MB ]
JSON: 831-1.0089509.json
JSON-LD: 831-1.0089509-ld.json
RDF/XML (Pretty): 831-1.0089509-rdf.xml
RDF/JSON: 831-1.0089509-rdf.json
Turtle: 831-1.0089509-turtle.txt
N-Triples: 831-1.0089509-rdf-ntriples.txt
Original Record: 831-1.0089509-source.json
Full Text

Full Text

STOP THE PRESS: A R E T H E N E W S M E D I A T A K I N G O V E R F O R E I G N P O L I C Y ? by E R I N T E T T E N S O R B.A., University of British Columbia, 1998 A THESIS SUBMITTED IN PARTIAL FULFILMENT OF T H E REQUIREMENTS FOR T H E DEGREE OF MASTER OF ARTS in T H E F A C U L T Y OF GRADUATE STUDIES (Department of Political Science) We accept this thesis as conforming to the required standard THE UNIVERSITY OF BRITISH COLUMBIA April 2000 © Erin Tettensor, 2000 In presenting this thesis in partial fulfilment of the requirements for an advanced degree at the University of British Columbia, I agree that the Library shall make it freely available for reference and study. I further agree that permission for extensive copying of this thesis for scholarly purposes may be granted by the head of my department or by his or her representatives. It is understood that copying or publication of this thesis for financial gain shall not be allowed without my written permission. Department of Tp) rtf Ca \ <>^t>jr\r The University of British Columbia Vancouver, Canada Date A^cWM 3fT7\ DE-6 (2/88) Abstract In recent years, discussion surrounding the role of the news media in foreign policy -specifically whether the media do in fact exert measurable influence over the character and direction of foreign policy - appears to have coalesced around analyses of the so-called 'CNN effect'. This phrase refers to the notion that news media - television in particular -are 'driving', 'influencing', or even 'formulating' foreign policy in times of crisis. The supposed salience of the CNN effect has, in fact, become a virtual truism in political science. Careful scrutiny of the theoretical underpinnings of the CNN effect, however, reveals that the theory, while not entirely without merit, has yet to be borne out by empirical evidence. This is because the mechanism by which the CNN effect operates will complete itself only under extremely rare circumstances. Namely, the media are most likely to influence public opinion in precisely those circumstances where public opinion is least likely to influence foreign policy. This study will examine the assumptions that underpin the CNN effect and show them to be lacking. Then, using counterfactual reasoning, the case of Somalia will be reviewed in an attempt to ascertain if the CNN effect operated in that case. The discussion will show that, despite .the fact that Somalia is most often cited as the crisis that best demonstrates the operation of the CNN effect, this portrayal is inaccurate. In sum, the study will show that the CNN effect has yet to materialize in international politics. ii Table of Contents Abstract 1 1 List of Figures ' v Acknowledgements v Introduction 1 Chapter I: Media Autonomy 12 A. The Autonomy of the Fourth Estate 13 A . l Finding the Story: Sources for media reports 14 A.2 Agenda-Setting 16 A.3 Media as Sources of Information 20 A. 4 In Search of a 'Spin': Framing and editorializing 22 A.4.1 Framing 23 A. 4.2. Editorializing 25 Chapter II: The Media, the Public, and Foreign Policy 32 B. The Media and Public Information 33 B. l Convincing the Masses: Media influence over public opinion 34 B. 1.1 Framing 34 B.l.2 Editorializing 36 B. 2 Media Information and Public Knowledge 40 C. Public Opinion and Foreign Policy 43 C l The Defenders: Reviewing the "Almond-Lippmann Consensus" 44 C. 2 The Challengers 46 C.3 Trends in Foreign Policy-Making: In search of the CNN effect 50 Chapter III: The CNN Effect in Somalia 61 D. Somalia: The exception to the rule? 65 E. Testable Implications of Counterfactuals 68 E. l Media Reports 70 E.2 Analogous Cases 76 E.3 Other Considerations 77 E.4 Processes and Rationalizations of Policy-Making 79 F. The CNN Effect in Somalia: Some final thoughts 83 Conclusion 85 Bibliography 89 iii List of Figures 52 Figure 1 54 Figure 2 56 Figure 3 71 Figure 4 iv Acknowledgements The author wishes to thank the following people for their insights, comments, and support: Brian Job Don Kossuth, Will Bain, and Don Blake. I am especially grateful to Kal Holsti, who has been with me through every step of the process, and who taught me that the right questions are always more important than the right answers. Introduction On the morning of December 9, 1992 former American statesman and influential academic George F. Kennan tuned his television into the morning news. He had just awakened, and his day was barely underway when the images that confronted him from his television screen imbued him with a sense of horror. United States Marines were rushing onto the beaches of Mogadishu, Somalia, apparently ready to greet hostile forces. Instead, the men were accosted by the glaring lights of television cameras in the dark. Kennan's horror upon viewing this sight stemmed from two sources. First, he feared that the use of force by Western powers was a venture which was questionable on both normative and instrumental grounds. To Kennan, the operation would likely prove costly in blood and treasure, but futile in its attempt to secure peace in the region. Second, and even more fundamentally, he dreaded the process that produced the intervention in the first place. Later that day, Kennan laid out his fears - in detail - in a diary entry that would later be submitted to the New York Times: "There can be no question that the reason for [this action] lies primarily with the exposure of the Somalia situation by the American media, above all, television. The reaction would have been unthinkable without this exposure. The reaction was an emotional one, occasioned by the sight of the starving people in question. ... But if American policy from here on out, particularly policy involving the uses of our armed forces abroad, is to be controlled by popular impulses, and particularly ones provoked by the commercial television industry, then there is no place - not only for myself, but for what have traditionally been regarded as the responsible deliberative organs of our government, in both executive and legislative branches."1 Dire warnings such as these have enjoyed an increasingly prominent place in the writings of George F. Kennan. "Somalia, Through a Glass Darkly," The New York Times, Sept. 30, 1993, A25. 1 both academics and policy-makers in recent years, as the nature and extent of the news media's role in foreign policy is examined with increasing interest. In part, this increased interest in media influence reflects the rather sudden and remarkable rise of the Cable News Network and its peers. The rapid proliferation of satellite technology and real-time news reporting has certainly captured the interest of elites and laypersons alike. The Gulf War, for example, marked the first major conflict in which policy-makers and their publics watched events unfolding at the same time, a development which in itself warrants considerable attention.2 In addition to this, however, a sharp increase in both the frequency and visibility of humanitarian crises abroad has also piqued the curiosity of many regarding the influence of mass media over foreign policy. This is because the end of bi-polarity has meant a weakening of consensus - both between states and within them - regarding foreign policy. Without the clearly defined parameters of the Cold War, where the Enemy was explicitly identified and where the national interest was articulated in terms of justice and necessity, a much wider array of opinions regarding foreign policy, particularly in times of crisis, has emerged. As an increasingly diverse chorus of voices struggles to be heard, it is not surprising that scholars have developed a keen interest in distinguishing which voices are the most influential, and to whom they belong. The literature on the role of the media in this chorus of voices - that is, discussion surrounding whether the media do in fact exert measurable influence over the character and direction of foreign policy - appears to have coalesced around analyses of the so-called 'CNN effect'. This phrase refers to the notion that news media - television in particular - are 'driving', 'influencing', 2Marvin Kalb. "A View from the Press," in W. Lance Bennett and David L. Paletz (eds.) Taken By Storm: The Media. Public Opinion, and US Foreign Policy in the Gulf War. (Chicago: University of Chicago Press 1994), 6. 2 or even 'formulating' foreign policy in times of crisis. The strength of the assertion depends on the individual advancing it; as such there exists a variety of definitions of the term 'CNN effect'. However, defenders of the idea share the belief that the news media are capable of changing both the process of foreign policy-making and the eventual product of that process - i.e. the resulting policy itself. In brief, then, the CNN effect refers to the belief that media pressure results in "a loss of policy control on the part of government officials supposedly charged with making [foreign] policy".3 Ideas surrounding the mechanism by which this media influence operates tend to be unique to the scholar who champions them. We can usefully divide thought on this matter into two categories: The 'direct' camp and the 'indirect' camp. In the first case, scholars such as Bernard Cohen have argued that journalists themselves, by engaging in enlightened discussion in newspaper columns or on network news, directly influence the policy-making elite by independently contributing valuable ideas to the debate. This process makes use of an extensive network of personal relationships which exists between seasoned foreign policy correspondents and the policy-makers whose activities they report. Theorists belonging in this camp, however, are in the distinct minority, and the currency of this idea has faded in recent years. This is because the expertise that, at the time of Cohen's seminal work, accompanied the position of 'foreign correspondent' - the education and experience which qualified journalists to contribute meaningfully to the debate occurring within the policy-making elite - is now largely absent. Discussion of this direct method of influence, therefore, is increasingly rare. 3WarrenP. Strobel. Late-Breaking Foreign Policy: The News Media's Influence on Peace Operations, (Washington: United States Institute of Peace Press 1997), 5. Emphasis in original. 3 Much more frequently, the mechanism involved is thought to be an indirect one, originating in the media and reaching policy-makers via the public. It is the tide of public opinion, in other words, that is the proximate cause of policy change; the media comprise the force that directs this tide. Brian Buckley describes the process in the following way: [A] politico-humanitarian emergency erupts in a distant land; Western governments, who may or may not have been aware of the impending crisis, are reluctant to spend their blood and treasure in areas where they have no major interests in play; the news media, by forcing the citizens of the comfortable 'North' to confront the awful reality of wholesale human suffering, generate an irresistible emotional public demand that ' something must be done'; governments - often against their better judgement - bow to these pressures and eventually intervene to halt or reduce the carnage, with unknown long-term results.4 This a succinct summary of the mechanism by which the CNN effect is generally thought to operate. Moreover, it matches many other descriptions in tone as well as content. For it is not difficult to detect in Buckley's depiction more than a hint of uneasiness regarding the implications of such an approach to policy-making. This is characteristic of the vast majority of observations regarding the CNN effect. The idea that "televised images, selectively conveyed, are increasingly inducing policy-makers ... to formulate their foreign policy decisions on the basis of capriciously defined agendas, and in response to emotional rather than rational criteria" is most often characterized as a 'fear' or a 'danger'.5 Observers from Walter Lippmann to Winston Churchill, from Madeleine Albright to Ted Koppel, have decried the perils of allowing public opinion to direct policy. Must we, students of international politics, likewise become fearful of the future? Are we 4Brian Buckley. "The News Media and Foreign Policy: An Explanation," (Centre for Foreign Policy Studies, Dalhousie University 1998), 24. 5Denis Stairs. "Foreword," in Brian Buckley. "The News Media and Foreign Policy: An Explanation," (Centre for Foreign Policy Studies, Dalhousie University, 1998), viii. Emphasis added. 4 truly entering an era of popular control where publics, spurred on by the nightly news, compel their governments to embark on any number of ill-destined paths? Should someone 'stop the press' before it is too late? These questions are at the heart of this analysis. In the discussion that follows, I will address the theoretical underpinnings of the CNN effect and examine some of the most frequently cited evidence used to support it. I will show conclusively that the theory of the CNN effect, while not entirely without merit, has yet to be borne out by empirical evidence. This is because the mechanism by which the CNN effect operates will complete itself only under extremely rare circumstances. Consider again the way in which the media supposedly direct foreign policy: A crisis is brought to the attention of the public via the news media. Continued reports of human suffering create a strong tide of public opinion, which in turn pressures government to adopt a particular course of action - one which it may or may not favour. Three core assumptions underpin this idea. First, we must assume that the media constitute an autonomous institution, one which conducts the business of journalism relatively unfettered by outside influences, and which engages in routine editorializing.6 Second, public opinion must be primed and galvanized by the media; press reports must be capable of creating a strong, coherent tide of public opinion that would not have existed in the absence of media coverage. Finally, we must assume that public opinion is factored significantly into the foreign policy decisions of statespersons. In other words, public opinion must be capable of producing a measured policy shift, one that departs from the course favoured by elites. Each of these assumptions is problematic on its own; I will explore this in greater detail 6 To suggest otherwise would require us to reduce the argument further, since this would imply the existence of a force which drives the media. 5 below. More critically, however, it is extremely unlikely that all three assumptions will obtain in any given case. This is because, as this analysis will show, news media are most likely to influence public opinion in precisely those circumstances where public opinion is least likely to influence foreign policy. The disjointed character of foreign affairs coverage, coupled with the relative disinterest of the public in such matters, ensures that when sudden and unexpected crises erupt, the vast majority of the public is caught off guard.7 As Yoel Cohen argues, in such circumstances, the public, "exposed to a sudden torrent of facts after years of not reading about the area, is at a loss in understanding the significance of the event".8 Therefore, public attitudes regarding a crisis situation will tend to be a great deal more malleable than opinions regarding more familiar and commonplace news items, such as taxes, crime, and the economy. In such circumstances, news media will have more opportunity to pursue their own agenda. This said, media cannot simply 'snap their fingers' and expect immediate public sympathy. It takes time to generate public interest in a foreign policy issue, even when dramatic crises, such as those in Rwanda or Somalia, erupt. The likelihood of significant media influence over the public, however, is inversely proportional to the likelihood of significant public influence over government decisions. This is because public opinion is most susceptible to media suggestion with respect to issues which are unfamiliar, and which bear less obvious consequences for the everyday lives of average people -namely, foreign policy. At the same time, however, foreign policy is an aspect of governance that 7 Most studies estimate that less than 15% of the public pays attention to international affairs on a day-to-day basis. See, for example, Michael Gurevitch. "The Globalization of Electronic Journalism" in James Curran and Michael Gurevitch (eds.) Mass Media and Society. (London: Arnold Publishers 1996), 204-224. 8Yoel Cohen. Media Diplomacy: The Foreign Office in the Mass Communications Age. (London: Frank Cass & Co. 1986), 46. 6 is traditionally regarded as resting primarily within the domain of elites. While one must be cautious of making careless generalizations about the role of public opinion in foreign policy, one can make two assertions without much controversy. First, because public interest in foreign affairs is far less than interest in domestic affairs, all things being equal, popular pressure on the government to adopt a particular course of action on any given international issue will be comparatively weaker than on any given domestic issue. As such, governments will tend to consult public opinion less regarding issues of foreign policy. Second, when all other things are not equal - i.e. in times of sudden crisis - the public will usually have far less time to attempt to influence government decisions. Public pressure does not manifest itself immediately; letters to Members of Parliament, Congresspersons, or heads of state take time to accumulate. Activists need time to coordinate with other like-minded individuals. Couple this delay with the interval necessary for public interest and attitudes to form, and we see a substantial time lag between the first reports of a crisis and any real opportunity to influence government (assuming such influence is possible). In addition, as we will see, it is usually government that brings major international issues to the attention of the news media, and not the other way around.9 Thus, before journalists even begin to deliberate about whether or not a given foreign event constitutes 'news' - indeed, often before the crisis actually breaks - statespersons are already in the process of forming plans to address probable outcomes and contingencies. In other words, governments usually have a 'head start' when it comes to deciding what action, if any, to take. This shortens the interval for potential public sway over the 9See, for example, Warren P. Strobel. Late-Breaking Foreign Policy, (Washington: United States Institute of Peace Press 1997); Shanto Iyengar & Adam Simon. "News Coverage of the Gulf Crisis and Public Opinion: A Study of Agenda-Setting, Priming, and Framing," in Shanto Iyengar & Richard Reeves (eds.) Do the Media Govern? Politicians, Voters, and Reporters in America, (Thousand Oaks: Sage Publications 1997), 248-257. 7 initial government response still further. Of course, none of this negates the possibility that public pressure could be factored into future decisions, or even reverse the initial policy implemented by the government. What is suggested by the time lag in public response, combined with the 'head start' that governments usually have, is that the shorter the interval between the onset of a crisis in a foreign country and the clear articulation of an initial government policy in response to that crisis, the less likely it is that public opinion will be a significant factor in the process leading to that policy. At this point, one might object that governments do not always respond to foreign crises in a timely way. What about those crises in which government response is sluggish, or where a clearly articulated policy is not manifest? In cases where there exists little consensus within elites about the most appropriate course of action, would not public opinion 'exploit the breach', thus becoming influential? The answer to this question depends heavily on timing. A lack of consensus within the policy-making elite is not exploitable if no force exists to take advantage of it. Since public opinion takes time to coalesce into a solid power, popular pressure may not build in time to have any real impact. The longer this lack of consensus obtains, however, the more likely it is that public attitudes will be taken into account. Does this mean that, in cases such as the war in Bosnia-Herzegovina, where the responses of Western governments were slow to materialize, the CNN effect will operate to the fullest? This is not necessarily the case. Recall my argument that the news media are most likely to influence public opinion in precisely those circumstances in which the public is least likely to influence foreign policy. The reverse is also true: Public opinion is most likely to influence foreign policy in those circumstances where media are least likely to influence public opinion. This is because, as a crisis 8 continues over time, media sway over public opinion plateaus and eventually begins to decline. As people pay closer attention to a given event, learning more about its history and subtleties, they begin to form their own opinions, as opposed to those planted in their minds by media accounts. This assertion ought to hold an instinctive appeal. Moreover, I will examine a substantial amount of evidence to indicate that these diminishing returns do in fact materialize. At what point this begins to occur will depend on a variety of factors, most notably the extent to which the public is familiar with analogous events. Put another way, if the public is able to draw analogies between a current crisis and a past one, some pre-formed opinions will exist. Thus, media influence will begin to drop off sooner than if the current crisis is viewed as completely unfamiliar. What this means is that, in order to attribute public activism to media influence, the power of news media over the public must be at or near its height at the same time as public sway over the government is high enough to affect foreign policy. However, the former tends to be at its zenith early in a crisis, while the latter usually reaches its apex at a much more mature stage. If strong popular pressure does not generally coincide with strong media influence, it will be difficult to witness the operation of the C N N effect. To summarize: The C N N effect currently enjoys a notoriety which is entirely disproportionate to its potential to enrich our understanding of international politics. This study will show that the theoretical underpinnings of the concept are flawed, and their successful application unlikely. The analysis will begin by discussing each of the three main assumptions that form the basis of the C N N effect, showing how each is controversial. Chapter One will examine the policy-making process, and the various ways in which news media could potentially intervene. Do the media form an autonomous institution? To determine this, I will explore the role of the media in 9 agenda-setting, information provision, and policy debate. In so doing, I will attempt to determine what path media influence is most likely to take. Chapter Two scrutinizes the complex relationships between the media and the public, and between public opinion and foreign policy. To what extent are the media capable of influencing public opinion, and under what circumstances is this most likely to occur? Similarly, to what extent is public opinion capable of altering foreign policy, and what conditions are most favourable for this influence? Drawing on conclusions reached from this discussion of the three core assumptions of the CNN effect, I will argue that the process by which the CNN effect operates is extremely delicate, and that there are many factors which militate against its successful completion. I will thus demonstrate why it is highly questionable simply to presume that all three assumptions obtain in any given case. 'Highly questionable' though it may be, however, this does not mean it is impossible for the CNN effect to function. Therefore, Chapter Three will explore a case study: The US response to the humanitarian disaster in Somalia. This crisis is among those most commonly cited as demonstrating the CNN effect in operation.10 I will examine the cogency of this argument by posing two counterfactuals. First, if the media had devoted less attention to the crisis in Somalia, the Bush Administration would not have launched Operation Restore Hope. Second, if the dramatic reports and graphic images of US casualties had not been broadcast all over the globe, US troops would not have been withdrawn when they were. Determining testable implications of these counterfactuals 10Warren P. Strobel. Late-Breaking Foreign Policy, (Washington: United States Institute of Peace Press 1997); See also Warren P. Strobel. "The Media and US Policies Towards Intervention: A Closer Look at the CNN Effect," in Crocker, Chester A. & Fen Osier Hampson with Pamela Aall (eds), Managing Global Chaos, (Washington: United States Institute of Peace Press 1996), 356-378; and Johanna Neuman. Lights. Camera. War. (New York: St. Martin's Press 1996). 10 will enable a careful analysis of the empirical evidence. In this way, I will address the likelihood that media influence was ultimately behind US policy in this case. Before delving into this discussion, however, an important admission is in order. The analysis that follows, while it does to a large extent apply to all industrialized democracies, will be overwhelmingly American. This is an unavoidable feature of any discussion of the CNN effect, for three main reasons. First, the concern that mass media drive policy is particularly relevant to the US in its capacity as the single remaining superpower. In the past decade, the United States has been at the forefront of most major humanitarian interventions involving the use of force, and this condition is likely to persist, at least into the near future. Second, research on the role of public opinion in foreign policy - a major component of this study - has been dominated by American authors. Finally, many would argue that the American media (including, of course, CNN) represent a special case. According to Brian Buckley, for example, the US media "still comprise the largest and most important component of the 'international press'".11 Therefore, while it would of course be desirable to avoid a national bias in this discussion, such a task would be impossible. Brian Buckley 1998, 39. 11 Chapter I: Media Autonomy To argue that the media drive policy is to suggest that the process of policy-making itself is somehow altered through media intervention. This belief is shared by all defenders of the CNN effect. Opinions diverge, however, on the issue of exactly how the process is ostensibly affected. For some, media influence manifests itself primarily in the form of agenda-setting and/or information-gathering. This more subtle form of guidance is not necessarily even intentional: The media, in the course of determining what constitutes 'news', create an implicit agenda which is passed on to policy-makers via the public. In addition, because the public receives nearly all its information through the media, the manner in which events are 'framed' by media reports will influence public opinion, and hence foreign policy. For some scholars, however, media influence penetrates even deeper. According to these observers, not only do the media, following Bernard Cohen's famous maxim, dictate what the public thinks about, they go so far as to tell the public what to think. These theorists argue that news media independently contribute to policy debate, editorializing to such an extent that people accept media 'opinions' as their own. We can make this distinction more clearly by simplifying decision-making into the following formula: The stimulus is introduced, relevant information is collected and reviewed, and a decision is reached which satisfies a particular minimum standard.12 According to the 'weak' account of the CNN effect, the news media intervene in the early stages, as a particular level of priority is being assigned to a given crisis, and information is being gathered and processed. In the 'strong' version, 12In some cases, of course, the 'minimum standard' may equal the optimal choice. In other cases, satisficing may occur. 12 the media not only become involved in these early stages, but they also intervene in the process during its more mature stages, as options are being weighed. Obviously, these two approaches to media influence are very different. While the end result they presume is more or less commensurate - public opinion is guided by news media - their respective estimations of intent are not alike. In the weak version, media influence is, for the most part, an unintentional product of journalists going about the business of providing news in an engaging fashion. In the strong version, the media are portrayed as actively pursuing an agenda and taking deliberate measures to get the public onside. The assumption shared by both the weak and the strong versions is that the news media constitute "an autonomous institution standing apart from the institutions of state power".13 The media must be portrayed as such in order for the CNN effect to make sense. For if the media were not autonomous, then we would be forced to trace the causal chain back yet further, since responsibility for altering policy would ultimately rest with whatever body guides the media. The question thus becomes: Is this a valid assumption? Are the media in fact the independent institution presumed by proponents of the CNN effect? A. The Autonomy of the 'Fourth Estate' According to supporters of the CNN effect, the news media seek out stories and prioritize them in a manner corresponding to their relative 'newsworthiness'. Proponents of the strong version further argue that the media develop policy preferences and engage in editorializing to promote these policy 1 3 Daniel C. Hallin. The Uncensored War: The Media and Vietnam. (Oxford University Press 1986), 7. 13 preferences to their reader- or viewership. It is this (highly American) conception of day-to-day journalism that prompts David D. Newsom to write: "[T]his free institution ... forms the crucible in which policy issues are shaped in the nation".14 However, characterizing the media as an essentially autonomous institution is problematic, for several reasons. First, it is relatively unusual for journalists independently to locate a story. Second, the context which the media provide - the 'spin' that accompanies the story - also originates largely from outside sources. Finally, addressing the additional assumptions of the strong version, editorializing in the media is rarer than many suggest. This is particularly the case with television, which is the medium most often credited with driving foreign policy. A . l . Finding the Story: Sources for media reports Journalists tend to rely quite heavily on outside sources for information, especially official sources.15 For example, in their study of television coverage during the Gulf War, Iyengar and Simon found that over 50% of all reports examined "emanated directly from official spokespersons".16 A study by Sigal reveals similar findings: Only 24% of all channels for news in the New York Times and the Washington Post - America's two most reputable newspapers for foreign news - could be attributed to reporter 'enterprise', (such as interviews or the reporters' own analyses). Four percent came 14David D. Newsom. Diplomacy and the American Democracy. (Indiana University Press 1988), 52. 1 5 Strobel 1997, 8; see also Robert M . Entman & Benjamin I. Page. "The Iraq War Debate and Media Independence," in W. Lance Bennett & David L. Paletz (eds.) Taken by Storm: The Media, Public Opinion, and US Foreign Policy in the Gulf War. (Chicago: University of Chicago Press 1994), 82-101. 1 6 Iyengar & Simon 1997, 256. 14 directly from other media sources. The remaining 72% were attributed to official proceedings, press conferences, press releases, "background briefings", and other government-initiated events.17 In sum, as Mermin argues, "most studies have found that American journalists turn to politicians and government officials for guidance in deciding what constitutes news".18 The lesson we must derive from this evidence is that media generally do not bring issues to the attention of policy-makers; rather, the reverse trend is the norm. If the current administration signals that a particular issue is a priority, it will be considered a 'story' by the media - provided, of course, that the issue is 'newsworthy'.19 Alternatively, an issue can be brought to the attention of journalists by lower-level government officials (such as Members of Parliament or Members of Congress), members of interest groups, concerned NGOs, or other activists seeking to bring an issue to the attention of the public or the administration. In sum, therefore, media cannot be characterized as wholly independent when it comes to locating potential 'stories'; instead, they are influenced -sometimes heavily - by other interested parties. This is not to suggest that media are merely the tools of government, reporting essentially what they are told to report. Media still exert considerable autonomy in deciding which issues, once brought to their attention, are actually newsworthy. As Strobel points out, "[n]o matter how hard they try, government officials will have difficulty persuading the networks to cover the Middle East 1 Clarice Olien, George A. Donahue & Philip J. Tichenor. "Conflict, Consensus, and Public Opinion," in Glasser, Theodore L. & Charles T. Salmon (eds.), Public Opinion and the Communication of Consent. (New York: Guildford Press 1995), 304. 1 8 Jonathan Mermin. "Television News and American Intervention in Somalia: The Myth of a Media-Driven Foreign Policy," Political Science Quarterly, 112:3 (Fall 1997), 387. 1 9 Strobel 1997, 64. 15 peace process, trade, or nonproliferation policy absent a dramatic crisis".20 Even the President of the United States cannot take air time for granted.21 Media are still, after all, fundamentally concerned with selling themselves, and will therefore highlight events and issues that are likely to attract the public's interest. It is this practice in the media - the prioritization of events according to their relative newsworthiness - that leads many to argue that the media independently "set the agenda", both for the public and for policy-makers. As Strobel puts it, "where the camera is and what it captures determines whether the crisis is on officials' agenda and among the public's collective foreign policy consciousness".22 The assertion that news media are instrumental in agenda-setting is not new; indeed, it is accepted by many observers as given. Like so many things that are taken for granted, however, the concept of media agenda-setting in the realm of foreign policy warrants closer attention. A.2. Agenda-Setting "The most readily evident power of the media in crisis," writes Brian Buckley, "is the capacity to force an item on to the policy agenda, or to give greater prominence to one already there."23 This supposed power of the media to set the agenda, despite its apparent status as one of the 'truths' about media influence, means a variety of different things to different people. To some, "the assertion is 2 0 Strobel 1997, 64. 2lIbid. 22Strobel 1996, 106. 23Buckley 1998, 41. 16 that the priorities of the media agenda influence the priorities of the public agenda".24 According to this conception, the more frequently a crisis appears as a 'news item', the more likely it is that the public will consider the event a priority. From this point, we are to assume that the public agenda will in turn affect the foreign policy agenda of elites. To others, the process is more direct. "If there is a crisis [somewhere] arid the newspapers and television networks are sending the bulk of their correspondents and crews to that area, that will be at the top of the foreign policy agenda. The resultant coverage will create the questions, the congressional hearings, and the public concern that will force the US government to comment or respond through action."25 According to this understanding, the media agenda directs the public agenda and the official agenda. In other words, depending on which definition of agenda-setting is being used, sometimes the agenda in question is the public's, and sometimes the government's. In some conceptions, the media independently add crises or events to the agenda; in other versions, they merely shuffle the hierarchy. While each of these rather different versions bills itself under the heading 'agenda-setting', however, it is a mistake to conceive of them as the same phenomenon. ' Some versions of agenda-setting are a good deal more probable than others. Take, for example, the assertion that if the news media are concentrating heavily on a particular crisis, it "will be at the top of the foreign policy agenda". This claim, quoted above, is particularly sloppy. A causal relationship between media coverage and 'the foreign policy agenda' is clearly implied, yet the exact nature of the relationship is unclear. Obviously, we are meant to accept prima facie that 24Maxwell McCombs & George Estrada. "The News Media and the Pictures in our Heads," in Iyengar, Shanto and Richard Reeves (eds), Do the Media Govern? Politicians. Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 237. 25Newsom 1988, 59. 17 the mere correlation of these two phenomena confirms the existence of a causal relationship. In other words, we are meant to assume that because a given crisis is important to the media, it is important to policy-makers as well. Such an inference, however, is not warranted by a mere illustration that if a crisis is on the agenda of media, it will tend also to be on the agenda of governments. The reason that some observers of the CNN effect feel comfortable making such casual and imprecise claims is that they purport to draw on the well-established literature of agenda-setting. What these scholars are actually doing, however, is condensing a sophisticated and nuanced body of work into an over-simplified series of catch-phrases. For what proponents of this version of agenda-setting fail to acknowledge is that most of the existing literature on media agenda-setting examines the relationships between media, the public, and the government in the context of domestic politics. This is important because the assumptions that form the foundation of domestic agenda-setting do not necessarily apply to the realm of foreign policy. In particular, once we wander out of the domestic sphere, the intimidating hydra that embodies the relationship between public opinion and foreign policy again rears its many heads. Therefore, we cannot be casual about asserting that news media "set the agenda" of elites in the context of foreign policy, relying on pre-existing literature as proof enough. For the pre-existing literature examines a very different environment, and as such it is not entirely analogous. Thus, in order to accept that such agenda-setting does indeed take place in the realm of foreign policy, we require a degree of substantiation that is absent in most accounts. Another definition of agenda-setting which appears in the context of the CNN effect - one that is infinitely more modest - claims that media priorities affect public priorities. What the public 18 is concerned about (what is on its agenda) is circumscribed by what the public is aware of. Therefore, exposure becomes the key issue. Since the vast majority of the public is exposed to international politics exclusively through the news media, the question of which events the media cover, and with what frequency (i.e. the media agenda), will condition the public agenda. Even in this more reasonable conception of agenda-setting, however, there are problems. For, as we have seen, most studies have found that journalists "turn to politicians and government officials for guidance in deciding what constitutes news".26 Therefore, we cannot accurately maintain that the media are ultimately responsible for what arrives on the public's foreign policy agenda, let alone extend the argument further to include the official foreign policy agenda. All we can claim is that news media, placing a premium on "the dramatic, the confrontational, and the violent", shuffle their own agenda in such a way as to appeal to their reader- or viewership.27 In so doing, media regulate the degree of public exposure to an issue, thereby influencing the priority accorded to that issue, relative to others in the public consciousness. Since public awareness is a function of exposure through the media, however, only those issues that remain as active stories for long enough will register as a priority to the public. But the media agenda is fickle, fluctuating continually as journalists seek out the next big story. Once a crisis has been a headline issue for a certain period of time, absent any major response from the government, it becomes "old news", losing its place of privilege. In many cases, therefore, the government itself, in deciding whether a crisis is a priority for the current administration, determines whether an issue remains on the media's agenda, and by extension the public's. 26Mermin 1997, 387. 27Buckley 1998, 44. 19 In sum, then, we cannot confer on the news media the status of autonomous institution by virtue of their power to "set the agenda" in foreign policy. For once this much-touted capability is carefully scrutinized, it begins to lose much of its potency. The news media are not independently responsible for what arrives on the public's foreign policy agenda, reliant as they are on official sources for stories. Nor are the media totally autonomous in deciding what priority an event will be accorded in the public consciousness. If the role of the media as agenda-setters is ambiguous, however, their role as providers of information is not. Once a story is brought to their attention, network news organizations -particularly satellite networks - quickly assume the capacity of information provider to the public. Indeed, the advent of 'real time' reporting has meant that news media often serve as information providers for statespersons as well. Scholars such as Hindell have argued that, since media are an important source of information for policy-makers, they therefore exert a kind of tacit influence over policy itself.28 It is thus important to examine the role of the news media in information provision. A.3. Media as Sources of Information In one of the early episodes of The West Wing, a television drama about the White House, the President of the United States (Martin Sheen) is preparing to address the nation following an American strike on Syrian military targets. He is pacing up and down the Oval Office, agitated because the CIA cannot yet confirm that US bombers have indeed achieved their objective. Finally, exasperated, he cries, "Will somebody call CNN and find out if we hit anything?!". Moments later, 28Keith Hindell. "The Influence of the Media on Foreign Policy," International Relations (1995), 78. 20 CNN informs the White House that, yes, the missiles did in fact hit their targets. Breathing a sigh of relief, Martin Sheen then goes live on television to declare the result. This scenario may sound like Hollywood taking liberties in its depiction of US politics. However, the scene was probably taken from an actual event that transpired shortly after Bill Clinton had been sworn in as President. Clinton's first military command was to order a Tomahawk strike on Baghdad's intelligence agency in retaliation for the planned assassination of George Bush. "At 6:30 pm, Tom Johnson, the president of CNN, was able to tell the President of the United States that the Tomahawks had indeed hit their target. The commander-in-chief then went live on CNN ... to announce the news to the people who had told him."29 This event was not an isolated incident. Many White House officials have confirmed that news media play an essential role in information-gathering. This is particularly true when crises erupt. As Marlin Fitzwater has explained, "[I]n most kinds of international crises now, we virtually cut out the State Department and the desk officers ... Their reports are still important, but they don't get here in time for the basic decisions to be made".30 Former President George Bush has gone so far as to claim, "I learn more from CNN than I do from the CIA". 3 1 It goes virtually without saying that the same is true for the public. Most of us do not experience international politics firsthand, at least not in a way that we perceive. However much or little we know about events in foreign countries, the information on which our knowledge is based 29Richard Reeves. "Overview," in Shanto Iyengar and Richard Reeves (eds.) Do the Media Govern? Politicians. Voters, and Reporters in America, (1997, 5. 30Quoted in Frank J. Stech. "Winning CNN Wars," Parameters, (Autumn 1994), 37-56. 31Ibid. 21 comes almost exclusively from indirect sources. For the vast majority of people, the primary source - indeed, the only source - of information about international affairs is the news, whether in print, on the radio, or on television. As such, public opinion "is not shaped by direct experience of politics, but through the images planted in our minds by news accounts".32 What this means is that the way in which news media present the information - that is, how the facts are 'framed' - may affect the way the public perceives a given event.33 And, since the media are critical sources of information for elites as well, how issues are framed may also exert a tacit influence on policy-makers. Whether or not the media can fairly be characterized as independent in placing a particular 'spin' on a story thus becomes a critical issue. A.4. In Search of a 'Spin': Framing and editorializing Although media are routinely supplied with stories, they exert more autonomy when it comes to deciding how to present those stories. This is due to the ability of media to place a particular 'spin' on the issue at hand. It is this ability that prompts proponents of the strong version of the CNN effect to argue that media intervene at the more developed stages of decision-making, when options are being advanced and weighed. This interference can arrive in either of two ways: It can occur as an inevitable result of the way in which the issue is 'framed', or it can be a result of editorialization. 'Editorializing' and 'framing' are often conflated, but they are not synonymous. 'Framing' refers to the manner in which the issue is characterized, while 'editorializing' occurs when journalists 3 2 W. Lance Bennett. "Cracking the News Code: Some Rules that Journalists Live By," in Iyengar, Shanto and Richard Reeves (eds), Do the Media Govern? Politicians, Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 103. 3 3McCombs & Estrada 1997, 237. 22 engage in comparatively open, subjective commentary on the issue. A.4.1 Framing Some argue that framing is the more effective of the two in swaying the public, since it operates under the guise of objectivity. As Entman explains, "frames are difficult to detect fully and reliably, because many of the framing devices can appear as 'natural', unremarkable choices of words or images".34 Audiences believe they are being exposed only to the "facts". However, framing is always present, for "[a]s storytellers, journalists inevitably frame the events they report".35 Domke et. al define framing as "highlighting some dimensions of issues while excluding other aspects".36 In deciding which elements to highlight, important dimensions of a particular issue are often omitted, either because they do not fit the 'angle' a journalist has chosen for the story, or because they remain on the "cutting room floor", sacrificed in the name of time constraints or consumer appeal. The result of this process is to produce a very specific conception of the issue at hand. It should be noted, however, that framing need not be intentional. In other words, certain dimensions of an issue may be highlighted without any conscious decision to do so. Indeed, the very nature of news reporting can in itself constitute a way of framing events. This is because news reports tend to be largely crisis-oriented, and often provide little context for the events they describe. 34Robert M . Entman. "Framing US Coverage of International News: Contrasts in Narratives of the K A L and Iran Air Incidents," Journal of Communication, 41:4 (Autumn 1991), 6. 3 5 Mermin 1997, 388. 36Daniel Domke, Dhavan V. Shah & Daniel B. Wackman. "Media Priming Effects: Accessibility, Association, and Activation," International Journal of Public Opinion Research, 10:1 (Spring 1998), 32. 23 Some believe that the effects of this type of unintentional framing can be significant. For example, Richard Nixon believed that the episodic reporting of the Vietnam war, devoid as it was of any theme or context, went a long way towards making Vietnam a political albatross.37 He wrote: In each night's T V news and each morning's paper the war was reported battle by battle, but little or no sense of the underlying purpose of the fighting was conveyed. Eventually, this contributed to the impression that we were fighting in military and moral quicksand, rather than toward an important and worthwhile objective".38 Thus, certain methods of reporting can frame issues in a particular way without any intention of doing so. Whether intentional or not, framing tends to be consistent throughout the various media; journalists tend to adopt similar approaches when reporting on particular issues.39 It is likely such consistency is in part a result of the inter-reliant nature of journalism; reporters often take cues from one another, both in terms of what constitutes 'news', and the way in which that news will be presented. However, the primary source for frames is - not surprisingly - the source of the stories themselves; i.e. political elites. In constructing frames, journalists have a "heavy dependence on elite sources"40 This is particularly the case with regard to international news. Framing, then, is not an activity which is free from government influence, as some analysts imply. Indeed, some policy-makers are extremely adept at 'feeding' frames to reporters, thus determining, to a large degree, the overall character of reporting on a particular issue. Some have 37Hallin 1986, 7. MIbid, 3. 39Entman 1991. "Ibid, 7-8. 24 suggested that during the GulfWar, for example, US General Colin Powell subjected the media "to the most sophisticated massage in the history of Pentagon salesmanship", resulting in an overall body of reporting that Sam Donaldson once described as dismayingly "jingoistic".41 The potential effects of such framing on the public, whether intentional or unintentional, are a source of some debate; I will return to this below. A4.2 Editorializing Events can also acquire a particular ' spin' as a result of editorializing in the media. Editorializing occurs when media deliberately inject subjective commentary into the presentation of an issue in order to convey a specific impression. In other words, when members of the media hold especially strong opinions on a given issue, they may feel compelled to contribute to the debate. This may take the form of explicit declaration - either supporting or criticizing current government policy - or it may manifest itself in more subtle ways, such as giving preferred airtime to like-minded activists. Editorializing, many argue, is a powerful agitator of public opinion. However, there is reason to dispute the supposed salience of editorializing, for two main reasons. First, such editorializing is not as common as some scholars would have us believe. For one thing, the media tend to be strongly committed, at least in theory, to the ethic of the "objective reporter".42 Editorializing is therefore not considered appropriate for the average journalist. Moreover, the public increasingly relies on television as its main source of news about foreign affairs; indeed "television is the more likely source for international news even for individuals who 4 1Kalb 1994, 3-4. 42Strobel 1997, 9. 25 regularly seek such information from newspapers".43 However, television is too "episodic and event-oriented" to provide much in the way of context or commentary. Thus, editorializing is not as common on television as it is in newspapers. What editorializing does occur on television is found mainly on network news, since local newscasts are comparatively brief of foreign affairs coverage. However, studies of American news ratings have consistently illustrated that the vast majority of the public does not rely on network news, turning instead to local newscasts as their primary source for news.44 The net result of these three facts is that a large section of the public is not exposed to a great deal of editorializing. Second, the number of permanent news bureaus abroad has been radically reduced, both by television networks and by print media.45 As a result, network news divisions are relying as never before on "freelance video footage and commentary from foreign stringers, some of whom have dubious connections. When regular correspondents are used, too often they are 'parachuted' into the latest strife to air knowing reports via satellite within hours".46 This tendency to utilize 'parachute journalism' creates a situation where journalists, many of whom will have little experience or knowledge of foreign affairs, arrive in the middle of a crisis with no sense of the history or context of the problem. This point - the way in which foreign affairs are covered by modern news organizations -4 3Guo, Zhongshi & Patricia Moy. "Medium or Message? Predicting Dimensions of Political Sophistication," International Journal ofPublic Opinion Research, 10: 1 (Spring 1998), 37. 44Hallin 1986, 107. 45See Strobel 1997; Neuman. Lights. Camera. War. 1996. 46Hoge, James R. Jr. "Media Pervasiveness," Foreign Affairs, 73:4 (July-Aug. 1994), 143. 26 is an extremely important one, and yet it is an issue which goes virtually unmentioned in almost all accounts of the CNN effect. Obviously, a journalist who is not familiar with foreign affairs in general, let alone the history or complexities of a particular crisis, is not in a very favourable position to contribute independently to policy debate. Those who advance the theory of the CNN effect do not seem overly eager to recognize that the days of Walter Lippmann are long passed; very few modern journalists - particularly television journalists - are informed enough to critique government policy.47 While many who write about media influence over policy still cite Bernard Cohen's seminal work, The Press and Foreign Policy, these scholars conveniently forget that the process identified by Cohen bears little resemblance to the mechanism by which the CNN effect supposedly operates today. In large part, this is because the foreign correspondent of Cohen's day was a very , different sort of character from today. According to Cohen, most of the foreign correspondents in 1963 (at the time of his writing) were well educated, usually in the fields of political science, economics, history, or international relations. Many had post-graduate degrees. "The foreign affairs reporter in Washington", he wrote, "is also a man of long experience in the field of international affairs".48 For Cohen, media influence stemmed mainly from the nationally syndicated columnists, since these columnists usually had extensive personal relationships with policy-makers, and commanded respect "as a by-product of their reputations ... for being well-informed".49 Clearly, there is a qualitative difference between the mechanism by which media influence ostensibly operates today and the process discussed by Cohen. ^Ibid.; see also Neuman. Lights, Camera, War, (New York: St. Martin's Press 1996). 48Bernard Cohen. The Press and Foreign Policy. (Princeton University Press 1963), 18. *9Ibid, 41-44. 27 Public opinion goes virtually unmentioned in Cohen's book, while the "well-informed man of long experience in the field of international affairs" does not appear in modern literature on the C N N effect. Meanwhile, the (not-so-well-informed and relatively inexperienced) parachute journalists who are the subject of CNN effect enquiry, lacking knowledge of their own, must turn to others to provide commentary on the events being reported. The commentary which is most available to the media tends to flow from official sources. As such, what editorializing does occur typically casts government policy in a favourable light. Thus Hallin writes: "[journalists tend to act as 'responsible' members of the political establishment, upholding the dominant political perspective and passing on more or less at face value the views of authorities assumed to represent the nation as a whole".50 Reeves agrees: "[MJajor news organizations routinely accept the assumptions and assertions of policy-makers, and these assumptions indelibly colour news reports".51 This is not only the case when the issues at hand are relatively low-profile or uncontroversial. Even when there exists considerable room for criticism, the media tend not to be adversarial. For example, in a study of news coverage of the US invasion of Panama in December 1989, Soderlund et. al found that media reports "tended to present a positive picture of the invasion in both text and visual material", even though the invasion itself was questionable on moral and legal grounds, and was subject to considerable international criticism.52 Not only did American media toe the line, but 50Hallin 1986, 10. 51Richard Reeves in Iyengar & Reeves 1997,101. 52Walter C. Soderlund, Ronald H. Wagenberg & Ian C. Pemberton. "Cheerleader or Critic? Television News Coverage in Canada and the United States of the US Invasion of Panama," Canadian Journal of Political Science, 27:3 (Sept. 1994), 596. 28 Canadian media as well tended to put a positive spin on the invasion.53 If the media always backed government policy, however, we would never witness criticism emanating from journalists. Obviously, this is not the case. The question thus poses itself: Why do media back official policy in some instances, but not others? Under what circumstances do the media stray from the 'party line'? The vast majority of scholars argue that the level of criticism in the media is inversely proportional to the degree of consensus within the policy-making elite.54 Even observers who are skeptical of the CNN effect agree that if policy is weak or amorphous, media scrutiny that casts government policy in a negative light tends to increase in salience. Most of these scholars admit, however, that the criticism or controversy which is found in the media tends to echo disputes among the policy-making elite, remaining well within the bounds of the debate going on within the political establishment.55 Media appear critical when there are divisions within the government because there is more than one version of the story to recount. In other words, opposing views are mainly being reported by the media, not generated by them. Even during the Vietnam war - arguably the most famous case of media dissention - media criticism could be traced largely to outside sources. During the war, the press drew primarily on officials and soldiers for information. But as the war went on, "these sources became much more divided. ... The news 'reflected' these divisions".56 Here again, therefore, the media cannot be characterized as totally autonomous, since the criticism they do advance tends to originate from elsewhere. "Ibid 54Hallin 1986; Strobel 1996, 1997; Hoge 1994. S5Hallin 1986, 10; Entman & Page 1994. 56Ibid. 29 In sum, therefore, editorializing within the media is not as prominent as many suggest; what editorializing does occur tends to be largely supportive of government policy. However, if official policy is weakly articulated, or if it is the subject of debate among policy-making elites, the media are much more likely to advance opposing opinions. Even so, the opposing opinions advanced by the media are usually drawn from outside sources, as opposed to having originated with the journalists themselves.57 What conclusions can we draw from this discussion of media autonomy? Clearly, there are major problems with characterizing the media as an essentially free-standing, autonomous institution. The media rarely come up with stories independently. Their role as independent agenda-setters is in doubt. While they do exhibit more independence in placing a particular 'spin' on the events they report, even this process is subject to considerable outside influence. Both the weak and the strong versions of the CNN effect are compromised by this result. Clearly, we cannot claim that media are ultimately responsible for the direction or character of foreign policy if the media themselves are subject to influence from other forces, particularly when the 'other sources' in question are mainly official sources. Why are proponents of both versions of the CNN effect seemingly oblivious to this fact? First, the theoretical underpinnings of the C N N •effect are sparse, to say the least, and have been subject to little real scrutiny. In the main, observers rely on purely anecdotal evidence to support their views. Second, the vast majority of commentators on the subject are journalists themselves; it is not surprising that they are less than sensitive to the possibility that the "Fourth Estate" is not independent. For while it is appealing to ascribe a high "The origin of criticism is not merely a semantic point; if criticism originates primarily from outside the media, then we cannot argue that the media as such are responsible for the reaction such criticism engenders. 30 level of importance to one's profession, it is considerably less appealing to recognize that one's activities are subject to guidance from outsiders. Third, and most fundamentally, the problem stems from an over-simplified and fundamentally inaccurate characterization of government - namely, that it is a unitary, homogenous body which experiences no significant divisions within itself. If the government is a unitary body, then the notion of a cycle of influence, from government to media to government, is nonsensical. However, once we recognize that the government, like any large body, is host to a diversity of opinion on any given subject, we see that media can be - and are - used as a forum for debate between different factions of government, or between the government and interest groups or NGOs. According to this conception of policy debate, the media (as the word originally implied) are primarily a platform for communication, not the source of it. Clearly, this is a vitally important role to play. But it is a far cry from being the initiators, or even the champions, of policy. One might object that, rare though it may be, journalists can be the instigators of discussion on a subject. For example, media may take an interest in a crisis in some foreign land that, from the perspective of the policy-maker, has no impact on the interests of the state. One might further argue that even if media are supplied with the story, and even if the 'spin' which accompanies that story is influenced by outside sources, it is not dictated by these sources. Therefore media are at least partially responsible for the resulting public opinion. This rebuttal is only, valid, however, if the media are in fact capable of measurably swaying public opinion on any given issue. This assumption will be the subject of enquiry in Chapter Two. 31 Chapter II: The Media, the Public, and Foreign Policy Before delving into the issue of media influence over the public, it would be prudent to make an important admission: The relationship between mass media and public opinion is dauntingly complex. In order to satisfy ourselves that we had exhausted every possible source of information on the subject, we would have to address decades of psychological research and reams of data on news reports, public opinion surveys, and the statistics derived therefrom. Obviously, one cannot accomplish all these things in a study of this sort. All one can do is draw on the tentative generalizations found within public opinion discourse in an effort to uncover trends which can help guide our discussion. With this disclaimer in mind, I turn to the second unspoken assumption which uhdergirds the CNN effect: Press reports must be capable of creating a strong, coherent tide of public opinion which would not have existed in the absence of media influence. The theory of the CNN effect depends heavily on the ability of the media to stir up public opinion, whether intentionally or unintentionally. In other words, in those rare cases where the media independently engage in criticism of official foreign policy, this criticism must be capable of 'moving' the public. In cases where news reports acquire their 'spin' in a more subtle fashion - i.e. from framing - public perception must be measurably affected by this portrayal. The media's actual ability to influence public opinion in this way is taken for granted by proponents of the C N N effect. But, as we shall see, this presumption in favour of media power may not be entirely warranted. 32 B. The Media and Public Opinion Strangely enough, the vast majority of analyses regarding the CNN effect lack any substantial discussion of media influence over the public. Instead, these analyses simply assume that the public will embrace a policy position advanced by the media. This supposition rests fundamentally on a very particular - and very common - characterization of public opinion. This portrait of public opinion depicts it as volatile, unstable, and subject to 'mood swings'. Furthermore, it is common to argue that "the political beliefs of the mass public lack a real structure or coherence"; these features combine to render public opinion fundamentally malleable.58 According to this view, it would take a relatively small nudge to guide the public in a certain direction. However, there is mounting evidence that this portrayal may be inaccurate. Recent studies of American public opinion have suggested that "mass opinion in the aggregate is in fact characterized by a good deal of stability, and that this is no less true of foreign policy than on domestic issues".59 Indeed, many of these studies not only characterize public opinion as "strong" and "stable", but also as "permissive" in the aggregate with respect to foreign policy.60 Studies examining public opinion at the individual level have revealed similar findings.61 In short, rather than portraying the public as a source of volatile mood swings, it is probably more accurate to 5801e R. Holsti. "Public Opinion and Foreign Policy: Challenges to the Almond-Lippmann Consensus," International Studies Quarterly, 36:4 (Dec. 1992), 442. S9Ibid.; Donald L. Jordan & Benjamin I. Page. "Shaping Foreign Policy Options: The Role of T V News," Journal of Conflict Resolution, 36:2 (June 1992), 227-241. 6 0Ole R. Holsti. Public Opinion and American Foreign Policy. (University of Michigan Press 1996), 40. "Christopher H. Achen. "Mass Political Attitudes and the Survey Response," American Political Science Review 69 (1975), 1218-1231. 33 "identify the public as a source of moderation and continuity"62 How do news reports affect this relatively stable body of opinion? In brief, "it is not clear how much, or in what ways, the news will affect [people's] opinions".63 Despite this, recent public opinion research has identified some trends, and it is to these that I now turn. B.l . Convincing the Masses: Media influence over public opinion As I have argued, there are two ways in which media can affect public opinion: Framing or editorializing. Framing is likely to be the more effective of the two, since it cloaks itself in the guise of objectivity. Still, studies of the effects of media framing have failed to turn up conclusive evidence that framing is capable of drastically altering public opinion in any detailed way. Indeed, what generalizations have been made indicate that this may not be possible, let alone likely. B.l . l Framing A study by Domke demonstrates typical findings regarding the impact of media framing on public opinion. They argue that "media framing of issues ... seems likely to foster priming effects" in consumers of mass media news.64 "Priming" refers to the process by which "activated mental constructs can influence how individuals evaluate other concepts and ideas."65 Individuals process information through what are called "cognitive structures"; these are mental frameworks that guide 62Holsti 1996, 45. Emphasis in original. 63Hallin 1986, 107. 6 4 D o m k e e £ al 1998, 32. 65Ibid.,3\. 34 the way in which we interpret events. These cognitive structures are based on past experience, and are relatively stable. Since judgements often depend less on the entire repertoire of people's knowledge and more on which aspects of their knowledge happen to come to mind, activation of a particular mental construct - priming - will affect the way in which we interpret a particular event.66 In short, "because of selective perception, the media will tend often to merely reinforce people's existing attitudes".67 For example, Hallin cites a 1968 study which found that 75% of respondents who considered themselves 'hawks' regarding the Vietnam war believed that Walter Cronkite was also a 'hawk'. Somewhat more thought Huntley and Brinkley were 'hawks'. However, a majority of 'doves' believed that each of the three anchors was a 'dove'. Thus, concludes Hallin, reports on the Vietnam war were "ambiguous and contradictory enough that both hawks and doves could easily have found material to support their own views of the war".68 This finding is especially interesting given the fact that the Vietnam war, as already noted, is the example most often cited to support the theory that media can drive public opinion. Framing, therefore, may have a relatively negligible effect on the substance of public opinion on a given issue. As a comparatively subtle form of placing 'spin' on a story, it is usually vague enough to allow people to "hear what they want to hear". This is not to over-state the case; it is likely that framing has some effect. What is considerably less likely is that, even if these effects could be quantified, they would be demonstrably influential enough to warrant being credited with 'driving' public opinion. Framing is most adept at placing issues in a particular normative light, 66Ibid, 33. 67Hallin 1986, 107. 6«Ibid. 35 characterizing actors or events as 'good' or 'evil', 'moral' or 'immoral'. This is not the same, however, as determining the public's policy preference. In other words, the press had little difficulty persuading the public that Serb aggression in Bosnia, for example, was 'wrong'. This is not the same as convincing the public that, say, American troops should be committed overseas, and it is this latter type of influence that the CNN effect assumes. Moreover, as I have already pointed out, it is problematic to characterize the framing of political issues - particularly international ones - as an activity in which the media are independent. Just as journalists rely heavily on political elites to supply their stories, so too do they rely on these elites to provide the context and frame for those stories. Therefore, even if we assumed that framing was capable of directing the policy preferences of the public, it would still be inaccurate to credit this direction to the media themselves. B.1.2 Editorializing What about editorializing, a far more obvious form of commentary on political issues? Again, and for many of the same reasons, research shows that this may have only a marginal effect on public perceptions of an issue.69 If this assertion seems to go against the prevailing wisdom on the subject, consider the 'evidence' on which the prevailing wisdom is based. Once again, the most famous piece of evidence used by those who argue that media drive public opinion (when these observers even bother to present any) is the case of American public opinion during the Vietnam war. What we appear to see in this case is a clear-cut example of media dissention provoking public dissention on a foreign policy issue. What we actually see is a clear-cut example of the post hoc ergo propter 69Holsti 1992. 36 hoc fallacy: After it, therefore because of it. There is no conclusive evidence that there exists a causal relationship between opinions expressed in media coverage of the Vietnam war and public opinion regarding that war. All that research on press reports regarding Vietnam has shown is that public opinion and media opinion were correlated. This alone does not confirm a relationship, let alone causation. Of course, it is likely that the two are related in some way. However, it is more reasonable to postulate that the same stimuli produced both phenomena, rather than simply to assume that one caused the other. In other words, it is more conservative to suggest that public anger over the Vietnam imbroglio was provoked by what was actually occurring in Southeast Asia, and not what the press was saying about it. After all, was it really necessary for the media to convince the American public that the number of US casualties was horrific? One might object that even if overt commentary from journalists was not needed to incite public disenchantment, media were still responsible for the reaction by virtue of their role in presenting shocking and bloody images of the conflict. Indeed, it is precisely this type of account of media influence that makes up the bulk of the argument supporting the existence of the CNN effect. Pictures are worth a thousand words, and they can move the public like no other medium. According to this view, it was these images of the atrocities in Vietnam that were the proximate cause of the outrage. Proponents of the CNN effect further argue that the media's conscious decision to dwell on such images constituted a form of editorializing; therefore the media were ultimately responsible for the public's anger. However, as Hoge points out, attributing public horror "to bloody televised images ignores findings that fewer than 2% of television presentations during the war 3 7 showed any blood.". A study by John E. Muller supports Hoge's claim. He contrasts poll data regarding public support for the Vietnam war, which was televised, and the Korean war, which was not. He found virtually identical patterns of decreasing public support over time. Daniel Hallin's research reveals similar findings: Support for the Korean war decreased as casualties increased.71 Obviously, we cannot attribute these results to shocking television images, since there were none. In short, careful scrutiny of public opinion data regarding both wars implies that "[t]he fact, not the image, of body bags was the important factor".72 Supporters of the CNN effect theory fall victim to the post hoc ergo propter hoc fallacy in their observations of modern conflicts as well. This is due in part to the fact that pro-CNN effect analyses of modern crises typically neglect to seek out other potential explanations for observed phenomena, satisfied as they are with their own. However, this is hardly a sound approach on which to base our conclusions. In the absence of any convincing evidence, therefore, it is unreasonable to claim that the news media alone direct the course of public opinion. All we can assert is that it is conceivable that media framing or editorializing can, in some circumstances, contribute to the character or direction of that opinion. Under what circumstances, then, is such media influence most likely? All other things being equal, media sway over public opinion is likely to be greatest in the early to medium stages of a crisis. The logic behind this assertion is simple: The less information or experience the public has 7 0Hoge 1994, 141. 71Hallin 1986. 7 2Hoge 1994, 141. 38 with a particular crisis, the greater the extent to which it will accept at face value opinions expressed in the news. As I have shown, the cognitive structures through which we interpret events are based on past experience. Therefore, the less exposure we have had to analogous events in the past, the more ill-defined and flexible those mental frameworks will be.73 By contrast, the more familiar the public is, either with a particular event, or with events that it considers analogous, the more pre-existing attitudes will exist. And because media framing, as we have seen, tends to "reinforce people's existing attitudes", public opinion will therefore be less malleable. A good example of this is the largely negative media attention directed at the recent sex scandal involving US President Bill Clinton. Despite negative framing (with coverage focusing overwhelmingly on the widespread condemnation in Congress), Clinton's approval rating actually rose in the midst of the media frenzy.74 The public clearly expressed a "mind of its own" in this case. The issues at stake were of a type with which the vast majority of people are reasonably familiar; individuals were thus able to make up their own minds on the subject without media guidance. In sum, as W. Phillips Davison argues, "[t]he mass media do not seem to have a very great effect on well-established attitudes. When people have strong political or religious preferences, even large amounts of information or propaganda are unlikely to shake them".75 Obviously, then, I am drawing a relationship between public information on and exposure to an issue, and the extent 7 3 It should be noted that it is probably not necessary that two or more events are actually analogous, only that a person considers them to be so. 74Khaled M . Batarfi. "Media Foreign Policy and Public Opinion," (Doctoral Thesis Chapter 1998), 7 5 W. Phillips Davison. Mass Communication and Conflict Resolution: The Role of the Information Media in the Achievement of International Understanding. (New York: Praeger Publishers 1974), 10. 39 to which public opinion will be pliable. What generalizations can be made about public information on foreign affairs?76 B.2. Media Information and Public Knowledge The general public's knowledge about foreign affairs can only be characterized as poor. Only about 15% of the public - the so-called "attentive public" - devotes its attention to foreign affairs on a regular basis.77 Couple this with the fact that most foreign affairs coverage (particularly in local newscasts) tends to be disjointed and crisis-oriented, and it becomes obvious that the vast majority of people will have little initial knowledge to draw on when a crisis breaks out. Once the media have had a chance to direct public interest towards a newly-erupted crisis, therefore, they will wield maximum influence over public opinion. This holds true unless the crisis is considered by the public to be a 'relapse' or an outgrowth of a previous conflict, such as the situations in Burundi in 1995, Kosovo in 1999, or Chechnya in the same year. In such cases, we can expect that public knowledge would at the outset be slightly higher than it was with the initial crises (in Rwanda, Bosnia, or Chechnya, respectively). Public knowledge would be only slightly higher because, as Bennet has shown, even in times of war, when coverage of one particular area is extensive, the public's knowledge of foreign affairs 76Because this analysis draws mainly on American literature, the section on public knowledge of foreign affairs will apply only to the American public. Most industrialized countries, including Canada, rank higher than the US when it comes to public knowledge of foreign affairs. Therefore, the generalizations made here should not be applied strictly to other countries, although they will apply to some extent. 7 7 01ien£* al. 1995. 40 - even with respect to a particular area - does not measurably increase.78 In other words, studies of public polls indicate that although the public may become more familiar with where and who over the course of a conflict, what and why, especially the latter, are largely missed. The main reason why the public fails to reach more than a superficial understanding of international crises is that an overwhelming majority of the public relies on television as its primary source of international news. Most research suggests that television is poorly suited to increasing people's knowledge of complex issues.79 A recent study by Guo & Moy attempts to discern the impact of television news on people's "political sophistication", which they define as a combination of three factors: Political salience (interest and attentiveness to political information); political knowledge; and political conceptualization (cognitive skill).80 Guo & Moy conclude that television news is only a valuable source of information for those who are already comparatively well-informed. Due to the brief, fragmentary nature of television news coverage, "viewers must devote greater mental activities in order to put the episodic pieces together into a meaningful whole. [Therefore], television news is likely to benefit only the 'initiated', or sophisticates equipped with appropriate schemas for filling in the missing information".81 However, Guo & Moy argue that television is highly effective in increasing "political 78Stephen Earl Bennet. "The Persian Gulf War's Impact on Americans' Political Information," Political Behaviour, 16:2 (June 1994), 179-201. 7 9 G u o & M o y 1998,29. mIbid, 26-27. nIbid., 30. 41 salience", or interest and attentiveness to political affairs.82 So while it may be a challenge for the public to glean more than superficial information from television coverage of a crisis, the longer the public is exposed not only to nightly news reports, but to reports in newspapers, magazines, on the radio and over the Internet, the more informed it will become. Quite obviously, if given sufficient time, the public learns enough to form a policy preference; otherwise we would never witness demonstrations or petitions regarding international affairs. How long it takes for this level of sophistication to be reached will depend on any number of factors, including the extent to which the public is able to draw analogies between the current crisis and previous events. As time goes by, public information increases. However, the longer a crisis stretches on, the less we are able to attribute public opinion to media influence. In other words, the media experience diminishing returns; their sway over public opinion declines as public information increases, however slowly. The more information and experience people possess with respect to a crisis, the more developed their cognitive structures will become. And, as I have shown, well-developed mental frameworks tend to be rather rigid. People thus become less susceptible to media framing, and they begin to "hear what they want to hear". In brief, people begin to make up their own minds. To summarize: No convincing evidence exists that either framing or editorializing in the news media have an impact that comes anywhere near "creating" or "driving" public opinion. Indeed, public opinion research suggests that such activities in the media have limited influence on public views. While it seems reasonable to assume that opinions broadcast in the media have some effect, those effects will reach their zenith in the early to mid-stages of a crisis. They will plateau and eventually decline as a crisis matures, since the public will begin to form its own opinions about «2Ibid, 25. 42 the issues involved - opinions which are not easily altered. Clearly, therefore, we cannot accept the second major assumption of CNN effect theorists, that media are responsible for the character or direction of public opinion. The most we can acknowledge is the likelihood that media contribute in some way to public views, along with many other sources of opinion, such as generation, family or peer groups, gender, socio-economic status, and the like. If there are serious difficulties associated with the first and second core assumptions of the CNN effect - those regarding media autonomy and the ability of news reports to create or direct public opinion - the third core assumption is even more problematic. Proponents of the CNN effect simply assume that public opinion is capable of influencing foreign policy in times of crisis to such an extent that policy-makers can be forced to adopt a course of action which they do not favour. As with the first two assumptions, the logical underpinnings of this claim go completely unaddressed. They will be addressed here, however, and we shall find them lacking. C. Public Opinion and Foreign Policy The assumption that public opinion exerts a powerful influence over foreign policy, even in times of crisis, is arguably the most critical plank on which the theory of the CNN effect is based. One would expect, therefore, that scholars who advance the theory would take great pains to establish the credibility of this assumption, vital as it is. Surprisingly, however, few discussions of the CNN effect attempt to wrestle this particular issue. Instead, the presumption is left more or less unspoken, sidestepping entirely an enormous body of research. Perhaps this avoidance is due to the complexity of the task; analysts have been trying to determine the degree to which public opinion affects foreign policy for decades. The issue is a 43 perplexing one, owing in part to "the great difficulty in untangling the extent of reciprocal processes. The effects of policy on opinion, for example, or officials' efforts to educate or manipulate the public".83 Whatever the reasons, the impact of public opinion on foreign policy is one of the most challenging questions in both international relations and public opinion research. It is therefore necessary, if anticlimactic, to state at the outset that a conclusion on this matter has not been reached. Quite the opposite, in fact; it would not be unfair to suggest that experts today are further from agreement on this issue than ever before. This is because the "mature consensus" that once existed is eroding. Assumptions that were previously taken for granted do not always appear to withstand empirical scrutiny.84 Nevertheless, it is still worthwhile to examine this erstwhile consensus, since many scholars still hold it to be true. C l The Defenders: Reviewing the "Almond-Lippmann Consensus" Interestingly, unlike proponents of the CNN effect - who assume that public opinion does influence foreign policy - most foreign policy analysts have traditionally concluded that it does not. Manheim, for example, argues that "[fjoreign policy is the more or less private preserve of a small group of so-called experts and, except for a few headline issues, the public is routinely and easily excluded from the process".85 This represents a fairly typical attitude. Since the public is relatively uninterested and uninformed regarding foreign affairs, international issues rarely become election issues. As 83Jordan & Page 1992, 227. 84Holsti 1992. 85Jarol B. Manheim. "Going Less Public: Managing Images to Influence US Foreign Policy," in Iyengar, Shanto and Richard Reeves (eds), Do the Media Govern?, (Thousand Oaks: Sage Publications 1997), 382. 44 such, policy-makers have no real motivation for including the public. The unqualified masses are thus quite content with letting statespersons make important foreign policy decisions, and statespersons are quite secure in the knowledge that they will not fall victim to the vagaries of emotionalist appeals. Ole R. Holsti refers to this perspective as the "Almond-Lippmann Consensus", after Gabriel Almond and Walter Lippmann, both of whom authored seminal works - in the 1950s and 1920s, respectively - regarding public opinion and foreign policy. Holsti identifies these works, and others like them, as being steeped primarily in the realist tradition, which is "intensely skeptical of the public's contribution to effective foreign policy".86 Realists have long argued that the domestic sphere and the international sphere are not analogous; thus the idea that a democratic state does not consult its public regarding international affairs is not inherently contradictory. In the realist view, [T]he public might be sufficiently informed to deal with local issues that impinge on their daily lives, but foreign affairs are too remote from their experience, and in any case they have little inclination to become more informed about such complex and remote issues. Finally, the effective conduct of diplomacy requires secrecy, flexibility, and other qualities that would be seriously jeopardized were the public to have a significant impact on foreign policy.87 This perspective is found many pivotal works of the inter-war and early post-war periods, such as Harold Nicolson's Diplomacy and Hans Morgenthau's Politics Among Nations. After World War II, a consensus began to develop in the field, informed largely by this attitude. According to Holsti, this consensus centred on three major propositions: (1) Public opinion is highly volatile; (2) public Holsti 1992, 440. 45 attitudes on foreign affairs are so lacking in structure or coherence that they might best be described as "non-attitudes"; and (3) public opinion has a very limited impact on the conduct of foreign policy.88 Clearly, if we accept the post-World War II consensus, the theory of the CNN effect collapses completely. For if the public cannot actually change the course of decision-making in foreign policy, the question of whether the media are actually responsible for public opinion becomes completely moot. This is not the whole story, however; if it were, we would end the discussion here. But there are a number of researchers who are increasingly skeptical of the Almond-Lippmann consensus. While I cannot hope to give these researchers their due in the context of this discussion, it is important to sketch briefly some generalizations based on their contribution to the study of public opinion and foreign policy. C.2. The Challengers The Almond-Lippmann consensus no longer commands exclusive loyalty from foreign policy analysts, if indeed it ever did. Another camp does exist, and it appears to be growing. In this camp, scholars point to new evidence which, while largely anecdotal, issues a strong challenge to many of the traditional suppositions.89 The first two of these suppositions, as identified by Holsti, have already been addressed. I have shown evidence suggesting that public opinion is in fact characterized by a good deal of stability with respect to both domestic matters and foreign affairs. I have also examined some elements of cognitive psychology which show that people interpret «*Ibid, 442. 89See, for example, Holsti 1992 and 1996; Jordan & Page 1992; Soderlund 1994. 46 events through frameworks that are structured on past experience. Indeed, public 'moods' have been shown to be, in the main, rather predictable responses to stimuli (i.e. real world events).90 The notion that voters are concerned exclusively with domestic issues has also come under fire. If foreign policy issues do not become electoral issues, as argued by realists, then statespersons can exclude the public with relative impunity. A study by Aldrich, however, challenges the accuracy of this idea. Their findings indicate that foreign policy issues did indeed have significant effects on presidential campaigns over a thirty-year period.91 Of course, these results might register aspects of voter behaviour which are no longer present. The team's study was conducted over the span of the Cold War, and as such major foreign policy issues would likely have revolved around relations with the Eastern Bloc, which was characterized quite plainly as 'the enemy' '. Since the end of the Cold War, in the absence of a proximate threat, foreign policy issues may not appear to have the immediacy that existed during the period being researched. Challenges to the third proposition, with which our enquiry should be most concerned, are the weakest. This is because none of these anti-consensus studies has thus far been successful in establishing a clear and systematic link between public opinion and the character of a particular foreign policy choice. Anecdotal evidence abounds, and incidents that appear to show a correlation are dutifully identified by scholars who wish to show that public opinion is factored into the decisions of statespersons. Unfortunately, this is not sufficient. For, as Holsti points out, even "[a] finding that major decisions seem to be correlated with public preferences does not, by itself, 90Holsti 1992, 446. 9 1 John H. Aldrich, John L. Sullivan & Eugene Borgida. "Foreign Affairs and Issue Voting: Do Presidential Candidates 'Waltz Before a Blind Audience?" American Political Science Review 83 (1989), 123-141. 47 establish a causal link".92 Attempts to move beyond spurious correlations have, thus far, not been entirely convincing. This is not to imply that, in the absence of conclusive evidence to the contrary, we should embrace the position that public opinion is not important to foreign policy-makers. This would be every bit as unwarranted as blindly asserting the opposite. Indeed, we are unlikely to glean from this discussion any real satisfaction on the subject, for it will be quite impossible to resolve the question here. A truly adequate examination of the emerging literature would be much more detailed. It is outside the scope of this study, however, to conduct such an examination; this would divert us from our purpose. The important point to recognize is that the mere existence of such widespread disagreement within the relevant literature is itself enough to suggest that theories which simply assume a causal relationship ought to be regarded with suspicion. Let us not be accused, however, of throwing the baby out with the bathwater. If there is even a possibility that public opinion can measurably influence foreign policy, then we are ill-advised to reject the CNN effect proposition outright. As there does appear to be such a possibility (since I have not proven otherwise), we should attempt to determine what conditions are most favourable for it to occur. In other words, when is public opinion most likely to affect the foreign policy decisions of statespersons? To put it succinctly, "all other things being equal, the more protracted the decision process, the more likely are policy-makers to be subjected to the impact of public opinion".93 Public pressure does not manifest itself immediately. Interested members of the public must first be exposed to an issue for a length of time that allows them to become sufficiently interested in it. (The length of time 92Holsti 1992, 453. 93Ibid, 444. 48 necessary will depend on any number of factors, ranging from the degree to which an individual feels personally affected by the issue to how heavily the subject is covered by the media.) Those who are inclined to become activists must then find and coordinate with like-minded individuals.94 Activists must write letters to government representatives, hold demonstrations and fund-raisers, and conduct a myriad of other ventures designed to get the attention of government officials. Finally, those Members of Congress or Parliament who are brought onside by such activities must then adopt tactics of their own in an attempt to persuade higher levels of government to adopt a policy change. An important point to note here is that public pressure is unlikely to be strong in the absence of any support from some sections of government. The public registers its disapproval largely through lobbying lower levels of government, persuading members of these lower levels to appeal to the Executive on its behalf. Thus, public sway is in part a function of consensus within elites. If private activists are unable to persuade lower levels of government to oppose a particular policy, their power will be blunted. Needless to say, the degree of consensus within elites may have little to do with the personal beliefs of politicians and much more to do with their estimation of potential ramifications at election time. In any case, each of these stages of public pressure takes time. Thus, a decision that is rendered relatively quickly is less likely to have been influenced by this process than one which was delayed. What does this mean for the CNN effect? Quite simply, since international crises are "usually characterized by short decision time", they are far less likely to have been influenced by 94It should be noted that media are critical in this capacity, providing a link for activists to find and learn about each other, and acting as a forum for them to broadcast their activities. 49 public lobbying.95 Pressure groups gain momentum over time, but since crises by their very nature usually require prompt (sometimes overly prompt) responses, the degree of influence at the time of the decision will usually be low. C .3. Trends in Foreign Policy-making: In search of the CNN effect It is time to recapitulate the argument thus far. I have shown that the news media cannot accurately be described as autonomous, reliant as they are on outside sources for both the stories they cover and, although to a lesser extent, the 'spin' that accompanies those stories. I have examined studies which suggest that public opinion is neither created nor driven by the media; indeed, opinions broadcast in the media may have a very limited impact on public opinion. Finally, I have argued that it is inappropriate simply to assume that public opinion affects foreign policy, and that this is especially true in times of crisis. These results are grave indictments against the validity of the CNN effect, since they indicate serious flaws with all three of its core assumptions. Still, improbabilities are not impossibilities. All I have done thus far is identify tendencies and salient trends. I have not foreclosed the possibility that in some rare cases, these inclinations will not apply. For example, it is not always the case that, in times of crisis, decisions are made in a timely fashion. When consensus within elites is low, what passes for policy may be nothing more substantial than rhetorical declarations made to keep critics at bay. In sum, then, all I have effectively demonstrated up to this point is that each of the three core propositions is a great deal less plausible than proponents of the CNN effect assume. The critical 95Holsti 1992, 444. 50 question thus becomes: Are those cases which constitute exceptions to one assumption likely to be exceptions to all of them? In order for the CNN effect to work, the same case must slip through the cracks all three times, bucking the trends identified in each of the core assumptions. In other words, the degree of media autonomy, their influence over the public, and public sway over foreign policy must all be high at the same time in order to attribute policy to the CNN effect. The main contention of this study, however, is that this is very unlikely ever to occur. To access the reasoning behind this assertion, one need only put the pieces of the argument together. First, media autonomy is related to the degree of consensus within elites. When consensus is low, media exert more autonomy, since they are able to choose between camps, rather than being forced to accept at face value the positions put forward by policy-makers. Second, public influence over foreign policy is also partly related to the level of consensus among policy-makers; the former is highest when the latter is lowest. Obviously, then, media autonomy and a relatively high degree of public influence over foreign policy will tend to coincide. If we were to stop the discussion right here, we would conclude that, when consensus within elites is low, the CNN effect will increase in salience. In fact, most scholars who have examined the CNN effect argue precisely this.96 Thus, Hoge writes, "[i]n the absence of a persuasive government strategy, the media will be catalytic".97 This conclusion is flawed, however, because the analysis leading up to it is incomplete. For public influence over foreign policy is also a function of time. The longer the decision process, the more likely it is that public pressure was a factor. However, turning to the third assumption, the degree of media influence over public opinion is also partly a function of time. The more protracted 96See, for example, Hallin 1986; Strobel 1996 and 1997; Hoge 1994. 9 7Hoge 1994, 138. 51 the crisis, the less we are able to attribute public opinion to media influence; as the public becomes more informed, it is less likely to accept at face value opinions broadcast in the media. In short, the news media are most likely to influence public opinion in the early to mid-stages of a crisis, when public opinion has not yet had a chance to weigh in to the policy-making process. Conversely, public opinion is most likely to be factored into foreign policy in the later stages of a crisis, when media influence over the public is already in decline. To see the point more clearly, refer to Figure Figure 1 A D 98It is critical to note that, like a demand curve in economics, the curves in Figure 1 do not represent quantitative measurements. The exact height or precise time is not important; significance lies instead in the shapes of these curves relative to the curves of the Figures that follow. 52 This figure illustrates the time line which must occur in order to witness the operation of the CNN effect. A crisis breaks at point 'A'. Elite consensus builds quickly, and remains steady for a time. For some reason, however - say, the state we are examining has sent a peacekeeping force, and that force suddenly sustains an unusually high number of casualties - the current policy comes under fire at a relatively early stage in the time line. Therefore, consensus begins to decline rapidly. Meanwhile, media influence over the public increases quickly in the early stages, as the public gains interest due to its exposure to media reports. In the mid-stages, this influence plateaus, remaining constant for some time. As the crisis matures still further, people begin to form their own opinions as their information and experience with the issues increase. Thus, the media curve eventually begins to decline, albeit slowly. As for public influence over foreign policy, this builds slowly through the early to mid-stages of the crisis as the cause gains notoriety and activists attempt to coordinate their activities. At line 'D', however, consensus within elites begins to drop. Since public influence over policy is related to the degree of consensus, the former increases sharply at approximately the same time. As we can see, therefore, the public curve, the media curve, and the consensus curve all come very close to intersecting with one another in zone 'B'. Media influence over the public is still high at the point where public opinion becomes a force so consequential that politicians ignore it at their own peril. The shapes of all three curves are instrumental in producing this result. However, it is the consensus curve which is most important to watch. In a more typical case, the consensus curve would not drop off so early in the crisis, and the decline would be much less abrupt. Therefore, since the public curve is partly dependent on the consensus curve, the former would not increase so sharply at such an early stage. In a more typical case, therefore, the three curves would not come anywhere 53 near intersecting, and the CNN effect would not operate. An illustration of this type of case can be seen in Figure 2. Unlike Figure 1, this time line represents an actual crisis; Kosovo in 1999. Several differences between these two figures should be immediately apparent. First, the media curve does not quite reach the height it achieved in Figure 1, and it begins to decline much earlier. This is because the public (not unreasonably) is likely to have viewed Kosovo as a conflict that was intimately related to the previous conflict in Bosnia. Since people were able to draw significant analogies between these two conflicts, a certain amount of pre-formed ideas and opinions surrounding the Kosovo situation will have existed. Therefore, media influence over public opinion will not have reached quite the same level as with other conflicts, and it will have begun to decline sooner. 54 The second readily apparent difference is that public influence over foreign policy increases more rapidly from the onset of the crisis in Figure 2. Once again, this is in part due to the relationship between the Kosovo situation and its predecessor(s). It is very likely that certain activists were able to draw on pre-existing organizations created to lobby in favour of intervention in Bosnia. Also, many activists were able to 'play the Bosnia card', exhorting their governments not to stand idly by a second time and allow a repetition of the disaster in Bosnia. Most importantly, however, the consensus curve is quite different in Figure 2. In this conflict, the degree of consensus both within and among NATO allies was an issue that was well-documented in the press. Consensus surrounding the bombing of Serb targets was quite high from an early stage, and remained high for.much of the duration. When consensus did begin to decline, it did so slowly. In short, those policy-makers (in North America) who believed that intervention was an inappropriate policy were in the minority from the beginning, and although their numbers did increase as time went on, they remained a subordinate force for the duration of the crisis. In the case of Kosovo, then, the CNN effect cannot have been responsible for the policy of intervention. For although the public curve intersects the media curve near its height (at point 'B'), it does not intersect the consensus curve until well after media influence has dropped off (at point 'C'). Of course, the intersection at point ' C is merely theoretical; the Kosovo crisis did not stretch on long enough for a policy change to become necessary. (Line 'E ' represents the actual termination of the crisis.) Whether the public would have desired such a change at point ' C (when it was theoretically capable of achieving it) will therefore remain a mystery. What about those crises where elite consensus is slow to materialize? According to the vast majority of observers, these types of cases are most likely to result in the CNN effect becoming 55 responsible for policy. As I have argued, however, this idea fails to take into account the shape of the media curve or the public curve, and as such it is inaccurate. Figure 3 demonstrates this point by representing the case of Bosnia. Figure 3 In this case, consensus is low from the outset, and remains so over most of the time line. In the later stages of the crisis (at a time which would correspond approximately to mid-1995), the curve eventually begins to increase at a steeper rate. Why consensus began to build when it did is a source of much controversy among scholars. Some attribute the sudden change of heart among Western governments to graphic television footage of the Sarajevo market bombing. For example, Hoge writes: "A CNN crew happened to be out and about the city that Saturday morning [when the 56 bombing took place]". Thus, he argues, television was largely responsible for the N A T O ultimatum that followed." Hoge appears to be another victim of post hoc ergo propter hoc. What is most bizarre about his contention regarding the Sarajevo market bombing is that he subsequently argues that it was "the fact, not the image, of body bags [that] was the important factor" in producing the American public's disenchantment with the Vietnam policy. One can only assume that it was the graphic nature of the Sarajevo images which prevents these two instances from being analogous in Hoge's mind. This would suggest that the footage from the Sarajevo bombing was more horrific than anything that had been shot in Vietnam. Perhaps we can accept that pictures of foreign civilians being murdered is more difficult to watch than pictures of one's own national soldiers being killed.100 However, it also suggests that the Sarajevo footage was more graphic than any previous images from Bosnia, and as Johanna Neuman argues, this is simply not the case. Without entering into a discussion about precisely which telecasts were graphic, and how much blood was shown, we can assert with confidence that there had been plenty of bloodshed on the nightly news in the years leading up to the N A T O ultimatum. It is therefore unlikely that television pictures of the Sarajevo bombing produced this result. As Neuman argues, "[w]hat is more likely is that it took years to build the political will to use muscle in the Balkans, despite the emotional pull of the pictures".101 Perhaps it was the power of public opinion that produced a higher level of consensus within "Hoge 1994,138. 1 0 0One might object that Hoge also points out the extreme infrequency of such bloody images from Vietnam. However, since he is crediting a change in America's Bosnia policy to a single bit of CNN footage, this objection is irrelevant. 101Neuman. Lights. Camera War. 1996. 117-118. 57 the policy-making elite. Certainly, public opinion had more time to materialize in this case than in most crises of its kind. By the time consensus began to rise, however, the crisis had drawn on for far too long to attribute public views to media framing or editorializing. In any case, the public was horrified by the death toll and human suffering in Bosnia; it is trite to assert that the media could have measurably contributed to or detracted from this sense of horror. The media did not have to persuade the public that something needed to be done. Leaving aside further debate surrounding what produced the rise in consensus, it is clear from this discussion that the shape of the consensus curve is key in determining whether or not the CNN effect exists in a particular case. For while there is some room for variation in the shape of both the media curve and the public curve, neither of these moves in an entirely different direction from case to case. The consensus curve, on the other hand, affected as it is by such a wide variety of variables including national interest, elections, partisan politics, international diplomatic relationships, and humanitarian concerns, can assume almost any shape. The curve that is most likely to result in the CNN effect is one in which consensus is relatively high at the outset, but drops off dramatically in the early to mid-stages, such as the one shown in Figure 1. Since the degree of consensus constitutes the key variable, we can make an important inference regarding the types of crises that are most likely to result in the CNN effect. Humanitarian disasters are more likely to pose consensus problems for governments than are crises that directly threaten the national interest of a state or its allies. When no clear national interest is at stake and no enemy is articulated, it is generally much more difficult to secure broad agreement.102 This is 1 0 2 A decline in consensus both within and between Western allies since the Cold War, for example, is a widely discussed phenomenon. 58 particularly true in those cases where some kind of peacekeeping or peace enforcement operation is proposed. To call for these types of operations, known as "Operations Other Than War" (OOTW), is often an almost knee-jerk reaction of the public and lower levels of government. However, because OOTW are, by their very nature, motivated by less clear-cut objectives - objectives for which the state has not traditionally been viewed as responsible for securing - consensus regarding whether this constitutes the most appropriate action is much more difficult to achieve. Is it worthwhile to expend blood and treasure in some distant land when there is no foreseeable material reward? This is a question that is being asked with increased regularity in the West. Humanitarian disasters are increasing in visibility, and likely in sheer numbers as well. Moreover, it is no longer possible to sublimate these international crises to more "pressing" Cold War concerns. This being the case, opportunities for the CNN effect to infiltrate the policy-making process are increasing. Moreover, while humanitarian crises pose problems for consensus, they are also likely to enhance the public's ability to influence policy. On the one hand, humanitarian disasters are, by their nature, mainly emotional issues, since they do not pertain to the strategic interests of most states. This makes these crises far easier for the public to understand and relate to than most matters of "high politics". Therefore, the public's lack of interest, one of the traditional reasons why the public is thought not to influence policy, is mitigated. At the same time, while governments may be willing to turn a deaf ear to public appeals when national interests are at stake, they may feel less justified in doing so when there appears to be little to lose by placating public demands. Considering all of this, it is not surprising that many have claimed to spot the CNN effect in action already. Major humanitarian interventions in Iraq, Somalia, Rwanda, Bosnia, and most 59 recently, Kosovo have prompted many to speculate that some or all of these initiatives were spurred on by the media. As I have argued, the CNN effect faces far too many obstacles to complete itself in all but the rarest of circumstances. Still, one of these five crises appears to be a more likely candidate than the others. In the case of Somalia, consensus within elites was initially relatively secure. Within less than a year, however, it had dropped off sharply.103 Couple this with extensive media coverage and a relatively low level of public experience with the issue, and we can re-construct Figure 1 with only minor variations. Indeed, Somalia's unique candidacy for the 'crisis that showed the CNN effect in action' has made it the most frequently-cited example for CNN effect theorists. This means that the Somalian crisis is also the most appropriate case study for this discussion. There are several key reasons for this, which will be discussed in Chapter III. 60 Chapter III: The C N N Effect in Somalia The 1990s opened with the promise of a new peace in the Western world; the Berlin Wall had come down, the superpowers were no longer enemies, and the United Nations appeared at last to be capable of living up to the ideals for which it was created. For East Africa, however, the decade began instead with one of the most horrific disasters in human history. Communal conflict and devastating famine were ravaging Somalia, and there seemed no end in sight. At one time, it was estimated that fully one third of Somalia's four and a half million to six million people would perish from the starvation and "random slaughter" that was destroying the country.104 The grim story is well known. Clan warfare was both a cause and a result of the complete collapse of all government in Somalia. Without any central authority, it became impossible for Somalians to combat the famine that was spreading throughout the country. Efforts by aid agencies to provide food and medical supplies were thwarted on a daily basis as supply routes were attacked by armed bandits. Food was looted and sold on the black market. In a country where those who were in the most dire need were the least able to secure relief, it seemed that Somalia would descend into permanent anarchy. The disaster continued unabated for over a year. The small United Nations contingent that was dispatched to ensure the safe delivery of supplies was simply not able to do so. Thus, on November 26, 1992, then-President George Bush announced that the US had no recourse but to respond more aggressively to prevent more deaths. The Bush Administration offered to send a 104Walter Goodman. "Why it took T V so long to focus on Somalia," The New York Times, Sept. 21 1992, C18. 61 contingent of US troops to aid the United Nations mission, provided that the American military could retain full command over its soldiers.105 Some 28,000 troops were thus dispatched in December as part of "Operation Restore Hope", with the aim of adding muscle to the relief effort.106 The enhanced U N mission enjoyed some early success. On June 5, 1993, however, things began to go sour for the international force. Twenty-four Pakistani peacekeepers were murdered by bandits loyal to a local warlord, General Mohammad Farah Aidid. In reprisal, US warplanes began to conduct daily assaults against Aidid's men. Soon, what had begun as a humanitarian mission had turned into a manhunt, and US involvement in Somalia very quickly began to resemble Beirut in 1983.107 American deaths followed, culminating in the October 3, 1993 disaster that claimed 18 American lives and wounded 78 others. It was the worst single firefight in US military history since the Vietnam War. 1 0 8 The American public was exposed to horrifying photos of an American prisoner beaten bloody, and to the truly disturbing images of a dead US Ranger being dragged through the streets of Mogadishu by jeering Somalis. Shortly thereafter, Bill Clinton, who had by then taken over as US President, set a deadline for US withdrawal. The timing of these events has led many observers to conclude that television was ultimately responsible for both the intervention and the subsequent withdrawal. As one scholar put it, "[b]y focusing daily on the starving children in Somalia, a pictoral story made for television, T V mobilized 105Clifford Krauss. "Washington seeks conditions on plan for Somalia force". The New York Times, Nov. 27 1992, A l . 106Strobel 1997, 137. 107"Drawing the Line in Somalia". The New York Times, June 20 1993, IV 16. 108Neuman, Lights. Camera. War. 1996, 14. 62 the conscience of the nation's public institutions, compelling the government into a policy of intervention for humanitarian reasons".109 Former ABC anchorman Ted Koppel takes the argument even further: Television brought home the images of starving Somalis, leading to the American-led United Nations rescue operation. That operation, in turn, led to the deaths of US servicemen. The image of one of those men in particular, the dead Ranger being dragged through the streets of Mogadishu, created a backlash that led to the withdrawal of most US forces. Television got us in, television got us out.110 There are two main implications attached to this view, both of which become clear when expressed in the form of counterfactuals.111 First, those who underscore the salience of the CNN effect in the case of Somalia imply that if the media had devoted less attention to the crisis, the Bush administration would not have launched Operation Restore Hope. Second, if the dramatic reports and graphic images of US casualties had not been broadcast all over the globe, US troops would not have been withdrawn when they were. There does exist a good deal of circumstantial evidence to support these assertions. On the face of it, therefore, both of these counterfactuals seem somewhat plausible. Unfortunately, however, scholars do not have the luxury of re-running history and splicing 109Bernard Cohen. "A View from the Academy," in W. Lance Bennett and David L. Paletz, eds. Taken by Storm. The Media, Public Opinion, and US Foreign Policy in the Gulf War. (University of Chicago Press 1994), 9-10. . 1 1 0Ted Koppel. "The Perils of Info-Democracy," in Chester A. Crocker & Fen Osier Hampson with Pamela Aall (eds). Managing Global Chaos. (Washington: United States Institute of Peace Press 1996), 356. 1 1 Counterfactuals , borrowing from Tetlock & Belkin, are "subjective conditionals in which the antecedent is known or supposed for the purposes of argument to be false". Philip E. Tetlock & Aaron Belkin. "Introduction," in Tetlock & Belkin (eds.) Counterfactual Thought Experiments in World Politics. (Princeton University Press 1996), 2. 63 out media influence. How, then, are we to assess the accuracy of these counterfactuals without engaging in useless conjecture? The task is daunting, but by no means impossible. For although we must by necessity base our analysis on a hypothetical situation, such a mental exercise can uncover many of the unspoken implications embedded in the notion of the CNN effect as it applies to Somalia. Indeed, the use of counterfactual reasoning, either in an implicit or explicit fashion, is necessary to test the validity of causal relationships such as the one suggested by proponents of the CNN effect. In other words, without turning Koppel's argument on its head and stating it in counterfactual form, we are relegated to dressing spurious correlations in the guise of causal inference.112 This fact can be illustrated through a fairly simple example. Consider the following proposition: The Soviet Union collapsed because Gorbachev led Soviet communism astray. Certainly, this seems a reasonable enough assertion; indeed, it is a view expressed by many. But regardless of how logical the theoretical underpinnings of this argument may appear, its veracity is ultimately held hostage by history. We cannot devise a controlled experiment to test the hypothesis. We can, however, explore its cogency by posing a counterfactual: If the Soviet Union had been led by someone other than Gorbachev in the late 1980s, it would not have collapsed. The implications of our hypothesis become instantly clear - namely, that there are no other factors that could have precipitated the fall of the Soviet Empire. Suddenly, the wisdom of our proposition becomes much less obvious. Thus, counterfactuals can "sensitize" us to possibilities we might otherwise have ignored; our awareness of these "other possibilities" allows us to derive plausible causal tetlock & Belkin 1996, 3. 64 11 -a • connections. Through a systematic application of such counterfactual reasoning, I will make some tentative conclusions about the US intervention in Somalia, and more specifically about the influence of the CNN effect in that initiative. In extracting the observable implications of the counterfactuals, I will show that claims of the media driving policy are significantly exaggerated. Before embarking on a search for testable implications, however, it is necessary to discuss at greater length the evidence that forms the basis of our counterfactuals. In other words, what is it about the case of Somalia that has led so many to argue, as Koppel does, that television was responsible for the intervention? D. Somalia: The exception to the rule? I have argued that the CNN effect could only complete itself in the rarest of circumstances. However, I have also argued that the crisis in Somalia most closely resembles the hypothetical timeline laid out in Figure 1. This is primarily because the shape of the consensus curve in the case of Somalia closely matches that of Figure 1: Initially, consensus is relatively high, and remains so for a time. At a fairly early stage, however, consensus begins to wane, and then to drop off sharply as outside events provoke a negative reaction within the elite. Consensus surrounding Operation Restore Hope was high at the outset, within both the elite and the public.114 In all likelihood, this was due to the erroneous evaluation that involvement in Somalia would be relatively cheap and painless, and probably because it would produce a favourable luIbid. 1 1 4 "Somalia: From Bad to Worse". The New York Times, Sept. 12, 1993. IV 18. 65 impression of American benevolence. Or perhaps it was because neither the public nor Congress was given a chance to object. Indeed, George Kennan has argued that there was absolutely no prior discussion in Congress of the undertaking, nor even a "proper public discussion".115 When the operation began to go sour in the fall of 1993, several members of Congress were to echo this belief.116 In any case, whether or not Bush consulted Congress or the public, approval among the decision-making elite at the time of the decision was high. Once the mission changed from one of providing food to one of hunting down a warlord, however, consensus among the various branches of government began to fade quite rapidly. This is not surprising; the United States has been keenly anxious to avoid "quagmire" since Vietnam. Interventions that appear to have no exit strategy and no definitive goal have been looked on with extreme disfavour. Thus, criticism of American policy began to emerge almost immediately after the 'manhunt' began, with some observers warning that the Administration needed to define more carefully the goals and scope of American involvement,117 and others blatantly arguing that the U N "shouldn't be gunning down Somalis from helicopters, no matter what warlord they support".118 The objections of lower level government officials from both parties were almost universally along these lines. "Having completed the original mission to feed the starving people of Somalia, 115Kennan 1993, A25. 116Clifford Krauss. "House vote urges Clinton to limit American role in Somali conflict," The New York Times, Sept. 29, 1993. A l . 1 ""Drawing the line in Somalia". The New York Times, June 20, 1993. IV 16. 118"Course correction needed in Somalia". The New York Times, July 14, 1993. A18. 66 we should bring our military forces home," said Democratic Senator Robert C. Byrd on September 25, 1993. "Without a legitimate purpose, we will be drawn futher into this quagmire, with a very real prospect for the continued loss of American lives."119 Republican Representative Benjamin A. Gilman agreed: "Despite our best intentions, United States policy in Somalia has failed. ... It's time to bring our forces home".120 In other words, critics charged either that the mission had been completed, or that it had metamorphosed into an unacceptable form. While the original assignment had been laudable and worthy of support, therefore, the situation after June 5, 1993 was so materially different as to warrant a re-examination of priorities.. In short, less than a year after the intervention was initiated, consensus began to decline sharply. This makes Somalia an excellent candidate for our study. On a more basic level, however, the mere chronology of events is sufficient to make one suspicious of a connection. For example, on December 4, 1992, when President Bush announced his intention to launch Operation Restore Hope, he began his speech by declaring, "Every American has seen the shocking images from Somalia...".121 And surely it cannot be a coincidence that President Clinton announced the withdrawal of US troops from Somalia on October 7, 1993, a mere three days after reports of the Mogadishu firefight began flooding in. 1 2 2 "Once again," The 1 1 9Eric Schmitt. "U.S. vows to stay in Somalia force despite an attack". The New York Times, Sept. 26, 1993.1. 120Krauss, "House vote urges Clinton to limit American role in Somali conflict," The New York Times, Sept. 29, 1993. A l . 121Quoted from Strobel 1997,141. 122Strobel 1997, 166. 67 Economist complained on October 9, "television images are shaping American foreign policy".123 In a similar vein, when US Secretary of State Madeleine Albright stated that "[television's ability to bring graphic images of pain and outrage into our living rooms has heightened the pressure both for immediate engagement in areas of international crisis and immediate disengagement when events do not go according to plan", she was almost certainly referring to Somalia."124 In sum, then, there exists a significant amount of anecdotal evidence to support the CNN effect theory. As I have argued, however, these anecdotes alone cannot truly convince us of the existence of a causal relationship. We must employ a more systematic review of the available evidence by attempting to capture and survey all of the potential observable implications of our hypothesis. In other words, if our hypothesis is correct - if media reports did in fact drive American foreign policy in Somalia - what else must be true? Which elements should be present, and which absent? Before reaching any conclusions, we must address these questions. E. Testable Implications of Counterfactuals In the opening chapter of their book, Counterfactual Thought Experiments in World Politics, Philip E. Tetlock and Aaron Belkin devise a set of criteria by which to judge counterfactual arguments. Probably the most important criterion among this set is what they refer to as "projectability". According to this principle, a researcher must "tease out testable implications of the connecting principles and determine whether those hypotheses are consistent with additional real-world The Economist, Oct. 9 1993, 22. Quoted from Neuman, Lights. Camera. War 1996, 14. 68 observations".125 In so doing, we address the crux of the issue. Namely, if our propositions are correct, what evidence should exist in the historical record? Recall our two counterfactuals regarding the US intervention in Somalia. First, had there been little or no media attention directed at Somalia, US troops would not have been dispatched. Second, if audiences had not been exposed to the horrifying images of dead peacekeepers, the US would not have withdrawn when it did. What observations should we be able to make if these two arguments are correct? The list that follows may not be exhaustive, but it will provide us with a strong framework through which to apply our analysis. If our hypothesis is valid, we should observe the following phenomena: 1. A significant number of media reports regarding the crisis in Somalia should have occurred prior to US government decisions, 2. Other reports of human suffering which were analogous in character and in frequency should have provoked a similar reaction from the public, and in turn the government, 3. Strong media pressure should have resulted in an immediate and measured shift in policy. This policy shift should not be readily attributable to other factors or considerations; and 4. The processes and rationalizations of US foreign policy should be demonstrably different in the 'information age' compared with pre-television history. I will examine each of these elements separately, in order to determine the extent to which some or all of them are in evidence in our case study. tetlock & Belkin 1996, 18. 69 £.1. Media Reports If the decisions of the US policy-making elite are to be traced back to the media, we must establish that significant media coverage of the crisis in Somalia preceded major government decisions: This assertion is not quite as obvious as it seems. Failure to identify properly all of the relevant 'major decisions' has been a notable source of misconception in several analyses of the C N N effect in Somalia. This is because some observers have treated the November 25, 1992 decision to launch Operation Restore Hope as the first major decision reached by the Bush Administration with regard to Somalia. This is not the case: Four months earlier, Bush had announced Operation Provide Comfort, an airlift of food and supplies to Somalia and northern Kenya. According to Walter Goodman of The New York Times, television reporting of the crisis prior to this decision was virtually non-existent. Writing in September of 1992, he complained: "NBC's 'Today' show carried two minutes of vivid pictures from a British freelancer as early as February, but sustained reporting by 'Today' began only last month ... In May, Brent Sadler of C N N sent back pictures of starving children and adults fighting for a few grains of rice, but coverage remained sporadic even after that."126 Even after Provide Comfort began, reporting on Somalia was nominal at best, at least until September. Warren P. Strobel's survey of media coverage regarding Somalia confirms this account. According to Strobel, the announcement of Operation Provide Comfort prompted a precipitous increase in media coverage of the crisis, from a total of nine reports on network evening news since January 1992, to thirty-six reports in the month of August alone (only two of which actually pre-date ;Goodman 1992, CI8. 70 the decision itself). Figure 4 represents Strobel's account of media coverage 128 Figure 4 175 150 125 100 75 50 2 5 T ! ] | Jan Feb Mar Apr May Jun Jul Aug Sep Oct Nov Dec Number of Stories on Network Evening News Mentioning Somalia in 1992 As in Goodman's account, media reports on Somalia clearly trickle off between the announcement of Operation Provide Comfort and the decision to launch Operation Restore Hope. Thus, it seems clear that with respect to the decision to intervene in Somalia, "[s]harp increases in the levels of television reporting tended to follow administration actions, rather than precede them".129 This fact alone, however, cannot speak to the actual influence of those reports that did pre-date the decision. It seems fair to assert that even a single news report, if sufficiently graphic and 'Strobel 1996, 360. 'Reproduced from Strobel 1996, 361. fbid., 360. Emphasis in original. 71 heart-wrenching, could stir up enough outrage within the public to provoke debate and ultimately action within the foreign policy elite of the United States. Therefore, it is desirable to know something of the content of these reports and the sources from which they originated. Strobel's analysis is also limited by its exclusive focus on evening reports. In contrast, what is particularly interesting about Goodman's discussion of daytime coverage is that it does offer some insight as to the content of the coverage, and this reveals some important clues as to the sources of some of the stories. For example, he traces "sustained" reporting by NBC back to an August interview on "Today", in which Senator Nancy Landon Kassebaum appeared as a guest.130 What is not mentioned by Goodman is that this appearance occurred on the very day a congressional delegation presented a report urging action on Somalia - a delegation headed up, in part, by Senator Kassebaum.131 Indeed, this was the second of two seperate occasions on which a congressional hearing prompted network reporting.132 It is evidence such as this which leads Mermin to argue that "television coverage of Somalia in the fall of 1992 did not originate in the independent actions of journalists but in the interaction of journalists engaged in routine newsgathering practices and sources in Washington who made efforts to get Somalia on the foreign policy agenda".133 Strobel agrees: "[t]he television coverage (and other media attention) that did take place was almost always a result, not of media initiative and agenda-setting, but of deliberate and successful attempts by others to stir up interest in Somalia in 130Goodman 1992, CI8. 1 3 1Mermin 1997, 394. 132/2>^. mIbid, 386. 72 order to move policy". Certainly, both of these assessments are in accord with our earlier findings regarding media autonomy and newsgathering practices. Indeed, Herman Cohen, then-Assistant Secretary of State for African Affairs, has gone so far as to claim that "[t]his is a clear example of the government using the media".134 However moving the images or facts contained in those reports, therefore, it would be irresponsible to state that the media were ultimately responsible for any effect that they may have produced. Since the stories themselves originated primarily from lower levels of government, the media are more accurately characterized as a link in the chain, rather than the catalyst, with respect to the decision to intervene. What is strange about Strobel's account is that he does not present analogous figures to indicate the levels of reporting in 1993, the year in which the decision to withdraw was reached. We can assume, however, given the notoriety of the image of the dead Ranger, that reports surrounding the October 3, 1993 tragedy were ubiquitous. Indeed, perusal of The New York Times from January 1991 on reveals that, with respect to that publication, the three days immediately following the battle (October 4, 5, and 6, 1993) had the highest level of reporting on Somalia since the crisis began. Clinton's announcement of the US withdrawal, as noted above, occurred on October 7, 1993. Thus, it appears that with respect to the decision to withdraw from Somalia, the major decision followed media reports, rather than preceding them. Appearances are deceiving, however; post hoc ergo propter hoc aside, we must be certain that the decision to withdraw was a deviation from the course leading up to it. As it happens, this is not the case. Once again, those who argue that television forced Clinton to withdraw have failed to identify prior developments which are of significant import. In brief, "the Clinton administration 134Quoted in Strobel 1996, 13 5. 73 was moving toward disengagement from the confontation in Somalia long before October and had taken active steps in that direction in the second half of September [of 1993]. " 1 3 5 In early September, the Senate asked the Clinton Administration to report on Somalia operations by October 15 and to get Congressional approval for any military role beyond November 15.136 On September 25, after the deaths of three American soldiers, Congress renewed calls for the immediate withdrawal of troops.137 On the 28th of that month, the House overwhelmingly passed a resolution (406 to 26) demanding that the Administration put limits on its mission or face the probability of a cutoff of funds.138 Impatience was growing as the mission changed from one aimed at feeding the starving to a manhunt for Aidid following the June 5 assassination of 24 Pakistani peacekeepers. "Liberals criticized the Administration for trying to serve as kingmakers, while conservatives characterized the deepening military involvement as a waste of American lives and money. Members of both parties evoked images of Lebanon and even Vietnam."139 In a tacit acceptance of that view, the Administration shifted its goal in late September from capturing Aidid, toward building new political structures that excluded him. Secretary of State Warren Christopher even sent a letter to the 135Strobel 1996, 182. 136MSomalia: from bad to worse". The New York Times, Sept. 12, 1993. IV 18. 137Schmitt 1993,1. 138Krauss, "House Vote Urges Clinton to Limit American Role in Somali Conflict". The New York Times, Sept. 29, 1993. A l . 1 3 9/2>^. 74 Secretary-General of the U N signalling such a change in US policy.140 Finally, on September 30, President Clinton announced that the United States was moving towards an exit strategy, stating that there must be a "date certain" for American withdrawal.141 From this evidence, it seems clear that the United States would have withdrawn its forces, and that an announcement to that effect would have been made sometime in the next few weeks. Therefore, while the "major decision" as such did follow the graphic reports of the October 3 firefight, it was by no means unexpected. This casts doubt on the premise that images of dead peacekeepers were responsible for the policy shift, if indeed the announcement of withdrawal can even be characterized as such. What is significantly more likely is that the deaths of an additional 18 Americans became 'the straw that broke the camel's back'. The patience of Congress was exhausted, and the executive branch could no longer deny that continued involvement smacked of quagmire. As one high-ranking Pentagon official put it, "I think we probably still would have gotten out. But it wouldn't have been considered the disaster it was".142 In short, since there were relatively few reports which occurred prior to major administration decisions regarding intervention, and since those that did occur did so primarily at the behest of official sources, we cannot say that television forced President Bush to go in. And since the Clinton Administration had sent many signals indicating its intention to withdraw US forces in the near future which pre-dated the October 3 disaster, we cannot say that television forced President Clinton mIbid. 141Elaine Sciolino. "U.N. chief warning U.S. against pullout of force in Somalia," The New York Times, A l . 142Quoted in Strobel 1997, 178. 75 to get out. Of course, it is possible that the horrifying images of the dead Ranger accelerated the process, but this is a far cry from actually initiating it. Still, the exact effect of these media reports remains unclear, and therefore we cannot claim that their role was of no consequence. E.2. Analogous Cases If our counterfactuals hold true, an examination of analogous cases and media attention should reveal similar patterns to the one we have proposed for Somalia. The most obvious parallel to be drawn is between the tragedy of Somalia and that of Bosnia-Herzegovina. As with Somalia, decisive foreign intervention seemed to be the only possible remedy to end the crisis unfolding in the Balkans. And, like Somalia, images broadcast during the height of the conflict were frequent and horrible. The analogy is particularly appropriate given the fact that both of these crises occurred within the same time frame; that is, both erupted during the Bush Administration and reached their climax during the Clinton Administration. This fact allows us, to an extent, to control for outside variables resulting from different policy-making environments. An examination of the Bosnian case, however, does not reveal the trend predicted by our hypothesis. For while media reports began flooding out of Sarajevo as soon as the bloodshed began in earnest in 1992, decisive US action did not occur until 1995. As Johanna Neuman notes, "[a] steady, three-year drumbeat in the news media, both in print and on the air, for intervention in Bosnia - complete with horrific images of concentration camps and bloodied civilians, of fleeing refugees and U N hostages - did not compel the West to go to war in the field".143 Former US Secretary of State Lawrence Eagleburger confirms this fact: "It made us damn uncomfortable. But this was a 143Neuman. Lights, Camera, War, 231. 76 policy that wasn't going to get changed no matter what the press said."144 Neuman also notes the example of Sudan, where although "the starvation was equally devastating, the pictures equally horrific, and, at first, equally as evident on CNN", the US failed to involve itself.145 Thus, even a brief examination of these cases significantly weakens the argument that media reports alone were sufficient to prompt intervention in Somalia. We must assume that other factors came in to play in the cases of Bosnia and Sudan which served to mitigate the impact of graphic reports. If we admit this possibility, however, we are forced to likewise admit that outside factors may have served to heighten the influence of media reports in the case of Somalia. E.3. Other Considerations An assertion that the news media drove policy in a particular instance implies that there were few, if any, other considerations that could have prompted a similar result. Failure to control for these possible variables prevents us from accurately determining which factor or combination of factors is responsible for the result, and thus prohibits valid causal inference. It is for this reason that we initially posed the counterfactual: If the media had devoted less attention to the crisis in Somalia, the Bush Administration would not have launched Operation Restore Hope. Unfortunately for our hypothesis, the search for other potential variables does not come up empty-handed. Indeed, there are several important considerations which may have exerted equal, if not greater, influence over Bush's decision to intervene in Somalia. First, as noted above, the war in Bosnia was raging concurrently with the conflict in Somalia. 144Quoted from Strobel 1996, 367. 145Neuman. Lights, Camera. War. 21. 77 The great powers found themselves under increasing pressure from many Eastern European countries, as well as from within their own jurisdictions, to get involved. The costs of such an intervention, however, were deemed to be prohibitively high. And yet the call to "do something" was insistent. Meanwhile, the costs of intervening in Somalia were assessed (incorrectly) as being relatively low. This has led many scholars to conclude that intervention in Somalia was a convenient, low-cost way of diverting attention from the uncomfortable situation in Bosnia.146 Edward N. Luttwak puts it succinctly: "The Bush Administration intervened in Somalia to make amends for its failure to act in Bosnia".147 Second, in 1992 the United States was on the defensive with respect to Islamic groups, a situation that could be traced back to the Gulf War. During Operation Desert Storm, Saddam Hussein had attempted to appeal to a "pan-Arabism" in order to marshal support against Coalition forces. While not entirely successful, Hussein may have been partially responsible for the apparent apprehension of some Islamic nations that the Bush Administration was undertaking to fabricate a "Green Menace" to replace the Soviet Threat. One short year after embarking on an aggressive campaign against a Muslim nation (Iraq), the US found itself being accused of 'looking the other way' while Bosnian Muslims were massacred by the Serbs. Thus, intervention in Somalia, a black Muslim state, may have been an expedient method of deflecting the charges of racism and ethnocentrism which emanated from Muslim nations as a result of the Balkan War, not to mention the vocal speculations by local aid organizations that the US was desensitized to pictures of "flies 146See, for example, Strobel 1997 and 1996; Neuman. Lights. Camera. War; and Edward N. Luttwak, "Wrong Place, Wrong Time". The New York Times, July 22, 1993. A23. 147Luttwak 1992, A23. 78 on sickly black faces".148 Third, the West was coming under increased attack from the United Nations, stemming from its inaction in Somalia. While the attention of the great powers (particularly Europe) seemed fixed on the Balkans, little or no notice was being taken of the starvation in Africa. Secretary-General Boutros Boutros-Ghali, himself an African, accused the Security Council of contemplating "a rich man's war in Yugoslavia while not lifting a finger to save Somalia from disintegration".149 Despite the fact that these potentially decisive variables remain outside the causal relationship posited in our counterfactuals, this does not mean that media influence played no role whatsoever in bringing about the decision to intervene in Somalia. It does not even imply that media sway was subordinate to other considerations. Indeed, as with any web of complex inter-relationships, it is difficult (if not impossible) to pinpoint which variables are more consequential than others in producing the observed result. Therefore, I will not overstate the importance of variables that rest outside the model. The existence of additional causal relationships does not totally negate the actuality of the one we have chosen to assess; it means that our relationship does not fully explain the result. We must therefore continue to probe our hypothesis from yet another angle. E.4. Processes and Rationalizations of Policy-Making An approach that may present an even greater challenge to our hypothesis rests within an examination of the extent to which US foreign policy-making in general can be shown to have undergone a qualitative change since the "information revolution". If, as hypothesized, the decision 148Goodman 1992, CI8. 149Quoted in Strobel 1996, 365. 79 to intervene in Somalia would not have been reached in the absence of media reporting on the crisis, this suggests that the process by which such decisions are reached has been altered by the introduction of the media variable. Thus, a convincing demonstration that American foreign policy-making has sustained such a transformation will greatly fortify our case for a causal relationship. One would expect to locate just such a demonstration within the ubiquitous literature on the C N N effect. As Buckley notes, however, one of the of the most striking features of this literature is that it is almost totally ahistorical.150 An assertion that 'pictures are worth a thousand words' seems to be all that CNN effect proponents feel compelled to point out when it comes to demonstrating that television has irrevocably altered policy-making. The abrupt transition from traditional newsgathering techniques to real-time satellite reporting is, according to many, the most profound revolution in communications history. This assertion is not as self-evident as it appears, however, once we move beyond hyperbolic statements and actually address the historical record. "One of the hazards of modernity" Denis Stairs cautions, "is that its packaging makes old things look new".151 Observers of the CNN effect would do well to mark this warning. For it is at best arguable that the temporal contraction involved in the shift to satellite news,~ or even from newspapers to television, was equal to that which occurred during the Industrial Revolution when the telegraph was introduced. As Neuman points out, The telegraph's impact was as revolutionary in the Industrial Age as that of satellite television or the computer in the Information Age. Even now, it is hard to comprehend the magnitude of the transition. In a world where communication had depended on the speed of a horse, or a sailing ship or a train, messages could 150Buckley 1998, 6. 151Stairs 1998, viii. 80 suddenly be received and answered almost instantly. This transition, from a leisurely pace of communication to almost instantaneous contact, most closely mirrors the changes in information technology that we are experiencing today.152 Complaints that a new technology discourages careful deliberation or reflection of an issue before a response is necessary are by no means new.153 Nor is the notion that policy-makers must rely on the media for the up-to-date information they require. Admiral Zacharias, chief of American naval intelligence during WWII, was reportedly fond of saying that "90% of all intelligence is derived from a careful reading of the press".154 What does all this mean? I have already argued that the news media play a limited role in agenda-setting, reliant as they are on official sources. I have also shown that media do not, except in rare circumstances, contribute independently to policy debate. Rather, they tend to offer a forum to hash out debate that is occurring within the elite. Finally, we cannot even say that media accelerate the process in such a way as to alter the policy-making environment to an unprecedented extent; the present 'Information Age' is not the first era in which decision-makers have had to adjust to a new technology. Therefore, we can only conclude that the process of foreign policy-making in the United States is not, due to the influence of television, qualitatively different than in pre-television times. Moreover, bringing the discussion back to Somalia, the issue of having to respond too quickly was almost certainly not relevant in the decision to intervene. In other words, there is no !Neuman. "The Media's Impact on International Affairs, Then and Now," (1996), 111. 'Buckley 1998, 18. ^Davison. Mass Communications and Conflict Resolution, (1974), 17. 81 evidence, nor is it even logical to speculate, that President Bush felt so pressured to respond immediately to Somalia's woes that he ordered an intervention which, had he had the luxury of deliberation, he would not have favoured. The mere fact that intervention occurred over a year after the eruption of the crisis all but eliminates this possibility. It is substantially more likely that 'almost-up-to-the-minute' footage of the October 3 debacle placed time pressure on Bill Clinton to withdraw, but, as we have seen, this decision was not a qualitative shift from the policy already favoured by his administration. This discussion of testable implications of counterfactuals has revealed some important details. On the one hand, it seems likely that graphic television images of suffering Somalis played some role in President Bush's decision to intervene. Indeed, senior policy-makers such as former Secretary of State Lawrence Eagleburger and former National Security Advisor Brent Scowcroft cited media pressure as one factor that went into their decision to intervene in Somalia. Likewise, Eagleburger's predecessor, James A. Baker III, has acknowledged that the US "probably wouldn't have" dispatched troops in the absence of media images.155 On the other hand, however, it is far from clear that media reports, or the CNN effect, were the deciding factor in provoking Operation Restore Hope, which is the relationship implied by most observers. In other words, while those who underscore the salience of the CNN effect may be correct in their assessment of the nature of media impact on foreign policy in the case of Somalia, they have thus far failed to create a compelling case for the extreme depth and salience which is claimed for the CNN effect. Counterfactual reasoning has proved to be a valuable tool in this discussion; it has enabled 155Strobel 1996, 364. 82 us to highlight the limitations of the hypothesis proposed by the CNN effect. Counterfactuals have some limitations of their own, however; because I have made use of hypothetical mental exercise, it is impossible to prove or disprove conclusively the hypothesis that we have examined. This said, through the use of counterfactuals I have pointed out several important angles from which cases thought to demonstrate the CNN effect must be examined if its existence is to be established in a compelling fashion. F. The CNN Effect in Somalia: Some final thoughts In one sense, it is obvious why Somalia is the case most frequently cited as displaying the CNN effect. As we have seen, it is not difficult to locate anecdotal evidence supporting the idea that graphic media reports created an undeniable tide of public opinion. In another very important sense, however, the notion that public pressure forced President Bush to intervene is a bit puzzling. Recall that, according to many, Bush sought no debate in Congress, nor any "proper public discussion". If Bush did not seek popular approval, how could public opinion have been responsible for his decision? One might object that it is absurd to suppose that a president would not seek the approval of Congress or the public in such a situation. Upon further reflection, however, we must ask ourselves: Why should there have been any consultation of the sort? In fact, why should we imagine that President Bush could have felt compelled at that time to undertake anything significant as a result of public pressure, as proponents of the CNN effect argue? After all, Bush's term as President was all but over. He had no prospect of personal re-election; similarly, he could not have been attempting to secure a Republican successor, since the elections were already decided. Bush was, in all practical terms, a 'lame duck' president, and these 83 are traditionally thought to be more capable of unilateral Executive action, since re-election is not an issue. Why, therefore, should public pressure have been responsible for the action? Kennan argues that Bush's motives most likely: [L]ay largely in his memories of the political success of the move into the Persian Gulf, and in the hope that another venture of this nature would arouse a similar public enthusiasm, permitting him to leave his Presidential office with a certain halo of glory as a military leader using our forces to correct deplorable situations outside our country.156 If this is even partially true, Bush did in some sense act with an eye to popular approbation. But this is not at all the same as acting because the public forced him to. To claim such would be akin to arguing that if I perform a good deed with the goal of receiving accolades, the true credit for the deed really belongs to whatever armchair Samaritans may applaud my efforts, despite the fact that they have themselves done nothing. Leaving this issue aside, however, a number of things have become clear in the course of this discussion. The case of Somalia has been instructive, and is indeed an ideal example to refer to when discussing the CNN effect, though not for the reasons imagined by proponents of the theory. The Somalia case does not embody the archetypal 'victim' of media influence. Still, it is a hive of activity for CNN effect theorists, and scrutinizing this hive has revealed a wealth of examples of the superficial analysis that is typical of these theorists. First, when evidence is cited, it is overwhelmingly anecdotal in nature, and is rather selectively conveyed. We are told that the decision to withdraw from Somalia, for example, came on the heels of graphic reports of American deaths. We are not told that the Clinton Administration had already articulated a desire to withdraw. We 156Kennan 1993, A25. 84 are told that, prior to the November decision to launch Operation Restore Hope, television journalists had questioned US Senators about what the government was doing to ameliorate the situation in Somalia. We are not told that these interviews were conducted at the behest of the Senators themselves. Second, causal connections are assumed rather than demonstrated. If x follows _y, we are expected simply to accept that y caused x. Public opinion supported the decision to intervene in Somalia. Therefore, public opinion caused the intervention in Somalia. Graphic images of casualties were cited by some government officials in criticizing the Administration policy. Therefore, these images created the criticism. Obviously, this type of inference is invalid if left unsupported. Yet such support is wanting in pro-CNN effect accounts; there is little effort to move beyond spurious correlations. Third, CNN effect theorists have failed to address the full implications of their arguments As such, they have omitted entirely large bodies of empirical data that are available to test the cogency of the theory, and which mitigate against its plausibility. In other words, the historical record is replete with important facts that bear, consequences for the credibility of the theory, but CNN effect theorists choose to ignore these facts altogether. Conclusion This discussion will not be closed with a series of flatly stated conclusions about the CNN effect. To do so would be to commit precisely those errors that I have pointed out on the part of the theory's proponents. This is because the ultimate validity of the theory is held hostage by a series of extremely murky relationships which we cannot responsibly claim to understand. Neither the precise 85 impact of the news media on public opinion, nor the influence of the public over foreign policy decisions, lends itself to easy analysis or steadfast conclusions. If we one day uncover the key to these puzzles, it will almost certainly not be in the near future. This caveat aside, the conclusions which have been reached in the course of this study are based on a great deal more evidence than the theory itself appears to be. As I have argued, the three core assumptions that underpin the CNN effect are problematic to say the least. First, the news media cannot reasonably be characterized as an autonomous, free-standing institution. Therefore, to claim that they are the source of any chain of events is suspect. Second, it is at best unclear that the media are capable of creating a strong tide of public opinion. While the exact nature of their influence over people's opinions is unknown, it is very unlikely that they are capable of dictating these opinions in any substantive sense. Third, public opinion analysts themselves are unwilling to claim without doubt that popular pressure is able to force foreign policy-makers to adopt a course that they do not favour. This claim on the part of CNN effect theorists, therefore, is unwarranted. There are exceptions to every rule, however, and so we cannot claim with certainty that these assumptions will never apply. But, as I have argued, it is extremely unlikely that all three will obtain in any given case. This is because news media are most likely to influence public opinion in precisely those circumstances where public opinion is least likely to influence foreign policy. This has meant that, despite claims to the contrary, we have yet to witness the CNN effect in action. Of course, this is not to imply that the CNN effect is like a switch that can simply be flipped 'on' or 'off. The assertion is that the media have not yet exerted enough influence over the foreign policy-making process - in the US or anywhere else - to be credited with 'driving', 'initiating', or 'creating' it, and that they are unlikely ever to do so. 86 To the best of the author's knowledge, the preceding study is the only one of its kind in the literature. Thus far, there appears to have been no attempt to examine systematically the theory of the CNN effect. This fact is no less disturbing than it is surprising. It is perhaps not so unusual that systematic criticisms have not yet emerged; the theory is, relatively speaking, a young one. What is more significant is that proponents themselves have not supported their conjecture with any real data. Hyperbolic claims take the place of careful analysis, and rather random anecdotes exist where empirical evidence ought to be. Perhaps this would not be a serious issue if the over-stated claims of journalists were recognized for being just this. But political scientists of all stripes are taking up the cause, seemingly without question. The fact that media are capable of driving foreign policy, in the United States and elsewhere, has become almost a truism, not unlike the democratic peace theory. The danger of this attitude is that it has already begun to seep into the minds of policy-makers. In the case of the US in particular, there is reason to fear this result. The American system of'checks and balances' has been known to 'check' itself from any action at all. If the Fourth Estate is thought to be so in more than name, a new kind of paralysis may ensue. Or, worse yet, the opposite may occur: The US may find itself committing to a myriad of ill-timed, ill-executed and ultimately ill-fated interventions all over the globe, in anticipation of public sympathy. The result of this would be an enormous expense of blood and treasure, and would likely leave areas of crisis in worse condition than if they had been left to their own devices (to say nothing of the deleterious effect on US prestige and the inevitable charges of global hegemony). This is not to be alarmist about the issue. The point is that whether or not the CNN effect is actually real, if policy-makers believe it to be so, the result may be the same. In other words, if 8 7 scholars continue to bemoan the encroachment of mass media into the policy-making process, it may become a self-fulfilling prophecy.157 1 This leaves aside, of course, the debate surrounding whether or not such a condition would actually be desirable. 88 Bibliography Achen, Christopher H. "Mass Political Attitudes and the Survey Response," American Political Science Review 69 (1975), 1218-1231. Batarfi, Khaled M . "Media Foreign Policy and Public Opinion," (Doctoral Thesis Chapter 1998), http ://www.uoregon. edu/~batarfi/cnn.html. Bennet, Stephen Earl. "The Persian Gulf War's Impact on Americans' Political Information," Political Behaviour, 16:2 (June 1994), 179-201. Bennett, W. Lance. "Cracking the News Code: Some Rules that Journalists Live By," in Iyengar, Shanto and Richard Reeves (eds.), Do the Media Govern? Politicians, Voters, and Reporters in America, (Thousand Oaks: Sage Publications 1997), 103-117. Bennett, W. Lance & David L. Paletz (eds.) Taken By Storm: The Media, Public Opinion and US Foreign Policy in the Gulf War, (University of Chicago Press 1994). Buckley, Brian. "The News Media and Foreign Policy: An Explanation," (Centre for Foreign Policy Studies, Dalhousie University 1998). Cohen, Bernard C. "A View from the Academy," in Bennet, W. Lance & David L. Paletz (eds.) Taken By Storm: The Media, Public Opinion and US Foreign Policy in the Gulf War, (University of Chicago Press 1994), 9-10. - The Press and Foreign Policy. (Princeton University Press 1963). Cohen, Yoel. Media Diplomacy: The Foreign Office in the Mass Communications Age, (London: Frank Cass & Co. 1986). Dachi, Stephen F. "The Making of American Foreign Policy," Pakistan Horizon, 47:2 (April 1994), 13-22. Davison, W. Phillips. Mass Communication and Conflict Resolution: The Role of the Information Media in the Achievement of International Understanding, (New York: Praeger Publishers 1974). Davison, W. Phillips & Frederick T.C. Yu (eds). Mass Communications Research: Major Issues and Future Directions, (New York: Praeger Publishers 1974). Dennis, Everette E. "The Media and Foreign Policy in the Post-Cold War Period," Freedom Forum Reports, 89 Domke, Daniel, Dhavan V. Shah & Daniel B. Wackman. "Media Priming Effects: Accessibility, Association, and Activation," International Journal of Public Opinion Research, 10:1 (Spring 1998), 31-52. Dorman, William A. "Press Theory and Journalistic Practice," in Iyengar, Shanto and Richard Reeves (eds), Do the Media Govern? Politicians. Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 118-125. The Economist, October 9, 1993. Entman, Robert M . & Benjamin I. Page. "The Iraq War Debate and Media Independence," in W. Lance Bennett & David L. Paletz (eds.) Taken by Storm: The Media, Public Opinion, and US Foreign Policy in the Gulf War. (Chicago: University of Chicago Press 1994), 82-101. Entman, Robert M . "Framing US Coverage of International News: Contrasts in Narratives of the K A L and Iran Air Incidents," International Politics and the Press, 41:4 (Autumn 1991), 6-37. George, Alexander L. "The Role of Force in Diplomacy" in Crocker, Chester A. & Fen Osier Hampson with Pamela Aall (eds), Managing Global Chaos. (Washington: United States Institute of Peace Press 1996), 209-222. Gergen, David R. "Diplomacy in a Television Age: The Dangers of Teledemocracy," in Serfaty, Simon (ed.), The Media and Foreign Policy. (New York: St. Martin's Press 1990), 47-64. Goodman, Walter. "Why it took T V so long to focus on Somalia," The New York Times, Sept. 21, 1992, C18. Guo, Zhongshi & Patricia Moy. "Medium or Message? Predicting Dimensions of Political Sophistication," International Journal of Public Opinion Research, 10: 1 (Spring 1998), 25-44. Gurevitch, Michael. "The Globalization of Electronic Journalism" in Curran, James & Michael Gurevitch (eds.) Mass Media and Society. (London: Arnold Publishers 1996), 204-224. Haass, Richard N. "Using Force: Lessons and Choices for US Foreign Policy," in Crocker, Chester A. & Fen Osier Hampson with Pamela Aall (eds.), Managing Global Chaos, (Washington: United States Institute of Peace Press 1996), 197-208. Hallin, Daniel C. The Uncensored War: The Media and Vietnam. (Oxford University Press 1986). Hindell, Keith. "The Influence of the Media on Foreign Policy," International Relations, Vol. 90 No. (1995), 73-85. Hoge, James R. Jr. "Media Pervasiveness," Foreign Affairs, 73 A (July-Aug. 1994), 136-144. Holsti, Ole R. Public Opinion and American Foreign Policy, (University of Michigan Press 1996). - "Public Opinion and Foreign Policy: Challenges to the Almond-Lippmann Consensus," International Studies Quarterly, 36:4 (Dec. 1992), 439-466. Iyengar, Shanto & Adam Simon. "News Coverage of the Gulf Crisis and Public Opinion: A Study of Agenda-Setting, Priming, and Framing," in Shanto Iyengar & Richard Reeves (eds.) Do the Media Govern? Politicians, Voters, and Reporters in America, (Thousand Oaks: Sage Publications 1997), 248-257. Jordan, Donald L. & Benjamin I. Page. "Shaping Foreign Policy Options: The Role of T V News," Journal of Conflict Resolution, 36:2 (June 1992), 227-241. Kalb, Marvin. "A View from the Press," in W. Lance Bennett and David L. Paletz (eds.) Taken By Storm: The Media, Public Opinion, and US Foreign Policy in the Gulf War, (Chicago: University of Chicago Press 1994), 3-7. Kelly, Lt. General W. Thomas (Ret). "A View from the Military," in W. Lance Bennett and David L. Paletz (eds.) Taken By Storm: The Media, Public Opinion, and US Foreign Policy in the Gulf War, (Chicago: University of Chicago Press 1994), 7-8. Kennan, George. F. "Somalia, Through a Glass Darkly," The New York Times, Sept. 30, 1993, A25. Kiser, Edgar & Margaret Levi. "Using Counterfactuals in Historical Analysis: Theories of Revolution," in Tetlock, Philip E. & Aaron Belkin (eds.), Counterfactual Thought Experiments in World Politics, (Princeton: Princeton University Press 1996), 178-210. Koppel, Ted. "The Perils of Info-Democracy," in Crocker, Chester A. & Fen Osier Hampson with Pamela Aall (eds), Managing Global Chaos. (Washington: United States Institute of Peace Press 1996), 355-356. Krauss, Clifford. "House vote urges Clinton to limit American role in Somali conflict," The New York Times, Sept. 29, 1993. A l . ~ "Washington seeks conditions on plan for Somalia force". The New York Times, Nov. 27, 1992, A l . 91 Lindt, Naomi. "Seeing is Believing: The Power of the Television Image over the American Public and the Political Agenda," http: //www. liunet. edu/cwis/bklyn/acadres/honorsprog/nyhs/naomi. htm. Luttwak, Edward N. "Wrong Place, Wrong Time". The New York Times, July 22, 1993. A23. Manheim, Jarol B. "Going Less Public: Managing Images to Influence US Foreign Policy," in Iyengar, Shanto and Richard Reeves (eds), Do the Media Govern? Politicians. Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 379-390. McCombs, Maxwell & George Estrada. "The News Media and the Pictures in our Heads," in Iyengar, Shanto and Richard Reeves (eds.), Do the Media Govern? Politicians, Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 237-247. Mermin, Jonathan. "Television News and American Intervention in Somalia: The Myth of a Media-Driven Foreign Policy," Political Science Quarterly, 112:3 (Fall 1997), 385-403. Neuman, Johanna. Lights. Camera. War. (New York: St. Martin's Press 1996). - "The Media's Impact on International Affairs, Then and Now," SAISReview, 16:1, 109-123. The New York Times. "Somalia: from bad to worse". Sept. 12, 1993. IV 18. - "Course correction needed in Somalia". July 14, 1993. A18. - "Drawing the line in Somalia". June 20 1993, IV 16. Newsom, David. D. Diplomacy and the American Democracy, (Indiana University Press 1988). Olien, Clarice, George A. Donahue & Philip J. Tichenor, "Conflict, Consensus, and Public Opinion," in Glasser, Theodore L. & Charles T. Salmon (eds.), Public Opinion and the Communication of Consent. (New York: Guildford Press 1995), 301-322. Reeves, Richard. "Overview" in Iyengar, Shanto and Richard Reeves (eds.), Do the Media Govern? Politicians. Voters, and Reporters in America. (Thousand Oaks: Sage Publications 1997), 3-8. Schmitt, Eric. "Clinton Reviews Policy in Somalia as Unease Grows". The New York Times, Oct. 6, 1993, A l . - "U.S. Aides Call for Pressing U.N. on Somalia Settlement". The New York Times, Oct. 2, 1993,1.3. 92 - "U.S. vows to stay in Somalia force despite an attack". The New York Times, Sept. 26, 1993.1. Schorr, Daniel. "How CNN Changed History," Nando Times News, Sciolino, Elaine "U.N. chief warning U.S. against pullout of force in Somalia". The New York Times, A l . Soderlund, Walter C , Ronald H. Wagenberg & Ian C. Pemberton. "Cheerleader or Critic? Television News Coverage in Canada and the United States of the US Invasion of Panama," Canadian Journal of Political Science, 27:3 (Sept. 1994), 581-606. Stairs, Denis. "Foreword," in Buckley, Brian. "The News Media and Foreign Policy: An Explanation," (Centre for Foreign Policy Studies, Dalhousie University, 1998), i-viii. Stech, Frank. J. "Winning CNN Wars," Parameters, (Autumn 1994), 37-56. Strobel, Warren P. Late-Breaking Foreign Policy, (Washington: United States Institute of Peace Press 1997). - "The Media and US Policies Towards Intervention: A Closer Look at the C N N Effect," in Crocker, Chester A. & Fen Osier Hampson with Pamela Aall (eds.), Managing Global Chaos. (Washington: United States Institute of Peace Press 1996), 356-378. Russet, Bruce. "Counterfactuals About War and Its Absence," in Tetlock, Philip E. & Aaron Belkin (eds.), Counterfactual Thought Experiments in World Politics, (Princeton: Princeton University Press 1996), 171-186. Tetlock, Philip E. & Aaron Belkin (eds.) Counterfactual Thought Experiments in World Politics, (Princeton University Press 1996). Tetlock, Philip E. "Political Psychology or Politicized Psychology: Is the Road to Scientific Hell Paved With Counterfactuals?" Political Psychology, 15:3 (Sept. 1994), 509-529. 93 


Citation Scheme:


Citations by CSL (citeproc-js)

Usage Statistics



Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            async >
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:


Related Items