Reporting War Journalism in Wartime
Reporting War Journalism in Wartime
Reporting War Journalism in Wartime
What is the role of the journalist in wartime? When faced with the responsibility
of bearing witness to the horror of modern warfare, how do reporters negotiate
the competing demands of their employers, of governments and military sources,
even of soldiers in the field of battle? How do they manage the tensions between
objectivity, patriotism, and sympathy with the suffering of local people caught up
in conflicts?
Reporting War examines the nature of contemporary war reporting in a range
of locales, including Africa, the Balkans, the Middle East—especially Iraq—and
southern Asia. The contributors, both practising and former journalists and
leading academics, consider issues including the influence of censorship and
propaganda, 24/7 rolling news, military jargon such as “collateral damage,”
embedded and unilateral reporters, and the visual imagery of war.
The book’s major focus on the Iraq war encompasses media coverage of the
run-up to war, the war itself, the anti-war movement in the UK, Europe, and the
US, and the role played by news sources outside the mainstream, including the
satellite channel Al-Jazeera and online reporting.
REPO RTING WAR
Journalism in wartime
INTRODUCTION
PA RT 1
War in the twenty-first century 23
1 Understanding: the second casualty 25
OLIVER BOYD-BARRETT
PA RT 2
Bearing witness 113
6 When war is reduced to a photograph 115
BARBIE ZELIZER
v
CONTENTS
PA RT 3
Reporting the Iraq war 245
13 Militarized journalism: framing dissent in the Gulf Wars 247
STEPHEN D. REESE
15 How British television news represented the case for the war
in Iraq 283
JUSTIN LEWIS AND ROD BROOKES
vi
CONTENTS
18 Big media and little media: the journalistic informal sector during
the invasion of Iraq 333
PATRICIA AUFDERHEIDE
Index 366
vii
CONTRIBUTO RS
Stuart Allan is Reader in the School of Cultural Studies, University of the West
of England, Bristol. He is the author of News Culture (Open University Press,
2nd edition, 2004) and Media, Risk and Science (Open University Press, 2002).
His previous edited collections include, with Barbie Zelizer, Journalism after
September 11 (Routledge, 2002).
Patricia Aufderheide is Professor and Co-director of the Center for Social Media,
School of Communication, American University, Washington DC. She is the
author of many works of cultural criticism and commentary, including The
Daily Planet: a Critic on the Capitalist Culture Beat (University of Minnesota
Press, 2000).
Oliver Boyd-Barrett is Professor of Communication at California State Poly-
technic University, Pomona. Recent books include Approaches to Media
(co-edited with Chris Newbold; Edward Arnold, 1995), The Globalization of
News (co-edited with Terhi Rantanen; Sage, 1998), and The Media Book (co-
edited with Chris Newbold and Hilde Van Den Bulck; Hodder Arnold, 2002).
Michael Bromley is Head of Journalism at Queensland University of Technology,
Australia, and has taught journalism in the UK and US. A former journalist,
he has published widely on journalism and the media, and is a founding co-
editor of the journal Journalism: Theory, Practice and Criticism.
Rod Brookes lectures at the Cardiff School of Journalism, Media, and Cultural
Studies, Cardiff University. He is the author of Representing Sport (Arnold,
2002).
Susan L. Carruthers is Associate Professor of History at Rutgers University in
Newark, New Jersey. She is the author of Winning Hearts and Minds: British
Governments, the Media and Colonial Counterinsurgency, 1945–60 (Leicester
University Press, 1995), and The Media at War: Communication and Conflict in
the Twentieth Century (Palgrave/Macmillan, 2000).
Nick Couldry is Senior Lecturer in Media and Communications at the London
School of Economics and Political Science. His books include The Place of
Media Power (Routledge, 2000), Inside Culture (Sage, 2000), Media Rituals: a
viii
CONTRIBUTORS
ix
CONTRIBUTORS
x
INTRODUCTION
RULES OF ENGAGEMENT
Journalism and war
“The very nature of war,” Kate Adie (1998) once observed, “confuses the role of
the journalist” (1998: 44). Confronted with the often horrific realities of conflict,
any belief that the journalist can remain distant, remote, or unaffected by what is
happening “tends to go out the window” in a hurry. Nevertheless Adie, at the
time the BBC’s chief news correspondent, offered no simple definition of what
the role of the journalist should be “when faced with the consequences of battle
and the muddle of war,” admitting instead that “I don’t have the answers, but I
keep on asking questions” (Adie 1998: 54).
War reporting, as Adie’s comments suggest, constitutes a litmus test of sorts for
journalism more broadly. While the role of war correspondent has long been asso-
ciated with a certain romantic lore, in actuality it is beset by an array of problems
associated with allegiance, responsibility, truth, and balance. Such problems arise
from time to time in the daily implementation of ordinary, everyday modes of
journalism, of course, but their apparent lack of easy resolvability in wartime
poses challenges that raise questions about the practice of journalism in more
forms than just reporting war. “We knew we were placing ourselves in the bull’s
eye of a war,” commented John Burns, chief foreign correspondent for The New
York Times, about the Iraqi conflict. “And I don’t think a journalist can sensibly
claim to have an exemption in a war zone. [By] the very nature of what’s about to
3
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
happen, there’s a significant risk that you can be killed” (CBC-TV, November 23,
2003).1 Journalists must be prepared to take that risk, but that does not make
them heroes, in his view. “When we find ourselves in very difficult situations, it’s
of our own choosing,” he added, but what is at stake is the need to “tell people as
much of the truth as you can.” Such observations, in our reading, suggest that tra-
ditional ways of thinking about the imperatives of war reporting and their
implications more broadly for journalism need to be examined anew.
The distinction to be drawn between patriotism and militarism is central to
the problems of reporting war. A reporter’s sense of national identity, however
defined, needs to be considered in a way that sheds light both on how it can
underpin journalism’s strengths while, simultaneously, recognizing the constraints
it can impose on the integrity of practice. Vital here is the (usually tacit) assump-
tion that covering the often harrowing nature of battle constitutes a higher order
of journalism. War journalists are thought to do what all journalists do, only in a
more heightened, vibrantly important fashion. To cover the story will entail,
more likely than not, encountering conditions of an entirely different order than
anything ordinarily associated with newswork. Images of the war reporter as
adventurer or risk-taker, in the optimum sense, or as daredevil, fortune-hunter, or
rogue, in the negative, help to fuel their celebration in novels, films, plays, and
other fictional treatments. Similarly implicit here, however, is the notion that
war reporters somehow “do journalism” better, that their experiences are more
authentic, engaged, and noteworthy than those of other kinds of journalists. And
yet, it is their very commitment to some rendering of national identity, even
patriotism, that is likely to engender a change in journalistic work. It may entail a
migration toward vague word choice, the absence of a broader perspective, the
lack of explicit images, even the wearing of flag pins. When journalists under
everyday conditions are seen to have strong sentiments for family, friends, or
community, they are often taken off the story that involves them. When their
sentiments for country are seen as strong in wartime, they are rarely removed
from the story; rather, the expectation, at least in some quarters, is that they will
simply change how they conduct themselves as journalists. The exceptional qual-
ity of this re-orientation deserves pause. It raises questions as to whether or not
our ways of thinking about war journalism, as a mode of practice, have fallen
short of the mark.
War reporting, in short, demands that notions of what constitutes good jour-
nalistic practice be realigned on the basis of different criteria than would typically
seem appropriate, criteria thrown into sharp relief—at times violently so—by
challenging circumstances. At the same time, war reporting’s positioning as a
litmus test for journalism also rests on an understanding of its capacity to influ-
ence public perceptions. Journalists are expected to function variously during
war: to be present enough to respond to what is happening, yet absent enough to
stay safe; to be sufficiently authoritative so as to provide reliable information, yet
open to cracks and fissures in the complicated truth-claims that unfold; to remain
passionate about the undermining of human dignity that accompanies war, yet
4
RULES OF ENGAGEMENT
impartial and distanced enough to see the strategies that attach themselves to cir-
cumstances with always more than one side. In these and related ways, then, war
reporting reveals its investment in sustaining a certain discursive authority—
namely that of being an eyewitness. “I was there to witness,” Adie recalled, “to
repeat what I hear, to observe the circumstances, note the detail, and confirm
what is going on with accuracy, honesty and precision” (Adie 1998: 46). Though
Adie admitted that she needed faith that what she sought to communicate would
not be in vain, it was not her mission to advocate particular forms of action. Nor
was it easy to counter the deceptive nature of the battlefield. Witnessing was “the
only way you can stand by your words afterwards, the only guarantee that you can
give your listeners, or viewers, or readers. You saw it, you heard it, you are telling
the truth as far as you know” (Adie 1998: 47).
This act of witnessing, of seeing for oneself the heart of the story, encapsulates
the larger problem of determining what counts as truth in the war zone. It forms
the basis of the journalist’s relationship with diverse publics, where any one
claim to truth will be interpreted in conjunction with varied—at times con-
flicting—normative criteria. Truth-telling, it needs to be acknowledged, is
necessarily embedded in a cultural politics of legitimacy; its authority resting
on presence, on the moral duty to bear witness by being there. Being there sug-
gests that the violence, devastation, suffering, and death that inevitably
constitute war’s underside will somehow be rendered different—more amenable
to response and perhaps less likely to recur—just because journalists are some-
where nearby. And yet the experience of a reporter’s being there, so important
for distant publics eager for news of the events of a war-torn region, is shaped
quite systematically by a weave of limitations—political, military, economic,
and technological, among others—that together may curtail the experience in
drastic ways.
Anecdotal evidence from a range of wars bears this out, whereby a journalist’s
capacity to be present was limited, undermined, or even denied altogether when
the battleground was placed off limits, the military barricade went up, the cam-
eras broke down, or sources refused to talk. Moreover, a journalist’s sense of
citizenship, even patriotism, may call into question his or her perceptions of how
best to conduct oneself as a reporter. All too often, journalists encounter those
who demand to know: are you with us, or are you against us? It is at this point
that individual journalists determine for themselves what their role should be,
knowing that their ad hoc decision may have profound implications for how their
audiences come to understand the nature of war and the consequences for its vic-
tims. In so doing, they know that even the most basic expectation of journalists
in wartime—being there—is rarely realized entirely in the way they may have
wanted, given the exigencies with which they must cope.
Perhaps nowhere did this receive as much attention as during the recent war in
Iraq. The decision to embed reporters with the US and UK military forces—as a
way of ensuring that journalists could “be there”—raised expectations that the
ensuing coverage would be better, more comprehensive, fuller, and more reliable.
5
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
6
RULES OF ENGAGEMENT
The big difference this time around is technology has got better, lighter,
easier to handle and it’s cheaper. It’s much more effective and so it’s not
unusual to go straight to live pictures out of, say, Basra. Often the pic-
tures are mundane, but you can cut to it at the flick of a switch . . . Last
Saturday we showed pictures of Iraqis down by the Tigris looking for
downed pilots and firing into the water. It was a live event and we went
with it—it was an incredible image, but in the end it didn’t amount to
anything.
(cited in Doward 2003)
7
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
Rolling news, its advocators insisted, proved far more effective in bringing the
war into people’s homes than more traditional newscasts. Sky News’ leading posi-
tion in the ratings was frequently attributed to its oft-repeated claim to be the first
to break stories. So-called “BBC insiders” refused to concede this point, however,
and also insisted that the corporation engaged in more thorough fact-checking
processes than its rivals, which took time.
All too often this incessant drive to be the first to break the story meant that due
care and accuracy were sacrificed in the heat of the moment. Examples from the
crisis in Iraq were all too plentiful, not least with respect to the array of claims
made regarding chemical and biological weapons of mass destruction (WMDs).
Reports broadcast on various rolling news networks in the US and UK also
included, for example, claims that Iraqi “drones” would be dispatched to spread
biological agents in the US; the Iraqi Republican Guard was planning to use
chemical weapons on coalition forces in the defence of Baghdad; a grenade attack
on a base in Kuwait was the work of terrorists (a US soldier was later charged); the
early fall of the port of Umm Qasr; various “bioweapon” caches, including missiles
containing sarin and mustard gas, had been found; the discovery of weapons-grade
plutonium; the seizure of various mobile “bioweapons” laboratories; the defection
of Tariq Aziz; the “uprising” in Basra, and so forth. These claims, among a myriad
of others, were later acknowledged to be false, if not necessarily by the network
which broadcast them in the first place (the Arab news network Al-Jazeera played
a vital role in challenging such claims, for example). A further, particularly con-
troversial example centred on the apparent “rescue” of Private Jessica Lynch from
an Iraqi hospital. Once again, the key claims made were eventually shown to be
untrue. “None of the absurd hype that surrounded the case came from her,” the
BBC’s John Simpson (2003) remarked afterwards; “it was all the invention of the
US Army spinners, and a credulous press desperate for some genuine heroics in a
war which seemed disturbingly short of gallantry” (Simpson 2003: 313).
Some officials responded angrily to allegations of “propaganda” and “spin”
where instances of false reporting occurred, insisting that such were the types of
mistakes typically made in the “fog of war.” Others blamed “over-enthusiastic
reporters” for getting carried away. Seldom acknowledged, however, were the
range of pressures brought to bear on journalists to influence the nature of the
coverage. Here remarks made by CNN’s Christiane Amanpour are telling. Inter-
viewed on CNBC by Tina Brown, she was asked to comment on assertions that
Bush administration officials had intimidated journalists, not least into feeling
unpatriotic if they gave voice to criticism or dissent. She stated:
I think the press was muzzled and I think the press self-muzzled. I’m sorry
to say but certainly television and, perhaps, to a certain extent, my sta-
tion [CNN] was intimidated by the administration and its foot soldiers
at Fox News. And it did, in fact, put a climate of fear and self-censorship,
in my view, in terms of—of the kind of broadcast work we did. I mean,
all of us should have . . .
8
RULES OF ENGAGEMENT
Here Brown interjected, asking: “And was there any story—was there any story
that you couldn’t do?” Amanpour replied:
All—it’s not a question of couldn’t do it, it’s a question of tone. It’s a ques-
tion of being rigorous. It’s really a question of really asking the questions.
All of the entire body politic in my view, whether it’s the administration,
the intelligence, the journalists, whoever, did not ask enough questions,
for instance, about weapons of mass destruction. I mean, it looks like this
was disinformation at the highest levels.
(CNBC transcript, “Topic A with Tina Brown,”
broadcast September 10, 2003)
9
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
are after cheerleading, they’re looking in the wrong place. It doesn’t mean we’re
not patriotic” (cited in Kurtz 2003).
A wide array of commentators have expressed their alarm about this apparent
blurring of news and comment, not least with regard to the implications for repor-
torial standards. Greg Dyke, Director General of the BBC at the time, focused
public attention on this problem in April 2003. Speaking at a journalism confer-
ence in London, he addressed the challenges confronting television news. “We
must temper the drama and competition of live, rolling news with the considered
journalism and analysis people need to make sense of events,” he argued (BBC
News Online, April 24, 2003). More than that, however, Dyke added: “Commer-
cial pressures may tempt others to follow the Fox News formula of gung-ho
patriotism but for the BBC this would be a terrible mistake.” In acknowledging
that Fox New’s partisan pro-Bush stance had helped it to overtake CNN in aver-
age daily viewer ratings, he insisted that such “unquestioning” support for the
White House was typical of the other US broadcast news media during the con-
flict. In a fragmented marketplace, he argued, no news operation was sufficiently
strong or brave enough to stand up to government or military officials. Attempts
to mix flag-waving patriotism with journalism, he feared, would inevitably under-
mine television news’ credibility in the eyes of the public. “Essential to the success
of any news organisation,” he stated, “is holding the trust of its audiences.”
Dyke returned to this theme when giving a speech at the International Emmys
in New York in November 2003, where he accepted an award for broadcasting
excellence. “News organisations should be in the business of balancing their cov-
erage,” he argued, “not banging the drum for one side or the other” (cited in BBC
News Online, November 25, 2003). Citing research suggesting that only a tiny
fraction of the commentators aired on US television were opposed to the war, he
said: “I have to tell you if that was true in Britain, the BBC would have failed in
its duty.” In making the point that television news has a responsibility to broad-
cast a range of voices, he observed that viewing figures in the US for BBC News
(namely BBC World and News 24) had effectively “doubled” in the last year. For
Dyke, these figures showed that there was a growing audience for “impartial”
news. “Telling people what they want to hear is not doing them any favours. It
may not be comfortable to challenge governments or even popular opinion, but
it’s what we are here to do.” BBC news reporting was hardly above criticism, of
course, being consistently assailed from both ends of the political spectrum.
Those on the right contended that it was “anti-war” in its coverage, while those
on the left believed that it relied much too heavily on pro-war sources of informa-
tion (echoes of this debate resounded throughout the Hutton inquiry and its
aftermath).2
Much of the criticism of the British 24-hour networks’ coverage of the Iraq
conflict centred on their widely perceived tendency to present images without
adequate context or explanation (effectively making them purveyors of “war
porn” in the judgment of some). Differences between the three networks—BBC
News 24, the ITV News Channel, and Sky News—in their coverage were most
10
RULES OF ENGAGEMENT
notable at a stylistic level, given that they each operate under strict conditions of
“due impartiality” imposed by broadcast regulators. In sharp contrast with the so-
called “Fox formula” (likened, by some, to an “either you’re with us or against us”
mentality), the three networks share similar space in the middle of the news spec-
trum. As Matt Wells (2003) has pointed out, not only do they have broadly
similar news agendas, with correspondents situated in the same places, they also
hold a “determinedly Western” perspective in common. During the conflict the
“most frequent location of live two-way interviews,” he observed, were “coalition
command in Doha, military headquarters in Kuwait, the White House in Wash-
ington and Downing Street in London. Jerusalem, Amman, Cairo and Riyadh
have barely had a look-in.” The extensive use of such interviews, needless to say,
underscored the networks’ commitment to meeting the near-constant demands
of live reporting (which appeared with less frequency on the terrestrial networks
as the war went on—one major exception being the capture of Saddam Hussein).
For some critics, however, this approach to two-ways not only narrowed the
range of views on offer, it also seemed to value immediacy for its own sake. “Tele-
vision has become a 24-hour slog,” argued veteran war reporter Jon Swain
(2003), “with the result that while many of today’s TV reporters may have all the
traditional dedication and intrepidness of their predecessors, they cannot use it.”
In effect, he added, “[t]hey are tied to the satellite dish on the hotel roof ready to
deliver ‘live spots’ and so are unable to explore in depth the stories they are sup-
posed to be reporting” (see also Adie 2002).
Jon Snow (2004), anchor of Channel 4 News, has coined the phrase “Rooftop
syndrome” to describe the ways “war correspondents are held hostage by the vora-
cious appetite of 24-hour television news.” One such victim, he contends, was
the BBC’s Rageh Omaar, “the face of the war in Baghdad,” but whose rooftop
appearances on the Information Ministry and the Palestine Hotel meant that he
seldom had the opportunity to use his skills (see also Omaar 2004). Similarly per-
tinent here are the insights of Martin Bell (2003), recently retired from the BBC
after more than 30 years in journalism. The 24-hour news services have special
responsibilities, he believes, which are defined by F-words:
They aim to be first and fastest with the news. Their nature, too often, is
to be fearful, feverish, frenzied, frantic, frail, false and fallible. Some mis-
takes are bound to be made, as they always have been, by journalists
seeking to discover the truth in the fog of breaking news; but those mis-
takes do not have to be as systemic as they have become in the rolling
news business, when rumour masquerades as fact, and networks compete
wildly with each other to get their speculation in first.
(Bell 2003: 71; emphasis in original)
Bell is particularly troubled where the reporting of war and terrorism is con-
cerned, but believes the point can be made more widely. In calling for a measure
of self-criticism among journalists, as well as a code of practice (long overdue, in
11
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
his opinion), he proceeds to issue an appeal for a return to first principles. The
test of excellence, he argues, is not “We got it first!” but rather “We got it right!”
(Bell 2003: 71).
One year after the invasion of Iraq commenced, answers to the question of
how best to ensure reportorial standards of excellence are proving increasingly
elusive. Nowhere are the imperatives shaping television news thrown into
sharper relief than where the interests of public service collide with the private
ones of shareholders. To the extent that the news agenda is determined by its
potential for generating audience ratings (and therefore advertising revenue, in
the case of commercial networks), newscasts will consistently prioritize stories
revolving around drama, conflict, and controversy over and above (expensive,
less “ratings-efficient”) investigative journalism. War reporting is no exception to
this general rule. The dynamics underpinning 24-hour news—especially when
“going live”—raise significant issues regarding public perceptions about the
nature of war.
Jack Straw (2003), the British Foreign Secretary, has even questioned whether
the First and Second World Wars would have been won, had they been covered
by these channels:
Had the public been able to see live reports from the trenches, I wonder
for how long the governments of Asquith and Lloyd George could have
maintained the war effort. Imagine the carnage of the Somme on Sky
and BBC News 24. But it is also worth speculating how much harder it
might have been to maintain the country’s morale after Dunkirk had
live reports confronted the public with the brutal reality of German tac-
tical and military superiority. Could the ‘spirit of Dunkirk’, so important
to national survival, have withstood the scrutiny of 24-hour live news?
(Straw 2003)
Straw himself was doubtful, but nevertheless made the important additional
point that “in a democracy, the benefits of hour by hour and day by day reporting
from the frontline far outweigh the disadvantages.” And as he rightly pointed out,
some correspondents on the frontline have “paid the ultimate sacrifice in pursuit
of the truth” (24 in Iraq at the time of writing). Still, there can be little doubt that
his interests as a Foreign Secretary necessarily contradict, at times, the interests of
journalists committed to investigating the realities of warfare. Precisely how
these contradictions are negotiated on the ground, of course, will have profound
consequences for public trust in both government and journalism in times of war.
The chapters
Reporting War: Journalism in Wartime addresses the tensions, contradictions, and
contingencies that shape the journalist’s role in wartime. Recognizing that the
aim of war coverage is frequently defined in relation to the urgent need to keep
12
RULES OF ENGAGEMENT
diverse publics adequately informed, Reporting War surveys the ways in which this
aim has been realized—or not—in the reporting of certain wars over recent years.
Our contributors, both practising and former journalists as well as academics with
a longstanding research interest in journalism, vary in perspective, academic dis-
cipline, and national origin, but they share the aim of recasting familiar
assumptions about war reporting from a distinctive perspective.
Interwoven throughout the book are two main threads in thinking about the
relative strengths and limitations of war reporting. First, the book tracks the
forms and practices of reporting war across time. Focusing in the main on con-
flicts since 1990, it explores how the nature of war itself has been transformed,
and how these changes have impacted upon the modes for journalistic engage-
ment. These modes—typically characterized too narrowly around evolutions
in technology—will be shown to prefigure a repository of adaptations by which
journalists constitute themselves as responsible for bearing witness. Second, it
considers the reporting of war in diverse locales. Ranging from Africa, the
Balkans, the Middle East—especially Iraq—and southern Asia, the book attempts
to discern both the general features and distinct specifics of reporting in compar-
ative contexts. In so doing it asks, among other questions, in what ways do the
US and British news media, in particular, bring to bear Western assumptions—
political, cultural, and moral—about how and why war is waged.
Reporting War identifies and critiques an array of pressing issues associated with
conflicts over recent years, always with an eye to what they can tell us about
improving journalism today. Such issues include the influence of censorship and
propaganda, “us” and “them” news narratives, access to sources, “24/7 rolling
news” and the so-called “CNN effect,” military jargon (such as “friendly fire” and
“collateral damage”), “embedded” and “unilateral” reporters, visual images of war,
tensions between objectivity, patriotism and humanitarianism, and online report-
ing, among others. Sustained attention is devoted to considering changes in
journalistic forms and practices in the war in Iraq, and the ways in which they are
shaping eyewitness reporting of that conflict. Taken together, then, the book’s
chapters raise important questions about how the exigencies of reporting war
have challenged the practice of journalism, and what they may portend for the
very future of journalism tomorrow.
13
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
14
RULES OF ENGAGEMENT
15
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
16
RULES OF ENGAGEMENT
In “Out of sight, out of mind?: the non-reporting of small wars and insurgen-
cies,” Prasun Sonwalkar draws attention to the conflicts routinely ignored by the
media on national as well as international levels. He points out that the news
media typically cover only a small share of the world’s conflicts, even though
terrorism and violence are usually defined as being inherently—and irresistibly—
newsworthy and draws upon the findings of a case study of reporting in India, the
two conflict zones of Kashmir and the northeast. A conflict is likely to receive
coverage, he suggests, only if journalists see it as affecting what they perceive to
be the “us” or “we” of news narrative, while a conflict revolving around “them”
may be routinely ignored or accorded ad hoc coverage. Sonwalkar’s interviews
with senior Indian journalists reveal that Kashmir is seen as deserving of sus-
tained coverage because it involves the Indian national “we” (Hindus, Muslims,
the idea of secularism in India’s constitution), while the northeast conflicts are
consistently accorded low status, primarily because they involve India’s socio-
cultural “they” (tribes, Christians, hill people).
Which war is represented in Australia’s media arguably owes more to the cul-
tural and historical location of war reporting than to the power of oligopolistic
media ownerships over journalistic performance. This is the opening thesis of
Michael Bromley’s “The battlefield is the media: war reporting and the forma-
tion of national identity in Australia—from Belmont to Baghdad,” which
focuses on how journalists have shaped what war has meant to Australians and,
in so doing, revalidated certain Australian national values originating in the
foundation of the nation-state. Bromley argues that while much was made of the
pro-invasion stand taken by media mogul Rupert Murdoch and his newspapers
before the war in Iraq; journalists could be relied on a priori to invoke and evoke
a mythology of war which tied military action seamlessly to Australian identity.
They were therefore largely supportive of military action. Notwithstanding con-
siderable public unease at Australia’s involvement in the invasion, once military
deployments were undertaken, journalists’ automatic reference point was the
ideal of the Australian “digger” (soldier) and the positive national characteristics
embedded in it.
17
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
dissent so as to pit it against a favored patriotic “support the troops” position. The
extent to which a military logic colors policy debate far from the war itself is con-
sidered in the light of more recent changes in the battlefield information
environment of Gulf War II in Iraq, and the location of audiences within more
globalized communities.
Nick Couldry and John Downey, in “War or peace?: legitimation, dissent, and
rhetorical closure in press coverage of the Iraq war build-up,” direct our attention
to British newspaper coverage of the onset of war. Analyzing the discursive posi-
tions and rhetorical strategies adopted by seven national newspapers during a
crucial week in the build-up to the 2003 Iraq war—the week of presentation of the
Blix report (beginning January 27) to the United Nations Security Council—they
show how these newspapers attempted to legitimize both war and peace. While
various notions such as the manufacturing of consent, the public relations state,
and the colonization of the public sphere emphasize the role of media in the repro-
duction of consent, the role of national newspapers in legitimating dissent was also
key. Couldry and Downey also examine the limits of this legitimation of dissent in
the build-up to the invasion of Iraq which they argue help to explain the pro-war
shift in public opinion discernible at the beginning of the war.
Justin Lewis and Rod Brookes offer what is in many ways a complementary
analysis, in “How British television news represented the case for the war in Iraq.”
Examining key features of the British television coverage of the war in Iraq, based
on an analysis of over 1,500 news reports broadcast by the British Broadcasting
Corporation (BBC 1), the Independent Television Network (for ITV), Channel
4, and Sky Television, they look at the overall shape of the coverage and then
explore the degree to which news channels accepted or challenged pro-war
assumptions during the war, notably about the threat of weapons of mass destruc-
tion and the enthusiasm of Iraq people for the US-led war. In so doing, Lewis and
Brookes explore a number of controversial issues. Was the BBC’s coverage anti-
war as many alleged? Conversely, were the “embedded reporters” effectively “in
bed” with the military? In answering these questions, they consider the relation-
ship between media coverage and public opinion.
When the first US missiles hit targets in Baghdad on March 20, 2003, news
organizations started hitting their targets around the world. Terhi Rantanen, in
“European news agencies and their sources in the Iraq war coverage,” discusses
how the war coverage affected the daily routines of European news agencies, such
as Reuters and Agence France-Presse. These agencies, she points out, are tradi-
tionally seen as the wholesalers of news to most media, and in a competitive global
media environment, have become one of the many sources used by the media. In
her chapter, Rantanen explores the measures the agencies took to cover the war
and the sources to which they subscribed. She also studies how news agency edi-
tors evaluated their news sources, suggesting that they share a professional
journalistic culture which goes beyond national boundaries. In the Iraq war, she
argues, sources outside Europe were seen as less reliable, and this applied to both the
US (such as the Associated Press) and Arab media (such as Al-Jazeera).
18
RULES OF ENGAGEMENT
Adel Iskandar and Mohammed el-Nawawy make Al-Jazeera the primary focus
of their chapter, “Al-Jazeera and war coverage in Iraq: the quest for contextual
objectivity.” Compared to its predecessors, Al-Jazeera represents a media revolu-
tion in the Arab world, the implications of which become particularly pronounced
at times of war. Indeed, Al-Jazeera’s success as a news organization with immense
popular appeal, they argue, is a product of its novel brand of journalism. To
describe the journalistic model within which the network operates, the authors
introduce the concept of “contextual objectivity.” Of particular interest in this
concept are the tensions embedded within it, along with the media’s reactions to
its application in war reporting—not least with regard to how it complicates the
work of journalists and media outlets alike. Iskandar and el-Nawawy argue that
this concept possesses the potential to serve as a useful standard by which all net-
works may be judged, a thesis which they explore via a close look at Al-Jazeera’s
coverage of the war in Iraq.
Attention turns in the last two chapters to the Internet, beginning with Patri-
cia Aufderheide’s “Big media and little media: the journalistic informal sector
during the invasion of Iraq.” Describing the flourishing of grassroots reporting,
opinion, and commentary on mainstream media during March and April 2003,
when a vigorous public-opinion minority in the US, along with international
voices, protested the US-led invasion of Iraq, she reports the findings of a
research project at American University’s Center for Social Media that surveyed
the journalistic “informal sector” during the war. Aufderheide contrasts this phe-
nomenon—which involved email, the Web, cable access, and public channels on
satellite TV—with the suppression of dissent on the part of large, entertainment-
oriented, US media companies. She proposes that the informal sector will
continue to grow and that relationships will develop between mainstream media
outlets and efforts in the informal sector, offering important evidence of a widely
shared demand for more public participation in decision-making and widespread
distrust of mainstream media reporting in crisis.
In drawing the book to a close, Stuart Allan’s “The culture of distance: online
reporting of the Iraq war” uses cultural theorist Raymond Williams’ notion of the
“culture of distance” to contextualize online journalism. Allan explores the
Web-based mediation of witnessing and the distant suffering of others by
addressing the use of the Internet as a news source during the conflict. He con-
siders both the online reporting of Al-Jazeera (www.aljazeera.net) and the rise of
warblogs, such as that belonging to Salam Pax, otherwise known as the “Bagh-
dad Blogger” (dear_raed.blogspot.com). Attention focuses on the ways in which
online reporting provides alternative spaces for acts of witnessing, a process
which is shown to be uneven, contingent, and frequently the site of intense
resistance. The culture of distance, Allan contends, is simultaneously a culture
of othering. At stake, in his view, is the need to deconstruct journalism’s “us
and them” dichotomies precisely as they are taken up and re-inflected in news
accounts where the structural interests of “people like us” are counterpoised
against the suffering of strangers.
19
S T U A RT A L L A N A N D B A R B I E Z E L I Z E R
Conclusion
All in all, Reporting War: Journalism in Wartime has at its heart a commitment to
exploring afresh the social responsibilities of the news reporter in public life
during wartime. It is not chronological, in that it leaps back and forth in time. It
is not comprehensive, in that it has not attended to every war that has occurred
over recent years. Neither has it conclusively raised all of the relevant issues in
the wars discussed in these pages. Even though the war in Iraq is still under way as
we edit this book, our hope is nevertheless to offer a timely intervention. In so
doing, we aim to contribute to a larger critical appraisal of both the guiding
imperatives of journalism in wartime and the ways in which these imperatives
impact more broadly across different modes of journalistic practice.
The echoes of war continue to resound long after the cacophony of battle has
been silenced. In this book, we have endeavored to create a forum of contem-
plation around war reporting—past and present—so as to help create spaces for
alternative voices to be heard. Our contributors, some of whom are former
reporters, have undertaken this work as academics committed to the study of news
and journalism, but also as citizens. “Citizenship,” as one of the last century’s finest
war correspondents, Martha Gellhorn (1988: 408), pointed out, “is a tough occu-
pation.” She believed that as citizens we are obliged to make our own informed
opinion and to stand by it. “The evils of the time change,” she observed, “but are
never in short supply and would go unchallenged unless there were conscientious
people to say: not if I can help it” (Gellhorn 1988: 408). Dissent, based on moral-
ity and reason, is at the heart of what it means to be a citizen. And while the
challenge of citizenship may be getting more difficult all of the time, there is nev-
ertheless always room for optimism. “There has to be a better way to run the
world,” Gellhorn insisted, “and we better see that we get it” (Gellhorn 1988: 409).
Notes
1 A transcript of this interview is available on the Canadian Broadcasting Corporation
webpage: http://www.cbc.ca/deadlineiraq/burns.pdf
2 The Hutton inquiry, named after Baron Hutton of Bresagh, was set up to investigate
the chain of events which led to the Government weapons expert David Kelly appar-
ently committing suicide after having been named as the likely source of a BBC report
by Andrew Gilligan. In the BBC report, Gilligan had maintained that an intelligence
dossier on Iraq had been altered under pressure brought to bear by the Prime Minister’s
office. The inquiry report largely exonerated the government, sparking allegations of a
“whitewash” by critics. A copy is available online at: http://www.the-hutton-inquiry.
org.uk/
References
Adie, K. (1998) “Dispatches from the front: reporting war,” Contemporary Issues in British
Journalism, the 1998 Vauxhall Lectures, Centre for Journalism Studies, Cardiff
University.
20
RULES OF ENGAGEMENT
21
Part 1
Oliver Boyd-Barrett
25
OLIVER BOYD-BARRETT
26
T H E S E C O N D C A S U A LT Y
Causes of war
Causes of warfare are typically problematic. I draw a distinction between proxim-
ate and deep-level strategic causes. At another level are broader, longer-term,
non-strategic causes unlikely to be decipherable in their entirety by the actors
involved, but which will be pried to the surface by historians. Proximate
causes have curious relationships to deep-level causes. The “aggression” of one
party, for example, may have been provoked by an apparent “victim,” represent-
ing pre-emptive retaliation against the “victim’s” maneuver toward its own,
unspoken, deep-level ambitions. Proximate causes are generally those cited by
27
OLIVER BOYD-BARRETT
the participants, often formulaic and often bogus. The opening gambits of war-
fare occur when journalists are most vulnerable to manipulation by official
sources. In contemporary warfare officialdom seeks to monopolize communica-
tion flows by limiting journalistic access to sanitized information from official
sources, by rationing transportation and communications facilities, excluding
non-approved journalists from military protection and facilities, keeping jour-
nalists out altogether—as in Grenada (1983) and Panama (1989), corralling
select numbers into press “pools”—as in the Falklands (1982), Grenada (1983),
Panama (1989), Afghanistan (2001), and Gulf War I (1991–2) (Thrall 2000), or
“embedding” them within military units—as in Gulf War II (2003).
Official controls are also most likely to work effectively in the opening stages of
conflict. Issues of legality and morality may be under-articulated or unclear. Cri-
tiquing the US administration while troops initiate combat attracts accusations
of lack of patriotism, even treason (especially if reporters attend diligently to
information from the opponents’ side), and endanger journalists’ already
restricted access to official sources and to the battlefield. This very moment when
proximate and deep-level causes of war should be open to rigorous analysis is also
the time when such analysis is most difficult and least popular. Dissection and dis-
putation of causes can be a heady, intellectual, wordy, controversial exercise, at a
time when the media are most of all preoccupied with covering the action, at the
behest, they would say, of audience interest. Wars can and often do start with
greatly insufficient attention to the possible multiplicity of real causes; if, when
war formally begins, journalists have sacrificed context, the chances that they will
recover it in time, before the issues become history, are not encouraging. The
problems are compounded when journalists and audiences are ignorant as to the
locations of battle, their cultures and societies, as was the case in Afghanistan in
2001, and Iraq, in both 1990–1 and 2003.
Confusion over the causes and preludes to war is a common failing in war
reporting, which makes impossible an assessment by citizens about the meaning-
fulness of the very combat that is its focus. This was as true of the second Gulf War
in 2003 as of the first in 1990. The intelligentsia was better informed the second
time: they had studied the lies of the first war; were aware of the controversial
impact of UN sanctions on the civilian population (especially children) and of
the “no fly zones” imposed by the US and Britain, and many doubted the claims
from Washington and London that Iraq posed an immediate threat to the US or
was in possession of significant stockpiles of chemical, biological, or nuclear
weapons. Such doubts fuelled unprecedented worldwide anti-war mass demon-
strations. A second difference was the speed with which the rationale for war
collapsed after initial combat, during the first weeks of occupation. Increasingly,
it appeared that the case had been built on exaggerations and outright decep-
tions. This was no surprise to anyone who had followed many of the “alternative”
news websites that provided day-to-day critical analysis drawn from a much wider
diversity of sources than available through the mainstream US media, and from
long before the war started as well as during and after the war. Among main-
28
T H E S E C O N D C A S U A LT Y
stream US media, however, the networks (terrestrial and cable news channels in
the US) uniformly banged the drums for war long before the onset of invasion.
During the war, television networks’ news frames complied fully with US admin-
istration policy: they represented war as a necessary response to an immediate
threat by Iraq, justifiable even without UN sanction, whose resulting conse-
quences of death and destruction, both military and civilian, were inevitable
sacrifices. Within that frame there was scope for mild disagreement among the
Pentagon-vetted ex-military network commentators as to whether sufficient
numbers of troops had been deployed and whether initial combat strategies were
appropriate. But no network acknowledged the appropriateness of an entirely
alternative frame—namely that Iraq did not pose a significant threat to the US,
that the invasion did not have UN sanction but constituted an aggressive breach
of Iraqi sovereign territory which the Iraqis were fully justified in defending, and
that the consequences of war in terms of death and destruction amounted to war
crimes attributable to the US and its allies (Boyd-Barrett 2003b). Unlike tele-
vision news, the mainstream press captured a broader range of perspectives in the
period leading up to the war and occasionally achieved significant exposures
(including Defense Secretary Donald Rumsfeld’s apparent acquiescence to Iraqi
use of chemical weapons in its war with Iran in the 1980s), but there was a cus-
tomary lack of outrage in face of growing evidence of subterfuge in the US
administration’s diplomacy and its case for war. Both before and during the war,
the press was slow to acknowledge the extent to which it had been routinely
duped by misinformation and fabrication (including the highly staged toppling of
the statue of Saddam Hussein, and the Hollywood-style make-believe attention
to the rescue of Private Jessica Lynch). Neither television nor the press seemed
capable of assessing and listening carefully to non-government expert sources,
including that of former UN weapons inspector Scott Ritter, who had argued that
the UN had indeed detected and supervised the destruction of most weapons of
mass destruction (WMD), and that any left in early 2003 would be inoperative.
29
OLIVER BOYD-BARRETT
military personnel, strategies, successes, and, less often, failures, and backed with
comments from (often vetted) Western military “experts.” This myopia might be
attributed to media reluctance to be seen as relying on “unreliable,” “censored,” or
“unverifiable” reports, a hypocritical position that is blind to media dependence
on government or military sources of their own side for their most regular, profes-
sionally scripted or staged, and above all safe information, disinformation or lies.
When Gulf War I broke in 1990–1, the White House, putting into practice the
lessons it had learned from the Falklands War and from the US invasions of
Grenada and Panama during the 1980s, knew how to implement its strategy of
“control by press pool.” Several pools were organized, although few saw combat.
Those journalists not selected for pools stayed in Dahrain, viewing televised brief-
ings. Pool allocations were organized by the media themselves; big media rigged
the process to suit themselves. Pool reporters were accompanied at all times by
military escorts, had limited access to troops in the battlefield, and so were often
unable to provide an independent first-hand view of combat. Military public
affairs escorts warned pool members against violating guidelines, barred them from
places “they should not be,” and reviewed press stories before transmission back to
the press corps, if necessary passing them up to the military command for approval.
Independent journalists risked getting detained, lost, captured, or killed.
Afghanistan in 2001 was very different. Here, correspondents had exceptional
freedom to roam the country, but they did so with little protection or other sup-
port from the military. The build-up of American and Alliance forces following
September 11 generally occurred without media participation, and when Amer-
ica unleashed its first wave of attacks in October, “only a handful of journalists
enjoyed a vantage point within Afghanistan” (Dalglish et al. 2003: 14). Forty
journalists were allowed to join military forces on the USS Enterprise and two
other warships, but these ships were incidental to the strikes and restrictions were
imposed on what could be published. “In effect, most American broadcasters
scratched out coverage from Pentagon briefings, a rare interview on a US aircraft
carrier or a humanitarian aid airlift, or from carefully selected military videos or
from leads.” Reporters seldom had interviews with troops or secured positions
near the front.
Gulf War II in 2003 introduced the Pentagon concept of the “embedded”
reporter. Schechter (2003) argued that the media became part of a psychological
warfare directed to the domestic US population. Its objective was to stifle dissent,
garner unquestioning support, and rally around a common symbol. On the
ground, the Pentagon (Dalglish et al. 2003) provided billeting, rations, and med-
ical attention and some assistance with communications if necessary to as many
as 1,000 journalists, 662 of whom were attached directly to armed service units.
Reporters were discouraged from going out on their own. For the bigger story, the
US administration set the agenda through its “messages for the day.” These
ensured consensus across the administration and set the media agenda. The work
of embedded reporters was subsidized by the Pentagon, overseen by “public
affairs” specialists, and linked to television networks that were dominated by mil-
30
T H E S E C O N D C A S U A LT Y
Covert war
Some conflicts do not acquire the status of “war” in media eyes, though they may
be as violent, devastating, and, above all, strategic, as formal military conflict.
Conflicts in which the West has non-legitimate or illegitimate interests (in par-
ticular relating to the extension of capitalist markets) tend to invite covert
Western intervention and escape critical media attention. Former CIA officer
Ralph McGhee (quoted by Scott 1985) illustrated this when he averred that
“where the necessary circumstances or proofs are lacking to support US interven-
tion, the CIA creates the appropriate situations or else invents them and
disseminates its distortions worldwide via its media operations” (McGehee 1981).
Pilger (2001) related the case of the overthrow of the nationalist leader of
Indonesia, President Achmed Sukarno, by General Suharto in 1965–6. Suharto
was dictator of Indonesia for the following 30 years. Sukarno’s overthrow had
unleashed a pogrom, supposedly against suspected Communists but claiming the
lives of between 500,000 and a million people, including at least 80,000 in the
island of Bali, a slaughter that the CIA described as one of the worst mass murders
of the twentieth century (1968 CIA report quoted by Pilger 2001: 25). Britain and
the US had agreed in 1962 on the need to “liquidate” Sukarno, who as founding
father of the non-aligned movement of developing countries, had sought a “third
way” between the interests of the two superpowers. He had nationalized US eco-
nomic interests and kicked out the International Monetary Fund and the World
Bank and used aid subsidies from the Soviet Union to confront the British in
Malaysia. The USA, under the executive leadership of Marshall Green, US
ambassador in Jakarta, with the help of Britain and Australia, spread anti-PKI and
anti-Sukarno propaganda, organized and propagandized the student movement,
equipped the army, handed over field communications networks, provided naval
escorts for Suharto military, and furnished lists of communist operatives for the
death squads (Kathy Kadane, quoted by Pilger 2001: 29). In covering Indonesia
then and later, Western media, including Murdoch’s The Australian, uncritically
framed the Suharto takeover as a positive outcome in the battle against commu-
nism. Many journalists celebrated the slaughter of up to a million people as good
news, while attacking those who accused Suharto of human rights violations.
The overthrow of Sukarno provided a model for American-directed death
squads in Operation Phoenix in Vietnam in 1967 which assassinated 50,000, and
for the US prosecution of civil war in Guatemala in 1968 from whence the model
31
OLIVER BOYD-BARRETT
spread to the rest of Latin America, including the overthrow of Salvador Allende
in Chile in 1973. More significantly yet, Suharto’s triumph was capped in 1967 by
a conference in Geneva sponsored by the Time-Life Corporation at which the
world’s leading transnational corporations (then as now, mainly American and
British) met with Suharto’s economic team. The conference planned Indonesia’s
transition, under a plan inspired by Ford Motor Co., in alliance with CIA front
organizations, to a “market economy,” and involved the parceling out of the
Indonesian economy to Western interests along with significant tax concessions.
The conference paved the way to the formation of the Inter-Government Group
on Indonesia, which Pilger (2001: 39) described as the “real and secret” center of
control of the Indonesian economy, and to the initiation of World Bank loans
that Sukarno had sought to avoid, sinking Indonesia into a $270 billion debt load
by 2000, even while the World Bank “lost track” of $10 billion of investments
that went to Suharto.
Issues of covert war were of considerable importance throughout the Cold War
and again during the “war on terrorism” (a war that, like the Cold War, also
resembles what I call the “designer” war below). Saunders (2003) noted that the
unofficial title for this war among top officials and ground-level operatives in the
US military and CIA was World War IV. He wrote (pp. 1–2):
Below the surface are dozens of operations, some secret and some simply
unnoticed, conducted by the CIA, the FBI, the diplomatic corps and
small, elite military squads . . . And much of the war is being fought by
foreign governments that are willing and able to do things Americans
wouldn’t or couldn’t . . . In some cases, that cooperation has led the
United States to endorse and enable activities that are deeply unsavory,
all in the name of stomping out terrorism.
Designer war
It is the “unofficial” causes and purposes of war that tend to get marginalized by
mainstream media in preference to the narratives relayed by official sources, the
sources that dominate all sourcing for mainstream media. This applies to conven-
tional wars and to conflicts or activities that may or may not be called wars.
The discourse over the “war on drugs” establishes an agenda in the Western
32
T H E S E C O N D C A S U A LT Y
media that focuses on: the scale of drug addiction in the US and elsewhere; the
efforts of legal authorities to enforce drug laws; whether the war is being “won”
or “lost;” and, less commonly, the impact of drug crime on prison populations
and control. The costs of the war, rising from $1 billion in the Reagan years to
nearly $20 billion a year by 2002 (Troshinsky 2002) are less readily caught by
this agenda, which also obscures how money earmarked for the “drug war” is
often spent in support of counter-insurgency and, not coincidentally, in the pro-
tection of oil pipelines, as in Colombia. The media agenda marginalizes darker
issues associated with collusion between US government agencies and the drugs
business, and the exploitation of the “war on drugs” as a pretext for different
goals, including the access to minerals, protection of oil interests in Third World
countries, financing covert operations and supporting unsavory but useful
“assets” (Scott and Marshall 1998; Webb 1998; Cockburn and St Clair 1999;
McCoy 2003). In Panama, for example, General Noriega was for many years
favored by Washington; he was a CIA asset, carefully protected, among others,
by George H.W. Bush senior when Bush directed the CIA from 1976 to 1978,
and again when he became Vice President to Reagan. Noriega was helpful to
Washington in the Iran/Contra scandal, when the planes flying personnel
and supplies to fight the dirty war in Nicaragua—financed by Iranian money
that paid for US supply of arms to Iran via Israel—returned to the US laden
with drugs (feeding the “crack” crisis in impoverished, ethnic, inner-city zones
such as South Central Los Angles). Noriega disappointed Washington when
he refused to withdraw the treaty that his predecessor had signed with President
Carter, under which control of the Panama Canal would revert to Panamanian
control in 2000. The ostensible reason for the US invasion of Panama in
1989 was to protect the lives of US citizens (although, apart from some dubious
incidents most likely provoked by US special forces, it is doubtful that any US
lives were really at risk before the invasion, and to “restore” democracy (critics
noted that Panama had never had a functioning democracy). The immediate
outcome of the war was that Noriega was sent for trial on drugs charges to
Florida. But the far more significant outcome was that the Panamanian military
was liquidated, and a new US puppet government, in tune with US foreign
policy and business interests, was installed.
In Afghanistan, the Taliban had been persuaded by the US administration to
eradicate the opium trade, by means of a ban on opium cultivation in early 2001,
which Washington rewarded with a $43 million grant. Yet following the US
bombing of Afghanistan in 2001 and the routing of the Taliban, Afghanistan
became the world’s leading supplier of opium. If the proximate reason for the “war
on drugs” is the volume of illegal drugs entering the US, two important “deep
level” causes have to do with oil, and with the financing of covert paramilitary
operations (as in Laos in the 1960s, and in both Nicaragua and Afghanistan in
the 1980s). Consistent involvement in drug operations by the CIA (and other
intelligence and para-military organizations) in all parts of the world have been
reliably charted (Cockburn and St Clair 1999); such operations finance and help
33
OLIVER BOYD-BARRETT
provide a cover for covert military and political operations, bypass legal con-
straints, and provide substantial illicit earnings to the participants and their
political sponsors.
34
T H E S E C O N D C A S U A LT Y
35
OLIVER BOYD-BARRETT
DU. Soldiers were ill-trained and ill-equipped to deal with radioactive munitions
in Gulf War I. It seems that during Gulf War II soldiers took considerable precau-
tions to avoid contamination, but Worthington (2003) noted that “any soldier
now in Iraq who has not inhaled lethal radioactive dust is not breathing.” Conta-
mination was even more likely to affect citizens. In addition to the US troops,
countless Iraqi and other civilians are believed to have suffered from a range of
health problems attributable to DU (Hecht 2003; Traprock Peace Center 2003).
Corroborating sources include the Royal Society, laboratories in Switzerland and
Finland, radiation expert Dr Helen Caldicott, a 1993 Pentagon report, the US
Army Environmental Policy Institute, the Uranium Medical Research Center in
Afghanistan (sources quoted in Worthington 2003). Dr Asaf Durakovic, profes-
sor of nuclear medicine at Georgetown University and a former army medical
expert found that 62 percent of sick veterans had uranium isotopes in their bodies
and even semen. A Christian Science Monitor report (Peterson 2003) found high
levels of radiation in the vicinity of DU bullets littered throughout Iraq. Few such
sites had warning notices. DU has also been used in military practice maneuvers
in many US states. The half-life of DU is 4.5 billion years, the age of our solar
system.
36
T H E S E C O N D C A S U A LT Y
depict wars as struggles within and between nation states, as media persistently
do, misses this fundamental dimension.
The metanarrative of hegemony takes different forms in different periods and
is currently, in the form of the National Security Strategy pronounced by Presi-
dent Bush in September 2002, as robust as it has ever been. Although presented
by the media as a new development in response to the events of September 11,
2001, much the same principles of preventive war supremacy had been enunci-
ated in the past by Dean Acheson in 1963 (Barsamian 2003) and by George
Kennan in 1948. Kennan noted that “we have about 60 percent of the world’s
wealth, but only 6.3 percent of its population. Our real task in the coming period
will be to maintain this position of disparity” (Gwyn 2003). Stromberg (2002)
noted that by the late nineteenth century, US leaders had “asserted a non-
negotiable right of the US government to be “secure.” US planning in 1943–4 for
the post-war period determined that future US defense would require effective
control of the Atlantic and Pacific Oceans, and that it could not allow rival
power to control the Eurasian land mass.
US hegemony is not simply about political or military hegemony nor even pri-
marily about these things. It is about economic hegemony and the US assertion to
its right to impose capitalism worldwide. Though terms such as “democracy” have
been frequently employed (as in William Kristol and Robert Kagan’s 1996 call for
a US “benevolent world hegemony”), it is “capitalism” that best explains the
intensification of US overseas interventions since 1990. US supremacy also
requires privileged access to the key resources on which its version of capitalism
depends, and for the past one hundred years or more this has included access to
cheap sources of oil. Western media typically fail to identify the relationship
between the wars in which the US engages and this metanarrative of US
supremacy. Most media coverage of the two Gulf Wars of 1990–1 and 2003 not
only failed to place them within the grander narrative of US hegemony but also
within the secondary narrative of oil supremacy. This secondary narrative is one
of at least 30 years’ duration and concerns US control of the Middle East and its
oil—not so much for the fuel, as for the power that control of the fuel bestows
over other regions and nations, notably Europe, Japan, and China.
The National Security Strategy, published in September 2002, claimed that
the attacks of September 11, 2001 had “opened vast, new opportunities” (p. 28).
This statement indicates a further narrative that the mainstream media have
been slow to recognize and prioritize—the narrative of the “neo-conservative”
influence in US foreign policy-making and, above all, what this influence signifies
(a rejection, among other things of Kissinger policies of co-existence). The core
of the neo-conservative movement came together with the formation in 1997 of
the Project for the New American Century. This group, together with the Com-
mittee for the Liberation of Iraq, a PNAC spin-off, and the American Enterprise
Institute, a conservative think tank that rented office space to PNAC, played a
highly significant role in pushing for the 2003 invasion of Iraq and in promoting
the cause of other interventions in the Middle East.
37
OLIVER BOYD-BARRETT
The metanarrative of US hegemony does not end with the acquisition of terri-
tory, military strength and trading power. Economic phenomena such as deflation,
stagnation, financial looting, over capacity, and massive economic imbalances are
reported by mainstream media as though these were quirky occurrences, either the
side-products of a “bubble” economy or the symptoms of a temporary economic
downswing. According to Beams (2003), they must be related to underlying pres-
sures within the capitalist system, in particular the declining rate of profit in the
period since World War II, a phenomenon that creates huge incentives for the
seizure and/or “privatization” of resources that are locked into national monopo-
lies or similar state-owned productive properties, even where these involve basic
amenities such as water, education, health, or power. The US push for global dom-
ination, argued Beams, is “driven on as it is by the crisis in the very heart of the
profit system, cannot bring peace, much less prosperity, but only deepening attacks
on the world’s people, enforced by military and dictatorial forms of rule.”
38
T H E S E C O N D C A S U A LT Y
refrain from the use of journalists. In fact, the CIA merely curtailed the practice
(Houghton 1996). The Reagan administration had no qualms about returning to
old habits and included an illegal CIA-administered domestic propaganda war
among covert operations in central America. A CIA memo quoted by Cockburn
and St Clair (1999: 32) explained that the agency maintained “relationships with
reporters from every major wire service, newspaper, news weekly and TV network”
and that in many instances “we have persuaded reporters to postpone, change,
hold or even scrap stories that could have adversely affected national security
interests or jeopardized sources or methods.” This was confirmed by the Guardian
newspaper in 1991 (quoted by Pilger 1998: 496), whose correspondent Richard
Norton-Taylor disclosed that some 500 prominent Britons were paid by the CIA
through the corrupt, now defunct, Bank of Commerce and Credit International,
including 90 journalists, many of whom were in “senior positions.” Nor was it
likely that such corruption would be reserved for non-US journalists. In 1996, the
Council on Foreign Relations suggested that the CIA be freed from some policy
constraints on covert operations, such as the use of journalists and clergy as cover.
CIA Director John Deutch argued that American journalists “should feel a civic
responsibility to step outside their role as journalists” (Cockburn and St Clair
1999: 90). A 1997 law, the Intelligence Authorization Act, actually permitted
reinstatement of the practice, subject to presidential approval; in any case, the
CIA had reserved the right to use the practice, noting, as Deutch stated before
Congress, that it already had the power to use US reporters as spies. Following the
mid-1970s, furthermore, many propaganda functions were transferred by the CIA
and Congress to privately funded organizations, through conduits such as the Ford
Foundation and similar bodies; examples included The Asia Foundation, Congress
for Cultural Freedom, and the National Endowment for Democracy (Brandt
1997), and many of these finance supposedly free media operations. Elsewhere
(Boyd-Barrett, forthcoming), I explore the implications of such penetration for
war reporting in Iraq of 2003 and other conflicts.
Conclusion
My purpose here has been to dissect the limitations of war-reporting-as-genre. My
examples are intended to demonstrate that with respect to some of the most
important international conflicts in the past half century or more, the media’s
reporting of war has been almost guaranteed to misinform and obfuscate. Herman
and Chomsky’s acclaimed “propaganda model,” which helps account for media
complicity with propaganda, does not sufficiently address evidence of direct state
penetration and the covert control of supposedly independent, privately-owned
media, a phenomenon that may provide additional explanation for the media’s
continued adherence to a patently faulty genre. For the media have typically
failed on numerous counts: to probe stated proximate causes of conflict, to prise
out deep-level causes, to avoid complicity with state propaganda machinery, to
follow and make accurate sense of strategic changes in the courses of war, and to
39
OLIVER BOYD-BARRETT
fully determine the factors explaining the ending of wars, and the aftermaths and
other implications of war. On each of these counts, war reporting remains in the
service of propagandistic purposes.
References
Alvarez, R. (2003, August 18) “The legacy of Hanford,” The Nation, vol. 277, no. 5,
pp. 31–5.
Barsamian, D. (2003, May) “An interview with Noam Chomsky,” Monthly Review,
retrieved August 30, 2003 from www.monthlyreview.org/0503chomsky.htm
Beams, N. (2003, July 10, 11) “The political economy of American militarism,” World
Socialist Web Site, retrieved August 30, 2003 from www.wsws.org
Bernstein, C. (1977, October 20) “The CIA and the media,” Rolling Stone.
Boyd-Barrett, O. (2003a) “Doubt foreclosed: US mainstream media and the attacks of
September 11, 2001,” pp. 35–49 in N. Chitty, R. Rush and M. Semati (eds), Studies in
Terrorism: Media Scholarship and the Enigma of Terror, Penang: Southbound Press.
Boyd-Barrett, O. (2003b) “Imperial news and the new imperialism,” Third World Resur-
gence, 151–2, pp. 44–8.
Boyd-Barrett, O. (forthcoming) “Casualties of genre,” paper prepared for the 2004 confer-
ence of the Center for Global Media Studies, Seattle.
Brandt, D. (1997) “Journalism and the CIA: the mighty wurlitzer,” NameBase NewsLine,
no. 17, April–June, retrieved August 30, 2003 from www.namebase.org/news17.html
Brzezinski, Z. (1998) The Grand Chessboard: American Primacy and Its Geostrategic Impera-
tives, New York: Harper Collins.
Chomsky, N. (2003) Middle East Illusions, Lanham, MD: Rowman and Littlefield.
Cockburn, A. and St Clair, J. (1999) Whiteout: the CIA, Drugs and the Press, New York:
Verso Books.
Dalglish, L., LaFleur, J., and Leslie, G. (2003) Homefront Confidential: How the War on Ter-
rorism Affects Access to Information and the Public’s Right to Know, 4th edition,
Arlington, VA: The Reporters Committee for Freedom of the Press.
Gwyn, R. (2003, January 1) “Dawn of imperial America,” Toronto Star.
Hecht, S. (2003, March 30) “Uranium warheads may leave both sides a legacy of death for
decades. Los Angeles Times accessed through Common Dreams News Center,” March
30, 2003, at http://www.commondreams.org/views03/0330-02.htm
Herman, E. and Chomsky, C. (1988) Manufacturing Consent: the Political Economy of the
Mass Media, New York: Pantheon Books.
Hitchens, C. (2001) The Trial of Henry Kissinger, New York: Verso.
Houghton, K. (1997) “Subverting journalism: reporters and the CIA,” Committee for the
Protection of Journalists, www.cpj.org/attacks96/sreports/cia.html
Kennan, G. (1948, February 24) PPS/23: “Review of Current Trends in U.S. Foreign
Policy,” in Foreign Relations of the United States, vol. I, pp. 509–29.
Kiley, S. (2003, August 17) “Chaos and cannibalism under Congo’s bloody skies,” The
Observer.
Knightley, P. (2002) The First Casualty: the War Correspondent as Hero and Myth-Maker
from the Crimea to Kosovo, Baltimore, MD: Johns Hopkins University Press.
Kristol, B. and Kagan, R. (1996) “Toward a neo-Reaganite foreign policy,” Foreign Affairs,
July/August, retrieved August 30, 2003 from www.ceip.org/people/kagfaff.htm
40
T H E S E C O N D C A S U A LT Y
Lokongo, A. (2002) “Media coverage of the Congo invasion: In the Footsteps of Western
Interests?,” Media Development, issue 1, www.wacc.org.uk/publications/md/md2002-
1/lokongo.html.
McCoy, A. (2003) The Politics of Heroin: CIA Complicity in the Global Drug Trade,
Lawrence Hill and Co.
McGehee, R. (1981, April 11) “The CIA and the White Paper on El Salvador,” The
Nation, vol. 255, no. 4, p. 423.
Meacher, M. (2003, September 6) “This war on terrorism is bogus,” The Guardian,
retrieved August 30, 2003 from www.guardian.co.uk
Peterson, S. (2003, May 15) “Remains of Toxic Bullets Litter Iraq,” Christian Science Mon-
itor, retrieved May 15, 2003 from Christian Science Monitor Online, http://www.
csmonitor.com/2003/0515/p01s02-woiq.html
Pilger, J. (1986) Heroes, London: Jonathan Cape.
Pilger, J. (1998) Hidden Agendas, New York: Random House.
Pilger, J. (2001) Heroes, Cambridge, MA: South End Press.
Pilger, J. (2002) The New Rulers of the World, London: Verso.
Raimondo, J. (2002, October 9) “Iraq: First Stop on the Road to Empire,” Antiwar.com,
retrieved October 30, 2003 from www.anti-war.com
Rampton, S. and Stauber, J. (2003) Weapons of Mass Deception: the Uses of Propaganda in
Bush’s War on Iraq, New York: Penguin.
Saunders, D. (2003, September 6) “The Fourth World War: gestation of a new American
Imperialism,” The Toronto Globe and Mail, Vancouver.indymedia.org.
Schechter, D. (2003) Embedded: Weapons of Mass Deception: How the Media Failed to Cover
the War in Iraq, New York: News Dissector/Mediachannel.org
Scott, P.D. (1985) “The United States and the overthrow of Sukarno,” 1965–7, Pacific
Affairs, 58, summer, pp. 239–64.
Scott, P.D., and Marshall, J. (1998) Cocaine Politics: Drugs, Armies, and the CIA in Central
America, Berkeley, CA: University of California Press.
Shah, A. (2003, January 21) “War, propaganda and the media” in Global Issues, retrieved
August 30, 2003, from www.globalissues.org/HumanRights/Media/Propaganda/Viet-
nam.asp?Print=True
Stromberg, J. (2002, October 12) “The old cause: what is ‘new’ in the new Bush doc-
trine?,” Anti-war.com, retrieved October 30, 2003 from www.antiwar.com
Thrall, T. (2000) War in the Media Age, Cresskill, NJ: Hampton Press Inc.
Traprock Peace Center (2003) “Gulf War Casualties and ‘Depleted’ Uranium: an Educa-
tional Campaign Providing Resources on Radioactive, Chemical and Biological
Weapons,” Traprock Peace Center, http://traprockpeace.org/depleteduranium.html
Troshinsky, L. (2002, August 15) “‘US drug war alliances futile,’ says expert,” United Press
International, United States Congress Senate Select Committee to Study Govern-
mental Operations with Respect to Intelligence Activities (1975) Final Report of the
Select Committee to Study Governmental Operations with Respect to Intelligence Activities,
United States Senate: Together with Additional, Supplemental, and Separate Views, 6 vols,
KF 31.5, G7 1976.
United States House Select Committee on Intelligence and Senate Select Committee on
Intelligence (2003) Report of the Joint Inquiry into the Terrorist Attacks of September 11,
Washington DC.
Vann, B. (2002, February 20) “US militarism targets South American oil,” World Socialist
Web Site, retrieved August 30, 2003 from www.wsws.org
41
OLIVER BOYD-BARRETT
Vann, B. (2003, June 27) “New York Times reporter Judith Miller accused of ‘hijacking’
military unit in Iraq,” World Socialist Web Site, retrieved August 30, 2003 from
wsws.org/articles/2003/jul2003/
Webb, G. (1998) Dark Alliance: The Contras and the Crack Cocaine Explosion, Seven Sto-
ries Press.
Worthington, A. (2003, May 3) “Death by slow burn: how America nukes its own troops,”
The Idaho Observer.
42
2
INFO RMATION WARFARE IN AN
AGE OF HYPER-MILITARISM
Richard Keeble
There was no war in the Gulf in 2003. Rather, a myth of heroic, spectacular war-
fare was manufactured, in large part, as a desperate measure to help provide a
raison d’être for the (increasingly out-of-control) military industrial complexes in
the US and UK—and to hide the reality of a rout of a hopelessly overwhelmed
“enemy” army. The links between mainstream journalists and the intelligence
services are crucial factors in the manufacture of the myth. But it is not essen-
tially a massive elite conspiracy. Rather, the myth’s origins lie deep within
complex military, historical, ideological, and political forces which it is crucial to
identify. Moreover, the manufacture of the “war” myth has profound implica-
tions for any study of the political and military origins of the conflict and press
representations.
43
RICHARD KEEBLE
44
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
lion (alongside a record White House deficit of $455 billion)—and that does not
include the extra billions expected for the occupation of Iraq. This represents
more than all the military spending of the rest of the world and more than twice
the spending of the next 15 of the world’s powers. Moreover, the US has military
bases in three-quarters of the countries of the world and 31 percent of all wealth.
Robert Harvey talks of the “United States of the World” (2003: 13–36). This is a
military colossus (backed by the UK) of a kind never before seen since the
Roman empire—and it is running out of control. As the late historian,
E.P. Thompson (1982), argued, there is a technological imperative driving the
US and UK toward warfare and testing new military systems.
The boom in military spending, begun during the Korean War years of the
1950s, continued relentlessly during the Cold War. By 1990, more than 30,000
US companies were engaged in military production, roughly 3,275,000 jobs were
in the defense industries and 70 percent of all money spent on research and
development was directed at defense work (Drucker 1993: 126). With the demise
of the Soviet Union, the United States became desperate in its search for new
enemies (Keeble 1997). Grenada (1983); Libya (1986); Panama (1989); Iraq
(1991, 1993, 1998 and 2003); Somalia (1992–3); Serbia (1999) (Hammond and
Herman 2000); and Afghanistan (2001) were all puny powers rapidly crushed by
the overwhelming firepower of the American colossus in a series of manufactured,
media-hyped militarist adventures (Webster 2003).
In the UK, the arms industry is worth more than £5 billion a year, amounting
to 20 percent of global weapons sales. It employs up to 150,000 people, the UK
standing as the world’s second largest manufacturer after the US with 32 percent
of the market. Yet arms deals remain remote from public scrutiny, being run by
the Defence Export Services Organization, “a secretive group within the Ministry
of Defence, controlled by the arms companies themselves and with a history of
actively conniving at bribery” (Leigh 2003) .
Thus the US/UK responses to the September 11 attacks, with the launch of
the endless “war on terrorism,” the attacks on Afghanistan and Iraq, and the
threats to the “rogue” states, Syria, Iran, and North Korea, are not distinctly new
strategies but accelerating long-standing strategies of military imperial adventur-
ism (Curtis 2003a, 2003b). Al-Qaeda, blamed for the September 11 atrocities
and a series of later attacks on Western interests, is a shadowy, elusive grouping
against whom traditional, war fighting strategies (involving major battle con-
frontations) are inappropriate. And so the US/UK is left manufacturing a
spectacle of traditional “warfare.” As US novelist Don DeLillo commented: “I’m
almost prepared to believe that the secret drive behind our eagerness to enter this
war is technology itself—that has a will to be realized. And that the administra-
tion is essentially a Cold War administration looking for a clearly defined enemy
which was not the case after September 11. Now there is a territorial entity with
borders and soldiers in uniform.”1
Moreover, given the integration of the media industries’ interests with those
of the military industrial complex and the importance of the media’s role in
45
RICHARD KEEBLE
46
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
x–xi) gave the media a crucial role. The heart of the secret state they identified as
the security services, the cabinet office and upper echelons of the Home and
Commonwealth Offices, the armed forces and Ministry of Defence, the nuclear
power industry and its satellite ministries, together with a network of senior civil
servants. As “satellites” of the secret state, their list included “agents of influence
in the media, ranging from actual agents of the security services, conduits of offi-
cial leaks, to senior journalists merely lusting after official praise and, perhaps, a
knighthood at the end of their career.”
Following the passing of the 1989 Security Service Act, links between the
media and MI5 and MI6 grew closer, according to James Adams (1994: 94–8).
Phillip Knightley, author of a seminal history of the intelligence services, even
claimed that at least one intelligence agent is working on every Fleet Street
newspaper.3
During the controversy that erupted following the end of the “war” and the
death of the arms inspector Dr David Kelly (and the ensuing Hutton inquiry) the
spotlight fell on BBC reporter Andrew Gilligan and the claim by one of his
sources that the government (in collusion with the intelligence services) had
“sexed up” a dossier justifying an attack on Iraq. The Hutton inquiry, its every
twist and turn massively covered in the mainstream media, was the archetypal
media spectacle that drew attention from the real issue: why did the Bush and
Blair governments invade Iraq in the face of massive global opposition? But those
facts will be forever secret. Moreover while the Gilligan affair might appear to
have reinforced the liberal notion of adversarial state–media relations, in fact, as
Rogers (1997: 64) argued, “this focus obscures the extent to which the media
have actually supported and colluded with the secret state.” Significantly, the
broader and more significant issue of mainstream journalists’ links with the intel-
ligence services was ignored by the inquiry.
Yet during the Hutton period, a myth emerged that the 2003 invasion of Iraq
was the first conflict to be justified on (dodgy) evidence supplied by the intelli-
gence services. Yet even during the Vietnam conflict, intelligence on the strength
of the Vietcong was faked to make the case for war more plausible (Ramsay
2003). Similarly, the US attack on Libya in 1986—deliberately aimed to effect
“regime change” by assassinating President Gaddafi—was justified by President
Reagan on dubious intelligence (dutifully reported in the mainstream media) of
Libyan responsibility for the bombing of a disco in West Berlin, frequented by US
servicemen. Intelligence misinformation before the 1991 Gulf massacres con-
stantly “over-sexed” Iraq’s alleged nuclear capability since opinion polls in the
States showed fears of President Saddam Hussein as a “nuclear monster” were
most likely to win support for the military option (Reich 2002). Even during the
1991 Iraqi conflict much of the reporting was based on intelligence-driven dis-
information. For instance, while Iraqi soldiers were deserting in droves and
succumbing to one massacre after another, all the British media highlighted intel-
ligence predictions of the “largest ground battle since the Second World War.”
Images of enormous berms, sophisticated Iraqi defences and trenches of burning
47
RICHARD KEEBLE
oil filled the media. But in the end there was nothing more than a 100-hour rout.
Colin Powell, in his account of the 1991 war, estimated that 250,000 Iraqi sol-
diers were eliminated (Powell 1995).
Similarly since the September 11 atrocities in the United States, the London-
based mainstream media have been awash with intelligence-inspired leaks
stressing the dangers of terrorist attacks in Britain. Even the Independent, most
critical of the US/UK rush to military action, gave credibility to dubious “intelli-
gence” sources. On September 16, 2001, for instance, Lashmar and Blackhurst
reported that at least three terrorist cells linked to Osama bin Laden were at large
in Britain. An “intelligence source” was quoted as saying: “There is no reason
why what happened in America couldn’t happen in Britain or any European
country.” This is terrifying stuff. But how much is fiction? (Bright 2002). Similarly
in September 2002 the Daily Express was awash in intelligence-inspired scare sto-
ries. “Nuclear attack in just months” it thundered on September 9; “Anthrax
threat on our streets: Britain on alert for Saddam suicide squads” it reported the
next day. A climate of fear is manufactured allowing the apparatus of the national
security state (surveillance cameras, email snooping, arrest without trials, and
demonization of asylum seekers) to expand. On September 15, 2002, drawing on
intelligence disinformation linking Iraq to nuclear weapons, the Sunday Express
editorialized: “War brings evil but we believe the country must not be frightened
from doing what we pray will save the world from the greater evil of nuclear
bombs. We see no alternative but to help demolish the Iraqi regime.”
On March 18, 2003, before the major air assault on Baghdad began, the Sun
typically reported: “According to intelligence reports Republican Guard units
have been equipped with chemical warfare shells to make a desperate last stand
south of Baghdad. A source said: ‘They clearly have given some chemical capabil-
ity to some forces.’” On April 2 the Sun “revealed” that Saddam Hussein had
issued a coded chemical attack on US/UK troops. Coalition intelligence chiefs, it
reported, interpreted a reference to “catching breath” in a speech by Saddam
Hussein “as a signal for lethal chemicals or nerve gas to be unleashed against US
forces massing south west of Baghdad.” There were similar reports throughout the
mainstream press.
48
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
The reality is that intelligence is the area in which ministers, and the
MI6 info ops staff behind them can say anything they like and get away
with it. Intelligence with its psychological invite to a secret world and
with its unique avoidance of verification, is the ideal means for flattering
and deceiving journalists.
The former foreign secretary, Lord Howe, told the Scott arms-to-Iraq inquiry:
“In my early days I was naïve enough to get excited about intelligence reports.
Many look, at first sight, to be important and interesting and significant and then
when we check them they are not even straws in the wind. They are cornflakes in
the wind” (Norton-Taylor 2003).
Another problem with intelligence is that anyone attempting to highlight its
significance is accused of lacking academic rigour and promoting “conspiracy
theory.” Certainly underlying the myth of “warfare” lie complex cultural, military,
and ideological forces. But given the close links between politicians, journalists,
and the intelligence services some conspiratorial elements have to be acknowl-
edged to be behind mainstream media’s coverage of the Iraqi crisis.
With the emphasis on intelligence, the focus of journalism shifts from objective,
verifiable “facts” to myth: in effect, there is a crucial epistemological shift. As Gen-
eral Richard Myers, chairman of the joint chiefs of staff, admitted: “Intelligence
doesn’t mean something is true. You know, it’s your best estimate of the situation.
It doesn’t mean it’s a fact. I mean, that’s not what intelligence is” (Stephen 2003).
Similarly, the historian Timothy Garton Ash stressed: “The trend in journalism as
in politics, and probably now in the political use of intelligence, is away from the
facts and toward a neo-Orwellian world of manufactured reality” (2003). The
assumption of Iraq possessing WMD was based entirely on unverifiable intelli-
gence reports as is so much of the reporting of the “war on terror.”
49
RICHARD KEEBLE
of the camera and the journalist’s eye-witness, with the war unproblematized, the
essential simulated, mythical nature of the conflict lay all the more subtly and
effectively hidden. Moreover, military censorship regimes always serve essentially
symbolic purposes—expressing the arbitrary power of the army over the conduct
and representation of “war.”
Significantly defence minister Geoff Hoon claimed: “I think the coverage . . . is
more graphic, more real than any other coverage we have ever seen of a conflict.”
Most of the critical mainstream coverage highlighted the information overload.
But, as David Miller (2003) commented: “It is certainly true to say that it is new
to see footage of war so up-close but it is a key part of the propaganda war to claim
that this makes it ‘real.’”
Some 600 US and 128 UK journalists, including journalists from the Western
Daily Press, Scotsman, Manchester Evening News, Ipswich Evening Star and Eastern
Daily Press, and one from the music network MTV, were “embedded” with military
units. According to Phillip Knightley (2003): “The idea was copied from the
British system in World War I when six correspondents embedded with the army
on the Western front produced the worst reporting of just about any war and were
all knighted for their services. One of them, Sir Phillip Gibbs, had the honesty,
when the war was over, to write: ‘We identified ourselves absolutely with the
armies in the field.’ The modern embeds, too, soon lost all distinction between war-
rior and correspondent and wrote and talked about ‘we’ with boring repetition.”
As the Times media commentator, Brian MacArthur, reported (2003):
“Embeds inevitably became adjuncts to the forces.” Audrey Gillan, with the
Household Cavalry for the Guardian, was one of the few to accuse the military of
censorship. She reported that soldiers complained of being like mushrooms—
kept in the dark with you know what shovelled on top of them—but she could
not use this phrase for fear of upsetting the brigade HQ.
Some 5,000 journalists were in the Gulf region to cover the hostilities. Two
thousand were in Kuwait and on ships with the US and UK naval task forces in
the Arabian Gulf; 290 were in Baghdad; 900 in northern Iraq with Kurdish fight-
ers: the rest were in Jordan, Iran, Bahrain, and at the Allied Central Command in
Doha, Qatar (Milmo 2003). Here there was little consistent challenge to the
dominant military agenda. On one occasion New York magazine writer Michael
Wolff (2003) dared to break ranks and ask the provocative questions: “Why are
we here? Why should we stay? What’s the value of what we’re learning at this
million-dollar press center?” He was soon to pay the price for his daring. Fox TV
attacked him for lacking patriotism and after right-wing commentator Rush Lim-
baugh gave out his email address, in one day Wolff received 3,000 hate messages.
Unprecedented access to the “front lines” was the carrot, but the stick was
always on hand. Fifteen non-Iraqi journalists were killed, two went missing and
many unilateral non-embeds were intimidated by the military. Had there been the
same death rate for journalists during the Vietnam war, there would have been
3,000 killed.4 As John Donvan (2003) argued, “coalition forces saw unilaterals as
having no business on their battlefield.” Unilateral Terry Lloyd, of ITN, was killed
50
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
by marines who fired at his car; Reuters camera operator Taras Protsyuk and José
Couso, a cameraman for the Spanish television channel Telecino, died after an
American tank fired at the fifteenth floor of the Palestine hotel in Baghdad while
Tarek Ayoub, a cameraman for Al-Jazeera, died after a US jet bombed the chan-
nel’s Baghdad office. Two journalists working for RTP Portuguese television, Luis
Castro and Victor Silva, were held for four days, had their equipment, vehicle, and
video tapes confiscated and were then escorted out of Iraq by the 101st Airborne
Division. How many Iraqi journalists perished in the slaughters we will never
know. For most of the Western mainstream media they are non-people.
51
RICHARD KEEBLE
52
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
Saddam’s record means that the coalition forces must be ready for any-
thing. This, after all, is a dictator who planned during the last Gulf War
to chain American PoWs to the front of his tanks; a murderer who—
uniquely in the history of depravity—has turned chemical weapons on
his own people.
On March 20, Julian Borger, in the Guardian, grappled with the contradictions.
On the one hand, he reported: “In terms of technology and sheer might, this
coming conflict is likely to be one of the most unequal in history.” Yet, to reaffirm
the myth of war, there is always the Saddam demonization card to play. So Borger
continued: “But the Iraqi leader’s proven readiness to embrace desperate and
unconventional measures makes him potentially a far more dangerous foe than
any the Pentagon has taken on in recent years.”
Significantly, the Observer, in outlining its support for military action in its
leader of January 19, framed its entire argument around the demonized personal-
ity of “Saddam.” First it referred to the “nature of Saddam Hussein’s regime and
the call by many Iraqi exiles and dissidents for him to be overthrown.” The war
was not about oil. “For the second motive for displacing Saddam is the danger he
poses to the wider world.” And it concluded: “If Saddam does not yield military
action may eventually be the least awful necessity for Iraq.”
53
RICHARD KEEBLE
54
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
55
RICHARD KEEBLE
Notes
1 Quoted in “Notes from New York, a profile of DeLillo,” by Duncan Campbell, The
Observer, May 4, 2003.
2 Quoted in Seamus Milne (1994) The Enemy Within: the Secret War Against the Miners,
London: Verso; reprinted by Pan in 1995: 262.
3 Phillip Knightley interviewed London, September 25, 2003.
4 Christiane Amanpour, chief international correspondent for CNN, quoted in Jessica
Hodgson (2003) “Mother of all war journos,” The Observer, November 2.
References
Adams, James (1994) Secret armies: the full story of SAS, Delta Force and Spetsnaz, London:
Hutchinson.
Almond, Mark (2003) “So how will he be judged?” The Guardian, May 15.
Baudrillard, Jean (1991) “The reality gulf,” The Guardian, January 11.
Beaumont, Peter and Graham, Patrick (2003) “Iraq terror spirals out of control as US
intelligence loses the plot,” Observer, November 2.
Beetham, David (2003) “The warfare state,” Red Pepper, June.
Bloch, Jonathan and Fitzgerald, Patrick (1983) British Intelligence and Covert Action,
London: Junction Books.
Bright, Martin (2002) “Terror, security and the media,” Observer online, July 21, http:
//observer.guardian.co.uk/libertywatch/story/0,1373,758265,00.html (accessed July 22).
Casey, Leo (2003) “Questioning Halabja: genocide and the expedient political lie,” Dis-
sent, New York, summer: 61–5.
Chomsky, Noam (1999) Lessons from Kosovo: the New Military Humanism, London: Pluto.
Cummings, Bruce (1992) War and Television, London: Verso.
Curtis, Mark (2003a) “Partners in imperialism: Britain’s support for US invasion,”
www.zmag.org May 10 (accessed May 11).
Curtis, Mark (2003b) Web of Deceit: Britain’s Real Role in the World, London: Vintage.
Donvan, John (2003) “For the unilaterals, no neutral ground,” Columbia Journalism
Review, May/June www.cjr.org/year/03/3/donvan.asp (accessed July 12, 2003).
56
I N F O R M AT I O N WA R F A R E I N A N A G E O F H Y P E R - M I L I TA R I S M
Dorril, Stephen (2000) MI6: Fifty Years of Special Operations, London: Fourth Estate.
Dorril, Stephen (2003) Spies and Lies, Free Press, April.
Dorril, Stephen and Ramsay, Robin (1991) Smear, London: Fourth Estate.
Drucker, Peter E. (1993) Post-capitalist Society, Oxford: Butterworth-Heinemann.
Epstein, Edward (2003) “How many Iraqis died? We may never know,” San Francisco
Chronicle, May 3.
Garton Ash, Timothy (2003) “Fight the matrix,” The Guardian, June 5.
Guillot, Clare (2003) “Nassiriya: le soldat Reeves face à la foule en colère,” Le Monde;
April 17.
Hammond, Philip (2000) “Reporting ‘humanitarian’ warfare: propaganda, moralism and
NATO’s Kosovo war,” Journalism Studies, vol. 1, no. 3, pp. 365–86.
Hammond, Philip and Herman, Edward S. (eds) (2000) Degraded Capability: the Media and
the Kosovo Crisis, London: Pluto.
Harvey, Robert (2003) Global Disorder, London: Constable.
Kampfner, John (2003) “The truth about Jessica,” The Guardian, May 15.
Keeble, Richard (1997) Secret State, Silent Press: New Militarism, the Gulf and the Modern
Image of Warfare, Luton: John Libbey.
Keeble, Richard (1998) The Myth of Saddam Hussein: Media Ethics, Kieran, Matthew (ed.),
London, Routledge, pp. 66–81.
Keeble, Richard (2001) “The media’s battle cry,” Press Gazette, October 5.
Keegan, John (2003) “Saddam’s utter collapse shows this has not been a real war,” The
Daily Telegraph, April 8.
Kellner, Douglas (1992) The Persian Gulf TV War, Boulder/San Francisco/Oxford: West-
view Press.
Knightley, Phillip (2003) “Turning the tanks on the reporters,” Observer, June 15.
Leigh, David (2003) “Greasy palms in pinstripe pockets,” The Guardian, September 16.
Luckham, Robin (1983) “Of arms and culture,” Current Research on Peace and Violence IV,
Tampere, Finland, pp. 1–63.
Lule, Jack (2002) “Myth and terror on the editorial page: The New York Times responds to
September 11,” Journalism and Mass Communication, Columbia, pp. 275–93.
MacArthur, Brian (2003) “Changing pace of war,” The Times, June 27.
McChesney, Robert (2002) “September 11 and the structured limitation of US journal-
ism,” in Stuart Allan and Barbie Zelizer (eds), Journalism after September 11, London:
Routledge, pp. 91–100.
McLaughlin, Greg (2002) The War Correspondent, London: Pluto Press.
Mailer, Norman (2003) “We went to war just to boost the white male ego,” www.timeson-
line.co.uk/article/0..482-662789.00.html (accessed August 12, 2003).
Mayrhofer, Debra (2003) “What’s in a name?” www.mwaw.org April 15 (accessed May 12,
2003).
Miller, David (2003) “Embedding propaganda,” Free Press, special issue, June.
Milmo, Cahal (2003) “Reporting for duty,” The Independent, March 18.
Morgan, David (2003) “Climate of fear,” Morning Star, June 12.
Norton-Taylor, Richard (2003) “The BBC row has been got up to obscure the ugly truth,”
The Guardian, June 28.
Peak, Steve (1982) “Britain’s military adventures,” London, The Pacifist, vol. 20, p. 10.
Pilger, John (1998) Hidden Agendas, Verso: London.
Pilger, John (2003) “The big lie,” Daily Mirror, September 22.
Plavsic, Dragan (2000) “NATO’s war: the truth comes out,” Socialist Review, September.
57
RICHARD KEEBLE
Powell, Colin (1995) Soldier’s Way, London: Hutchinson (with Joseph Persico).
Ramsay, Robin (2003) “Lying about Iraq,” Lobster, no. 45, summer.
Reich, Stephanie (2002) “Slow motion holocaust,” Covert Action Quarterly, no. 72,
pp. 22–8.
Rogers, Ann (1997) Secrecy and Power in the British State: a History of the Official Secrets
Act, London: Pluto Press.
Said, Edward (2003) “A window on the world,” The Guardian Review, August 2.
Stephen, Andrew (2003) “America,” New Statesman, August 4.
Swain, Jon (2003) “Why the reporter is the last bastion of truth,” Observer, March 16.
Thompson, Edward P. (1982) Notes on Exterminism: the Last Stage of Civilisation, Extermin-
ism and the Cold War, London: Verso, pp. 151–63.
Webster, Frank (2003) “Information Warfare in an Age of Globalization,” Daya Kishan
Thussu and Des Freeman, eds, Information Warfare in an Age of Globalization: War and
the Media, London: Sage, pp. 57–86.
Wolff, Michael (2003) “I was only asking,” The Guardian, April 14.
58
3
A MO RAL IMAGINATION
The media’s response to the war on terrorism
Susan D. Moeller
Once upon a time there lived a mandarin in China. Sequestered in his palace his
sleep was untroubled, even when a calamity befell a hundred million of his
people. Destruction of that immense multitude held little interest for him; he
snored “with the most profound security.” Then, late one evening his lapdog
nipped him on his littlest finger. The court doctor came to minister to him and
told him that if the injury grew worse over the night, he might well have to
amputate the finger. Oh, how his sleep was disturbed then. The dull pain of the
bite and the greater agony of wondering whether he would have to suffer the loss
of his little finger kept him awake through all the dark hours.
The eighteenth-century French philosopher Denis Diderot and the eigh-
teenth-century Scottish political economist Adam Smith both told versions of
this tale of the Chinese mandarin (Smith 1976; Ginzburg 1994). The inherent
drama of the fable is in the mandarin’s complete moral disregard of the value of
the lives of his subjects. Yet, as Diderot and Smith and generations of moral
philosophers since them have explored, there are ways—there must be ways—to
make the lives of distant strangers of value to us.
One plausible manner in which to do so is through what has been called
“enlightened self-interest.” A broader and more long-term understanding of self-
interest, for example, might have influenced that self-centered despot to
comprehend that the deaths of so many of his subjects could not but negatively
affect his own luxurious lifestyle. His sleep might then have been troubled as he
contemplated his future loss of living standards and he might have applied him-
self to figuring out what could best ameliorate the situation. Perhaps among those
possible remedies might have been some that would not just protect his own
resources, but that might prevent future losses of life: the building of a dam, if the
calamity was a catastrophic flood; the institution of mandatory grain reserves, if
the calamity was caused by a plague of locusts.
When contemplating action, noted the Nobel Prize-winning physicist Richard
Feynman, the core political question is: “Should I do this?” Yet, in order to deter-
mine whether a certain action “should” be taken, Feynman argued that another
59
SUSAN D. MOELLER
question has to first be addressed: “If I do this, what will happen?” The first ques-
tion, “what will happen?” may have an answer that can be arrived at empirically.
The answer may be knowable. In an example that Feynman gives, for instance, it
may be evident that the enactment of a certain economic policy will cause a
depression. But it may not inevitably follow that a depression is something to be
avoided at all costs. The answer to the question “Should I do this?” is a matter of
judgment.
You see, only knowing that it is a depression doesn’t tell you that you do
not want it. You have then to judge whether the feelings of power you
would get from this, whether the importance of the country moving in
this direction is better than the cost to the people who are suffering. Or
maybe there would be some sufferers and not others. And so there must
at the end be some ultimate judgment somewhere along the line as to
what is valuable, whether people are valuable, whether life is valuable.
(Feynman 1998: 44–6)
60
A M O R A L I M A G I N AT I O N
factor in the formation of behavior, but by itself is not sufficient cause for action.
A higher motivation must be claimed—such as President Bush’s “crusade” against
terrorism or his commitment to “rid the world of the evil-doers,” as he called for
in the days after September 11. Values must lie alongside interests. Moral inten-
tion, in other words, is assumed to be essential to Americans’ high regard of
themselves. Philosophers, politicians, and the public alike assume a tie between
public policy and moral obligation.
At any given point in history, however, certain policies which have been
deemed to be moral, in hindsight look less so.1 The impetus for action which
appeared to be concern for others has been revealed to be a mandarinesque self-
interest, with very little of it “enlightened.” During the nineteenth century, for
example, journalist John L. O’Sullivan wrote an editorial supporting the annexa-
tion of Texas in the July/August issue of the United States Magazine and Democratic
Review. The article persisted through history, for in it O’Sullivan argued that it is
“the right of our manifest destiny to over spread and to possess the whole of the
continent which Providence has given us for the development of the great exper-
iment of liberty and federaltive [sic] development of self government entrusted
to us.”2
The doctrine of “manifest destiny” helped generations of Americans rationalize
the institution of policies that were racist at best and genocidal at worst. Yet the
argument for “manifest destiny” was couched in moral language and cited Provi-
dence as the ultimate originator of the doctrine—territorial expansion was not
only inevitable and divinely ordained, but would bring “liberty, civilization and
refinement” to the “masses of mankind.”
Although the phrase “manifest destiny” was coined in the mid-nineteenth
century, the nation—and before that, the colonies—had always thought of itself
as a place apart, not just geographically, but morally. The new world was to be in
the words of John Winthrop in 1630, a “city upon a hill,” and those who lived in
it were to be “blessed” by “the Lord our God . . . in the land whither we go to pos-
sess it.” Those who came to America, and those who have helped to define these
United States have incorporated into their visions a destiny not imitative of the
destinies of other nations or other empires. As O’Sullivan wrote six years before
his “manifest destiny” editorial,
Our national birth was the beginning of a new history, the formation
and progress of an untried political system, which separates us from the
past and connects us with the future only; and so far as regards the entire
development of the natural rights of man, in moral, political, and
national life, we may confidently assume that our country is destined to
be the great nation of futurity.
(O’Sullivan 1839)
“America,” he argued, “is destined for better deeds.” Here, in an essay that early
laid out his thoughts on “manifest destiny,” we hear not only the call of duty but a
61
SUSAN D. MOELLER
People of conscience must be morally affronted by the abuses and injustices sanc-
tioned by “monarchies and aristocracies,” dictators, and warlords, O’Sullivan
argued. Americans have a responsibility to protect the oppressed and to enforce
the basic rights of all human beings. But in what cause were these words applied?
Minus the more flowery rhetorical flourishes, this passage could have been
seamlessly included in President George W. Bush’s address to the joint session of
Congress on September 20, 2001, when he outlined “our war on terror.” That war
began, he said,
with al-Qaeda, but it does not end there. It will not end until every ter-
rorist group of global reach has been found, stopped and defeated . . . This
is not, however, just America’s fight. And what is at stake is not just
America’s freedom. This is the world’s fight, this is civilization’s fight, this
is the fight of all who believe in progress and pluralism, tolerance and
freedom.
Bush ended his speech with the same invocation of God that O’Sullivan had made:
The course of this conflict is not known, yet its outcome is certain. Free-
dom and fear, justice, and cruelty, have always been at war. And we
know that God is not neutral between them. Fellow citizens, we’ll meet
violence with patient justice, assured of the rightness of our cause and
confident of the victories to come. In all that lies before us, may God
grant us wisdom and may he watch over the United States of America.
(US Newswire, September 20, 2001)
Why such language? It offers moral justification. Historian Jackson Lears wrote in
a New York Times opinion:
62
A M O R A L I M A G I N AT I O N
The proposed war against and rebuilding of Iraq has brought the senti-
mental, self-satisfied sense of Providence back into fashion. One might
have supposed that an attack on our country would have rendered
utopian agendas unnecessary—as it did for most Americans during
World War II. But while a war on terrorism may not need Providence to
justify it, a war to transform the Middle East requires a rhetoric as
grandiose as its aims. The providentialist outlook fills the bill: it pro-
motes tunnel vision, discourages debate and reduces diplomacy to
arm-twisting.
(Lears 2003: A25)
63
SUSAN D. MOELLER
on terrorism became the window through which all international events were
viewed—a situation that emphasized places and events that had (or were pur-
ported to have) connections to “global terrorism,” but that left others that didn’t
neatly fit the terrorism frame out of public view.
Newsrooms scrambled to cover both domestic and foreign terrorist-related
events—a scramble made all the more ungainly because all but a few media out-
lets were woefully understaffed with reporters expert in international affairs, a
consequence of the past years’ closing of overseas bureaus and cutbacks on time
and space devoted to foreign news in order to save money and boost profits.
Understaffing—and the prior undervaluing of international coverage—made it
more difficult for news organizations to cover the assumptions behind the “war on
terror” frame, and parenthetically made it more difficult for them either to nimbly
cover the changing terrorism story (from Osama bin Laden and al-Qaeda, to
anthrax, to the axis of evil, to Iraq and Saddam Hussein, to WMD, etc.), or to
cover foreign stories unrelated to the terrorism arc.
Three months after the attacks on the World Trade Center, on the 60th
anniversary of the attack on Pearl Harbor, President Bush defined his conception
of terrorism before more than 8,000 sailors and marines and their families
assembled on the vast deck of the nuclear-powered aircraft carrier, the USS Enter-
prise. “We’re fighting to protect ourselves and our children from violence and fear.
We’re fighting for the security of our people and the success of liberty,” Bush
said. “We’re fighting against men without conscience but full of ambition to
remake the world in their own brutal images” (US Newswire, December 7, 2001).
Bush described these men as “a movement, an ideology that respects no bound-
ary of nationality or decency . . . They celebrate death, making a mission of
murder and a sacrament of suicide.” And he compared the 9/11 “terrorists” to the
enemies of World War II: “They have the same will to power, the same disdain for
the individual, the same mad global ambitions. And they will be dealt with in
just the same way . . . Like all fascists, the terrorists cannot be appeased; they must
be defeated” (US Newswire, December 7, 2001).
Naming not just the 9/11 conspirators, but a much larger conception of “the
enemy” as “terrorists” and naming America’s cause as a “global war against terror-
ism”—rather than a more limited effort to eradicate al-Qaeda or to capture Osama
bin Laden—was an attempt by President Bush to forestall and even pre-empt
media and public criticism of such ancillary agendas as overthrowing Saddam. The
Bush administration succeeded at labeling its foreign policy objectives as part of a
war against terrorism, thus making it very difficult for political opponents or media
commentators to challenge the president without coming off as not only “soft” on
defense, but as cavalier about the lost American lives of 9/11.
The Bush administration’s tactics that effectively proscribed dissent were
familiar ones from the Israeli–Palestinian conflict. The only way to be “for” ter-
rorism is to redefine terrorism as something else, to change not just the language
but the focus and context of the debate. In the Israeli–Palestininan skirmishing
for the moral high ground there, Israeli officials have long cited Palestinian “ter-
64
A M O R A L I M A G I N AT I O N
rorism” as their rationale for their policies, which have included helicopter mis-
sile attacks against Palestinian buildings, “targeted assassinations,” closures of
Palestinian towns and villages, and razing of the homes of alleged militants.
Hamas and Islamic Jihad spokespeople have claimed that Israeli “state terror” has
forced them into such strategies as suicide bombings and other attacks on Israeli
soldiers and civilians. Both sides emphasize context. Both sides assert that their
actions are retaliatory; that they are not the perpetrators, but instead the victims.
Both sides also contest who are the civilians—in other words who are the legiti-
mate targets of attacks: Israel argues that the settlers are civilians, while many
Palestinian militants consider the settlers who have built homes on land seized in
the 1967 war to be combatants in effect, because they have been the leading edge
of the Israeli occupation and have been protected by the Israeli military. New
York Times reporter Serge Schmemann wrote,
One side invokes the murder of Israeli innocents by human bombs, the
other speaks of the injustices suffered by the Palestinians . . . This min-
gling of acts of murder with a desire for freedom has enabled the leaders
on both sides to befuddle their people and their supporters, portraying
suicide bombers as martyrs in a just struggle of national liberation, or
casting the destruction of the institutions and symbols of Palestinian
statehood as a war on terrorism.
(Schmemann 2002: section 4: 1)
The challenge with such terms as “innocence” and “terrorism” is that their mean-
ing is in the eye of the beholder. President Bush’s post-9/11 call for a broader war
on terrorism had its contemporary antecedents with similar rationales: in this
case Bush characterized those who perished on September 11 as “innocents” and
those who killed them as “terrorists.” And the media responded as they always
have at the start of a national crisis—they rallied in support of the president and
appropriated his characterization of the situation. At the end of October 2001,
for example, the then-CNN chairman Walter Isaacson wrote a memo to his staff
that ordered them to balance the broadcast images of civilian devastation in
Afghanistan with reminders of the American lives lost at the World Trade
Center and the Pentagon. Isaacson suggested language for his anchors, including:
“The Pentagon has repeatedly stressed that it is trying to minimize civilian casu-
alties in Afghanistan, even as the Taliban regime continues to harbor terrorists
who are connected to the September 11 attacks that claimed thousands of inno-
cent lives in the US.” It “seems perverse,” Isaacson said, “to focus too much on
the casualties or hardship in Afghanistan” (Kurtz, October 31, 2001: C1).
But like the Israeli and Palestinian labeling, pro-al-Qaeda and pro-Saddam
forces familiar to viewers of Al-Jazeera also described as “innocent civilians”
those who died in Afghanistan or Iraq as a result of American bombing. Post-
September 11, there is a larger recognition that innocence is conferred, rather
than inherent, that innocence needs to be asserted; it is not unequivocally
65
SUSAN D. MOELLER
self-apparent. During World War II, those who were killed in Nanking or in the
Blitz or in the concentration camps were just called “civilians.” Their innocence
was implicit in their status as civilians. Saying “innocent civilians” would have
been redundant. Today the innocence of the victims on each side of the war on
terrorism is loudly proclaimed—just as it has been in the Israeli–Palestinian con-
flict. Affirming one’s innocence may confer no protection, but it allows one to
lay claim to the moral high ground.
Chosen words
“Terrorist” and “terrorism” have long been charged words because the terms are
used as political epithets, but they are rarely defined. There are political sensitivi-
ties about the usage of these terms and phrases that the media rarely
acknowledge. When Secretary of State Donald Rumsfeld says that North Korea is
a “terrorist regime,” for example, few journalists define that phrase—or contrast
the kind of terrorism attributed to North Korea with the terrorism attributed to
al-Qaeda. All terrorism is conflated, a conflation that has helped leave the
impression in American minds, for example, that Saddam Hussein was in part
responsible for the attacks of 9/11.
Americans want to believe that they hold the moral high ground. And they
are abetted in their belief by media that do not always explain that that ground is
contested—and that the language used to make the moral case is typically loaded.
“After 10 months of strife,” reporter Cameron Barr of the Christian Science Moni-
tor wrote a month and a half before September 11, “the Israeli–Palestinian
conflict is increasingly defined by ‘terrorism’—both the act and the epithet” (Barr
2001: 1). But, he continued, “perhaps no word in modern political usage is more
controversial than ‘terrorism.’ The United Nations spent 17 years trying to come
up with a universally accepted definition, and failed.” Walter Laqueur, in a recent
article “Left, right and beyond: the changing face of terror,” observed that terror
“is not an ideology or a political doctrine, but rather a method—the substate
application of violence or the threat of violence to sow panic and bring about
political change” (W. Laqueur 1996; Purdum 2002: section 4: 1). In his earlier
book Terrorism, Laqueur noted that “it is not the magnitude of the terrorist opera-
tion that counts, but the publicity.” The terrorist is inseparable from his or her
beholder. As reporter Melvin Maddocks has said, “A terrorist without an audi-
ence is inconceivable” (Maddocks 1980).
“Terrorism is theater,” wrote terrorism expert Brian Michael Jenkins.
66
A M O R A L I M A G I N AT I O N
After September 11, 2001, a number of media outlets took it upon themselves to
give their readers and viewers a hasty history course in terrorism. “The word orig-
inated during the French Revolution when enemies of the state were guillotined
in the Reign of Terror,” reporter Jim Auchmutey reminded the readers of the
Atlanta Journal-Constitution. “‘Those hellhounds called terrorists . . . are let loose
on the people,’ British politician Edmund Burke wrote in one of the earliest
usages cited by the Oxford English Dictionary” (Auchmutey 2001: 1D).
Geoff Nunberg, on National Public Radio’s interview program “Fresh Air,”
(Bogaev and Nunberg 2001) traced the changing meaning of the term, noting
that in the late nineteenth century some like the Russian revolutionaries who
assassinated Tsar Alexander II in 1881 viewed it as “a justified political strategy,”
and even Jack London believed that terrorism could be a “a powerful weapon in
the hands of labor.”
By the 1980s, Nunberg noted, terrorism began to be applied more generically “to
all manners of political violence,” including acts such as the fatwah against
Salman Rushdie, which Nunberg observed “was far from an act of indiscriminate
violence; more like state-sponsored contract killing.” New York Times editor
A.M. Rosenthal even attacked Christopher Hitchens for “refusing to describe the
fatwah against Salman Rushdie as terrorism.” “By then,” Nunberg commented,
“the word ‘terrorism’ had acquired a kind of talismanic force, as if refusing to
describe something as terrorism was the next thing to apologizing for it” (Bogaev
and Nunberg 2001).
But these thoughtful attempts to locate the meaning of the word “terrorism”
did not typically wrestle with the consequences of the Bush administration’s
usage of the term “terrorism” and its choice to aggregate its foreign policy deci-
sions under the rubric “war on terrorism.” Anne-Marie Slaughter, the dean of the
Woodrow Wilson School at Princeton University, and the former J. Sinclair
Armstrong Professor of International, Foreign, and Comparative Law at Harvard
67
SUSAN D. MOELLER
Law School, was one of the few to early point out the effect of that choice. “From
a legal perspective, the difference between calling what has happened war and
calling it terrorism is considerable,” she wrote in the Washington Post the Sunday
after 9/11. “Terrorism is a matter for the courts and prosecutors. War is up to our
military forces. But which best describes what we face now?” (Slaughter 2001).
Those rare media institutions which did caution about the usage of the word
“terrorism” were pilloried for their lack of patriotism. Media columnist Howard
Kurtz reported on a Reuters internal memo written by Stephen Jukes, the wire
service’s global head of news, after the 9/11 attack. “We all know that one man’s
terrorist is another man’s freedom fighter and that Reuters upholds the principle
that we do not use the word terrorist,” said Jukes in a conversation with Kurtz.
“To be frank, it adds little to call the attack on the World Trade Center a terrorist
attack.”
“We’re trying to treat everyone on a level playing field,” Jukes said, “however
tragic it’s been and however awful and cataclysmic for the American people and
people around the world . . . we don’t want to jeopardize the safety of our staff.
Our people are on the front lines, in Gaza, the West Bank and Afghanistan. The
minute we seem to be siding with one side or another, they’re in danger.” “We’re
there to tell the story,” Jukes told Kurtz. “We’re not there to evaluate the moral
case.” Despite the reasonableness of Jukes’ responses, Kurtz chided him for his
(and Reuters’) “value-neutral approach”—as if fairness and balance were not
vaunted journalistic values, especially in the coverage of international issues
(Kurtz, September 24, 2001: C1).
Before 9/11, a number of media outlets in addition to Reuters had a policy in
place about the use of the words “terrorism” or “terrorist.” The Associated Press,
according to its spokesman Jack Stokes, uses a variety of terms and does permit
the use of the word “terrorist” for those in non-governmental groups who carry
out attacks on the civilian population (Gelfand 2002). Other news organizations
have shunned the words in reference to the Israeli–Palestinian conflict, aware of
the politicization of the words, but have had no compunctions about using the
terms in other circumstances. Roger Buoen, the assistant managing editor of the
Minneapolis Star Tribune, for example, explained in a pre-9/11 statement to his
paper’s ombudsman that the paper tried to avoid “characterizing the subjects of
news articles” in order to let readers make up their own minds on the evidence
presented. “In the case of the term ‘terrorist,’” he wrote, “other words—‘gunman,’
‘separatist,’ and ‘rebel,’ for example—may be more precise and less likely to be
viewed as judgmental.” And Buoen also noted that the paper was especially con-
cerned about perceived bias in its Mid-east coverage: “We also take extra care to
avoid the term ‘terrorist’ in articles about the Israeli–Palestinian conflict because
of the emotional and heated nature of that dispute. However, in some circum-
stances in which nongovernmental groups carry out attacks on civilians, the term
is permitted” (Gelfand 2002).
The dilemmas of talking about “terrorism” do not stop with problems of defini-
tion. A few news outlets observed problems of labeling as “terrorist” even those
68
A M O R A L I M A G I N AT I O N
69
SUSAN D. MOELLER
“It’s a question of being rigorous. It’s really a question of really asking the
questions. All of the entire body politic in my view, whether it’s the
administration, the intelligence, the journalists, whoever, did not ask
enough questions, for instance, about weapons of mass destruction.”
And then Amanpour added that part of the cause of the journalists’ fail-
ures and/or reticence was that they had to contend with “disinformation
at the highest levels.”
(www.mediaweb.co.za)
A hundred and fifty years ago, in the era when “manifest destiny” was first enun-
ciated, there was no precedent of an investigative press that could critically
examine the statements and contentions of public officials, that could challenge a
foreign policy that seemed to have little basis in fact or that seemed geared only
to rally the audience for partisan political ends. Now, in the first decade of the
twenty-first century there has been little incentive to do so, despite the rich
precedents of journalists challenging the powers-that-be in decades past. In the
current profit-driven deadline news business there has been little inclination to
examine the operating assumptions of foreign policy when the effects of that
policy have been so demanding of coverage. It has made sense that over the last
several years, coverage of breaking news—such as the US troop engagement in
Iraq—is what blankets the news, while coverage of the motivations behind for-
eign policies has been minimal. It has been easier to embed with the troops, and
give a-day-in-the-life style coverage, than do enterprise reporting.
As a result, during the Iraq war and the immediate “aftermath,” many journal-
ists toed the administration’s and Pentagon’s line in Iraq, foregoing their own
investigation of events. New York Times Pulitzer Prize winning reporter John
Burns noted his disgust over the media coverage of Iraq in an interview included
in an edited collection of journalist interviews, Embedded: the Media at War in
Iraq, an Oral History. “Now left with the residue of all this, I would say there are
serious lessons to be learned. Editors of great newspapers and small newspapers
and editors of great television networks should exact from their correspondents
the obligation of telling the truth about these places,” Burns said. “It’s not impos-
70
A M O R A L I M A G I N AT I O N
sible to tell the truth.” More information could have been gotten out about the
Bush administration’s stated rationale for American involvement, and even
about the Iraqi regime, he argued (Burns 2003).
Increasingly more journalists and pundits have been speaking out, charging
that during the pre-war period and the “hot” conflict in Iraq most of “the media
largely toed the Bush administration line in covering the war and, by doing so,
failed to aggressively question the motives behind the invasion,” as USA Today
media critic Peter Johnson wrote (Johnson 2003). “For some reason or another,
Mr Bush chose to make his principal case on weapons of mass destruction, which
is still an open case,” noted New York Times reporter Burns (Burns 2003: 161).
But by the end of the Iraq conflict in the late spring of 2003, it was not so much
that the media accepted the administration’s rationale for its policies—in fact by
then many journalists, especially editorial and opinion writers were downright
skeptical of the linkages Bush had drawn between other nations’ purported
WMD and the concomitant terrorist threat to the US, for example. By then, it
was not that journalists remained uncritical of the statements emanating from the
White House, the Pentagon, and the State Department, as that they still were
too accepting of how international issues and events were prioritized.
Too often the recognition that the administration gave to terrorism and pur-
portedly related issues, such as WMD, received comparable recognition from the
media—even if journalists criticized the administration’s “spin.” In effect the
media continued to confirm the Bush administration’s political and diplomatic
agenda-setting. Through overly stenographic reporting on the president, the
media amplified the administration’s voice—so when Bush said to the country
that Americans are vulnerable to WMD in the hands of terrorists, the media
effectively magnified those fears, even while they challenged the prior assump-
tion that Iraq had been an “imminent threat.” “The nation is being trained to
consider terrorism only in its most apocalyptic forms,” Mark Leibovich of the
Washington Post noted in May 2003. “Many sociologists, scenario planners and
counterterrorism experts believe the government and the media are too focused
on extreme menaces—namely the terrorist attacks that involve weapons of
mass destruction” (Leibovich 2003). Again, it had been President Bush who set
the tone for the apocalyptic approach. As his carrier speech on May 1, 2003
detailed:
Home Security Secretary Tom Ridge underlined that fear when he moved the US
terror alert back to Orange later that same month. Ridge noted that while there
71
SUSAN D. MOELLER
was no specific threat against the US that had prompted the heightened status,
“weapons of mass destruction, including those containing chemical, biological or
radiological agents or materials, cannot be discounted,” he said (Langley 2003).
72
A M O R A L I M A G I N AT I O N
73
SUSAN D. MOELLER
actively adversarial. They held the political actors liable by being eyewitnesses to
events, by communicating bluntly what they saw and heard, by publicly juxtapos-
ing a statement made previously with a comment made later. In those instances,
the media not only cultivated a moral imagination in their audience but also
understood that they had a responsibility to do so—the human rights of distant
strangers were reported on because the well-being of those strangers was under-
stood to be relevant to Americans.
If the media were delegates to the mandarin, representing the concerns of his
subjects, what arguments would they have at their disposal to convince him to
become engaged by the trials of people he had never seen? They could try to cul-
tivate a moral imagination in him sufficient to acquaint him with others’ pain
and torments. In his awakened awareness of the vulnerability of his own comfort
he could find cause to alleviate the suffering of those others. The virtue in empa-
thy is that it is both a welcome gloss over the cold habit of self-interest as well as
a meaningful and practical end in itself.
In the real aftermath of 9/11, challenging the terrorism frame is itself an act of
moral imagination. The “if-you’re-not-with-us-you’re-against-us” attitude of Pres-
ident Bush’s war on terrorism is a new variant of “manifest destiny”: American
“destiny” is “manifest.” Or else, just as the “Cold War” label seduced the media
and entire nations into believing that it explained everything, even though it
omitted much and distorted more, so too the terrorism frame threatens a nuanced
understanding of the world. Both dangerous simplifications of the world order
were articulated by politicians who then accused those who didn’t believe of
being unpatriotic—even those who only wanted to challenge the application,
not the need for such policies.
The war on terrorism substitutes moral rhetoric for a moral imagination. “His-
tory has driven us toward moral enlightenment—and then left the final choice to
us,” wrote author Robin Wright. “Religious motivation isn’t necessary. Simple self-
interest will do” (Wright 2003). Enlightened self-interest. Self interest of the kind
that would convince a mandarin to care for those “hundred millions of his
people.”
Notes
1 Of course there also remain campaigns that have retained their moral flavor, even in
historical hindsight—such as the American Civil Rights campaign, as articulated by
such Christian preachers as the Rev. Martin Luther King, Junior.
2 In a recent book, author Linda Hudson argued that it was not O’Sullivan, but a nine-
teenth-century political writer named Jane McManus Storm Cazneau who wrote the
anonymous “Annexation” article (Hudson 2001).
References
Auchmutey, J. (October 21, 2001) “The power of language,” The Atlanta Journal and Con-
stitution, p. 1D.
74
A M O R A L I M A G I N AT I O N
Barr, C.W. (July 31, 2001) “In Mideast, one weapon of choice is a loaded word,” Christian
Science Monitor, p. 1.
Bilgrami, A. (2002) “Accountability: the philosophical background,” ICRC’s Forum: War
and Accountability, Geneva, p. 15.
Bogaev, B. and Nunberg, G. (October 5, 2001) “History of the term terrorism,” Fresh Air,
NPR.
Burns, J. (2003) “The moral compass of Iraq” in Bill Katovksy and Timothy Carlson (eds),
Embedded: the Media at War in Iraq, an Oral History, Guilford, CT: Lyons Press, pp. 160,
161.
Feynman, R. (1998) The Meaning of It All: Thoughts of a Citizen-Scientist, Reading, MA:
Perseus Books, pp. 44–6.
Friedman, T. (May 14, 2003) “Two right feet,” The New York Times, p. A25.
Gelfand, L. (February 3, 2002) “Newspaper careful in use of label ‘terrorist,’” Minneapolis
Star Tribune, p. 27A.
Ginzburg, C. (Autumn 1994) “Killing a chinese mandarin: on the moral implications of
distance,” Critical Inquiry, vol. 21, no. 1, pp. 46–60.
The Guardian leading article (May 20, 2003) “Self-fulfilling prophecy: US threats make
matters worse in Iran,” The Guardian, London, p. 21.
Hudson, L.S. (2001) Mistress of Manifest Destiny: a Biography of Jane McManus Storm
Cazneau, 1807–1878, Austin: Texas State Historical Association.
Johnson, P. (September 15, 2003) “Amanpour: CNN practiced self-censorship,” USA
Today, p. 4D.
Kurtz, H. (September 24, 2001) “Peter Jennings, in the news for what he didn’t say,” The
Washington Post, p. C1.
Kurtz, H. (October 31, 2001) “CNN chief orders ‘balance’ in war news; reporters are told
to remind viewers why US is bombing,” The Washington Post, p. C01.
Langley, A. (May 20, 2003) “US terror alert raised to second highest level: intelligence
assessments indicate al-Qaeda in operational period,” State Department.
Laqueur, T. (2001) “The moral imagination and human rights” in Michael Ignatieff,
Human Rights as Politics and Idolatry, Princeton, NJ: Princeton University Press,
pp. 131–5.
Laqueur, W. (September/October 1996) “Postmodern terrorism,” Foreign Affairs.
Lears, J. (March 11, 2003) “How a war became a crusade,” The New York Times, p. A25.
Leibovich, M. (May 1, 2003) “Fear factoring: nervous about the terrorist threat, people
imagine the worst,” The Washington Post, p. C1.
Los Angeles Times editorial (March 14, 2003) “The right war in Iraq,” Los Angeles Times,
p. 14.
Maddocks, M. (January 17, 1980) “‘Terrorist’: a bomb-thrower of a word,” Christian Science
Monitor, p. 23.
Moeller, S. (1999) Compassion Fatigue: how the Media Sell Disease, Famine, War and Death,
New York: Routledge.
O’Sullivan, J.L. (November 1839) editorial, United States Magazine and Democratic Review.
O’Sullivan, J.L. (July/August 1845) “The great nation of futurity,” United States Magazine
and Democratic Review.
Purdum, T. (April 7, 2002) “What do you mean, ‘terrorist’?,” The New York Times, Week in
Review Desk, section 4: p. 1.
Schmemann, S. (April 7, 2002) “The world: beyond reason, the method of this madness,”
The New York Times, Week in Review Desk, section 4: p. 1.
75
SUSAN D. MOELLER
Slaughter, A. (September 16, 2001) “A defining moment in the parsing of war,” The Wash-
ington Post, p. B04.
Smith, A. (1976) Theory of Moral Sentiments in A.L. Macfie and D.D. Raphael (eds),
Oxford: Clarendon Press, vol. 3, no. 3.4, pp. 136–7.
United Press International (May 2, 2003) “Bush speaks to nation aboard warship.”
US Newswire (September 16, 2001) “Transcript of President’s September 16 remarks upon
arrival on South Lawn.”
US Newswire (September 20, 2001) “Text of President Bush’s September 20 speech as pre-
pared for delivery to Congress.”
US Newswire (November 29, 2001) “Transcript of remarks by the President to US Attor-
neys Conference in Washington.”
US Newswire (December 7, 2001) “Transcript of remarks by President Bush on USS Enter-
prise.”
Wright, R. (September 11, 2003) “Two years late, a thousand years ago,” The New York
Times, p. A25.
www.csmonitor.com/specials/terrorism/lite/expert.html
www.mediaweb.co.za/ArticleDetail.asp?ID=2572
76
4
THE PR OF TERRO R
How new-style wars give voice to terrorists
Media coverage of terror has slid in and out of the coverage of war, largely because
recent wars have been preceded and followed by episodes of terror. A loosely
linked chain was created in the wake of September 11, whereby al-Qaeda’s attack
led the US to declare war against the regime in Afghanistan, and then against
Iraq, motivated, in part, by the allegation that Saddam Hussein had supported
the September 11 terrorists. Similarly, the outbreak of Palestinian violence in
2000 and the bloodshed which followed, led to the so-called “a-symmetrical” or
“low-intensity” war which the Israeli army had declared against the terrorists. In
all three wars, the lines between terrorists and “harborers” were blurred, with the
two becoming interconnected symbols in an “axis of evil.”
The logic of fighting evil regimes with the aim of abolishing terror is fraught
with internal contradictions not only for politicians and generals, but also for jour-
nalists. Consider how the inability of winning against terror in Afghanistan and
Iraq altered the framing of war itself. Starting as an almost naïve proclamation of
the triumph of good against evil, the aftermath has altered the story and the ways
in which it is covered. Terror, seeping in, stained the victory with daily doses of
bloodletting, re-framing reality as an explosive, here-to-stay routine. The high-
spirited coverage of war, now seen as unfinished, has transformed into the nagging
coverage of the confused, chaotic, violent routine of the aftermath of war.
The process reminds us once again that distinct, relatively short, media
events—such as operation “shock and awe,” and, mutatis mutandis, the Twin Tower
attack—invite journalists to mobilize and join in the crisis, or the high-powered
action, as players rather than spectators—embedded with the army in Iraq, and
connecting with the traumatized public after September 11. But once the glamour
of the event fades, its overriding power exhausted, the media are left to themselves
to search for and compete over the best follow-up stories.
We argue that the unfinished story of September 11, reinforced by the in-
conclusive aftermath in Afghanistan and Iraq, has sent media off on their own
search for terrorists. Uncontrolled and undefeated, they remain the mysterious,
evasive power, which may raise its head at any time, anywhere around the globe,
77
TA M A R L I E B E S A N D Z O H A R K A M P F
remaining a constant threat. This multi-branched story, with its significant serial
character, plays on the public’s deep seated anxiety, on the prurient desire to get
acquainted with these behind-the-scene heroes, but no less on the ambition of
journalists and their sense of adventure. While terrorists have to hide from the
military forces pursuing them, interaction with the media is of mutual interest,
resulting in various versions of hide-and-seek. Thus, in parallel to the (widely
criticized) patriotism that characterized US media coverage of the wars in
Afghanistan and in Iraq, the same media have developed new genres of covering
terrorists, quite apart from the coverage of terror attacks, which may even be
characterized as “unpatriotic.”
This chapter argues that September 11 changed the image of terror and the
journalistic approach to terror. Thus, following September 11, which effectively
demonstrated the power of new-style international anarchic terror, terrorists have
come to be labeled enemies, rather than criminals. And, if until September 11 ter-
rorists could capture media attention only via violent action, following it the
terrorists have been given voice. We argue that there has been a gradual shift
whereby terrorists have become regular, sought after sources, achieving a status in
which they speak for themselves, are listened to, explain their motives, and, to
some extent, set the news agenda. Rather than a bloody blot on the front page rep-
resenting the merciless suffering inflicted on innocent victims, they now make
statements, give interviews, and negotiate with governments, while opinion edi-
torial writers and commentators, following their trail, speculate about their plans,
their ideologies, and their psychological makeup.
In what follows we analyze this change, bring evidence to substantiate it, pro-
pose possible causes, and discuss some of the ethical issues it raises. First we point
to the processes that changed the character of both terrorist and media organiza-
tions during the 1990s, and the ensuing new genres of reporting terror. Live
disaster marathons signal the beginning of journalists relinquishing control to ter-
rorists, followed by the media’s regular pursuit of terrorists as legitimate news
personas, disconnected from specific violent acts. We bring initial evidence for
this transformation from television news channels and from the inside pages of the
printed press. Second, we offer four possible reasons for the upgrading of terrorists
to the cultural status of celebrities. Third, we define and illustrate new-style genres
of covering terror. These quasi-news soft formats include unmediated showing of
“home movies,” interviewing by proxy stories of search and find missions, center-
ing on the journalists heroic endeavors; and psychological profiles of (live or dead)
terrorists, provided by families, friends, and colleagues. In both the coverage of
war, and in the new formats of covering terror, journalists end up compromising
professional norms, as access is controlled by the sources (the army/ the terrorists).
78
THE PR OF TERROR
fear, it relies on the media for realizing its aims. Until recently, most media
researchers agreed on the way in which news media cover demonstrators, rioters,
and protest movements, and, all the more so, terrorists. They argued that whereas
politicians and representatives of the elite are free to address the media at any
time (crossing the threshold through the “front door”), the only chance of radical
groups to invade the screens is via the “back door,” that is, by the use of violence
(Gitlin 1980; Wolfsfeld 1991). Hence, a trap emerged in which groups such as
the anti-Vietnam movement in the US, Aldo Moro’s kidnappers in Italy, and the
Palestinians during the 1980s found themselves. The more violence they created,
the greater the chance of crossing into the screens and being viewed by the
public. The chance, however, was also greater for the coverage to be more nega-
tive, and therefore act as a boomerang.
Two processes during the 1990s radically changed the character of both terror
itself and the way in which it was covered. Until that time, terrorist groups such
as the IRA, the Red Brigade or the PLO, even if they had international connec-
tions, were perceived as an internal national phenomenon, handled in the
domestic field. Governments treated terrorists by controlling their appearance—
more precisely, their non-appearance1—on local media. This was done by
legislation, by putting pressure on national media, or by negotiating and reaching
an understanding with media institutions over limiting exposure to terrorists.
Media institutions followed suit, adopting internal directives for limiting cover-
age (Weimann 1999).2 The situation in which both terror and media were
operating within the nation’s boundaries made it possible for governments to
limit the terrorists’ capacity to exploit the media for creating public anxiety,
enhancing their bargaining power while holding hostages, or communicating
with their own supporters.
This contained situation changed when connections among terrorist organiza-
tions in various countries tightened, making terror into an international network.
At the same time, revolutionary communication technologies created a new
media ecology, transforming the journalistic profession and its locus. Thus, the
(insufferable) ease of transmitting live from various sites disintermediated editors
by interrupting news editions with “breaking news.” The realization of the
public’s right to know by the accepted journalistic practices suffered another blow
by the establishment of new competing media channels, broadcasting around the
clock and viewed beyond the state’s borders, with each quoting everyone else,
and with every channel doing its best to keep zapping viewers from escaping. A
significant example is the taking root of the marathonic live format throughout
and following terrorist attacks, which in effect cancelled the editor’s role, and,
with it, the striving for fair, precise, and responsible news, and the obligation for
not publishing, knowingly or carelessly, untrue or inexact information. The genre
of disaster marathons caused public criticism of journalists for playing into terror-
ists’ hands, that is, for inadvertently doing them a service (Liebes 1998).3
Whereas live marathonic broadcasts make for controversial journalism, their
format does not deviate from the principle of covering terror only when it acts.
79
TA M A R L I E B E S A N D Z O H A R K A M P F
80
THE PR OF TERROR
responsible for the most destructive (physically and symbolically) terrorist attack
directed at the US, became a superstar for whom major ethical principles of jour-
nalism were being compromised.
81
TA M A R L I E B E S A N D Z O H A R K A M P F
historiography heralds the end of the grand narratives of the postcolonial era,
narratives well known for their being a “white mythology,” aimed at repressing
various others. Its demise points to a need to offer alternative grand narratives to
the repressed (Weinrib 2003). The first step is raising the heretofore depressed
awareness of those whose voice is not heard. In this spirit, media scholars such as
Nick Couldry (2001) see in the World Trade Center attack a statement of the
repressed, those pushed to the margins by the institutions of media, and left in a
state of symbolic inferiority. According to this perception it is possible to prevent
future attacks by listening to the voice of the repressed, and thus contributing to
the creation of more equality between the worlds. Based on the notion that a
genuine act of communication emerges out of interruption rather than out of har-
mony, thinkers such as Levinas, Derrida, and Lyotard insist on the importance of
communicating with alterity rather than with sameness (Pinchevski 2003). By
talking to people who are like oneself only reinforces one’s own perceptions by
their reactions. Levinas argues that attention has to shift from the content of
conversation to the communication act or, in other words, from what was said to
the fact of saying. The spirit of these perceptions softens the traditional view,
according to which journalists should be careful not to appear to be conducting
negotiations with terrorists, or, worse, giving them a platform for propaganda.
Last, the increasing personalization of politics (of which the media are a main
culprit) and, with it, the incessant search for new superstars (Kellner 2003), cre-
ates the right atmosphere for the protagonists of terror to become international
celebrities. Status conferral, as Lazarsfeld and Merton were aware of, is indifferent
to whether whoever appears is an honorable hero or a lowly villain. Villains,
notably gangsters, are well known heroes (albeit tragic) in the heart of American
popular culture (Warshow 1979). Moreover, in the era of multiculturalism, the
judgment of celebrity as villain or hero itself increasingly lies in the eye of the
beholder (as seen in the split reactions of the American public in the Clarence
Thomas affair, or in the O.J. Simpson trial). And regardless of what the judgment
is, curiosity only speeds up when a celebrity is discovered as a villain at heart.
82
THE PR OF TERROR
83
TA M A R L I E B E S A N D Z O H A R K A M P F
Interviewing by proxy
Among the forms of the coverage of terrorists discussed, interviews, even if
second hand, still present an attempt to carry out journalistic work, however
constrained. Before examining the question of interviewing by proxy, the funda-
mental issue of whether terrorists should be given voice at all needs to be
confronted. Whereas a case can be made for distinguishing between terrorists
acting for the legitimate cause of fighting occupation, it is difficult to justify inter-
views with a hiding terrorist, promoting acts of mass destruction, based on the
belief that there is no possibility of compromise between Islam and the infidels
(the Americans, capitalism, Western culture).10 Whereas news reports on terror-
ists within the mandate of surveillance (on clues to the identity of perpetrators,
on progress of the chase, etc), devoting television prime time to the pre-
advertised exclusive scoop of listening to bin Laden, means that the system to
whose eradication he dedicates his life is giving him the stage (Blondheim and Liebes
2002). Arguments such as “know your enemy” or the need to “release pressure” in
order to avoid an explosion, relevant to local conflicts with limited goals (the
Palestinians in Israel, the IRA in Britain), do not hold in the case of diffuse and
total terror (Blum-Kulka, et al. 2003). Second, even if the White House claim
that bin Laden makes use of being interviewed for transmitting operational orders
is not particularly convincing, there is nothing like appearing on American net-
works for strengthening his power among his supporters. And third, for CNN’s
original target audience, the fact of incorporating bin Laden in familiar formats
(with its careful distribution among “different voices”) means more than the
unmediated broadcast of his repetitive mutterings to his cohorts in a foreign lan-
guage, and it may even create an illusion of lessening the threat, while contributing to
its increase. And last, taking into account his consistent effort never to incrimi-
nate himself directly or be discovered, and his use of Western channels to
strengthen himself vis à vis his supporters, there is no hope that bin Laden (or his
cohorts) would cooperate in the way in which other interviewees are expected to
do.
But even if the principle according to which the interviewing of terrorists
is a legitimate journalistic mission is accepted, the professional compromises
that media are pushed to make exactly in such ethically borderline cases are ques-
tionable. Note that the more dangerous the terrorist, the more painful the
concessions.
As stated above, Osama bin Laden’s appearance on CNN on February 1, 2002
in what the network called “an interview” is a good example of demonstrating
the risk to journalistic ethics in interviewing by surrogate. The show features
segments of an interview granted to Tayseer Alouni, an Al-Jazeera reporter,
three months earlier, to which CNN was allowed to add written “omnibus”-style
questions. That CNN’s bosses were aware of the problematics of interviewing by
proxy can be seen by their avoidance of the issue in their (misleading) introduc-
tion to the program: “Accused terrorist mastermind Osama bin Laden said ‘the
84
THE PR OF TERROR
battle has moved to inside America’ in the only television interview he’s granted
since the September 11 attacks—now airing for the first time.”
Missing in this framing is the fact that the “scoop” was by then old news (also
reported by The New York Times in December 2001), and that Wolf Blitzer (or
any other CNN reporter) was not holding the microphone.
The professional issues demonstrated by this case study allow for the reexami-
nation of the definitions of terms such as “scoop,” “news value,” and “news
interviews.” Specifically, the dubiousness of the interview under scrutiny can be
put in the form of a number of ethical issues:
First, as the Los Angeles Times pointed out (October 18, 2001), the interview
was planned “even as (CNN) executives added that they don’t know whether he
is dead or alive,” that is, they were giving up on the fundamental practice of being
able to grant their audiences the basic knowledge of whether or not the inter-
viewee is the person CNN claims he is. The question—is this the real bin
Laden?—takes on particular relevance after the known practice of the appear-
ance (mostly on audio tapes) of “doubles” of Saddam Hussein and Mullah Omar,
following the Iraq war. As no Western journalists could ever meet with bin Laden
following September 11 (and as very few saw him before), there would always be
a doubt as to the authenticity of tapes emerging from somewhere.
Reactions of the competing news channels to the interview expressed mixed
feelings of “criticism, curiosity and envy,” making it quite clear that given the
opportunity they would do the same. Expressing the belief that at war journalists
should be patriotic (a belief reaffirmed by US journalists during the Iraq war), and
voicing a softer version of Condoleezza Rice’s concern with the channel’s hand-
ing airtime to bin Laden for communicating with his own supporters, CBS News
President Andrew Heyward stated that “CNN should not be seen bargaining
with terrorists” or “as providing a platform for propaganda.” In a similar spirit, the
interview was called “a slap in the face of the American people” and was com-
pared to “interviewing Adolf Hitler or Emperor Hirohito, who ordered the attack
on Pearl Harbor.”11
Further, competing news channels criticized the interview’s newsworthiness,
illustrating the confusion around this term (ironically the first critic was Al-
Jazeera’s Director General, who decided not to air the interview for reasons of
lacking “newsworthiness”). Strictly, the criterion for judging an interview as
“newsworthy” consists of whether any new, significant information was gained by
talking to the interviewee. CNN’s own definition for the journalistic scoop they
were offering viewers can be found in Blitzer’s opening promo: “Late October, in
the only television interview with Osama bin Laden since the September 11
attacks, broadcast here for the first time, he makes clear that the war of terror is
not finished.” As what is highlighted as the new information in the quote could
not be of great surprise for CNN audiences, the interview’s news value must lie
in the first part of Blitzer’s introduction. These words hold the promise of seeing
bin Laden for the first time since the September 11 attack. The offer to look at
bin Laden’s face connects directly with the viewers’ awed curiosity about the
85
TA M A R L I E B E S A N D Z O H A R K A M P F
86
THE PR OF TERROR
become the best show in town. This means that journalists are prepared to take
personal risks to seek them out and talk to them in person.
In case the point about bin Laden’s sacrificing his comfortable life for a
cause was missed, Miller goes even further, comparing bin Laden to a no
87
TA M A R L I E B E S A N D Z O H A R K A M P F
88
THE PR OF TERROR
89
TA M A R L I E B E S A N D Z O H A R K A M P F
growing up with the present, touching on the political: “In all my teenage time
my symbols were body-building and movie stars, and singers, and people like that.
Then it changes . . . the guerrilla, the fighter; then it was the stone thrower, and
today it is the martyr.”
A slightly different variation of journalistic missions aimed at uncovering the
roots of the phenomenon takes reporters into the field in an attempt to investi-
gate the home roots of celebrity terrorists (now dead, in hiding, or in prison) by
visiting their “natural” environment, and meeting the families in which they
grew up. One such story is based on interviews with family members of Zacarias
Moussaoui, al-Qaeda topnotch, now in prison (The New York Times, February 9,
2003). Under the title (reminding us that) “Everybody has a mother,” Mous-
saoui’s brother offers the familiar psychological interpretation to the “causes of
(his brother’s) anger.” He also happens to be promoting a book he is finishing on
the subject. The abundance of profile stories of Muhammad Atta offered by the
press, disclose dubious scoops such as an interview with his father entitled “he
never had a kite” (Newsweek, September 25, 2001), psychological descriptions of
the “double life of a suicide pilot” as “the shy, caring, deadly fanatic” (Guardian,
September 23, 2001), and the familiar motive of “the seeds of rage,” suggesting an
ideological explanation.17
The symmetry between villain and victim, the strong element of arbitrary fate,
and the shifting responsibility to external sources, on the one hand, and the pro-
ject of getting acquainted with the terrorist’s personal biography, on the other, all
contribute to creating empathy between the reader/viewer and the story’s protag-
onist. The focus on psychological and environmental interpretations effectively
cut off the perpetrator from his or her action and, worse, from the suffering
caused. Moreover, these profile stories, usually reserved for the stars of popular
culture, gradually upgrade terrorists to a new status in the exhibitionist culture of
spectacle (Kellner 2003). The paparazzi-style pursuit of terrorists may be seen in
the attempts to interview anyone who knew them (recall interviews with Mullah
Omar’s personal driver, Newsweek, January 21, 2001), and in stories equating the
notorious fame of terrorists with that of other popular culture stars. One example
of the latter is the extensive coverage of the Hebron “invincible” football team,
in which six members became suicide bombers, causing a situation in which “the
team has started to lose” (Newsweek, July 7, 2002). The interlacing in the story’s
narrative of the engagement in popular sport and the engagement with terror–
from its being the best team, through its losing, and ending with the wish of a
father of one suicide bomber that “the Jihad soccer team will one day be born
again”—blurs the boundaries between the protagonists’ identity as football play-
ers, with whom it is easy to identify, and their identity as suicide bombers, with
whom it is more difficult. Described by the way they look (one has “a baby face,”
another has “brooding eyes”), their mobilization by a mosque called “Jihad,” their
participating in the Hammas terrorist movement, and their participation in a
football team (“squad” in the story) with the same name, there is no mention that
“Jihad” means “a sacred war,” and the translation of Shaheeds (suicide bombers)
90
THE PR OF TERROR
into “martyrs” (“the Jihad mosque became an factory of martyrs”) refers only to
the Muslim meaning of “saint” without the action involved.
Common to in-depths interviews with hidden terrorists (as in Newsweek’s fea-
ture “Muhammad’s Army”) or with their relatives (as in interviewing Muhammad
Atta’s father), or in the drawing up of psychological profiles with the help of pro-
fessionals, is looking at terrorists as disconnected from their acts, thus creating a
distance between the character responsible for terror and the suffering he causes.
In his book Modernity and the Holocaust, Zygmunt Bauman (1989) explains the
way in which all the moral scruples of Nazis carrying out orders during the holo-
caust were pushed aside by the distance created between the act—executed within
the framework of the bureaucratic system and facilitated by modern technology—
and the devastation it caused. This distancing cancels the moral meaning of the
killers’ actions and contributes to the erasing of the humanity of the victims by
their remaining invisible. We argue that the genre of interviewing, and other per-
sonalizing formats used—ones that lay terrorists on the psychologist’s couch, or
that venture into the field to talk to their families—are aimed at exposing their
fundamental humanity. These genres also lead readers and viewers to empathize
with the story’s heroes, creating a short circuit between the person we get to know
intimately and his/her responsibility for the suffering he/she caused and is plan-
ning to cause. In other words, positioning viewers vis à vis bin Laden, or likewise
protagonists, as having a face, and making them understand their motives and
their world view, disconnects terrorists in time and place from the acts they had
committed. It also neutralizes journalists’ capacity to play “devil’s advocate.” They
end up allowing their interviewees to create ambiguity regarding their responsibil-
ity for executing the atrocities which make them into sought after media stars.
Whereas Bauman points to the strategy of erasing the humanity of the victims
(to facilitate their being treated inhumanely), current television creates an oppo-
site problem by showing the human faces of the perpetrators of mass killings who
cannot be confronted with their crimes or their victims because journalists
relinquish their basic tools to do so. The fundamental conflict of interest that
exists between interviewer and interviewee, with subjects maneuvering between
a rhetoric of avoiding responsibility, directed at the channel’s target audiences,
and another rhetoric of “aleihoom,” disintermediating that audience directed to
their own fans.18 Thus, distancing reinforces the status of terrorists as a new kind
of celebrity of popular culture. In bin Laden we watch an authentic and charis-
matic visionary who, in spite of not taking any direct responsibility for
September 11, makes use of the opportunity to “balance” the picture by justifying
the mass killing and promising more. Appearing on CNN he can show his status
in the eye of the enemy to millions of potential fans, watching directly or from
the wings.
Finally, to return to the question of the paradox of media enthusiastically sup-
porting the government and the military at war (and accused of exaggerated
patriotism) that switch roles to undermine government policy by turning the
enemy into a super star. First, what is common to an embedded journalist and one
91
TA M A R L I E B E S A N D Z O H A R K A M P F
who chases terrorists is the sense of being there. Conveying to viewers the
authenticity of the experience has long overtaken ideals such as objectivity, accu-
racy, and responsibility. Second, the timing and the rhythm of covering war is
dictated by the government and the military. Stories of getting acquainted with
the persona of terrorists (in between attacks), however, are initiated by journal-
ists, who are (at least in part) in charge of the timing and the pace. Third,
pursuing terrorists provides journalists with an opportunity to prove that they are
not always patriotic. Once left on their own, they can return to their journalistic
role.19 It may also be the result of obeying the internal logic of popular exhibi-
tionist culture, which results in the treating of different subjects by similar
formats. Thus, as from “our” side the story of war and disaster is told as personal-
ized, human experience (tragic, heroic, or both), the coverage of terrorists is
also motivated by the search for heroes, villains, and victims. True, the terrorist
is seen from “our” perspective (as contestants or victims), but the enemy’s
courage, daring, and determination can be admired. And in the context of an
unfinished story, in which the scoring of points continues, it will always have
high ratings.
Notes
1 According to Kern, Just, and Norris (2003), the importance of the local angle in news
reporting on terrorism during the 1970s and the 1980s led to paying little attention in
the case of the American media to most of the terror attacks around the world. Citing
Weimann and Winn (1993) they reveal that between 1968–80 less than 20 percent of
terror attacks were reported by the three main American television networks. The
numbers declined further in the 1990s, when international terror attacks occurred less
frequently.
2 Thus in the case of the IRA (from the 1970s until the mid-1990s), the use of laws pro-
hibiting the interviewing of terrorists on media in Britain created a constant rift
between the government and media organizations (Wilson 1997); likewise, in the
Israeli case, interviewing Palestinians was legally prohibited until the beginning of the
Oslo process.
3 One case in point is the public criticism in Israel following a 72-hour long live broad-
cast of Israel’s public television channel in the wake of a series of terrorist attacks on
buses, carried out by the Hamas in March 1996. Academics and left-wing politicians
accused the channel of exacerbating the impact of the attacks, and playing into the
hands of the opposition to the Oslo peace process, both by the obsessive occupation
with the events and by the “whining” delivery style of its anchor.
4 The years in which this option is accessible in the New York Times Internet site.
5 In stories of journalists’ pursuit of terrorists, we did not include journalistic missions
aimed at exposing essential information for investigative reporting (such as the one
conducted by Daniel Pearl), concerned with raising a warning about environmental
risks.
6 Overall, the word “terror” appeared 301 times—156 between January 1, 1996 and
September 10, 2001, and 145 times between September 11, 2001 and August 31,
2003. The word “terrorist” appeared 250 times—76 between January 1, 1996 and Sep-
tember 10, 2001, and 174 between September 11, 2001 and August 31, 2003. In the
first period, only one item may be considered fitting for the category of the personal-
ization of terror. It is Jeffery Goldberg’s story of his voyage to Haqqania Madrasa (The
92
THE PR OF TERROR
New York Times, June 25, 2000), a militant Muslin seminary in Pakistan, in which he
interviewed future Taliban soldiers and leaders.
7 A recent example of serving terrorists’ causes is the Israeli media’s launching of a live
marathon following a suicide bomber’s attack in an Ultra Orthodox neighborhood in
Jerusalem on August 2003. In the attack’s aftermath, television could fatalistically
foretell Israelis that this is “the end of the Hudna (a reconciliation period between
Israel and the Palestinians),” and, we add, the victory of terror.
8 Whether Al-Jazeera’s allocating “so much airtime” to bin Laden, and making him into
a celebrity, is due as claimed by el-Nawawy and Iskandar (2002) to his promoting
advertising sales, or whether it is also the result of empathy for his cause, there is unan-
imous agreement that bin Laden is the channel’s star.
9 Tapes recorded by suicide bombers on the eve of their mission, designed for the
recruitment of new candidates, and for blocking the way for last minute regrets.
10 The basic claim that “terror” against occupation is legitimate, while terror against ide-
ology or for ideology (as in the bin Laden case) are not, can be found in the debate
over the definition of terror, in the US media and in Arabic countries such as Egypt,
Jordan, and Saudi Arabia (Haaretz, September 11, 2003). This distinction was
adopted also by the Washington Post for the labeling of members of the Palestinian
Hamas group as “militants” and members of al-Qaeda as “terrorists” (a distinction crit-
icized by pro-Israeli Media Watch camera.org). The explanation of the Washington
Post’s Ombudsman (September 21, 2003) for the differences was that “Hamas con-
ducts terrorism but also has territorial ambitions, is a nationalist movement and
conducts some social work. As far as we know, al-Qaeda exists only as a terrorist net-
work.”
11 Voiced by the President of the Media Research Center in Alexandria Va, this criticism
could be reinforced by bin Laden’s first statement in which he announces that “the
battle has moved inside America. We will work to continue this battle, God permit-
ting, until victory, or until we meet God.”
12 As Eric Sorenson, president of NSNBC, said sending out questions “takes out the ele-
ment of surprise, and rehearsed answers are not as honest as spontaneous ones.” In the
same spirit, Fox channel’s spokesman announced “the only way we would do it is if we
could have a sit down interview with bin Laden” (Los Angeles Times, October 18,
2001).
13 The claims against Alouni’s interview (for example, the insult from a governmental
source “he looked like a wimp,”) concentrated on his failure to make bin Laden
answer the crucial questions (The New York Times, December 12, 2001). Further,
Alouni’s extra deferential attitude may be evident in his repeatedly addressing bin
Laden with the differential “Sheikh.” A typical example of his representing bin
Laden’s interests might be seen in the following question: “al-Qaeda is facing now a
country that leads the world militarily, politically, technologically. Surely, the al-
Qaeda organization does not have the economic means that the United States has.
How can al-Qaeda defeat America militarily?”
14 Sometimes, an invitation is extended by the terrorists, as in the case of a Daily Tele-
graph reporter in Pakistan, who received a fax one week before the first anniversary of
September 11 on the attack, and was brought, eyes tied up, via a number of safe places,
to a flat in which he was introduced to a suitcase stuffed with al-Qaeda correspon-
dence. A few days later the source was arrested.
15 From a discourse perspective, Miller changes his footing (Goffman 1981) from claim-
ing authorship for his question to taking the role of animator in which he represents a
third party.
16 Johnson managed the scoop, he tells us, through the assistance of “a well-connected
intermediary,” asked to “help in connecting some fighters.” It is clear that in the com-
petition between the military and the media in pursuit of terrorists, the media wins.
93
TA M A R L I E B E S A N D Z O H A R K A M P F
For obvious reasons intermediaries for arranging interviews are easier to find than col-
laborators (i.e. intermediaries for arranging capture).
17 “In the streets where the terrorists came of age, many young Egyptians profess admir-
ation for America. But the seeds of rage are here too” (“Muhammad Atta’s
neighborhood,” Newsweek, December 16, 2001).
18 Recall Dan Rather’s interview with Saddam Hussein, heavily criticized by the govern-
ment, or Blitzer’s interview with bin Laden analyzed above.
19 This of course reminds us that interviews are based on the assumption of the inter-
viewee’s willingness to cooperate with the interviewer. Interviews with criminals are
unusually carried out in the law courts that have their methods for trying to extract
true answers from defendants.
References
Bauman, Z. (1989) Modernity and the Holocaust, Cambridge, England: Polity Press.
Blondheim, M. and Liebes, T. (2002) “Live television’s disaster marathon of September 11
and its subversive potential,” Prometheus, 20: 3.
Blum-Kulka, S., Kampf, Z., and Liebes, T. (2003) “Talking with the enemy?: interviews
with Palestinians during the second Intifada” in Y. Shlesinger and M. Muchnik (eds),
Lamed Le ILASH (Hebrew), Studies in Modern Hebrew, Jerusalem: Tzivonim Pub-
lishing.
Carey, J. (2002) “American journalism on, before and after September 11” in B. Zelizer
and S. Allan. (eds), Journalism after September 11, London and New York: Routledge.
Clayman, S.A. (1992) “Footing in the achievement of neutrality: the case of interview dis-
course” in P. Drew and J. Heritage (eds), Talk at Work, Cambridge: Cambridge
University Press, 163–98.
Couldry, N. (2001) A Way Out of the (Televised) Endgame, online, available www.open-
democracy.net/debates/article-8-39-11.jsp (accessed August 20, 2000).
el-Nawawy, M. and Iskandar, A. (2002) Al-Jazeera: How the Free Arab News Network
Scooped the World and Changed the Middle East, MA: Westview, Cambridge.
Ezrachi, Y. (2002) “There is no objective journalism,” Hayin Hashvi’it (Hebrew), vol. 39,
pp. 38–40.
Gitlin, T. (1980) The Whole World is Watching: Mass Media in the Making and Unmaking of
the New Left, Berkeley, CA: University of California Press.
Goffman, E. (1981) “Footing” in Forms of Talk, Philadelphia, PA: University of Pennsylva-
nia Press, pp. 124–60.
Kellner, D. (2003) Spectacle Culture, New York: Routledge.
Kern, M., Just, M. and Norris, P. (1993) “The lesson of framing terrorism” in P. Norris,
M. Kern, and M. Just (eds), Framing Terrorism: the News Media, the Government and the
Public, New York: Routledge.
Liebes, T. (1998) “Television’s disaster marathons: a danger for democratic processes” in
T. Liebes and J. Curran (eds), Media, Ritual and Identity, London: Routledge, pp. 71–84.
Pinchevski, A. (2003) “Interruption and alterity: dislocating communication,” PhD dis-
sertation: McGill University.
Warshow, R. (1979) “Movie chronicle: the Western” in G. Mast and M. Cohen (eds), Film
Theory and Criticism, New York and Oxford: Oxford University Press.
Weimann, G. and Winn, C. (1993) The Theater of Terror: the Mass Media and International
Terrorism, New York: Longman Publishing/Addison-Wesley.
94
THE PR OF TERROR
95
5
RESEARCHING
US MEDIA–STATE RELATIONS
AND TWENTY-FIRST
CENTURY WARS 1
Piers Robinson
For those of us seeking to understand the relationship between the state and
media during wartime, the Iraq war is potentially a unique case. The unprece-
dented scale of domestic and international public opposition, the failure to
obtain unequivocal UN authorization and the distinctive justifications used to
“sell” the war, including the alleged WMD threat and humanitarian justifications
related to Saddam Hussein’s record of tyranny, created an ambiguous and contro-
versial political environment. In addition, the presence of 24-hour news, the
Internet and satellite broadcasting appeared to destabilize wartime media–state
relations. An important question raised for those who have attempted to assess
the relative balance of power between the US media and the state during
wartime is whether we have witnessed an empowered wartime media or, con-
versely, continued domination of the news agenda by official viewpoints.
With this question in mind, I start by setting out an overview of our empirical
and theoretical understanding of media–state relations during previous wars.
There are reasons for thinking why we may be witnessing a more powerful and
influential media, including the impact of real-time and 24-hour news, trans-
national media organizations (so-called “global” media) and the Internet as well as
post-Cold War geopolitical circumstances, but there are also many reasons to sug-
gest that the impact of such changes is far less than is often assumed to be the
case. Setting out the research imperatives necessary to test the thesis that there
has been a significant change in the balance of power between government and
media during wartime, I argue that in the absence of appropriate research and
substantial empirical evidence, we cannot yet claim that there has been a signifi-
cant change in the balance of power and influence between the US media and
the state during wartime.
96
U S M E D I A – S TAT E R E L AT I O N S
Five reasons are frequently put forward in the literature on media–state relations
for explaining media deference to the state in wartime: reliance upon govern-
ment information sources, Cold War ideology, the “rally round the flag” effect,
political flak, and the corporate nature of mainstream US news media. I will out-
line each in turn.
First, journalists rely overwhelmingly on government sources when construct-
ing the news (Hallin 1986: 63–70; Herman and Chomsky 1988; Bennett 1990:
109; Wolfsfeld 1997; Mermin 1999). The need to supply a steady and rapid flow
of “important” news stories, combined with the vast public relations apparatus of
government and powerful interests more broadly, means that journalists tend to
become heavily reliant on public officials when defining and framing the news
agenda.
Second, during the Cold War the ideology of anti-Communism acted as a
control mechanism by providing journalists with a template with which to
“understand” global events, as well as providing political elites with a powerful
rhetorical tool with which to criticize as unpatriotic anyone who questioned US
foreign policy (Hallin 1986: 8–58; Herman and Chomsky 1988: 29).
97
PIERS ROBINSON
Third, owing to nationalism and the desire to “support our troops,” the so-
called “rally round the flag” effect (Mueller 1973) occurs during wartime. This
phenomenon can be understood to limit critical reporting through journalists’
and editors’ own patriotic response to military action as well as the desire among
media outlets to reflect the patriotism displayed by the public.
Fourth, when controversial material is aired it generates a disproportionate
degree of “flak” from individuals connected with powerful interests including gov-
ernment “spin doctors.” Such criticism serves to caution editors and journalists
against putting out news stories that are “too” controversial.
Fifth, the “size, ownership and profit orientation of mass media” (Herman and
Chomsky 1988: 3), and their shared “common interests . . . with other major cor-
porations, banks, and government” (Herman and Chomsky 1988: 14) creates a
clash of interest between the media’s supposed role as a watchdog of the govern-
ment and the interests of government and business elites (Bennett 1990: 109;
Herman and Chomsky 1988: 3, 14). Via institutional factors including strategic
interventions by owners of media conglomerates, recruitment processes that
select and reward journalists who see the world in a way congenial and relatively
unchallenging to elite interests, routine self-censorship on the part of journalists
and editors, and, perhaps most importantly, the internalization of a US elite ideo-
logical framework, news stories that run contrary to vested interests are, on
balance, less likely to surface than those consistent with the worldview of major
conglomerates.2 While the importance of economic factors for the shaping of
media output are sometimes downplayed or dismissed as conspiratorial, there does
exist remarkable proximity between US foreign policy objectives and US business
interests, as shown by Jonathan Mermin in his 1999 study of US media coverage
of post-Vietnam US military interventions:
One powerful interest that has a major stake in US foreign policy and
does have access to the news is business. But business, as a rule, has
found US foreign policy to be quite consistent with its interests. In the
Cold War, Washington supported anti-Communists against Commu-
nists—real and imagined—the anti-Communists being the side more
interested in economic engagement with the United States in terms
favourable to American business. In the post-Cold War era, a major
organizing principle of US foreign policy has been to secure investment
opportunities, market access, and oil for American business. The objec-
tives of US foreign policy therefore continue to match the interests of
American business.
(Mermin 1999: 28–9)
98
U S M E D I A – S TAT E R E L AT I O N S
media coverage of international affairs.3 Also, the research discussed should not
be taken to suggest that media deference to the government in wartime is
absolute, only that, on balance, news media tend to reflect the government line
during war.
Technological developments
At the forefront of debate during the 1990s was the apparent ascendancy of the
24-hour news channel and the associated proliferation of advanced news-
gathering equipment facilitating real-time television coverage,4 the rise of
transnational media organizations (global media) and the Internet. The net effect
of these technological developments, according to many analysts, was govern-
ment loss of control over the information environment and a news media that
was, at least potentially, more likely to be adversarial and “off-message” (e.g.
Annis 1991; Nye 1999; Rothkopf 1999; Shapiro 1999; Volkmer 1999; Deibert
2000; and Herrera 2002). The working assumption here was that media became
less dependent upon government information sources when defining and framing
the news agenda and more likely to include alternative viewpoints. For example,
Nye claimed that:
The free flow of broadcast information in open societies has always had
an impact on public opinion and the formulation of foreign policy. But
now the flow has increased in volume and shortened news cycles . . .
The so-called CNN effect makes it hard to keep items that might other-
wise warrant a lower priority off the top of the public agenda. Now, with
99
PIERS ROBINSON
Specifically, with respect to the 24-hour news cycle, James Hoge claimed that
today’s pervasive media increase the pressure on politicians to respond promptly
to news accounts that by their immediacy are incomplete, without context and
sometimes wrong (Hoge 1994). In relation to real-time television coverage, Pre-
ston argued that:
The key assumption underpinning these claims was that the ability to report in
real-time, 24 hours a day enabled the media to report events without mediation
by government press briefers. With respect to the transnationalization of media
organizations, Brown argues that it “challenges the ability of states to control
information flows” (Brown 2003a: 88) and, according to Shaw (2000), this “diffu-
sion of information through the increasingly global media cannot be contained
within the bounds that even the most powerful state leaders would prefer” (Shaw
2000: 33). Volkmer (1999: 4–5) took this analysis one step further by arguing that
CNN International is actually facilitating the emergence of a global public
sphere. Finally, the Internet was argued by many to have both expanded the
availability of alternative information for journalists and the public and to have
empowered non-elite voices. For example, analyzing the David and Goliath
struggle between the indigenous Chiapas guerrilla army and the Mexican state,
Kellner claimed that “From the beginning, the peasants and guerrilla armies
struggling in Chiapas, Mexico used computer data bases, guerrilla radio, and
other forms of media to circulate news of their struggles and ideas” (Kellner 1998:
182). The problem with such claims, however, is that there are many reasons to
suppose that technological advances have had little impact on the range of
sources used and substantive viewpoints expressed in mainstream US media.
100
U S M E D I A – S TAT E R E L AT I O N S
this hourly cycle only when a significant event occurs at which the media are pre-
sent. So, for example, during the 2003 Iraq war the hourly bulletin cycle would
often be interrupted for live coverage of “US/UK” press briefings and statements
by leading political figures. Furthermore, as Thussu argued, the drive to deliver
audiences to advertisers in a fiercely competitive market can lead 24-hour news
“to sensationalization and trivialization of often complex stories and a temptation
to highlight the entertainment value of news” (Thussu 2003: 118). For example,
the fall of Baghdad to US forces was epitomized by live coverage of the statue of
Saddam Hussein being pulled to the ground by a group of Iraqi men. The event
was enthusiastically reported by some Western journalists through comparison
with the collapse of Communist regimes in Eastern Europe. These comparisons
were, however, dubious for two reasons. First, the now seminal images generated
by coverage of the fall of Communism throughout Eastern Europe showed mas-
sive numbers of people protesting/celebrating as opposed to the several hundred
Iraqis seen pulling down the statue of Saddam in central Baghdad. Second, the
overthrow of the Communist regimes was via popular public protest, not armed
intervention by a foreign power. More significantly, the statue was pulled down
with practical help from US soldiers. Hence it is a matter of controversy whether
the pulling down of the statue was orchestrated by US soldiers. Certainly, sober
review of the footage shows that, beyond the several hundred Iraqi men working
with the US marines to pull down the statue, the surrounding area was sparsely
populated with Iraqi civilians who remained passive when the statue fell. More
generally live coverage of the fall of Baghdad was often at the expense of more
measured assessments regarding both Iraqi support for the US forces and the
actual degree of control that US forces had achieved. As we are now all too well
aware, the ability of US/UK forces to maintain even minimal levels of security
within Iraq has caused a sober reassessment among journalists and policy-makers
regarding initial claims of victory and liberation. In short, 24-hour news and real-
time reporting may create the impression of greater transparency, accuracy, and
diversity; but the superficial nature of such coverage can actually limit the overall
depth and quality of reporting.
101
PIERS ROBINSON
The Internet
While the Internet appears to facilitate a more plural media sphere, its impact in
terms of influencing both mainstream news coverage and pressure group mobiliza-
tion has been far from clear. In terms of influencing mainstream news coverage,
the extent to which on-line information sources have been integrated into
processes of news-gathering has received insufficient attention from empirically
based academic research8 and it might well be the case that journalists and editors
remain predisposed to rely upon what are perceived to be more “credible” sources
of information such as government officials and other elite voices. This is
arguably even more likely to be the case during wartime when misinformation
and propaganda on the World Wide Web is likely to be at a peak. In respect of
the mobilization of interest groups, there is reason and evidence to indicate that
the empowering potential of the Internet is far less than is often argued to be the
case. For example, with respect to the quality of on-line political expression (a
much vaunted component of arguments surrounding the positive impact of the
Internet on deliberative democracy), Wilhelm (2000: 98) found that:
102
U S M E D I A – S TAT E R E L AT I O N S
103
PIERS ROBINSON
104
U S M E D I A – S TAT E R E L AT I O N S
ment during wars and crises have accelerated. Coalition military operations in
Kosovo, Afghanistan, and the 2003 Iraq war have been accompanied by sus-
tained and highly organized attempts to influence media agendas by promoting
coverage of some issues rather than others and to influence the framing of stories
in ways that support the government’s cause. As Brown (2003a: 93) argued, at
least some of the impetus for these attempts during the 2002 war in Afghanistan
came from the UK government’s Director of Communications and Strategy, Alis-
tair Campbell:
Other activities inherent in the “war room” model include the use of press
releases, media appearances, press conferences, and speeches. In strategic terms,
these activities seek to encourage the development of common media frames over
time. In tactical terms they serve to minimize coverage of damaging or hostile
stories and to discredit oppositional counter-narratives. Some of the routine
consequences during the 2003 Gulf War of such attempts at “perception manage-
ment” would appear to include the repeated and premature announcements
concerning suspected WMD sites as the US/UK military advanced into Iraq,
over-optimistic announcements concerning a popular uprising against Saddam’s
forces in Basra and premature claims regarding the taking of the border town of
Umm Qasr during the first weeks of the conflict. A more dramatic and excep-
tional example of media-management was the recovery of the US soldier Jessica
Lynch that involved a military operation coordinated and exploited in order to
facilitate dramatic and positive news coverage. Ruthlessly exploited by the US
military at the time as a tale of daring heroism, press briefers at Central Com-
mand (Centcom) in Doha, Qatar, presented the rescue as occurring under hostile
conditions and exaggerated the circumstances under which Jessica Lynch was
captured to create the impression she had fought to the last bullet. In reality, as
Lynch herself has admitted, she did not fire a single shot during her capture and
she was in fact well cared for by Iraqi doctors and nurses.10
Overall, the jury is still out as to the extent to which the ascendancy of these
coordinated and sustained approaches to wartime media management have
counteracted the potential effects of new media technologies discussed earlier.
Accordingly, there is need to identify the key government information manage-
ment strategies and actual outputs (press briefings, etc.) over the course of
recent wars, including strategic attempts to develop common framings over time,
tactical activities designed to minimize damaging coverage and/or to discredit
counter-narratives, the techniques used to coordinate US and UK information
management strategies and the involvement of embedded journalists. Mapping
105
PIERS ROBINSON
106
U S M E D I A – S TAT E R E L AT I O N S
107
PIERS ROBINSON
The Vietnam War was complicated by factors that never before occurred
in America’s conduct of a war . . . More than ever before, television
showed the terrible human suffering and sacrifice of war. Whatever the
intention behind such relentless and literal reporting of the war, the
result was a serious demoralization of the home front, raising the ques-
tion whether America would ever again be able to fight an enemy
abroad with unity and strength of purpose at home
(Nixon 1978: 350)
As we now know from Hallin’s (1986) Uncensored War, the notion of an anti-war
media during Vietnam is largely a myth. The same exists with respect to the
CNN effect of the 1990s. At the time of the interventions in Iraq (1991) and
Somalia (1992) the claim that the media were responsible for these “humanitar-
ian” wars was widespread throughout foreign policy and humanitarian circles.
Again, such claims were made in the context of an ongoing debate over the rise
of 24-hour global news media and changing geostrategic realities. And yet subse-
quent research has demonstrated that the CNN effect was largely exaggerated
and that the actual influence and power of mass media was far more circum-
scribed than suggested by widely held assumptions regarding the CNN effect
(Gowing 1994; Strobel 1997; and Robinson 2002). Substantive military inter-
vention, even during the 1990s, remained driven by geostrategic reasoning rather
than any kind of media pressure or CNN effect.
Such examples should be sobering for those who make claims of significant
change in media–state relations. So too should be the fact that only two of the five
factors put forward to explain media deference during war time—Cold War ideol-
ogy and media reliance on government sources—could be argued to have changed.
Even with respect to these factors, however, there are many reasons to suggest con-
tinuity rather than change. In particular, it seems likely that any empowering
effects of new media technology may well have been counteracted by reinforced
government media management strategies. Little change in the areas of political
108
U S M E D I A – S TAT E R E L AT I O N S
flak, nationalism, and the corporate nature of the US news media suggests that the
media have “served the military rather well” (Carruthers 2000: 271–2), acting
now, as in previous wars, in support of US government war objectives.
Yet, as I have discussed, there remains a persistent discourse surrounding new
communication technology and the post-Cold War political environment, which
suggests that the contemporary media are more powerful and influential than
their predecessors. Perhaps more than a few academics and commentators are
guilty of wishful thinking, or at least engaging in research that is predicated upon
the assumption of change rather than continuity.17 There is need, then, to focus
more closely on the reasons that change continues to haunt our discussions of war
reporting, even when evidence suggests the contrary.
Notes
1 Thanks to the editors Stuart Allan and Barbie Zelizer, as well as Peter Goddard and
Marion Lloyd, for feedback on earlier drafts. Some of the material, in particular that
which identifies future research agendas, is drawn from an ESRC research grant pro-
posal, award reference RES-000-23-0551, developed by the author, Peter Goddard
(University of Liverpool) and Philip Taylor and Robin Brown (Institute of Communi-
cations, University of Leeds).
2 For useful introductions to the processes by which institutional factors limit critical
journalism see Klaehn (2002) and Herman (2000).
3 For a detailed review of the overlap between the claims of Herman and Chomsky
(1988) and other key scholars see E. Herring and P. Robinson (2003), “Too polemical
or too critical? Chomsky on the study of the news media and US foreign policy,” forum
on Chomsky, Review of International Studies vol. 29, no. 4: pp. 553–68.
4 The phrase “real-time” should be understood to refer both to the actual live reporting
of events and the use of “on the spot” footage obtained through the use of electronic
news gathering equipment (ENG) and other such portable newsgathering equipment.
This equipment allows journalists to report live or almost live without relying upon
government sources.
5 For example see US-based CNN, UK-based Sky News, and BBC News 24.
6 According to Volkmer (1999: 160) “World report” is run for 15 minutes daily as part
of the show International Hour, at 6.30 and 9.30 GMT Tuesday to Friday and has two
omnibus editions on a Sunday.
7 See “The untouchables,” The Guardian, March 31, 2003.
8 As a guide, to date I have reviewed well over 50 academic texts that analyze the
impact of new information communication technologies (NICTs) and have yet to find
an examination of the impact of online information sources upon news-gathering.
9 See, for example, Bruce Bimber (2003), Information and American Democracy, Cam-
bridge University Press: Cambridge.
10 For further details see “The war we never saw: the true face of war,” Channel 4, June 5,
2003, “The truth about Jessica,” The Guardian, Thursday, May 15, 2003, and “Lynch:
military played up rescue too much,” CNN, Saturday, November 8, 2003: downloaded
from http://edition.cnn.com/2003/US/11/07/lynch.interview/index.html, download
date November 12, 2003.
11 Inferences developed from research would need to be triangulated with primary inter-
views with journalists, policy makers and press briefers, and secondary sources (such as
memoirs, public statements and academic accounts) in order to build up a thorough
picture of media–state relations during the war.
109
PIERS ROBINSON
12 See in particular the speech by British Prime Minister Tony Blair to the Economic
Club of Chicago, Hilton Hotel, Chicago, USA, Thursday, April 22, 1999, available
online at www.fco.gov.uk
13 For further details on this case of intervention see Robinson 2002: pp. 93–110.
14 For opinion poll data on US beliefs in an al-Qaeda/Hussain link and Bush’s linking of
the two see “Hussein link to 9/11 lingers in many minds” by Dana Milbank and Clau-
dia Deane, Washington Post, Saturday, September 6, 2003, p. A01.
15 David Clark, “Iraq has wrecked our case for humanitarian wars,” The Guardian,
Monday, August 11, 2003.
16 For further reading on humanitarian war and intervention see O. Ramsbotham and
T. Woodhouse (1996) Humanitarian Intervention in Contemporary Conflict, Cambridge:
Polity Press and Blackwell.
17 For further details on problems surrounding theorizing change and the tendency to
exaggerate media impact see Philip Hammond (2000) “Lessons of the first draft of his-
tory,” Media Culture and Society, vol. 22, pp. 847–51; Piers Robinson (2000) “Media
power and world politics: problems of research design,” Media, Culture and Society, vol.
22, no. 2, pp. 227–32; and Robin Brown (1998) “It’s got to make a difference, hasn’t
it?: communication technology and practice in world politics,” paper presented at the
British International Studies Association Annual Conference, University of Sussex,
December 14–16, 1998.
References
Annis, S. (1991) “Giving a voice to the poor,” Foreign Policy, 84, pp. 93–106.
Baudrillard, J. (1991) “La Guerre du Golfe n’a pas eu lieu,” Libération, March 29, 1991. For
an English translation, see Baudrillard (2000) The Gulf War Did Not Take Place, trans-
lated with an introduction by P. Patton, Sydney: Power Publications.
Bennett, W.L. and Paletz, D.L. (eds) (1994) Taken by Storm: the media, Public Opinion and
US Foreign Policy in the Gulf War, Chicago, IL: University of Chicago Press.
Bennett, W.L. (1990) “Toward a theory of press state relations in the United States,” Jour-
nal of Communication, vol. 40, no. 2, pp. 103–25.
Bimber, B. (2003) Information and American Democracy, Cambridge: Cambridge University
Press.
Brown, R. (1998) “It’s got to make a difference, hasn’t it? Communication technology and
practice in world politics,” paper presented at the British International Studies Associ-
ation Annual Conference, University of Sussex, December 14–16, 1998.
Brown, R. (2002) “The US and the politics of perception management,” Journal of Infor-
mation Warfare, vol. 1, no. 3, pp. 40–50.
Brown, R. (2003a) “Spinning the war: political communications, information operations
and public diplomacy in the war on terrorism” in D. Thussu and D. Freedman (eds),
War and the Media: Reporting Conflict 24/7, London: Sage.
Brown, R. (2003b), “Clausewitz in the media age: rethinking the military media relation-
ship” in M. Just, M. Kern and P. Norris (eds), Framing Terrorism, London: Routledge.
Carruthers, S.L. (2000) The Media at War: Communication and Conflict in the Twentieth
Century, Macmillan: London.
Cleaver, H.M. Jr (1998) “The Zapatista effect: the Internet and the rise of an alternative
political fabric,” Journal of International Affairs, vol. 51, no. 2, pp. 6,211–40.
Deibert, R. (2000) “International plug ‘n’ play: citizen activism, the internet and global
public policy,” International Studies Perspectives, vol. 1, no. 3, pp. 255–72.
110
U S M E D I A – S TAT E R E L AT I O N S
Entman, R. (2000) “Declarations of independence: the growth of media power after the
Cold War” in B.L. Nacos, R. Shapiro, and P. Isernia (eds), Decision Making in a Glass
House: Mass Media, Public Opinion and American and European Foreign Policy in the 21st
Century, London: Rowman and Littlefield Publishers, pp. 11–26.
Froehling, O. (1997) “The cyberspace ‘war of ink and internet’ in Chiapas, Mexico,” Geo-
graphical Review, vol. 87, no. 2, pp. 291–307.
Glasgow University Media Group (1985) War and Peace News, Milton Keynes: Open Uni-
versity Press.
Gowing, Nik (1994) “Real-time coverage of armed conflicts and diplomatic crises: does it
pressure or distort foreign policy decisions?” Harvard: Joan Shorestein Barone Center,
working paper, pp. 91–4.
Hallin, D. (1986) The Uncensored War, Berkeley, CA: University of California Press.
Hammond, P. (2000) “Lessons of the first draft of history,” Media Culture and Society, p. 22.
Hammond, P. and Herman, E. (eds) (2000) Degraded Capability: the Media and the Kosovo
Crisis, London: Pluto Press.
Herman, E. (2000) “The propaganda model: a retrospective,” Journalism Studies, vol. 1,
no. 1, pp. 101–12.
Herman, E. and Chomsky, N. (1988) Manufacturing Consent, New York, NY: Pantheon.
Herrera, G.L. (2002) “The politics of bandwidth: international political implications of a
global digital information network,” Review International Studies, vol. 28, no. 1,
pp. 93–122.
Herring, E. and Robinson, P. (2003) “Too polemical or too critical? Chomsky on the study
of the news media and US foreign policy,” Forum on Chomsky, Review of International
Studies, vol. 29, no. 4, pp. 553–68.
Hill, K.A. and Hughes, J.E. (1998) Cyberpolitics: Citizen Activism in the Age of the Internet,
Rowman and Littlefield Publishers INC: Lanham, Boulder, New York, Oxford.
Hoge, J. (1994) “Media pervasiveness,” Foreign Affairs, vol. 73, pp. 136–44.
Kellner, D. (1998) “Intellectual, the new public spheres, and techno-politics” in C. Tou-
louse and T.W. Luke (eds), The Politics of Cyberspace, Routledge: New York and
London, pp. 167–86.
Kennan, G.F. (1993) “Somalia, through a glass darkly,” The New York Times, Septem-
ber 30.
Klaehn, J. (2002) “A critical review and assessment of Herman and Chomsky’s ‘propa-
ganda model,’” European Journal of Communication, vol. 16, no. 4, pp. 147–82.
Mandelbaum, M. (1994) “The reluctance to intervene,” Foreign Policy, vol. 9, pp. 3–8.
Mermin, Jonathan (1999) Debating War and Peace, Princeton, NJ: Princeton University
Press.
Mueller, J. (1973) War, Presidents and Public Opinion, New York: John Wiley.
Nacos, B.L., Shapiro, R., and Isernia, P. (eds) (2000) Decision Making in a Glass House:
Mass Media, Public Opinion and American and European Foreign Policy in the 21st Cen-
tury, London: Rowman and Littlefield Publishers.
Nixon, R. (1978) The Memoirs, New York: Grosset and Dunlap.
Nye, J. (1999) “Redefining the national interest,” Foreign Affairs, vol. 78, no. 4, pp. 22–35.
Paletz, D.L. (1994) “Just deserts” in W.L. Bennett and D.L. Paletz (eds), Taken by Storm:
the Media, Public Opinion and US Foreign Policy in the Gulf War, Chicago, IL: University
of Chicago Press, pp. 277–92.
Philo, G. and McLaughlin, G. (1993) The British Media and the Gulf War, Glasgow: Glas-
gow University Media Group.
111
PIERS ROBINSON
Pickerell, J. (2000) “Environmentalists and the Net: pressure groups, new social move-
ments and new ICTs” in R. Gibson and S. Ward (eds), Reinvigorating Democracy? British
Politics and the Internet, Ashgate: Aldershot, pp. 129–50.
Preston, A. (1996) “Television news and the Bosnian conflict” in J. Gow, R. Paterson and
A. Preston (eds), Bosnia by Television, London: British Film Institute.
Ramsbotham, O. and Woodhouse, T. (1996) Humanitarian Intervention in Contemporary
Conflict, Cambridge: Polity Press and Blackwell.
Robinson, P. (2000) “Media power and world politics: problems of research design,” Media,
Culture and Society, vol. 22, no. 2.
Robinson, P. (2001) “Theorizing the influence of media on world politics: models of media
influence on foreign policy,” European Journal of Communication, vol. 16, no. 4,
pp. 523–44.
Robinson, P. (2002) The CNN Effect: the Myth of News, Foreign Policy and Intervention,
Routledge: London and New York.
Romano, D. (2002) “Modern communications technology in ethnic nationalist hands: the
case of the Kurds,” Canadian Journal of Political Science, vol. 35, no. 1, pp. 127–49.
Rothkopf, D. (1999). “The disinformation age,” Foreign Policy vol. 114, pp. 82–96.
Shapiro, Andrew (summer, 1999). “The Internet,” Foreign Policy, no. 115, pp. 14–27.
Shaw, M. (2000) “Media and public sphere without border,” in B.L. Nacos, R. Shapiro,
and P. Isernia (eds), Decision Making in a Glass House: Mass Media, Public Opinion and
American and European Foreign Policy in the 21st Century, London: Rowman and Little-
field Publishers, pp. 27–40.
Strobel, W. (1997) Late Breaking Foreign Policy, Washington DC, WA: United States Insti-
tute of Peace.
Taylor, P.M. (1997) Global Communications, International Affairs and the Media Since 1945,
Routledge: London and New York.
Thrall, A. (2000) War in the Media Age, Cresskill, NJ: Hampton.
Thussu, D.K. (2000) “Legitimizing ‘humanitarian intervention’? CNN, NATO and the
Kosovo crisis,” European Journal of Communication, vol. 15, no. 3, pp. 345–61.
Thussu, D.K. (2003) “Live TV and bloodless deaths: war, infotainment and 24/7 news” in
D.K. Thussu and D. Freedman (eds), War and the Media, London: Sage, pp. 117–32.
Vickers, R. (2000) “Blair’s Kosovo campaign: political communications, the battle for
public opinion and foreign policy,” Civil Wars, vol. 3, no. 1, pp. 55–70.
Volkmer, I. (1999) News in the Global Sphere, Luton: University of Luton Press.
Wallis, R. and Baran, S. (1990) The Known World of Broadcast News: International News
and Electronic Media, London: Routledge.
Wilhelm, A. (2000) Democracy in the Digital Age: Challenges to Political Life in Cyberspace,
Routledge: London and New York.
Wolfsfeld, G. (1997) The Media and Political Conflict, Cambridge: Cambridge University
Press.
112
Part 2
BEARING WITNESS
6
WHEN WAR IS REDUCED
TO A PHOTOGRAPH
Barbie Zelizer1
Journalism’s images of war disturb. Among the most powerful visuals known to
humankind, they are haunted by the stubborn inevitability and proximity of
death. Combining the cool mechanics of the camera with the hot passions of
wartime, they offer visual statements about circumstances much of the world
prefers not to see: shattered bodies and mangled buildings, triumphant soldierly
gestures, hopes and broken spirits nestled inside devastation somehow too deep
for the camera to record. Journalism’s images of war both show what has been and
offer glimpses of what might be. And for those who have never experienced war
firsthand, journalism’s images of war provide what may be the only depiction of
what sadly has become one of the reigning circumstances of the contemporary
age.
But journalism’s images of war do not emerge from a vacuum. They are shaped
through a turn to the visual—a journalistic emphasis on images that typically
takes place during wartime. Crafted through a maze of practices and standards,
both explicit and implicit, by which photographers, photographic editors, news
editors, and journalists decide how war can be reduced to a photograph, journal-
ism’s images of war have come to represent an elaborated template for imagining
and assessing the wars of the twenty-first century. Such images reflect what the
camera sees by projecting onto that vision a set of broader assumptions about how
the world works. The ways in which this happens, however, raise questions about
the degree to which images, particularly photographs, are well suited to shaping
the public’s encounter with war.
When war is reduced to a photograph, the camera provides images that show far
more than just the scene at hand. Images that are composite, more schematic than
detailed, conventionalized, and simplified work particularly well in this regard.
Used as pegs not to specific events but to stories larger than can be told in a simple
news item, journalism’s images become a key tool for interpreting the war in ways
consonant with long-standing understandings about how war is supposed to be
waged—notions about patriotism, sacrifice, humanity, the nation-state, and fair-
ness that come as much from outside journalism as from within. War is presented
115
BARBIE ZELIZER
as often heroic and reflective of broader aims associated with nationhood, clean
and at times antiseptic, and involving human sacrifice for a greater good. Whether
or not this actually reflects how war is waged is beyond the mission of journalism’s
images of war. For a range of visual cues, or familiar templates for using images,
helps journalists depict circumstances that are fraught with unpredictability,
stressful judgment, emotionalism, inconsistency, and high stakes, facilitating jour-
nalism’s depictions of war but not necessarily in ways that support the image’s
workings as a tool of news relay. Relying on particular images, cues, and themes
that have proven themselves over time, journalism’s images of war gravitate
toward the memorable—as established through frequently depicted, aesthetically
appealing, and familiar images—as much as toward the newsworthy. What does
this mean for the health of the body politic, which depends on images of war to
provide presumably reliable visual depictions of war-stricken zones?
When compared with times of peace, wartime often produces both more pic-
tures, more varied pictures, and pictures whose display is justified on a wide range
of attributes—such as memorability, drama, or vividness. The use of pictures
depends not so much on explicit and articulated standards but on informal strate-
gies among journalists about how best to use them (Zelizer 1993; Rosenstiel 2003;
Glaser 2003). As reporters, editors, photo-editors, and photographers, journalists
agree on a myriad of interpretive strategies—regarding who depicts the war, what
is depicted, how it is depicted, and in which ways—that are shaped against the
larger restrictions through which pictures can be displayed. In wartime, informal
strategies among journalists thereby shape the turn toward the visual, whereby it
is agreed, though not necessarily articulated, that photographs depict certain
aspects of the war but not all.
For only certain aspects of war are ever seen in the images of war. Lacking
depiction are those sides of war which do not fit the prevailing interpretive
assumptions about how war is to be waged. In most wars, there are few or no
images of human gore, one’s own war dead or POWs, military operations gone
badly, or the effect of one’s own war on civilians of the other side. In certain con-
flicts, such as those in which the West is involved, this is more pronounced than
in others. Lacking depiction too have been a range of conflicts that missed being
seen almost totally—wars in East Timor, Sudan, Somalia, Liberia, and Zaire, to
name a few.
And yet wartime’s turn to the visual offers a way for news organizations to
accommodate the prominence of visual representation as a way of engaging with
war. The turn to the visual has a number of predictable attributes in war journal-
ism which have come to characterize our expectations of war as we begin the
twenty-first century. Examined here primarily in the US media and largely in
conjunction with the most reported wars of the past five years, the turn to the
visual nonetheless can be found in a far broader range of military conflicts and
geographic contexts, all of which have come to inhabit public depictions of the
contemporary era. In suggesting that journalists regularly gravitate toward certain
kinds of visual depictions in their visualization of recent wars, the turn to the
116
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
visual raises questions about the ways in which images function in wartime. More
specifically, it queries the role played by visuals in relaying a strategically crafted
depiction of war and in undermining the maintenance of a healthy body politic.
117
BARBIE ZELIZER
118
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
were not much better, generally portraying British triumphs, placid views far from
the frontlines, and circumstances in which everything looked “ship-shape and
everyone happy” (Knightley 1975: 15). The US Civil War brought with it a stark
realism associated with photography and moved away from the romanticized tenor
of drawings, introducing what came to be known as the “first living room war”
(Goldberg 1991: 20) and couching war’s representation in terms of realist dis-
course and authenticity rather than imagination and interpretation. World War II
was documented by more photographers than any previous military conflict, and
the rapid transmission of their depictions of battlefields by wire helped make the
war-news reader during the 1940s “far closer to events on the battlefield than his
[sic] grandfather, regardless of the actual distance he might be removed from active
fighting” (Mathews 1957: 193). A peculiarly autonomous sense about the process
of capturing a scene through photographs existed in these pre-television times. As
World War II photographer Margaret Bourke-White said of a shot she made of a
dead family killed in a German air raid, “a protecting screen draws itself across my
mind and makes it possible to consider focus and light values and the techniques of
photography, in as impersonal a way as though I were making an abstract camera
composition” (cited in Goldberg 1986: 245).
The introduction in later wars of more sophisticated imaging technologies—in
particular, the moving image and its evolving technologies of documentary, tele-
vision, cable, and digitalization—raised questions as to whether or not
photographs would continue to matter in war’s visualization. Accounts of news-
gathering practices widely held that the Korean War was the first covered by
television, that the Vietnam War introduced graphic images of war, that the Gulf
War of 1991 advanced cable images 24 hours a day, and that the 2003 war in Iraq
embraced broadband Internet and its associated digital devices—websites con-
taining live video, photography collections, audio reports, animated weaponry
displays, and interactive maps—that offered a “more intimate and multi-faceted
view of the war than . . . possible ever before” (Harmon 2003: C4). Through it all,
the still photograph persisted as a viable mode of recording, even though funda-
mental questions lingered regarding its use.
In each war, photographic depiction flourished, though it depended on what
the military and political forces, and sometimes editors, allowed the public to see.
During the Spanish American War, the Journal put out as many as 40 editions in
one day, much of which were characterized by a “lavish use of photographs and
illustrations” (Moeller 1989: 68). The photographs of Larry Burrows during the
Vietnam War earned him the title of “the greatest war photographer there has
ever been” (Hopkinson cited in Knightley 1975: 408), despite the fact that his
much-publicized doubts about the role of the photographer and the war itself
were at odds with his capacity to jump into settings and take photographs with-
out hesitation. The photographs of Eddie Adams, Ron Haeberle, and Malcolm
Browne produced still shots far more memorable than the corresponding nightly
television footage, with Haeberle’s images of the massacre at My Lei displayed as
still visuals on television.
119
BARBIE ZELIZER
120
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
the first war photographers really didn’t photograph war at all. Because
of the bulk of their equipment and the length of time it took to make an
121
BARBIE ZELIZER
122
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
noted that the photographs of Afghani children were “a glorious testament to the
timeless beauty of innocence and to the common admiration of this human con-
dition among Earth’s inhabitants” (Takase 2001: A12). At a time in which the
incursion into Afghanistan was only a month or so under way and thousands of
refugees had already gone homeless and hungry while others had lost their lives
in brutal battles, the laudation of photography’s aesthetic appeal gives pause to
the image’s role in documenting the conflict.
During the war in Iraq, color remained at the forefront of the photograph’s aes-
thetic appeal. The Chicago Tribune featured a front-page image of the back profile
of a soldier outlined against an amorphous orange background, making it appear
as if he were disappearing into an unidentifiable abyss (DiNuzzo 2003). As the
US forces moved into Basra in March of 2003, The New York Times featured a
half-page portrayal of a woman carrying potable water (“Hauling water” 2003:
B1). The picture, striking for that same peculiar orange haze that was typical of
many images from the region owing to the impact of dense sand in the air, was
significantly lacking in definitive detail: the Times mentioned nothing about
where precisely the photograph was taken other than on the outskirts of Basra,
nothing about the women other than that they carried potable water, and noth-
ing about what lurked behind the orange haze.
In fact, the orange haze appeared to be the striking reason for choosing the
photograph and others like it. On a day in which accompanying news stories doc-
umented attacks against Iranian opposition groups in Iraq, exchanges of fire in
Mosul, and the plundering of the Iraqi National Museum, this artistic choice to
lead the section on the war displaced an informationally rich one. In the words of
the Chicago Tribune’s art critic:
In this sense, “eye-popping colour draws us from the truth” (Artner 2003: 4). As
news, then, the aesthetic appeal of images works against the information that
such depictions can provide.
At times, the aesthetic appeal of war images turns too to explicitly non-
journalistic modes of visual representation. One article in the Philadelphia Inquirer
in March of 2003 about the false surrender of Iraqi troops was illustrated by a
reprint of the Goya painting, The Third of May 1808, in which he depicted the
execution of Madrid residents at the hands of Napoleonic forces (pictures
appended to Lubrano 2003: C1). With the decision to include it came another
decision not to include a photograph that depicted more directly what was hap-
pening in Iraq, suggesting that the image’s aestheticization works too against its
newsworthiness in wartime.
123
BARBIE ZELIZER
124
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
125
BARBIE ZELIZER
words as directed cues for interpretation is built into the ways in which journalists
create familiar standards for describing war to the public. Calling East Timor
“Indonesia’s Tiananmen” (“The Tiananmen in East Timor,” 1992: A20) or
“another Cambodia” (Kondrake 1979) explains complicated wars through associ-
ation. In a column titled “The real echoes from Vietnam” alongside an image of
Saigon, Newsweek correspondent Jonathan Alter addressed the circumstances of
war-torn Iraq in 2003 by discussing the implications of having “a moment early
last week [when] it felt like 1967” (Alter 2003: 41). Some images are contextual-
ized against an unspecific past, couched in broad terms, such as a shot from The
Economist—depicting weeping women in Bosnia—that was entitled “Out of the
past” with no details that identified the shot as contemporary or tracked which
past was being referenced (“Weeping women” 1991). Other times the past is
invoked more specifically, as when The New York Times asked “Does the world
still recognize a holocaust?’ in an article about a 1993 execution of Muslims in
the Balkans (pictures appended to Darnton 1993). Other parallels are struck
through recalling the circumstances of a photograph’s taking, as in likening the
image of Mohammad Aldura, a 12-year-old Palestinian boy shot to death in the
midst of a battle between Israeli soldiers and Palestinians during the Intifada of
2000 (Abu-Rahma 2000), and, once again, Eddie Adams’ iconic shot (Adams
1968) of the shooting of a Vietcong soldier by the South Vietnamese Chief of
Police in 1968 (Roskis 2000). Parallels are also shaped by region or locality. Con-
flicts in Burundi were labeled as “the next Rwanda” (Gillett and des Forges
1994), while battles in Russia and Moldova were called “the next Bosnia” (“The
next Bosnia” 1992: 43). On March 31, 2003, in its first issue after the recent war
in Iraq began, Time magazine titled its cover photograph “Gulf War II.” In each
case, the implication that earlier wars could offer an appropriate backdrop against
which to interpret contemporary battles sets up an associative framework by
which the more recent wars are seen through the filter of the earlier ones. The use
of words as guides, then, pushes audiences to a position by which they are
required to engage with the war that came before while dealing with the more
recent military conflict. As Time put it, “the shock of recognition is acute . . .
Surely these pictures . . . come from another time” (Walsh 1995: 46–7).
A second way of cueing the past through images is by simply repeating the the-
matic portrayal of earlier wars, visually marking the past but offering no verbal
cues. Ranging from the predictable—the heroic gestures of soldiers—to the
grotesque—stark depictions of human corpses, journalism’s images of war rework
recognizable and familiar themes of representation in visually depicting new
wars. Pictures are used as parallels in ways that mark the associations surrounding
certain wars as antecedents to later ones. The Tianenmen Square shot of a man
stopping a cordon of Chinese tanks in June 1989 with nothing more than a shop-
ping bag symbolized individual courage standing firm before the armed might of
the state (Widener 1989), but it also recalled a 1968 shot of a plumber in Prague,
who, on coming face to face with the Soviet tanks about to overrun the Czech
city, stood in front of the tank, tore open his shirt, and shouted at the tank com-
126
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
127
BARBIE ZELIZER
Figure 6.2 US marine doctor with Iraqi girl, photograph by Damir Sagolj, April 2003
(courtesy of Reuters)
128
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
the fact that other historical parallels with other events might have been more
suitable (pictures appended to Infield 2001: A11; discussed in Zelizer 2002). A
Time column on the so-called victory in Afghanistan was accompanied by two
photographs—one of Afghanistan and one of Somalia (pictures appended to
Krauthammer 2001: 60–1). A Boston Globe article about the toppling of the
statue of Saddam Hussein in Baghdad in 2003 was illustrated with additional
photographs depicting other seemingly similar moments from earlier points in
time—the razing of the Berlin Wall, the lone protester in Tienanmen Square, and
workers jumping off a statue of Lenin in Riga before Latvia seceded from the
Soviet Union (pictures appended to Gilbert and Ryan 2003: D1). The New York
Times ran one essay entitled “Watching Iraq, and Seeing Vietnam,” in which pic-
tures of Saigon and Baghdad were positioned side by side (Whitney 2003: section
4: 1; Figure 6.3) and another in which pictures of dead soldiers in the streets of
Baghdad and Mogadishu were lined up side by side alongside the suggestion that
the image would help turn the tide against the later war as it had done in the ear-
lier one (Kifner 2003 section 4: 1). Other times the photograph of the past
completely displaces the contemporary photograph. A New York Times piece
about civilian casualties, written at the onset of the war in Iraq in 2003, was illus-
trated by one image of dead civilians in Kosovo in 1999 and two photographs of
structural devastation in Kosovo in 1999 and Kabul, Afghanistan in 2001, but no
image at all of what was then already transpiring in Iraq (Eviatar 2003: D7, D9).
A Time article republished a picture of a Kurdish refugee fleeing Iraqi persecution
Figure 6.3 “Watching Iraq, and Seeing Vietnam,” The New York Times, November 9,
2003 (courtesy of The New York Times)
129
BARBIE ZELIZER
in 1991 as a bridge between one article about Saddam Hussein in 2003 and
another querying the US betrayal of Iraqis during the earlier Gulf War (picture
appended to McGeary 2003: 57), the picture offering a visual bridge by which the
more recent article could be understood. An article about urban warfare in Bagh-
dad during the war in Iraq positioned a drawing of what soldiers in the street
would look like alongside a photograph of a tank unit patrolling Mogadishu,
Somalia, ten years earlier (“A nation at war” 2003: B12). Newsweek featured an
image of Saigon in a column on Iraq (Alter 2003: 41). In all of these cases, the
distance between the war of today and that of earlier wars is collapsed via its
visual representation. The image of then substitutes for the image of today.
130
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
wartime? With the turn to the visual serving up the material by which this is
accomplished, there is need to think more critically about what such ends suggest
for the viability of a healthy and contested public sphere.
The ends that are at stake are numerous and varied. One might only think
about the pictures not seen—of the recent war in Afghanistan and, at the time of
writing, the ongoing war in Iraq. The fact that in times of war, journalism shifts to
a mode of photographic relay that proceeds basically without directives other
than informal impulses toward frequency, aesthetic appeal, and familiarity allows
for unusual, even faulty news judgment. It allows for pictures of the memorable,
frequent, dramatic, aesthetically appealing, powerful, and familiar over the news-
worthy, and it encourages pictures in a way that facilitates faulty comparisons
across events. Most important, it allows journalists to strike parallels between
wars for no better reason than the surrounding mandates for interpreting them
resemble each other. Those faulty comparisons are problematic. How alike are
the wars positioned here as visual parallels? They are not very alike at all. Yet
depicting new wars in ways that link them with old practices of depiction forces a
spectator position on the public that may have less to do with the war itself and
more to do with the parallel being struck. The public sees many pictures of war,
but what it sees is not necessarily what it needs to see. Seen are shards of wartime
presented in a way that forces certain public responses and mutes questions about
the ultimate value of both the depictions and the parallels being displayed, for
whom and to which ends.
When war is reduced to a photograph, then, its usage depends on journalism
being less journalistic than it needs to be. What clearer note is needed to signal
the dangers that photographs pose to the maintenance of a healthy body politic
in wartime?
Notes
1 Thanks to Lauren Feldman and Bethany Klein for collecting photographs studied for
this chapter. Parts of the chapter were presented at the Shorenstein Center on the
Press, Politics, and Public Policy at Harvard University, University of Southern Cali-
fornia, the Conference on the Media at War at the University of California at
Berkeley, Middle Tennessee State University, the New Hampshire Humanities Coun-
cil, the New School University, Grand Valley State University, and Wabash College.
Special thanks to the Annenberg School for Communication at the University of
Pennsylvania and to Harvard University’s Shorenstein Center, where I was a fellow
during the spring of 2004, for technical help in completing this chapter.
References
Abu-Rahma, Talal, “Shooting of Mohammad Aldura,” September 30, 2000, image taken
as a video still for French TV2.
Adams, Eddie, “General loan shooting of Vietcong soldier,” February 2, 1968, Saigon,
Vietnam, image taken for AP/World Wide Photos.
Alter, Jonathan, “The real echoes from Vietnam,” Newsweek, April 14, 2003, p. 41.
131
BARBIE ZELIZER
“An American servicewoman watched images from Al-Jazeera . . . ,” The New York Times,
March 24, 2003, p. B6, image taken for Getty Images.
Artner, Alan G, “The war, in color,” Chicago Tribune, April 10, 2003, section 5: pp. 1, 4.
Barthes, Roland, “The rhetoric of the image,” in Image/Music/Text, New York: Hill and
Wang, 1977.
Barthes, Roland, Camera Lucida, London: Hill and Wang, 1981.
Bielik, Ladislav, “The end of the Prague spring,” August 21, 1968, Bratislava, Czecho-
slovakia, image taken for Smena.
Bouju, Jean-Marc, “Iraqis detained,” Chicago Tribune, March 27, 2003, p. 3, image taken
for the Associated Press.
Brown, John, “‘They got it down’: the toppling of the Saddam statue,” Counterpunch, April
12, 2003, posted at http://www.counterpunch.org/brown04122003.html
Browne, Malcolm, “Reverend Quang Duc in flames,” June 12, 1963, Saigon, Vietnam,
image taken for AP/World Wide Photos.
Capa, Robert, “Death of a loyalist soldier,” September 15, 1936, image taken for Vu.
Capa, Robert, “Dead German soldier,” May 14, 1945, image taken for Life.
Chennelly, Joseph, “New way,” Time, December 24, 2001, p. 61, image taken for US
Marine Corps.
Cohen, Roger, “In Bosnia, the war that can’t be seen,” The New York Times, December 25,
1994, p. E4.
Coonradt, Peter, “Afghan coverage” (letter to the editor), Los Angeles Times, November
22, 2001, p. B12.
Darnton, John, “Does the world still recognize a holocaust?” TheNew York Times, April 25,
1993, section 4: p. 1.
Delay, Jerome, “Statue of Saddam Hussein falls,” The New York Times, April 10, 2003, p. 1,
image taken for Associated Press.
DiNuzzo, Nuccio, “An army soldier clears a building,” Chicago Tribune, March 27, 2003,
p. A10, image taken for Chicago Tribune.
Douglas, Susan, “A three way failure,” Progressive, July 1994, p. 15.
Eviatar, Daphne, “Civilian toll: a moral and legal bog,” The New York Times, March 22,
2003, pp. D7, D9.
Faddoud, Joseph, “Rescuer cradles injured baby,” May 22, 1985, image taken for Agence
France-Presse.
Fisher, Ian, “Threat gone, Iraqis unearth Hussein’s nameless victims,” The New York Times,
April 25, 2003, pp. A1, A10.
Fisk, Robert, “Baghdad: the day after,” The Independent, April 11, 2003, p. 1.
Franklin, Thomas E., “Flag raising after September 11,” Bergen Record-AP, image taken
September 2001.
Gilbert, Matthew and Suzanne C. Ryan, “Snap judgments: did iconic images from Bagh-
dad reveal more about the media than Iraq?” Boston Globe, April 10, 2003, p. D1.
Gillan, Audrey, “‘We came here not to get out alive: we’re ready for anything’; Kurdish
protesters besieged in London say they will die for the cause,” The Guardian, February
17, 1999, p. 3.
Gillet, Eric and Alison des Forges, “The next Rwanda,” The New York Times, August 11,
1994, p. A23.
Glaser, Mark, “Photojournalism gets boost online,” Online Journalism Review, posted
August 7, 2003.
Goldberg, Vicki, Margaret Bourke-White: a Biography, New York: Harper and Row, 1986.
132
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
Goldberg, Vicki, The Power of Photography, New York: Abbeville Press, 1991.
Halbwachs, Maurice, The Collective Memory, Chicago, IL: University of Chicago Press,
1980.
Hall, Stuart, “The determinations of news photographs,” in Stanley Cohen and Jock
Young (eds.), The Manufacture of News, London: Sage 1974.
Harmon, Amy, “Improved tools turn journalists into a quick strike force,” The New York
Times, March 24, 2003, pp. C1, C4.
“Hauling water,” The New York Times, April 17, 2003, B1, image taken for Reuters.
Hicks, Tyler, “Statue’s head dragged through Baghdad streets,” The New York Times, April
10, 2003, p. 1, image taken for The New York Times.
Hilbrand, David and Gail Shister, “A flood of images into homes,” Philadelphia Inquirer,
March 27, 2003, pp. A1, A20.
Hill, James, “Iraqi soldier killed,” The New York Times, March 27, 2003, image taken for
The New York Times.
Hinds, Julie, “Photographic memory,” Detroit Free Press, April 11, 2003, pp. H1, H4.
Howe, Peter, Shooting Under Fire: the World of the War Photographer, New York: Artisan,
2002.
“India suicide,” January 11, 2000, image taken for Associated Press/Press Trust of India.
Infield, Tom, “Tuesday’s terror evoked echoes of Pearl Harbor,” Philadelphia Inquirer, Sep-
tember 13, 2001, p. A11.
“The Intifada,” West Bank, December 12, 1987, image taken for Dpa.
Kakutani, Michiko, “Shock, awe and razzmatazz in the sequel,” The New York Times,
March 25, 2003, pp. E1, E5.
Kifner, John, “Good as a gun: when cameras define a war,” The New York Times, November
30, 2003, section 4: pp. 1, 5.
Knightley, Philip, The First Casualty, New York and London: Harcourt Brace and
Jovanovitch, 1975.
Kondrake, Morton, “Another Cambodia,” New Republic, November 3, 1979, pp. 13–14.
Krauthammer, Charles, “Only in their dreams,” Time, December 24, 2001, pp. 60–1.
Lane, Charles, “When is it genocide?” Newsweek, August 17, 1992, p. 27.
Leroux, Charles, “Still images, moving words,” Chicago Tribune, March 27, 2003, section 2:
p. 1.
Lichfield, John, “Single pictures that capture history’s turning points,” The Independent,
April 10, 2003, p. 10.
Lubrano, Alfred, “Even war has rules,” Philadelphia Inquirer, March 30, 2003, section c:
p. 1.
Luckhurst, Tim, “Altered images,” The Independent, March 16, 2004, np.
Lyons, Nathan, Photographers on Photography, Englewood Cliffs: NJ: Prentice-Hall, 1966.
McCoy, Glenn, “Victory flag raising” (political cartoon), Philadelphia Inquirer, April 21,
2003, p. A2.
McCullin, Don, “Notes by a photographer,” in Emile Meijer and Joop Swart (eds), The
Photographic Memory: Press Photography—Twelve Insights, London: Quiller Press, and
the World Press Photo Foundation, 1987, pp. 11–26.
McGeary, Johanna, “Target: Saddam,” Time, April 14, 2003, pp. 51–7.
Marshall, Tyler, “Faced with a TV war, Europeans switch off,” Los Angeles Times, March
19, 1994, p. A16.
Mathews, J.J., Reporting the Wars, Minneapolis: University of Minnesota Press, 1957.
“Moments remembered,” Life, December 26, 1960, pp. 91–102.
133
BARBIE ZELIZER
Moeller, Susan D., Shooting War: Photography and the American Experience of Combat, New
York: Basic Books, 1989.
Nalkur, Sonal, “Images of liberation in the coverage of Afghan women,” unpublished mas-
ters’ thesis, Annenberg School for Communication, 2002.
“A nation at war,” The New York Times, March 27, 2003, p. B12.
“The next Bosnia,” The Economist, June 6, 1992, pp. 43–4.
“Portraits of war: conflict, capitulation,” Philadelphia Inquirer, March 23, 2003, p. A21.
Rich, Frank, “The spoils of war coverage,” The New York Times, April 13, 2003, section 2:
pp. 1, 15.
Robins, Kevin, “The war, the screen, the crazy dog, and poor mankind,” Media, Culture
and Society 15 (1993).
Rohde, David, “Evidence indicates Bosnia massacre,” Christian Science Monitor, August 18,
1995, p. A1.
Rosenstiel, Tom, “War in Iraq: covering the news,” Washingtonpost.com, posted April 10,
2003.
Rosenthal, Joe, “Flag raising atop Mount Suribachi,” image taken on February 23, 1945.
Roskis, Edgar, “Replaying the pictures,” Le monde diplomatique, December 2000.
“Rwanda killers leave a village of the dead,” The New York Times, May 14, 1994, p. C3.
Sagolj, Damir, “US marine doctor with Iraqi girl,” Detroit Free Press, April 11, 2003, p. H1,
image taken for Reuters.
Sekula, Alan, “On the invention of photographic meaning,” in Photography Against the
Grain, 1974; reprint, Halifax: Press of the Nova Scotia College of Arts and Design,
1984.
Smith, W. Eugene, “Baby found in cave,” August 28, 1944, location unknown, image
taken for Life magazine.
“Somalia: the mire,” The Economist, September 11, 1993, pp. 62–3.
Stepan, Peter (ed.) Photos that Changed the World: the 20th Century, Munich, London and
New York: Prestel, 2000.
Swain, Nigel, Hungary: The Rise and Fall of Feasible Socialism, London, 1992.
Takase, Richard M., “Refugee children” (letter to the editor), Los Angeles Times, Novem-
ber 26, 2001, p. A12.
Taylor, John, War Photography: Realism in the British Press, London: Routledge, 1991.
“The Tiananmen in East Timor,” The New York Times, January 21, 1992, p. A20.
van der Deen, David, “Watching for a judgment of real evil,” The New York Times, Novem-
ber 12, 1995.
Voss, F.S., Reporting the War, Washington, DC: Smithsonian, 1994.
Vulliamy, (ed.) Seasons in Hell: Understanding Bosnia’s War, New York: St Martin’s Press,
1994.
Walsh, James, “Unearthing evil,” Time, January 29, 1995, pp. 46–7.
“Weeping women,” The Economist, August 10, 1991, image taken for WHOM.
Whitney, Craig R., “Tunnel vision: watching Iraq, seeing Vietnam,” The New York Times,
November 9, 2003, section 4: p. 1.
Widener, Jeff, “Massacre in Beijing,” June 5, 1989, People’s Republic of China, image
taken as a video still.
Zelizer, Barbie, “CNN, the Gulf War, and journalistic practice,” Journal of Communication
42, winter 1992, pp. 66–81.
Zelizer, Barbie, “Journalists as interpretive communities,” Critical Studies in Mass Commu-
nication, vol. 10, no. 3, 1993, pp. 219–37.
134
W H E N WA R I S R E D U C E D T O A P H O T O G R A P H
Zelizer, Barbie, “Journalism’s ‘last’ stand: wirephoto and the discourse of resistance,” Journal
of Communication, vol. 45, no. 2, 1995, pp. 78–92.
Zelizer, Barbie, Remembering to Forget: Holocaust Memory through the Camera’s Eye,
Chicago, IL: University of Chicago Press, 1998.
Zelizer, Barbie, “Photography, journalism, and trauma,” in Barbie Zelizer and Stuart Allan
(eds), Journalism after September 11, London and New York: Routledge: 2002,
pp. 48–68.
135
7
THE PERSIAN GULF TV WAR
REVISITED
Douglas Kellner
The 1991 war against Iraq was one of the first televised events of the global village
in which the entire world watched a military spectacle unfold via international
satellite and cable networks.1 In retrospect, the Bush administration and the Pen-
tagon carried out one of the most successful public relations campaigns in the
history of modern politics in its use of the media to mobilize support for the war.
The mainstream media in the United States and elsewhere tended to be a compli-
ant vehicle for the government strategy to manipulate the public, imperiling
democracy which requires informed citizens, checks, and balances against exces-
sive government power, and a free and vigorous critical media (see Kellner 1990,
1992, and 2001). Indeed, if the media do not adequately inform citizens, provide a
check against excessive government power and corruption, and adequately debate
the key issues of the day, democracy is undermined.
Moreover, the US media, especially CNN, completely dominated global cover-
age of the event. CNN had cameras and reporters in Baghdad throughout the
war, a large crew in Israel, and live coverage of all US military and government
press conferences. Thus its images, discourses, and material tended to shape
global coverage of the event. This meant that the Bush administration and the
Pentagon were able to control the flow of representations and framing of events,
and thus to manage the global media spectacle of Gulf War I.2
In this chapter, I first discuss the production of the text of the “crisis in the
Gulf” and then “the Gulf War.” This will involve analysis of disinformation and
propaganda campaigns by the Bush administration, the Pentagon, and their
allies, as well as dissection of the constraints produced by the so-called pool
system. I also indicate how the political economy of the media in the United
States facilitated the manufacturing of consent for US government policies.
Then I critically engage the meanings embedded in the text of the war against
Iraq and the reception of the text/event by the audience. The latter process will
take on the question of why the Gulf War was popular with its audiences and how
the Bush administration and the Pentagon marshaled public support for the war.
136
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
137
DOUGLAS KELLNER
138
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
defeat, even by foreign forces, will be greeted as deliverance by his own nation
and by much of the Arab world.” As it turned out, both Iraq and the Arab world
were deeply divided over Hussein and the sweeping generalities that Hoagland
proclaimed were totally off the mark.6
Thus, the Bush administration and Washington Post disinformation and propa-
ganda concerning the Iraqis’ readiness to invade Saudi Arabia worked effectively
to shape media discourse and public perception of the crisis and to legitimate
Bush’s sending US troops to Saudi Arabia. In particular, Patrick Tyler’s front-page
story concerning Hussein’s meeting with Joe Wilson and Iraq’s alleged refusal to
negotiate a solution or leave Kuwait provided the crucial media frame through
which debate over the advisability of sending US troops to Saudi Arabia was dis-
cussed.7 On August 7, in a PBS McNeil–Lehrer discussion of the proper US
response to Iraq’s invasion of Kuwait, co-anchor Judy Woodruff stated: “Iraq’s
leader Saddam Hussein was quoted today [in the Post story—D.K.] as saying the
invasion of Kuwait was irreversible and permanent.” Later on in the same show,
former National Security Adviser (and Iran/Contra felon) Robert McFarlane
quoted the story as evidence that Hussein was not going to leave Kuwait, and that
therefore US military intervention in Saudi Arabia was necessary. And in a dis-
cussion with Arab-American leaders as to whether a US military intervention was
justified, Woodruff interjected: “the US chargé in Baghdad did have a two-hour
meeting with Saddam Hussein yesterday which by all accounts was very unsatis-
factory as Saddam Hussein insisted that he was going to stay in Kuwait and made
what were reported to be veiled threats against other nations in the area”—all dis-
information that Bush administration officials fed to the Post, which was then
disseminated by other mainstream media.
In his early morning television speech on August 8, which announced and
defended sending US troops to Saudi Arabia, Bush claimed that “the Saudi gov-
ernment requested our help, and I responded to that request by ordering US air
and ground forces to deploy to the kingdom of Saudi Arabia.” This was false as
accounts of the Saudi–US negotiations later indicated that the United States
pressured the Saudis to allow US military intervention into their country (Wood-
ward 1991: 241ff; and Salinger and Laurent 1991: 110ff). Bush repeated the
dubious claim that “Iraq has massed an enormous war machine on the Saudi
border,” and his administration emphasized this theme in discussion with the
media, which obediently reproduced the argument. At 9:24 a.m. on August 8, for
instance, Bob Zelnick, ABC’s Pentagon correspondent, dutifully reported that
the Pentagon informed him that Iraqi troop presence had doubled since the inva-
sion of Kuwait, that there were now more than 200,000 Iraqi troops in Kuwait
with a large force poised to invade Saudi Arabia.
Yet it is not at all certain how many troops Iraq actually deployed in Kuwait
during the first weeks of the crisis. All pre-invasion reports produced by the Bush
administration indicated that Iraq had amassed over 100,000 troops on the
border of Kuwait. Initial reports during the first few days after the invasion sug-
gested that Iraq actually had between 80,000 and 100,000 troops in Kuwait, more
139
DOUGLAS KELLNER
than enough for an occupation, as the Bush administration liked to point out and
as the mainstream media diligently reported; once the US forces were on their
way to Saudi Arabia, the Iraqi forces allegedly doubled and reports claimed that
there were at least 100,000 Iraqi troops amassed on the border of Saudi Arabia.
But these figures invariably came from Bush administration or Pentagon sources,
and sources critical of the US claims concerning the number of Iraqi troops
deployed revealed a quite different figure.
St Petersburg Times reporter Jean Heller published two stories (November 30
and January 6) suggesting that satellite photographs indicated far fewer Iraqi
troops in Kuwait than the Bush administration claimed (the January 6 story was
republished in In These Times, February 27, 1991: 1–2). Heller’s suspicions were
roused when she saw a Newsweek “Periscope” item that ABC’s “Prime time live”
had never used several satellite photographs of occupied Kuwait City and south-
ern Kuwait taken in early September. Purchased by ABC from the Soviet
commercial satellite agency Soyez-Karta, the photographs were expected to
reveal the presence of a massive Iraqi troop deployment in Kuwait, but failed to
disclose anything near the number of troops claimed by the Bush administration.
ABC declined to use them and Heller got her newspaper to purchase the satellite
photographs of Kuwait from August 8 and September 13 and of Saudi Arabia
from September 11. Two satellite experts who had formerly worked for the US
government failed to find evidence of the alleged build-up. “‘The Pentagon kept
saying the bad guys were there, but we don’t see anything to indicate an Iraqi
force in Kuwait of even 20 percent the size the administration claimed,’ said Peter
Zimmerman, who served with the US Arms Control and Disarmament Agency
during the Reagan administration” (Heller 1991: 2).
Both satellite photographs taken on August 8 and September 13 showed a sand
cover on the roads, suggesting that there were few Iraqi troops on the Saudi
border where the Bush administration claimed that they were massed, threaten-
ing to invade Saudi Arabia. Pictures of the main Kuwaiti airport showed no Iraqi
planes in sight, though large numbers of US planes were visible in Saudi Arabia.
The Pentagon refused to comment on the satellite photographs, but to sugges-
tions advanced by ABC (which refused to show the photographs) that the
pictures were not high enough quality to detect the Iraqi troops, Heller responded
that the photograph of the north of Saudi Arabia showed all the roads swept
clean of sand and clearly depicted the US troop build-up in the area. By Septem-
ber, the Pentagon was claiming that there were 265,000 Iraqi troops and 2,200
tanks deployed in Kuwait which posed a threat to Saudi Arabia. But the photo-
graphs reveal nowhere near this number and, so far, the US government has
refused to release its satellite photographs.
Interestingly, Bob Woodward (1991) noted that the Saudis had sent scouts
across the border into Kuwait after the Iraqi invasion to see if they could detect
the Iraqi troops that the United States claimed were massed for a possible inva-
sion of their country. “The scouts had come back reporting nothing. There was
no trace of the Iraqi troops heading toward the kingdom” (Woodward 1991:
140
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
258–9). Soon after, the US team arrived with photographs of the Iraqi troops
allegedly massed on the Saudi border, and General Norman Schwarzkopf
explained to the Saudis that the Iraqis had sent small command-and-control
units ahead of the mass of troops, which would explain why the Saudi scouts
failed to see them (Woodward 1991: 268). Former CIA officer Ralph McGehee
told journalist Joel Bleifuss: “There has been no hesitation in the past to use doc-
tored satellite photographs to support the policy position that the US wants
supported” (In These Times, September 19, 1990: 5). Indeed, Emery (1991)
reported that King Hussein of Jordan was also sent pictures of tanks moving along
roads near the Saudi/Kuwaiti border which had been shown to the Saudis, and
that King Hussein claimed that the Saudis had “pressed the panic button” when
they saw the photographs. King Hussein was skeptical and “argued that if Saddam
Hussein had wanted to invade the Saudis, he would have moved immediately,
when the only thing between him and the Saudi capital was a tiny and
untested—if expensively equipped—Saudi army” (Emery 1991: 25).
This was how the disinformation campaign worked to legitimate US deploy-
ment of troops in Saudi Arabia: high Bush administration officials called in
journalists who would serve as conduits for stories that Iraq refused to negotiate a
withdrawal from Kuwait and that they had troops stationed on the borders of
Saudi Arabia, threatening to invade the oil-rich kingdom. The Pentagon and
Bush administrations also released information at press conferences concerning
the Iraqi threat to Saudi Arabia and Iraq’s unwillingness to negotiate, and these
“official” pronouncements supplemented the unofficial briefings of reporters. In
turn, editorial writers and commentators on television networks took up these
claims which they used to bolster arguments concerning why it was necessary for
the US to send troops to Saudi Arabia.
141
DOUGLAS KELLNER
invented Iraqi atrocities in Kuwait, such as the killing of premature babies who
were allegedly taken out of incubators and left to die on the floor. In October
1990 a tearful teenage girl testified to the House of Representatives Human
Rights Caucus that she had seen Iraqi soldiers remove 15 babies from incubators
and leave them to die on the floor of the hospital. The girl’s identity was not
revealed, supposedly to protect her family from reprisals. This story helped mobi-
lize support for US military action, much as Bush’s Willie Horton advertisements
had helped him win the presidency by playing on primal emotions. Bush
mentioned the story six times in one month and eight times in 44 days; Vice-
President Dan Quayle referred to it frequently, as did General Norman
Schwarzkopf and other military spokespersons. Seven US senators cited the story
in speeches supporting the January 12 resolution authorizing war.
In a January 6, 1992, opinion editorial piece in The New York Times, John
MacArthur, the publisher of Harper’s magazine, revealed that the unidentified
congressional witness was the daughter of the Kuwaiti ambassador to the US. The
girl had been brought to Congress by Hill and Knowlton, which had coached her
and helped organize the congressional human rights hearings. In addition, Craig
Fuller, Bush’s former chief of staff when he was Vice-President and a Bush loyalist,
was president of Hill and Knowlton and was involved with the PR campaign, as
were several other former officials for the Reagan administration who had close
relations with the Bush administration.
Thus, the Kuwaiti government developed a propaganda campaign to manipu-
late the American people into accepting the Gulf War and the Bush administration
used this campaign to promote their goals. Hill and Knowlton organized a photo-
graph exhibition of Iraqi atrocities displayed at the United Nations and the US
Congress and this was widely shown on television. They also assisted Kuwaiti refu-
gees in telling stories of torture, lobbied Congress, and prepared video and print
material for the media.
On January 17, 1992, ABC’s “20/20” disclosed that a “doctor,” who testified that
he had “buried fourteen newborn babies that had been taken from their incubators
by the soldiers,” was also lying. The “doctor” was actually a dentist and later admit-
ted to ABC that he had never examined the babies and had no way of knowing
how they had died. The same was true of Amnesty International which published
a report based on this testimony. (Amnesty International later retracted the
report, which had been cited frequently by Bush and other members of his admin-
istration.) ABC also disclosed that Hill and Knowlton had commissioned a “focus
group” survey which brought groups of people together to find out what stirred or
angered them. The focus group responded strongly to the Iraqi baby atrocity sto-
ries, and so Hill and Knowlton featured them in its PR campaigns for the Free
Kuwait group.
In addition to carrying out a massive propaganda campaign, the US govern-
ment also instituted a sustained effort to control information and images through
the pool system (see above) and few of the US broadcasting networks sought out
critical or alternative views. Few significant anti-war voices were heard in the
142
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
mainstream media during the first months of the troop build-up in Saudi Arabia.
A study by the media watchdog group FAIR reported that, during the first five
months of television coverage of the crisis, ABC devoted only 0.7 percent of its
Gulf coverage to opposition to the military build-up. CBS allowed 0.8 percent,
while NBC devoted 1.5 percent, or 13.3 minutes for all stories about protests,
anti-war organizations, conscientious objectors, and religious dissenters. Conse-
quently, of the 2,855 minutes of television coverage of the crisis from August 8 to
January 3, FAIR found that only 29 minutes, or roughly 1 percent, dealt with
popular opposition to the US military intervention in the Gulf.9
The small amount of images of anti-war demonstrators and opinion that
appeared in the mainstream media during the crisis in the Gulf often juxtaposed
anti-American demonstrations in Arab countries that frequently burned US flags
with images of US demonstrations. Such a juxtaposition coded anti-war demon-
strators as Arabs and as irrational opponents of US policies. US demonstrators
were portrayed as an unruly mob, as long-haired outsiders; their discourse was
rarely cited and coverage focused instead on the chanting of slogans, or images of
marching crowds, with media voice-overs supplying the context and interpreta-
tion. Major newspapers and newsmagazines also failed to cover the burgeoning
new anti-war movement. Thus, just as the media symbolically constructed a neg-
ative image in the 1960s of anti-war protesters as irrational, anti-American, and
unruly, so too did the networks present the emerging anti-war movement of the
1990s in predominantly negative frames.
Not only was the discourse of the anti-war movement ignored, but also “none
of the foreign policy experts associated with the peace movement—such as
Edward Said, Noam Chomsky or the scholars of the Institute for Policy Studies—
appeared on any nightly news program” (FAIR press release, January 1991). A
Times-Mirror poll, however, that was recorded in September 1990 and January
1991 discovered “pluralities of the public saying they wished to hear more about
the views of Americans who oppose sending forces to the Gulf” (Special Times-
Mirror News Interest Index, January 31, 1991). Furthermore, soldiers who were
alarmed at their deployment in the Saudi desert and objected to the primitive
living conditions there were silenced, in part by Pentagon restrictions on press
coverage and in part by a press corps unwilling to search for dissenting opinions.
And yet on the eve of the war, more than 50 percent of the American public
opposed a military solution to the crisis. Perhaps images of families being sepa-
rated and young troops being sent to the Saudi desert produced a negative
response to the possibility of a war in the region that could take many US lives.
Quite possibly, despite the lack of critical discourse on the media, many individu-
als could still think for themselves and produce anti-war opinions against the
grain of the dominant pro-military solution government and media discourse.
And maybe the memory of Vietnam and US military misadventures produced
apprehensions over a war in the Persian Gulf. But the disinformation and propa-
ganda campaigns were successful in that they persuaded the majority of nations in
the UN and the US Congress to support a declaration legitimating the use of
143
DOUGLAS KELLNER
force to expel Iraq from Kuwait. And once the war began, the Bush administra-
tion was quickly able to mobilize support for its positions.
144
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
justification for the war were stressed and the narrative was oriented toward a suc-
cessful conclusion that was presented as a stunning victory.
It was obviously in the television networks’ interests to attract the audience to
their programming, and competition revolved around presenting the most patri-
otic, exciting, and compelling coverage. To properly explicate this dimension of
the text of the Gulf War, one needs to focus on the production of the text within
the framework of the political economy of commercial television. First, the
sources of the news on the mainstream media were generally limited to the Bush
administration and the military. This was partly the result of the pool system that
restricted media access to the theater of battle and that exercised censorship over
every image and report filed. Yet the networks themselves also restricted the
range of voices that appeared. A survey by FAIR of the television coverage of the
first two weeks of the war revealed that of the 878 news sources used by the three
major commercial networks, only 1.5 percent were identified as anti-war protest-
ers—roughly equivalent to the amount of people asked to comment on how the
Gulf War disrupted their travel plans. In the 42 nightly news broadcasts, only one
leader of a peace organization was interviewed, while seven Super Bowl players
were asked their views of the war.10
On the other hand, in report after report, television portrayed pro-war rallies,
yellow ribbons, and the wave of patriotism apparently sweeping the country. The
networks also personalized the US troops and their families, thus bonding the
public to the troops in the desert, helping manufacture support for the US mili-
tary policies. In these ways, the audience was mobilized to support every move of
the Bush administration and the Pentagon and, as the war went well and rela-
tively fast, the country was swept along in a victory euphoria, as if it was winning
the Super Bowl of wars and was thus number one in the world. Such imagery and
discourse helped create support for a war that barely 50 percent of the public and
Congress desired on the eve of Bush’s bombing of Baghdad.
Furthermore, the audience was terrorized into support for the US troops by a
series of propaganda campaigns, masterfully orchestrated by the Bush administra-
tion and the Pentagon. Early in the crisis, reports were leaked that Iraqi chemical
weapons were being brought to the field of battle, and throughout the war there
were many reports of the threat of Iraqi chemical weapons. In addition, there were
almost daily reports on the threats of terrorism manipulated by the Iraqis. When
the Iraqis paraded US POWs on television, there were claims that they were tor-
turing coalition troops. Such reports created a mass hysteria in sectors of the
audience, which were positively bonding with the troops. Moreover, after the Iraqi
Scud attacks on Israel and Saudi Arabia, there were reports of thousands of people
buying gas masks and vignettes of families producing sealed rooms in their home in
the case of chemical attack. Obviously, such hysteria helped mobilize people
against the Iraqis and make them desire their military defeat and punishment.
On the whole, television and the mainstream media arguably served as propa-
ganda arms for US government policy. The media endlessly repeated Bush
administration “big lies,” such as its alleged efforts to negotiate a settlement with
145
DOUGLAS KELLNER
the Iraqis when it was actively undermining the possibility of a diplomatic settle-
ment. And the media repeated every propaganda line of the day, amplifying Bush
administration claims concerning alleged torture and mistreatment of US POWs
(later revealed to be highly exaggerated), that an Iraqi infant formula milk fac-
tory destroyed by US bombing was really a military installation producing
chemical/biological weapons, that a civilian sleeping shelter was really a military
command and control center, or that Iraqi “environmental terrorism” was respon-
sible for the Persian Gulf oil spill and other ecological devastation.11
The US mainstream media projected the image of the war most desired by the
Pentagon and the Bush administration; i.e. that it was fighting an eminently
clean and successful hi-tech war. From the beginning, the bombing of Iraq was
portrayed as efficient and humane, targeting only military facilities. Over and
over, despite pictures from Iraq that revealed the contrary, the Pentagon and the
Bush administration stressed the accuracy of their bombing strategies and the oft-
repeated images of the precision bombs, with video cameras built into their
heads, presented an image of pinpoint accurate bombing. Likewise, the frequent
pictures of Patriot missiles apparently knocking out Iraqi Scud missiles created
the impressions of a highly effective hi-tech war. Later, the Pentagon itself admit-
ted that only 7 percent of the bombs used were so-called “smart bombs” and
admitted that over 70 percent of its munitions missed their targets, but the domi-
nant images of a hi-tech war presented an impression of a highly efficient
technowar. It was also later revealed that a large percentage of US casualties
resulted from “friendly fire,” from the bombing of one’s own troops.
The lack of significant critical voices in the mainstream media during the crisis
in the Gulf and then the Gulf War also can be explained by reflection on the
political economy of the media and the system of media production in the United
States. The broadcast media are afraid to go against a perceived popular consen-
sus, to alienate people, and to take unpopular stands because they are afraid of
losing audience shares and thus profits. Because US military actions have charac-
teristically been supported by the majority of the public, at least in their early
stages, television is extremely reluctant to criticize what might turn out to be
popular military actions.12
The broadcast media also characteristically rely on a narrow range of estab-
lished and safe commentators and are not likely to reach out to new and
controversial voices in a period of national crisis. The media generally wait until
a major political figure or established “expert” speaks against a specific policy and
that view gains a certain credibility as marked by opinion polls or publication in
“respected” newspapers or journals. Unfortunately, the crisis of democracy in the
United States is such that the Democratic party has largely supported the conser-
vative policies of the past decades and the party leaders are extremely cautious
and slow to criticize foreign policy actions, especially potentially popular military
actions. The crisis of liberalism is so deep in the US that establishment liberals
are afraid of being called “wimps” or “soft” on foreign aggression, and thus often
support policies that their better instincts should lead them to oppose.
146
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
Warrior nation
The result of the propaganda blitz and war hysteria was a warrior nation that turned
many in the television public into fanatic supporters of the Bush administration war
policy. The pro-war consensus was mobilized through a variety of ways in which the
public identified with the troops. Television presented direct images of the troops to
the public through “desert dispatches” that produced very sympathetic images of
young American men and women “in harm’s way” and serving their country. Tele-
vision news segments on families of the troops also provided mechanisms of
identification, especially because many of the troops were reservists, forced to leave
their jobs and families, making them sympathetic objects of empathy and identifi-
cation for those able to envisage themselves in a similar situation. There were also
frequent television news stories on how church groups, schools, and others adopted
US troops in Saudi Arabia as pen pals, thus more intimately binding those at home
to the soldiers abroad. People were also bound to troops through rituals of display of
yellow ribbons, chanting and waving flags in pro-war demonstrations, and entering
into various pro-war support groups.
The media generated support for the war, first, by upbeat appraisals of US suc-
cesses and then by demonizing the Iraqis that made people fervently want a
coalition victory. Initial support was won for the war effort through the media-
generated euphoria that the war would be over quickly, with a decisive and easy
victory for the US-led coalition. Then, the audience got into the drama of the
war through experiencing the excitement of the Scud wars and the thrills of
technowar war with its laser-guided bombs and missiles and videotapes of its suc-
cesses. The POW issue, the oil spills and fires, and intense propaganda campaigns
by both sides also involved the audience in the highly emotional experience of a
television war. The drama of the war was genuinely exciting and the public
immersed itself in the sights, sounds, and language of war.
The media images of the hi-tech precision bombing, (seeming) victories of
Patriot over Scud missiles, bombing of Iraq, and military hardware and troops
helped to mobilize positive feelings for the US military effort in much of the audi-
ence. Military language helped normalize the war, propaganda and disinformation
campaigns mobilized pro-war discourse, and the negative images and discourses
against the Iraqis helped mobilize hatred against Iraq and Saddam Hussein. Polls
during the first weeks of the war revealed growing support for the war effort, dis-
playing a widespread propensity to believe whatever the media and military were
saying. A Times-Mirror survey of January 31, 1991, indicated that 78 percent of
the public believed that the military was basically telling the truth, not hiding
anything embarrassing about its conduct of the war, and providing all of the
information it prudently could. Seventy-two percent called the press coverage
objective and 61 percent called it for the most part accurate. Eight out of ten said
the press did an excellent job. Fifty percent claimed to be addicted to television
watching and said that they could not stop watching coverage of the war. Fifty-
eight percent of adults under 30 called themselves “war news addicts” and
147
DOUGLAS KELLNER
21 percent of these “addicts” said they had trouble concentrating on their jobs or
normal activities, while 18 percent said they were suffering from insomnia.
It was, however, the total media and social environment that was most respon-
sible for mobilizing support for the US war policies. From morning to evening,
the nation was bombarded with images of military experts, vignettes of soldiers at
home and abroad, military families, former POWs, and others associated with the
military. Military figures, images, and discourse dominated the morning talk
shows, the network news, discussion programs, and the 24-hour-a-day CNN
war coverage, as well as saturation coverage on C-Span and many other cable
networks. On home satellite dishes, the channels were saturated with live trans-
missions concerning the war, as the networks prepared or presented their reports
from the field, and one satellite transponder provided hours per day of live mili-
tary pool footage from Saudi Arabia for use by the networks—propaganda
provided by the military free of charge. Television news pre-empted regular pro-
grams for weeks. The result was a militarization of consciousness and an
environment dominated by military images and discourses.
148
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
149
DOUGLAS KELLNER
Moreover, the fact that Saddam Hussein continued to rule with an iron fist in
Iraq and that his neighbors continued to feel threatened after the Gulf War raised
questions as to the success of Bush senior’s policy and whether the war really
accomplished any significant long-term goals, other than temporarily boosting
Bush’s ratings in the polls and producing a positive image of the US military after
the shame of defeat in Vietnam. In the chaotic aftermath of the Gulf War in
which the US failed to intervene to helped the uprising against Hussein’s regime
and in the light of his reassertion of his power and domination, the extreme
hyperbole of the construction of Saddam Hussein and his regime as absolute evil
to some extent backfired since the Evil One was not removed from power in the
aftermath of the war. Although Bush urged the Iraqis to overthrow Hussein, once
the US declared an end to the fighting and Iraqi rebels rebelled against Hussein’s
regime, the US remained on the sidelines. General Schwarzkopf himself stated in
a PBS television interview on March 27, 1991, that he had preferred to continue
fighting to “annihilate” completely the Iraqi military which was violently sup-
pressing the insurgent forces against Hussein as Schwarzkopf spoke. The
continuation of Saddam Hussein in power, the destructive environmental effects
of the war, the so-called Gulf War syndrome that has disabled over 160,0000 US
troops, the need for Bush junior to “finish the job,” and the current chaos and
instability of Iraq reveal Gulf War I to have been in retrospect a Pandora’s box of
evils that produced a brief euphoric high with a long hangover.15
Consequently, saturation television coverage of dramatic political events is a
two-edged sword: it might shape public opinion into supporting the US interven-
tion, as it obviously did during the Gulf War. But repeated images of a drawn-out
stalemate, or images of death and destruction in a fighting war, or representations
of protracted suffering as long-term effects of the war, could be turned against the
political group and its leaders who produced such destruction. The very ubiquity
of the broadcasting media and now the Internet, and the central role of the media
in contemporary politics, renders media spectacle a complex and unpredictable
political force (Kellner 2003a). Lust for pictures to attract audiences led the net-
works into a race to get into Iraq and to interview its leaders and to show its
people. Images of continual and increased suffering of the Iraqi people and others
150
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
in the area as a result of US military interventions in the 1991 and 2003 Iraq wars
have helped produce negative images of the US throughout the world and might
ultimately lead people to see that war is no way to solve political conflict, and
that it produces overwhelming destruction, suffering, and death.
Hence, a multiperspectival approach that captures different aspects of a com-
plex phenomenon like mainstream media coverage of US interventions in the
Middle East should also analyze the contradictions of audience reception of the
media texts and potentially contradictory images and effects, as well as analyzing
media texts and their conservative, systems-maintenance functions. Although
my analysis has focused primarily on the ways that US media coverage of the
1991 Gulf War supported the policies of the Bush administration and the Penta-
gon, analysis of the reception by audiences of the multiple Middle East crises, the
Gulf War, and its aftermath might have ultimately helped undermine Bush senior
and the conservative hegemony of the Reagan/Bush senior years, contributing to
the latter’s defeat in 1992. Perhaps Bush senior went overboard in demonizing
Hussein, and his regime’s continued rule of Iraq served to rob Bush senior of
claims of genuine victory. Likewise, although Bush junior’s administration
appeared to have won a smashing military triumph in the 2003 war against Iraq,
its consequences and aftermath could undo the regime of the son, as the contra-
dictory aftermath of Gulf War I helped to undo the presidency of the father (see
Kellner 2004).
In any case, the effects of television and the mainstream media are contradic-
tory and may have unintended consequences. While in the spring of 1991, the
Gulf War constituted a tremendous victory for the Bush administration and the
Pentagon, the event did not save Bush senior’s presidency and eventually raised
questions as to whether he was really an effective president. Its short-term effects
in temporarily boosting Bush senior’s popularity and the sudden shift in public
opinion concerning the war and Bush also point to the fickleness of audiences in a
media-saturated society, who soon forget the big events of the previous year.16
And yet the woefully one-sided coverage of the Gulf crisis and war by the
mainstream media calls attention once again to the need for alternative media to
provide essential information on complex events like the Gulf War. During the
war, those of us who opposed it got information from computer databases, such as
PeaceNet, or progressive publications like The Nation, In These Times, The Village
Voice, and Z Magazine. Locally, in addition to holding teach-ins at universities,
critics of the war attempted to make use of public access television and radio to
criticize the Bush administration’s war policy and refusal to negotiate a diplo-
matic solution. Democratizing our media system will require a revitalization of
public television, an increased role for public access television, the eventual
development of a public satellite system, and the production of progressive com-
puter databases, websites, list-serves, and weblogs (Kellner 1990 and 1995).
Because politics are more and more acted out on media screens, without the
reconstruction of television and the mass media, the prospects for democratiza-
tion of US society and polity are dim.
151
DOUGLAS KELLNER
Notes
1 In this chapter, I draw on my book The Persian Gulf TV War (Kellner 1992), and later
studies of the event in Kellner 1995 and Best and Kellner 2001.
2 Following the publication of my Persian Gulf TV War (Kellner 1992), I was invited all
over the world to lecture on the event and interviewed officials from British, French,
Belgium, German, Finnish, Swedish, and many other state television networks; all of
the people I interviewed affirmed that CNN tended to control the media flow and
frames in depiction of the war. Things were quite different, however, in the 2003 war
on Iraq that had many Arab television networks present in Iraq, as well as many Euro-
pean and other global networks that framed the event quite differently from the US
networks.
3 By the “mainstream media” in the United States, I mean the major national television
networks, including ABC, CBS, CNN, NBC, and CBS; the national weekly news
magazines Time, Newsweek and US News and World Report; and national newspapers
such as The New York Times, The Wall Street Journal, USA Today, and The Washington
Post. These mainstream media are corporate media, owned by an ever-shrinking and
more powerful number of transnational corporations. See the contrast between main-
stream and alternative media that I develop in Kellner 1990 and the analysis of
corporate media in Kellner 2003a.
4 Following the example of British censorship of the press during the Falkland Islands/
Malvinas war, the US tightly controlled press access during the Grenada foray and
instituted the pool system during the Panama invasion; for detailed analysis of how it
worked during the Gulf War, see Kellner 1992: pp. 80ff.
5 See the discussion below and Kellner 1992, pp. 17ff for documentation of this claim.
6 Hoagland’s remarkable misreading of Iraq was duplicated by the Bush administration’s
neo-conservatives who claimed prior to the US/UK war against Iraq in 2003 that Iraqis
would welcome their “liberators” with open arms and that the US would easily be able
to reconstruct the country and lead it to democracy (see Kellner 2004).
7 Through computer data base searches, I discovered how this story was taken up by the
television networks and most major newspapers, and was used in many later sum-
maries of the story to explain why Bush had to send US troops to Saudi Arabia; see the
documentation in Kellner 1992.
8 A study undertaken by the Gannett foundation indicated that there were over 1,170
articles linking Hussein with Hitler (LaMay et al. 1991, p. 42). This comparison obvi-
ously presupposes a false analogy in terms of the military threat to the region and the
world from the Iraqi army—whose threat was hyped up from the beginning. Iraq’s 17
million population can hardly compare with Germany’s 70 million and its military was
significantly less threatening than Hitler’s military machine, which was the most pow-
erful in the world in the 1930s. Nor could Iraq, which depends on oil for over 95
percent of its exports, be compared with an industrial powerhouse like Germany. It is
also inappropriate to compare a major imperialist superpower with a regional power,
Iraq, that itself is the product of colonization.
It might also be noted how the Bush administration and media personalized the
crisis, equating Iraq with its leader. Whereas in coverage of the 8-year war between
Iran and Iraq, in which the US covertly supported Iraq, references were to “Baghdad”
and “Iraq,” during the Gulf crisis and war it was usually “Saddam Hussein” who was
referred to as the actor and source of all evil (I am grateful to Richard Keeble for this
insight.)
9 FAIR, press release, January 1991.
10 Cited in Joel Bleifuss, In These Times, March 20, 1991, p. 5.
11 In fact, allied bombing was also responsible for much of the ecological damage; see the
documentation in Kellner 1992 and Clark 1992.
152
T H E P E R S I A N G U L F T V WA R R E V I S I T E D
12 There was also organized right-wing pressure against networks or mainstream publica-
tions that criticized the Bush administration or the Pentagon; such pressure was
increased in the Afghanistan and 2003 Iraq war, so that during these events the US
broadcasting media were even less critical of the second Bush administration and Pen-
tagon (see Kellner 2003b).
13 Muller documented that within a month and a half after the Iraqi surrender only a
small majority of 52 percent believed that the US intervention was largely or totally
successful and that, by June 1991, the number fell by another 8 percent (1994, p. 277).
Other polls which tracked public opinion between the end of the war and summer
1991 showed declines of 14 percent and 19 percent in support for the proposition that
the war had been worth fighting (Mueller 1993, p. 214).
14 See Murray Waas, Los Angeles Times (May 7, 1992). Another Los Angeles Times story
by Murray Waas and Douglas Frantz, “Bush tied to ’86 bid to give Iraq military
advice,” described how: “As Vice President during the Ronald Reagan administration,
George Bush acted as an intermediary in sending strategic military advice to Iraqi
President Saddam Hussein at a critical point in the Iran–Iraq war, according to sources
and classified documents.” A recent book by Chambers Johnson The Sorrows of Empire
(2004: 223ff) documents how the CIA helped install Saddam Hussein into power,
provided military assistance to Iraq during the 1980s war with Iran, and provided loans
and material that enabled him to develop his weapons programs, including chemical
and biological weapons.
15 On the Gulf War syndrome, see Hersh 1998. See also Phil Hirschkorn and Richard
Roth, “Gulf War veterans suing companies for chemical exports” (CNN, January 17,
2003) who report: “About 209,000 Gulf War veterans have filed claims with the Vet-
erans administration, and 161,000 of them are receiving disability payments.”
16 David Halberstam claimed that the Gulf War “was a war without real resonance . . .
when it was over, it was over, leaving remarkably little trace” (2001, pp. 16). In some
ways, this may be true domestically, but it helped create resentment of US military
power that had resonance in enraging Jihadist forces who have targeted the US with
terrorism attacks, and it also failed to eliminate the regime of Saddam Hussein,
tempting the second Bush administration into what is now appearing as a major
disaster.
References
Best, Steven and Kellner, Douglas (2001) The Postmodern Adventure, London and New
York: Routledge and Guilford Press.
Chomsky, Noam (1989) Necessary Illusions, Boston: South End Press.
Clark, Ramsey (1992) War Crimes: A Report on United States War Crimes Against Iraq,
Washington: Maisonneuve.
Emery, Michael (1991) “How Mr Bush got his war: deceptions, double-standards and dis-
information,” Westfield, NJ: Open Magazine pamphlet series; originally published in
the Village Voice, March 5, 1991, pp. 22–7.
Eveland, W., McLeod, E., and Signorielli, N. (1994) “Conflict and public opinion: rallying
effects in the Gulf War,” Journalism Quarterly, 1994, vol. 71, no. 1, pp. 20–31.
Friedman, Alan (1993) Spider’s Web, New York: Bantam Books.
Halberstam, David (2001) War in a Time of Peace: Bush, Clinton, and the Generals, New
York: Scribner.
Heller, Jean (1991) In These Times, February 27: 1–2.
Herman, Edward and Chomsky, Noam (1988) Manufacturing Consent, New York: Pantheon.
153
DOUGLAS KELLNER
Hersh, Seymour (1998) Against All Enemies: Gulf War Syndrome: the War between Amer-
ica’s Ailing Veterans and Their Government, New York: Simon and Schuster.
Johnson, Chalmers (2004) The Sorrows of Empire, New York: Metropolitan Books.
Kellner, Douglas (1990) Television and the Crisis of Democracy Boulder, CO: Westview.
Kellner, Douglas (1992) The Persian Gulf TV War, Boulder, CO: Westview.
Kellner, Douglas (1995) Media Culture, London and New York: Routledge.
Kellner, Douglas (2001) Grand Theft 2000, Boulder, Co.: Rowman and Littlefield.
Kellner, Douglas (2003a) Media Spectacle, London and New York: Routledge.
Kellner, Douglas (2003b) From 9/11 to Terror War: Dangers of the Bush Legacy. Lanham,
MD.: Rowman and Littlefield.
Kellner, Douglas (2004) “9/11, Spectacles of terror, and media manipulation: a critique of
Jihadist and Bush media politics” in Critical Discourse Studies, vol. 1, no. 1.
LaMay, Craig, et al. (1991) The Media at War, New York: Gannett Foundation Media
Center.
Mueller, J. (1993) “American public opinion and the Gulf War” in S.A. Renshon (ed.),
The Political Psychology of the Gulf War: Leaders, Politics and the Process of Conflict, Pitts-
burgh, PA: University of Pittsburgh Press, pp. 199–226.
Mueller, J. (1994) Policy and Opinion in the Gulf War, Chicago, IL: University of Chicago
Press.
Rogin, Michael (1987) Ronald Reagan: the Movie, Berkeley, CA: University of California
Press.
Salinger, Pierre, and Laurent, Eric (1991) Secret Dossier: the hidden agenda behind the Gulf
War, New York: Penguin Books.
Solop, F.I. and Wonders, N.A. (1991) “Reaction to the Persian Gulf crisis: gender, race,
and generational differences,” paper presented at American Association of Public
Opinion Research.
Woodward, Bob (1991) The Commanders, New York: Simon and Schuster.
154
8
TRIBALISM AND TRIBULATION
Media constructions of “African savagery” and
“Western humanitarianism” in the 1990s
Susan L. Carruthers
155
SUSAN L. CARRUTHERS
has chosen to wage its pre-emptive and retaliatory “war on terror.” In recent
months, the failed UNOSOM mission has haunted accounts not only of how
Rwandan lives were subsequently squandered but also how America rendered itself
vulnerable through its undignified scurry from Somalia (Cain 2003). Refusing to
sustain even limited casualties in the interests of nation-building in Africa, Amer-
ica found itself losing over 2,000 lives on home soil. These two phenomena were
not unconnected, or so the new received wisdom goes—one fount of which may
be traced to Black Hawk Down, Mark Bowden’s bestseller from which Ridley Scott
dramatically reconstructed Hollywood’s “Battle of Mog” (Bowden 2002). Where
the book intimates, the movie more boldly underscores chains of connection
between the horn of Africa and the Twin Towers, between Aidid and bin Laden,
implying that the premature US pull-out inspired those seeking to spook a timor-
ous Uncle Sam (Bowden 2002: 428; Carruthers 2003: 178–9).
This lesson in the perils of pusillanimity now underpins the Bush administra-
tion’s public account of how terrorists are emboldened—not, as some critics of
US global policy might have it, by the arrogance of American power wielded
overwhelmingly and unilaterally, but rather by the sole superpower’s squeamish-
ness when it comes to incurring casualties. Hence cut-and-run must be replaced
with “shock and awe.” Reading from a script that might have been ghosted by
Bowden, President Bush bolstered his September 7, 2003, call for $87 billion to
prosecute the “war on terror” by summoning the spectre of another scuttle—this
time from Baghdad rather than Mogadishu: “We have learned that terrorist
attacks are not caused by the use of strength. They are invited by the perception
of weakness. And the surest way to avoid attacks on our own people is to engage
the enemy where he lives and plans” (Bush 2003). Present in all but name, Soma-
lia is clearly the case in point as Bush seeks to animate American energies for
combating another “problem from hell.” Rwanda and Somalia are not, then, fin-
ished stories safely laid to rest in the files and photographic archives—to the
extent that the past ever is beyond retrieval by hungry hunters for serviceable
lessons or salutary warnings.
This chapter concentrates on journalism from the first half of the 1990s. More
particularly, it examines appraisals of media coverage offered from a range of acad-
emic and other vantage points (Livingston and Eachus 1995; Murison 1996; Wall
1997; Philo et al. 1998; Livingston and Eachus 1999). These critiques fall into
two broad camps: one focused on the conditions under which the West “inter-
venes”; the other on the inadequacies of what the media offers as knowledge
about Africa. In this division of labor (somewhat schematically overdrawn here),
media studies, NGO activists, and international relations scholars tend to
concentrate on Western policy processes and outcomes, while Africanists,
anthropologists, and political geographers deconstruct the poverty of media
imagery and analysis of complex crises. Juxtaposing these two distinct scholarly
mappings of the terrain, I propose an alternative, more bi-focal, lens through
which to view media representations of Africa and their wider significance for
identity politics and practical policy alike.
156
T R I B A L I S M A N D T R I B U L AT I O N
157
SUSAN L. CARRUTHERS
158
T R I B A L I S M A N D T R I B U L AT I O N
vision images’ denial of space “for what have traditionally been regarded as the
responsible deliberative organs of our government, in both the executive and leg-
islative branches” (Kennan 1993). His criticism explicitly rejected any notion
that this development might be technologically determined: a function of short-
ened news-cycles and real-time footage that are the necessary stocks-in-trade of
global news channels. Rather, Kennan and others (including the former UK
Conservative Foreign Secretary Douglas Hurd) have lambasted what they see as
declining journalistic abidance by professional codes of objectivity and neutral-
ism in favor of a crusading approach, simultaneously adversarial and advocatory: a
lamentable “do something” journalism that has flourished as post-Cold War con-
flicts—in which the West ostensibly has no particular stake—have proliferated
(Seib 1998: 44–5).
The critique of media “power without responsibility” is well-worn, dusting
down old charges against American journalists in Vietnam of abandoning objec-
tivity, espousing anti-war partisanship, and showing rather too much graphic
footage of what modern weapons do to human flesh to permit the war’s sagacious
managers to prosecute it to a successful conclusion (Hallin 1989). While many
reporters (Halberstam 1972; Arnett 1995; and others) have disputed the accusa-
tion of back-stabbing, some nevertheless endorse the notion that unvarnished
reportage did indeed play a determining role in undermining popular support for
the war in Vietnam—but by “showing it as it was” rather than through actively
oppositional reporting (Carruthers 2000: 112). In a similar fashion, some of those
who have found themselves the more recent target of attacks on “do something”
journalism have been keen to corroborate a CNN effect of sorts, while investing
it with humanitarian potential and hence with a different moral valence than
those who identify a “curve” they are keen to decry. Former BBC reporter and MP
Martin Bell, for instance, has made a number of interventions on behalf of a
“journalism of attachment” that takes specific aim at Hurd’s revulsion for “real-
time” television: “The mandarins’ objection is not just to the power but to the
impertinence of the upstart medium, which challenges their monopoly of
wisdom, and rushes in where the pinstripes fear to tread” (Bell 1995: 137–8,
1997, 1998).
A more laudatory reading of the CNN effect, in at least its initial catalytic
phase, unites such journalists with many advocates of “humanitarian inter-
vention”: a second group (including both non-governmental organization
workers and international relations scholars) that has been attentive to media
reporting of “distant crises” (Rotberg and Weiss 1996; Shaw 1996; Ignatieff 1998;
Wheeler 2000). Here again, the specificities of conflict in Africa—the character
and consequences of UN interventions—have often been less salient than broad
claims about the emergence of a “global civil society” that mitigates the reluc-
tance of states to contravene the principle of sovereign inviolability when
confronted with instances of grave humanitarian disaster in foreign locales
(Clarke 2002: 94–5; Shaw 1996). For some, television is to be congratulated for
constituting “macropublics of hundreds of millions of citizens . . . nurturing public
159
SUSAN L. CARRUTHERS
controversies beyond the boundaries of the nation-state” (J. Keane 1996: 172–3).
Both the immediacy and indelibility of images of human suffering lead to a con-
traction of sympathies confined to local kinship communities, while expanding
our capacity for empathy with “distant suffering” wherever it occurs. According
to civil society optimists, such as John Keane, however questionable some forms
of media fascination with violence may be, television’s focus on human distress
nevertheless “contains a hidden, potentially civilizing dialectic” (1996: 172–3,
182–3; Boltanski 1999).
Those who credit television with enhancing Westerners’ eagerness for “saving
strangers” find both encouraging and dispiriting evidence in Africa (Wheeler
2000). To the extent that Operation “Restore Hope” was animated by this new
cross-border solidarity with human suffering, and insofar as the benevolence of its
intentions is accepted, then Somalia may offer some cause for optimism. But the
disastrously conflicted character of the mission, which quickly “crept” from a mil-
itarized food assistance operation to a manhunt for General Aidid in which all
Somalis were treated as potential enemies, squeezed public enthusiasm for the
deployment in the US and beyond (Dowden 1995: 93; Stech 1994). Hence wea-
ried or confused would-be humanitarians were all too susceptible to images of
“their” casualties being dragged through the streets of Mogadishu, seizing upon
the shocking images aired in early October 2003 as decisive evidence that it was
time for US troops to depart. Moreover, the disaffection engendered by a mission
that palpably destroyed hope directly contributed to US failure to intervene in
order to prevent a thoroughly foreseeable genocide in Rwanda in April/May
1994. After the stunning events of October 2003, purportedly the “worst day” of
Clinton’s life (at least pre-Lewinsky), post-Vietnam aversion to “quagmires” was
compounded by fear of another “Mog” (Gowing 1994: 67; Dauber 2001). The
most immediate manifestation of this hybrid “Vietmalia syndrome” was Presiden-
tial Decision Directive no. 25, which established restrictive criteria governing,
and indeed tightly circumscribing, any future US intervention overseas (Shat-
tuck 1996: 173). Accordingly, when Rwanda’s genocide began in April
1994—after months of well-documented preparatory propaganda and planning—
Clinton’s response involved a scrupulous avoidance of the descriptor “genocide”
(which would have mandated an international response), and a steadfast desire to
station US troops as far from the site of slaughter as possible (Destexhe 1995;
Prunier 1997).
During the genocide’s initial phase, the UN’s chief concern was to rescue its
own (predominantly white) personnel, and effect as speedy an exit from Rwanda
as possible. Many journalists—to the extent that they had hitherto populated the
“boringly” placid country—followed suit (Hilsum 1995). Six weeks of killing of
the most intimate face-to-face variety, went largely unreported, or certainly
barely televised, by Western media (Lorch 1995; Livingston and Eachus 1999).
Live satellite broadcasting facilities were not established in the border town of
Kigali until late May (Minear, Scott, and Weiss 1996: 64). By then the story had
moved on—or at any rate a second story had emerged which became confused in
160
T R I B A L I S M A N D T R I B U L AT I O N
much press reporting with the insufficiently told story of genocide: namely, a mas-
sive exodus of refugees, fleeing the anarchic compound of genocide and civil war,
as the Rwandan Patriotic Front advanced into the country from Ugandan exile to
replace the government of assassinated President Habyarimana. Latching onto a
now-familiar icon of wounded innocence—the refugee—the international media
began to cover this story intensively, focusing on the relief operations of various
NGOs in establishing a city-sized encampment at Goma, and the onset of a
deadly cholera epidemic among its rapidly swelling population (Rosenblatt
1996). At this point, Washington announced a contribution to the UN’s relief
effort, arguably impelled by images of humanity in extremis that cried out for
action (Minear, Scott, and Weiss 1996: 65).
Given this dilatory response on the part of both media and policy-makers who
preferred to extract their bodies from harm’s way and avert their gaze from geno-
cide, advocates of “intervention” question how television’s affective properties
can be enhanced such that quickly animated sympathies and salvific impulses do
not wither prematurely, or collapse into self-protectionism at the first sign of
humanitarianism claiming “our” lives while trying to save “theirs” (Girardet
1995; Rotberg and Weiss 1996). Where some take issue with the calibre of West-
ern reporting—its ready recourse to visual and verbal cliché—others critique the
patchiness as well as the paucity of coverage. While certain “emergencies”
become constituted as media spectacles (as the Rwandan post-genocide relief
operation did) others, sometimes just as costly to human life, unfold unobserved.
Why, for example, was a devastating famine in Sudan during 1992, which claimed
more lives than its counterpart in Somalia during the same period, almost wholly
ignored (Livingston 1996; Natsios 1996)? Why is media attention to distant dis-
tress so selective and so short-lived? Do news organizations wrongly presume
indifference, apathy, or “compassion fatigue” on the part of their consumers
(Moeller 1999)? How far do, or should, humanitarians’ and journalists’ agendas
converge?
Communications scholars share an interest in at least some of these same ques-
tions, albeit with a different—and less practical—stake in the answers. Media
research has long concerned itself (in a certain branch, at least) with mapping
the multivalent flows of influence between media, state, and audience; and with
questions of how news is selected, filtered, and framed. Over the past decade, sev-
eral scholars have devoted themselves to a more systematic analysis of the CNN
effect than the frequently impressionistic formulations of those often holding an
immediate, personal investment in asserting that new(s) media either do, or
emphatically do not, lead policy-making. In order to dissect the purportedly
“pushing” and “pulling” properties of real-time news, analysts have not only scru-
tinized apparent instances of the effect’s efficacy (Somalia, in particular, but also
Rwanda, former Yugoslavia and the 1991 Kurdish crisis) but have also examined
why CNN fails to cover many similarly “meretricious” instances of grave human-
itarian catastrophe around the world. In so doing, scholars have sought to dispel
an often implicit assumption of CNN-effect proponents: that the news media
161
SUSAN L. CARRUTHERS
162
T R I B A L I S M A N D T R I B U L AT I O N
163
SUSAN L. CARRUTHERS
If Somalia forms “exhibit A” in the CNN effect debate then Rwanda provides
a more pressing point of concern for Africanists anxious to deconstruct the
impoverishment of Western news analysis. Nowhere did the ready resort to
“ancient ethnic hatreds” more powerfully pre-script coverage than in Rwanda.
The genocide of 1994 was reduced to a simple tale of Hutu slaughtering their
Tutsi neighbors: the latest iteration of an ongoing cycle of bloodletting since time
immemorial (Murison 1996; Myers, Klak, and Koehl 1996; Wall 1997; McNulty
1999). This prevailing narrative assigned clear-cut moral and ethnic identities to
Tutsi victims and Hutu perpetrators. Violence may have been reported as “black-
on-black” but media-made morality was etched in black and white: good Tutsi,
evil Hutu. Paradoxically, however, in their eagerness to maintain these categories
and to affirm the purity of refugee victimhood, many journalists reporting the
post-genocide exodus into Tanzania and Zaire failed to appreciate that those fill-
ing the camps around Rwanda’s borders included several thousand who had
enacted the genocide—not, as reporting often implied, an indistinguishable mass
of “innocents” fleeing in fright (de Waal 1994: 25; F. Keane 1996: 186). Concen-
trating on the “big story” of Western humanitarianism, the selfless altruism of
relief workers, and the spareness of human suffering at its most raw and seemingly
undifferentiated, much reportage failed to note how the camps became havens for
the exiled interahamwe militia, which would (in time) form a serious challenge to
the new RPF regime (de Waal 1998; Fair and Parks 2001). Rather, refugees were
required to serve as mute totems: “Good people to whom bad things happen,” as
one journalist later characterized the role (Minear, Scott, and Weiss 1996: 64).
As a corrective to such ill-informed, inaccurate or downright tendentious
accounts, many Africanists have sought to dismantle the attribution of violence
to “ethnic hatreds”—whether with reference to Rwanda’s genocide, the anarchic
conditions of Somalia, the marauding gangs of machete-wielding child soldiers in
Liberia, Sierra Leone, or other locations in what Madeleine Albright dismissively
denoted as the “hopeless continent.” Such scholars (and advocates of a more self-
reflective humanitarianism) stress the constructedness of social identities in
Africa, as elsewhere, together with the complex local and global determinants of
violence that give the lie to any suggestion that tribalism underlies African crises.
Some question, with regard to Rwanda’s genocide, whether Hutu and Tutsi con-
stitute distinct ethnicities at all (Mamdani 1996; McNulty 1999). All, however,
share an understanding of ethnicity’s potency as a manipulable resource—a rich
and renewable seam of energy to be mined—in struggles over power and privi-
lege, representation, and resources. In the words of Johan Pottier:
164
T R I B A L I S M A N D T R I B U L AT I O N
extremists killed Rwanda’s Tutsi and sent their bodies “back to Ethiopia”
via the Nyabarongo and Akagera rivers. The imagined origins of “the
Tutsi”, along with their (poorly understood) migrations and conquest of
Rwanda, were evoked by power-crazed Hutu politicians to instill “ethnic
hatred” in the very people they themselves oppressed: the victims of
oppression were spurred on to kill a minority group which the oppressors
had labelled the “real enemy”.
(Pottier 2002: 9)
The point here is more profound than that an intense barrage of hate propa-
ganda—transmitted largely by radio—primed Rwanda’s genocide, turning
neighbours into killers (Chrétien 1995; F. Keane 1996; Kirschke 1996; Des Forges
1999). While analyses of radio’s role may say something significant about how
genocide is organized, and how dispersed the locus of responsibility (with broad-
casters included among those subsequently indicted for crimes against humanity at
Arusha), they do not necessarily suffice to dislodge “ethnic hatred” from its
explanatory pedestal. What is required, Pottier and others suggest, is an account
that does more than substitute “manufactured” hatreds for “authentic” (and sup-
posedly ancient) ones, leaving matters at that (Appadurai 1998). As Jan
Nederveen Pieterse puts it “Ethnicity, although generally considered a cause of
conflict, is not an explanation but rather that which is to be explained. The termi-
nology of ethnicity is part of the conflict and cannot serve as the language of
analysis” (1997: 71).
What constitutes sufficient explanation—and which interests are inflected
through the idiom of ethnicity—will be returned to later. Since most media
deliver only stories saturated in essentialism, several commentators probe the roots
of this explanatory impoverishment. Why do Western journalists so routinely “get
Africa wrong”? Should this be put down to narrowly professional shortcomings
within news organizations or ascribed to media’s broader ideological situatedness?
For some critics, the answer lies primarily in the long-term erosion of foreign news
services, which has hit Africa particularly severely since agency staffers and perma-
nent correspondents were already so over-stretched and thinly spread. As a result
of this depletion of accumulated knowledge, more news organizations rely—at
least for crisis coverage—on “parachute” journalism: star reporters simply airlifted
into and out of the latest location of humanitarian disaster (Pedelty 1995). Wholly
ignorant of local conditions, and harried by over-abundant deadlines (with rolling
news increasing journalists’ on-camera airtime and correspondingly diminishing
their opportunities for off-camera investigation), these parachutists predictably
plunder a stock of well-worn clichés, stereotypes, and pre-scripted storylines: of
African tribalism, implacable enmities, unspeakable evil, maniacs with machetes,
and benefactors in blue berets.
In journalists’ telling, this reliance on certain easily grasped conventions arises
less because it conserves their time than because it fails to make a drain on ours:
with Western audiences conceived as intolerant of complexity, especially in
165
SUSAN L. CARRUTHERS
places of which they know, and care, rather little (Behr 1992). Hence CNN’s
insistence that Rwanda’s genocide could not have been covered while it unfolded
in April/May 1994, as that would have required simultaneous coverage alongside
Nelson Mandela’s inauguration as South African premier. Rather, it had to
appear sequentially—after the massed press corps moved from Johannesburg to
Kigali en route “home” (without which coincidence, by implication, the refugee
situation might have received considerably less attention). Why not both simul-
taneously? “[S]howing two African topics at the same time . . . might confuse
their audience”: viewers presumed incapable of grasping that “good news” from
Africa might unfold synchronously with something conforming to the more
familiar catastrophic mold (Gassman 1995: 157). Similarly, journalists are often
wont to attribute their eagerness to establish clear-cut points of identification for
sympathy and targets for blame to a deficiency in viewers’ ability to grasp com-
plexity, or a deficit in public patience to fathom it. Circumstances not being
obligingly simple but attention-spans being so truncated, journalists must neces-
sarily reach for stock characterizations. Hence the concentration in
post-genocide Rwanda on the figure of the refugee: a late twentieth-century icon
of innocence—as long, of course, as refugees remain encamped at Goma and not
Sangatte (“over there” and not, like the fin de siècle hate figure—the “bogus
asylum seeker”—making more pressing claims on/of us “over here”).
According to Pottier, journalists’ ignorance also renders them susceptible to
manipulation by African elites eager to promote their own particularist versions
of contemporary events, together with an appropriately (re)configured account of
history (2002). But perhaps this “tainting” is a less significant phenomenon
(since rather few journalists seriously trouble to seek out indigenous sources, and
some insist on their skepticism toward “native informants” (Meisler 1992)) than
journalists’ embeddedness within particular cultural, political, and economic
matrices which endlessly replenish the reservoir from which easy “ethnic” expla-
nations are retrieved, while filtering out alternative accounts. After all,
journalists who reflect retrospectively on their African days tend not to do much
better a job of explaining the deep roots of the continent’s crises, however much
remission from deadlines they may enjoy for rumination. Hence in journalistic
memoirs such as Scott Peterson’s Me Against My Brother, “Heart of Darkness”
motifs are just as evident as in more instantaneous analyses (and equally of a
piece with the tropology of Hollywood’s imperial “rescue fantasies,” such as the
recent Bruce Willis vehicle, Tears of the Sun). Whether, then, in news reports,
movies, memoirs, or travelogue, Africa remains cast as a continent of epic odyssey
and spiritual quest—a place that has “always known violence and war”—result-
ing in epiphanic but invariably bathetic lessons about humanity, and the capacity
for evil lurking in all of us (Peterson 2001: xiii). “[I]n Africa,” Peterson muses
with an audacity that would surely be inconceivable applied to any continent
other than Africa, “there is a Jungian balance between remarkable good and
intense evil. But it may be more of a Manichean battle between the forces of light
and dark, because as worthy of spiritual celebration as the good may be, the
166
T R I B A L I S M A N D T R I B U L AT I O N
degree of evil is also extraordinary” (2001: xiv). His account of Somalis’ bloodlust
as a function of these “ancient nomadic warriors” having been catapulted “by
default into a new era” could have been lifted straight from accounts of Kenya’s
Mau Mau “emergency” in the 1950s, which similarly explained Kikuyu “atavism”
as a crisis of modernization as backward tribes struggled with, and against, the dis-
ruptive impact of “progress” on ossified traditions and irrational beliefs (Peterson
2001: 6; Carruthers 1995: 128–93).
To note the persistence of egregious Othering in Western constructions of
Africa is not to highlight anything either new or neglected (Hawk 1992; Ned-
erveen Pieterse 1992). Africa—as dark continent—has long functioned as foil to
the West’s virtuous self-conception as cradle of civilization and progenitor of the
Enlightenment. Too often, however, emphasis is placed on correcting Western
“misperceptions” of Africa. Crucially, we need to go beyond a dissection of these
enduring Western figurations, probing the ideological work of the identity-
practices involved. How does the West constitute itself in relational terms to
Africa? What politics are enabled by construing “evil” as a qaat-chewing warlord,
a kleptocrat outfitted in leopard-skin, an ululating child toting an AK-47?2 What
functions, in other words, do particular representations of Africa perform for “us”?
If decontextualization and excessive ethnicization in news reporting result more
from media’s structural position within a particular set of power relations than
from narrowly procedural shortcomings, then making sense of how Africa is—and
is not—covered requires an account of news-media’s role in simultaneously nor-
malizing and obscuring the current global order. The task is thus to make visible
the ways in which news-media efface the systemic underpinnings of local crises
such that the West effectively “disappears” from reporting on the roots of African
“failure,” while Western virtuousness is affirmed as the civilized correlative to bar-
barism (Salter 2002).
167
SUSAN L. CARRUTHERS
present such situations will always threaten to collapse back into barbarism. So,
why “go there” when intervention promises to be protracted, or, if temporary,
then recurrently required to staunch the inevitable reanimation of murderous
passions? For liberals of this stripe, the issue is how to build support for sustained
overseas commitments, and for timely interventions that do not simply exacer-
bate dependence, corruption, autocracy, or population displacement (de Waal
1998). The proposed solution is a media that builds support for militant human
rights interventionism, cultivating Western willingness to expend blood and
treasure in the name of human solidarity—an agenda that facilitated liberal
enthusiasm for “humanitarian bombing” of Serbia during NATO’s Kosovo cam-
paign, recently leading some (notably Michael Ignatieff) into highly supportive
postures toward the “liberationist” pretensions of current US foreign policy
(Kaldor 1999; Ignatieff 2000, 2003; Wheeler 2000).
For some on the left, on the other hand, the trouble with Western journalism is
that it legitimizes intervention too readily. Mel McNulty, for example, argues that
“Misinterpretation of these crises [in Rwanda and Zaire] as ethnically-driven
facilitated Western interventionary responses, the rationale for which may be
summarized as ‘they are mad, we are sane, we must save them from themselves,’
and served, whether deliberately or accidentally, to make a bad situation worse”
(1999: 268). She goes on to impute a fair degree of deliberation to the West,
which while making a show of its charity being imposed upon, calculatedly threw
a “cloak of humanitarian concern” over an intervention that served “military-
strategic interests” in the Great Lakes region (McNulty 1999: 268). The precise
nature of these interests—and the extent to which they may be ascribed to others
besides the French (whose Opération Turquoise was widely criticized as an inter-
vention on behalf of the génocidaires)—remains unclear, however.
What, then, is the character of this interventionism? Does not a concentration
on sporadic, short-term militarized operations, which may be fairly readily attrib-
uted to specific national neocolonial agendas, obscure the wider and more
enduring role of Western institutions in establishing, maintaining, and adjusting
the global neoliberal framework within which Africa is deeply embedded? “Inter-
vention”—whether invoked by its advocates or its detractors—perpetuates an
assumption that the West’s typical posture toward Africa is one of either benign
or malign neglect, punctuated by brief intrusions of intense activity, altruistic or
self-interested depending on the interpreter’s perspective (Feher 2000; Chandler
2002). Missing here is more than acknowledgment of the deep colonial backstory.
After all, several accounts (including contemporaneous media reports) have
inserted Belgian imperial “divide and rule” tactics into their analyses of Rwanda’s
genocide: with the consolidation of antagonistic ethnicities attributed to Belgian
machinations in instituting a kipende system that crystallized ethnic difference
(giving it both documentary shape and somatic form), and materially privileging
the minority Tutsi population.
Excavation of the colonial roots of current crises is undoubtedly necessary. But
historical correctives must not occlude the constant influence in (and over)
168
T R I B A L I S M A N D T R I B U L AT I O N
Notes
1 “The West” is a problematic descriptor on at least two counts. The geographic catch-
all implies a degree of homogeneity that fails to capture the diversity of the region so
labeled, while implying grid coordinates that do not fully map the location of the
global north. These qualifications notwithstanding, a dispersed but powerful entity,
“the West,” continues to use that self-descriptor, and to construe itself in “civiliza-
tional” terms under this banner. I use it here (henceforth without quotation marks) as
both a shorthand convenience but also to invoke—and critique—that self-
conception. Similarly, “Western media” functions as an abbreviation for mainstream
English language journalism of primarily the United States and United Kingdom (and
their respective “global” news organizations), which of course also homogenizes the
variety of opinion and outlets to be found in both states and beyond.
2 Take, for example, the following passage from William Shawcross’ Deliver Us from
Evil: Warlords and Peacekeepers in a World of Endless Conflict, which, while critical of
the corruption and confusion surrounding certain UN interventions, reserves a partic-
ular vernacular of opprobrium for its African “devils”: “The Nigerian presidential
Gulfstream was furnished in dictator-chic style, with beige leather seats, gold-plated
seat buckles and other fittings, and gold taps in the lavatory . . . There was a small VIP
169
SUSAN L. CARRUTHERS
section at the front of the plane where Annan sat, a large television screen before him.
I wondered how many really bad people had sat in that seat” (Shawcross 2000, p. 271).
References
Appadurai, A. (1998) “Dead certainty: ethnic violence in the era of globalization,” Public
Culture, vol. 10, no. ii, pp. 225–47.
Arnett, Peter (1995) Live from the Battlefield, London: Corgi.
Behr, E. (1992) Anyone Here Been Raped and Speaks English? A Correspondent’s Life Behind
the Lines, Harmondsworth: Penguin.
Bell, M. (1995) In Harm’s Way: Reflections of a War Zone Thug, London: Hamish Hamil-
ton.
Bell, M. (1997) “TV news: how far should we go?,” British Journalism Review, vol. 8, no. i,
pp. 6–16.
Bell, M. (1998) “The truth is our currency,” Press/Politics, vol. 3, no. i, pp. 102–9.
Boltanski, L. (1999) Distant Suffering: Morality, Media and Politics, Cambridge: Cambridge
University Press.
Bowden, M. (2002) Black Hawk Down: a Story of Modern War, New York: Signet.
Bush, G.W. (2003) “In Bush’s words: ‘we will do what is necessary’ in the fight against
terror,” The New York Times, September 8, pp. A10.
Cain, K.L. (2003) “The legacy of Black Hawk down,” The New York Times, October 2,
p. A27.
Carruthers, S.L. (1995) Winning Hearts and Minds: British Governments, the Media and
Colonial Counter-insurgency, 1945–60, Leicester: Leicester University Press.
Carruthers, S.L. (2000) The Media at War: Communication and Conflict in the Twentieth
Century, Basingstoke: Macmillan/Palgrave.
Carruthers, S.L. (2003) “Bringing it all back home: Hollywood returns to war,” Small Wars
and Insurgencies, vol. 14, no. i, pp. 167–82.
Chandler, D. (2002) From Kosovo to Kabul: Human Rights and International Intervention,
London: Pluto Press.
Chrétien, J.-P. (1995) Rwanda: les médias du génocide, Paris: Karthala.
Clarke, J.N. (2002) “Revisiting the new interventionism,” Peace Review, vol. 14, no. i,
pp. 93–100.
Dauber, C. (2001) “‘The shots seen ‘round the world’: the impact of the images of
Mogadishu on American military operations,” Rhetoric and Public Affairs, vol. 4, no. iv,
pp. 653–87.
Des Forges, A. (1999) Leave None to Tell the Story: Genocide in Rwanda, New York: Human
Rights Watch.
Destexhe, A. (1995) Rwanda and Genocide in the Twentieth Century, London: Pluto Press.
de Waal, A. (1994) “African encounters,” Index on Censorship, vol. 6, pp. 14–31.
de Waal, A. (1998) Famine Crimes: Politics and the Disaster Relief Industry in Africa, Bloom-
ington, IN: Indiana University Press.
Dowden, R. (1995) “Covering Somalia—recipe for disaster” in E. Girardet (ed.), Somalia,
Rwanda and Beyond: the Role of International Media in Wars and Humanitarian Crises,
Dublin: Crosslines Global Report.
Fair, J.E. and Parks, L. (2001) “Africa on camera: television news coverage and aerial
imaging of Rwandan refugees,” Africa Today, vol. 48, no. ii, pp. 35–57.
170
T R I B A L I S M A N D T R I B U L AT I O N
171
SUSAN L. CARRUTHERS
McNulty, M. (1999) “Media ethnicization and the international response to war and
genocide in Rwanda” in T. Allen and J. Seaton (eds), The Media of Conflict: War
Reporting and Representations of Ethnic Violence, London: Zed Books.
Meisler, S. (1992) “Committed in Africa: reflections of a correspondent” in B. Hawk, (ed.)
Africa’s Media Image, Westport, CT: Praeger.
Minear, L., Scott, C., and Weiss, T. (1996) The News Media, Civil War and Humanitarian
Action, Boulder, CO: Lynne Reinner.
Moeller, S. (1999) Compassion Fatigue: How The Media Sell Disease, Famine, War and
Death, New York: Routledge.
Murison, J. (1996) Fleeing the Jungle Bloodbath: the Method in the Madness, Edinburgh:
Centre of African Studies, occasional paper no. 65.
Myers, G., Klak, T., and Koehl, T. (1996) “The inscription of difference: news coverage of
the conflicts in Rwanda and Bosnia,” Political Geography, vol. 15, no. i, pp. 21–46.
Natsios, A. (1996) “Illusions of influence: the CNN effect in complex emergencies” in
R. Rotberg and T. Weiss (eds), From Massacres to Genocide: the Media, Public Policy and
Humanitarian Crises, Washington, DC: Brookings Institution.
Nederveen Pieterse, J. (1992) White on Black: Images of Africa and Blacks in Western Popu-
lar Culture, New Haven, CT: Yale University Press.
Nederveen Pieterse, J. (1997) “Sociology of humanitarian intervention: Bosnia, Rwanda
and Somalia compared,” International Political Science Review, vol. 18, no. i, pp. 71–93.
Neuman, J. (1996) Lights, Camera, War: Is Media Technology Driving International Politics?
New York: St Martin’s Press.
Pedelty, M. (1995) War Stories: the Culture of Foreign Correspondents, London: Routledge.
Peterson, S. (2001) Me Against My Brother: at War in Somalia, Sudan and Rwanda, New
York: Routledge.
Philo, G., Hilsum, L., Beattie, L., and Holliman, R. (1998) “The Media and the Rwanda
crisis: effects on audiences and public policy” in J. Nederveen Pieterse (ed.), World
Orders in the Making: Humanitarian Intervention and Beyond, Basingstoke: Macmillan,
1998, pp. 211–29.
Pilkington, E. (1993) “Shots that shook the world,” The Guardian, October 11, 1993.
Pottier, J. (2002) Re-imagining Rwanda: Conflict, Survival and Disinformation in the Late
Twentieth Century, Cambridge: Cambridge University Press.
Power, S. (2002) A Problem from Hell: America and the Age of Genocide, New York: Basic
Books.
Prunier, G. (1997) The Rwanda Crisis: History of a Genocide, London: Hurst.
Richards, P. (1996) Fighting for the Rain Forest: War, Youth and Resources in Sierra Leone,
Oxford: James Currey.
Robinson, P. (1999) “The CNN effect: can the news media drive foreign policy?,” Review
of International Studies, vol. 25, no. 1, pp. 301–9.
Robinson, P. (2001) “Operation Restore Hope and the illusion of a news media driven
intervention,” Political Studies, vol. 49, pp. 941–56.
Robinson, P. (2002) The CNN Effect: the Myth of News, Foreign Policy and Intervention,
London and New York: Routledge.
Rosenblatt, L. (1996) “The media and the refugee” in R. Rotberg and T. Weiss (eds), From
Massacres to Genocide: the Media, Public Policy and Humanitarian Crises, Washington,
DC: Brookings Institution.
Rotberg, R. and Weiss, T. (eds) (1996) From Massacres to Genocide: the Media, Public Policy
and Humanitarian Crises, Washington, DC: Brookings Institution.
172
T R I B A L I S M A N D T R I B U L AT I O N
Salter, Mark (2002) Barbarians and Civilization in International Relations, London: Pluto
Press.
Seib, P. (1998) Headline Diplomacy: how News Coverage Affects Foreign Policy, Westport,
CO: Praeger.
Sharkey, J. (1993) “When pictures drive foreign policy,” American Journalism Review, vol.
15, no. x, pp. 14–19.
Shattuck, J. (1996), “Human rights and humanitarian crises: policy-making and the
media” in R. Rotberg and T. Weiss (eds), From Massacres to Genocide: the Media, Public
Policy and Humanitarian Crises, Washington, DC: Brookings Institution.
Shaw, M. (1994) “Civil society and global politics: beyond a social movements approach,”
Millennium: Journal of International Studies, vol. 23, no. iii, pp. 421–34.
Shaw, M. (1996) Civil Society and Media in Global Crises: Representing Distant Violence,
London: Pinter.
Shawcross, W. (2000) Deliver Us from Evil: Warlords and Peacekeepers in a World of Endless
Conflict.
Stech, F. (1994) “Preparing for more CNN wars” in J. Petrie (ed.), Essays on Strategy XII,
Washington, DC: National Defense University Press.
Strobel, W. (1997) Late-Breaking Foreign Policy: the News Media’s Influence on Peace Opera-
tions, Washington, DC: United States Institute of Peace.
Wall, M.A. (1997) “The Rwanda crisis: an analysis of news magazine coverage,” Gazette:
the International Journal for Communication Studies, vol. 59, no. ii, pp. 121–34.
Wheeler, N.J. (2000) Saving Strangers: Humanitarian Intervention in International Society,
Oxford: Oxford University Press.
173
9
HUMANIZING WAR
The Balkans and beyond
Philip Hammond
A notable feature of the propaganda surrounding the war on terrorism has been
the tendency of coalition leaders to fall back on what Michael Ignatieff calls the
“dominant moral vocabulary” of the 1990s (The New York Times, February 5,
2002): the discourse of humanitarianism and human rights. Washington report-
edly spent hundreds of thousands of dollars hiring advertising and public relations
consultants to “humanise the war” in Afghanistan (Channel 4 News,
November 6, 2001), though the results were bizarre. President George W. Bush
invited American children to donate a dollar to the Red Cross while his airforce
deliberately bombed the organization’s facilities in Kabul and Kandahar; and
Afghan children had trouble distinguishing the aid packages from the cluster
bombs, both dropped by US planes. By the time the invasion of Iraq started in
March 2003 political leaders seemed to have all but forgotten about searching for
WMD, let alone combating international terrorism, instead promising to “liber-
ate” the Iraqi people. It is as if the war on terrorism cannot be justified on its own
terms, and has to be invested with some higher “moral” purpose in the form of
humanitarianism or upholding human rights.
This adoption of humanitarian rhetoric by the Bush administration has dis-
mayed many former enthusiasts of ethical interventionism. In Europe, criticism is
tinged with anti-Americanism and distaste for the Republican Party, as in Timo-
thy Garton Ash’s concern that “the association with Bush’s America is tarnishing
[the] liberal internationalist project” (Guardian, September 19, 2002). Yet there
are similar worries in the US. Richard Falk, for example, laments a fall from the
“golden age” of the 1990s: since 9/11, humanitarianism has been used to provide
“a cover for imperial objectives,” offering “post hoc rationalizations for uses of
force otherwise difficult to reconcile with international law” (Nation, July 14,
2003). David Rieff is also bothered by the cynical instrumentalism with which
humanitarianism is treated, noting that Secretary of State Colin Powell described
NGOs in Afghanistan as “an important part of our combat team” (2002: 236).
More broadly, having advocated military intervention in the Balkans throughout
174
H U M A N I Z I N G WA R
the 1990s, Rieff is now worried that talk of humanitarianism provides an auto-
matic justification which elides the human costs of war (Rief 2002: 284). Yet
while many critics have challenged the high-flown moral claims made about the
war on terrorism, the possibility of a more benign imperial relationship between
the Great Powers and weaker states is usually still assumed.
The discourse of humanitarianism and human rights was promoted throughout
the 1990s by journalists and commentators as an organizing principle for a post-
Cold War world order (Herman and Peterson 2002), nowhere more conspicuously
than in media coverage of the former Yugoslavia. As they sought to encourage
Western intervention in Bosnia, reporters and intellectuals developed the “moral
vocabulary” which was later given an official stamp of approval by NATO during
the 1999 Kosovo conflict, and which has since been used as a standby justification
for intervention anywhere from Afghanistan to Liberia. This chapter looks back at
the pre-9/11 “golden age” and suggests that those who now complain that US
imperialism is destroying international order should reflect on how 1990s humani-
tarianism and human rightsism did the same. The moralistic media consensus
which developed in favor of intervention in the Balkans was premised on the
notion that Western action to uphold human rights should override established
principles of international law, particularly that of non-interference in the inter-
nal affairs of sovereign states. This development has been driven by the felt need of
Western societies to discover some new moral purpose in the post-Cold War
world, despite the disastrous consequences of intervention for those on the receiv-
ing end of their benevolence.
175
PHILIP HAMMOND
176
H U M A N I Z I N G WA R
“to create clarity . . . as to which side was the cause of the problem . . . and that
meant the Kosovar Albanians agreeing to the package and the Serbs not agreeing
to the package” (Moral Combat, March 12, 2000). Embarrassingly, the ethnic
Albanians initially refused to sign up, since the agreement did not offer full inde-
pendence for Kosovo; whereas the Yugoslav side accepted the political
agreement, though arguing that it should be implemented by UN rather than
NATO troops. Since the “negotiations” were designed to produce exactly the
opposite outcome, the terms were changed in order to make them impossible for
Yugoslavia to accept: Kosovo’s future as part of Serbia was left uncertain, a
NATO force was insisted on, and an appendix was inserted giving NATO troops
unrestricted access to the whole of Yugoslavia, including territorial waters and
airspace, and immunity from local law. At the talks, British Foreign Secretary
Robin Cook openly invited the ethnic Albanian delegation to sign so that
airstrikes could be carried out (BBC Ceefax, February 21, 1999).
At least some journalists were fully aware of America’s intentions at Rambouil-
let: a senior State Department official told them that “the bar was set too high for
the Serbs to comply” because “they need some bombing” (Kenney 1999). Yet no-
one saw fit to report this deliberate provocation of war, instead maintaining that
Yugoslavia had been offered what NATO’s then Secretary General, Javier Solana,
called a “balanced and fair peace agreement.” Most British journalists proved
incapable of describing the Rambouillet process accurately until months after the
war ended, when it no longer mattered. It was not until March 2000 that the
BBC’s up-to-the-minute online news service reported that the negotiations had
been “designed to fail” (Mason 2000). Even as the Rambouillet talks were under-
way, Western states were using the OSCE mission as cover for their own
intelligence operations. Predictably, this too was not reported until a year later,
when the US Central Intelligence Agency admitted its agents had been among
the OSCE monitors, “developing ties with the KLA and giving American mili-
tary training manuals and field advice on fighting the Yugoslav army and Serbian
police” (Sunday Times, March 12, 2000). Far from being “exhausted,” diplomacy
had not been attempted. Rather, there was a public show of diplomacy which
provided a pretext for war.
In a speech a year after the bombing, NATO Secretary General Lord George
Robertson asked “was the intervention in Kosovo moral?” He suggested that “the
only possible answer is ‘yes,’” because “By March of 1999, Serb oppression had
driven almost 400,000 people from their homes.” According to Robertson, “this
was ethnic cleansing—plain and simple”: “before the air campaign . . . the atroc-
ities being committed by Serb forces against the Albanians were organized,
systematic, and dictated by a centrally directed strategy” (Robertson 2000).
Robertson’s claims rest on a narrative developed during the war by NATO pro-
pagandists and compliant journalists which entails an extreme distortion of
events.
It is true that the conflict had created around 400,000 refugees and “internally
displaced persons” during the year preceding the bombing, although by March
177
PHILIP HAMMOND
1999 many had returned home “in places where there is no violence, and es-
pecially where KVM has a continuing presence” (UNHCR 1999a). In contrast to
Robertson’s retrospective assertion that “this was ethnic cleansing—plain and
simple,” contemporaneous assessments suggested different reasons for the flight of
refugees, including “clashes between Government security forces and the KLA,
kidnappings, street violence and, more recently, military exercises by the
Yugoslav army” (UNHCR 1999a). German Foreign Ministry reports in early
1999 stated that:
The refugees included around 55,000 who fled to other parts of Serbia or to Mon-
tenegro (UNHCR 1999b), the vast majority of whom were undoubtedly Kosovo
Serbs. Particularly since the start of the one-sided ceasefire, the main cause of vio-
lence was KLA activity. Just before the bombing the US Committee for Refugees
(1999) reported: “Kosovo Liberation Army . . . attacks aimed at trying to ‘cleanse’
Kosovo of its ethnic Serb civilian population.” UNHCR said, “Over 90 mixed
villages in Kosovo have now been emptied of Serb inhabitants and other Serbs
continue leaving, either to be displaced in other parts of Kosovo or fleeing into
central Serbia.” The Yugoslav Red Cross estimates there are more than 30,000
non-Albanian displaced currently in need of assistance in Kosovo, most of whom
are Serb.
At the time, even NATO privately acknowledged the real cause of continuing
conflict. According to minutes of the North Atlantic Council, the KLA was “the
main initiator of the violence,” and had “launched what appears to be a deliber-
ate campaign of provocation” (Moral Combat, March 12, 2000). This was at the
very moment when Western intelligence agencies and private US military train-
ing companies were providing assistance to the KLA.
NATO did not go to war because of “systematic” atrocities and large numbers
of refugees. Rather, NATO promised that the bombing would prevent a refugee
crisis. On the first day of the war, James Rubin insisted that if NATO had not
acted, “you would have had hundreds of thousands of people crossing the border.”
The following day Blair declared: “fail to act now . . . and we would have to deal
with . . . hundreds of thousands of refugees” (BBC, March 25, 26, 1999). It was
only after the bombing began, and hundreds of thousands did indeed flee, that
the war was quickly re-presented as a response to the refugee crisis. As a senior
NATO official later explained: “Following the fiasco of the lightning strikes, the
refugees provided us with a new objective for the war” (Le Nouvel Observateur,
July 1, 1999). This was possible because journalists proved quite willing to reverse
the chronology of events to claim: “This is why NATO went to war—so the
refugees could come back to Kosovo” (BBC, June 16, 1999).
178
H U M A N I Z I N G WA R
This was when the claims about “ethnic cleansing” and “genocide” became
important: it had to be maintained that the huge refugee exodus from Kosovo was
due to a premeditated campaign of atrocities which the Serbs would have carried
out regardless of whether NATO intervened. Bang on cue, documents outlining a
secret Serbian plan for “ethnic cleansing”—codenamed Operation Horseshoe—
were revealed by Germany’s Foreign Minister, Joschka Fischer, on April 6, 1999.
After the war, German Brigadier General Heinz Loquai, a former OSCE adviser,
exposed the supposed blueprint for genocide as a fake concocted by the German
intelligence services (Sunday Times, April 2, 2000). At the time, however, the
theme of “genocide in Kosovo” was taken up with enthusiasm by the media.
Every UK news organization discovered “echoes of the Holocaust” (Sun, April 1,
1999), relishing each atrocity story. In an article which her own newspaper, the
Guardian, did not publish until after the war, Audrey Gillan recounted in the
London Review of Books (May 27, 1999) how UNHCR spokesman Ron Redmond
spoke to the press of bodies being desecrated, eyes being shot out. The
way he talked it sounded as if there had been at least a hundred murders
and dozens of rapes. When I pressed him on the rapes, asking him to be
more precise, he reduced it a bit and said he had heard that five or six
teenage girls had been raped and murdered. He had not spoken to any
witnesses.
Gillan went on to describe how a BBC reporter then “reeled off what Ron Red-
mond had said, using the words ‘hundreds,’ ‘rape,’ and ‘murder’ in the same
breath.” It would be extraordinary if no crimes or atrocities were committed in
Kosovo, since it was engulfed by civil war as well as under NATO bombardment.
The suggestion of exceptional “genocidal” violence, however, was simply West-
ern propaganda designed to justify the further violence of NATO bombing.
Media hysteria fed off wild official statements estimating the death toll in
Kosovo at anything from 10,000 to 100,000 or higher. However, post-war forensic
investigations have so far failed to corroborate such numbers. By February 2003,
after more than three years of investigations, a total of 4,019 bodies had been
found, according to the UN’s Kosovo Office of Missing Persons and Forensics. In
addition, there are 164 people who the UN classifies as missing “considered
dead,” plus around 800 bodies exhumed from sites in Serbia where they had
apparently been concealed during the war. This would bring the total to around
5,000, although it should also be borne in mind that these numbers do not distin-
guish between civilians and soldiers, nor between Albanians and Serbs. More
bodies will probably be found, but in 2003 the UN suggested that a “generous
estimate for the mortal remains still unaccounted for and to be recovered during
this year in Kosovo, would be between 500 and 700.”3 There were many tragic
deaths, but there was not a new Holocaust.
Despite all the evidence to the contrary, the belief persists that Kosovo was a
successful and “moral” war. Hence in 2003 many of those who opposed war with
179
PHILIP HAMMOND
Iraq were at pains to emphasize their prior support for NATO bombing. “We sup-
ported the war in Kosovo,” announced the Independent in an anti-war editorial
(February 7, 2003), and the Guardian’s Jonathan Freedland criticized the build-up
to war while writing that “US power can sometimes be a force for good in the
world: that’s why I supported the Kosovo campaign” (February 5). Repelled by
Bush, liberal commentators demanded more of the moralizing that had justified
the Kosovo campaign. Johann Hari, for example, wrote that:
Similarly, James Rubin suggested that: “if Tony Blair had been the explainer of
the rationale for war in Iraq . . . as opposed to some of the comments coming out
of the Pentagon, we’d be in much better shape in terms of global public opinion”
(Iraq: A Just War?, Channel 4, February 28, 2003).
Yet while the propaganda for war with Iraq was undoubtedly clumsy, it was no
more inept than in Kosovo, which also featured bogus diplomacy and dodgy doc-
uments. In 1999 the propaganda tended to be taken at face value because many
journalists and commentators were predisposed to welcome war. It fitted the
“moral” worldview which developed in the 1990s, epitomized by Western percep-
tions of the former Yugoslavia.
180
H U M A N I Z I N G WA R
racies struggling to escape Communist tyranny, then it made sense for Western
governments to offer support, as local politicians were quick to recognize. Croat-
ian President Franjo Tudjman, for example, was angling for international backing
when he described the conflict as “the same that has been going on in Eastern
Europe for the past three years: democracy against communism” (European,
August 18, 1991).
Yet Western triumphalism following the Cold War was short lived: the unity of
the West could no longer be taken for granted in the absence of the Soviet
enemy, and the sense of cohesion and purpose which the old politics of Left and
Right had given to domestic politics was now lost. In 1991 Croatia’s historic ally
Germany led the European Community in recognizing Slovenian and Croatian
independence, effectively challenging US policy. In 1992 America sought to re-
establish its leadership, pushing for recognition of Bosnian independence in a
reversal of its previous support for the integrity of the Yugoslav federation. In
both Croatia and Bosnia, as had been predicted, the result was bitter civil war.
While the relatively painless secession of Slovenia may have looked like another
former Communist nation making a transition to democracy, these protracted
and bloody conflicts demanded a different explanation. An influential essay by
Samuel Huntington recast the East/West division as a “clash of civilisations,”
arguing that “the Velvet Curtain of culture has replaced the Iron Curtain of ide-
ology as the most significant dividing line in Europe.” The world was riven by
civilizational “fault lines,” one of which ran “almost exactly along the line now
separating Croatia and Slovenia from the rest of Yugoslavia.” Those on the wrong
side—Orthodox Christians and Muslims, in the case of Yugoslavia—were “much
less likely to develop stable democratic political systems” (1993: 30–1). Again
this was a theme taken up by local leaders, Tudjman suggesting that Serbs and
Croats were “not just different peoples but different civilisations” (European,
August 18, 1991), and Slovenian minister Petar Tancig contrasting the “violent
and crooked oriental-bizantine heritage” of Serbia and Montenegro with the
“more humble and diligent Western-catholic tradition” of Slovenia and Croatia
(quoted in Johnstone 2002: 137). Media coverage echoed these ideas. Peter Jenk-
ins wrote, for example, that “There were two Europes for many centuries before
the Cold War was thought of: Western Christendom, Catholic and baroque, and
Eastern Orthodox Europe which, in the Balkans, merged into the Ottoman
Empire and the world of Islam” (The Independent, November 12, 1992). This per-
spective suggested that Yugoslavia was torn by “ethnic” conflicts which had a
long history, but that some ethnic groups were close to, if not part of, the West.
Other proponents of “ethnic” explanations for Yugoslavia’s break-up, however,
made no such distinctions. Robert Kaplan’s (1994) development of Huntington’s
idea drew the division, not between different civilizations, but between the civil-
ized world and various zones of anarchy: “places where the Western Enlighten-
ment has not penetrated,” which are populated by “re-primitivized man” and
under constant threat of “cultural and racial war.” This was the spirit in which
the Telegraph’s Defence Editor, John Keegan (1993: xi), argued that:
181
PHILIP HAMMOND
That the Balkan tribes had been killing each other for centuries and
that there was nothing that could be done. It was nobody’s fault. It was
just, somehow, the nature of the region. It was a lie that Western
governments at that time liked. It got the Western world off the hook.
(Little 2001)
182
H U M A N I Z I N G WA R
a clear aggressor and clear victim” (quoted in Ricchiardi 1996). For Little, the
war was “very simple and straightforward”: it happened because “Milosevic was a
megalomaniac dictator” (Guardian, January 6, 2003). If events did not fit this
simplistic framework, they were distorted or disregarded. Western journalists con-
sistently downplayed or ignored attacks by Croats and Bosnian Muslims, so that
Serbian attacks appeared to be evidence of a one-sided war of aggression. Nik
Gowing (1994: 35) points out, for example, that the Croat siege of Mostar was
“virtually unreported,” despite suffering at least as bad as the plight of Sarajevo.
Indeed, when Western politicians, journalists, and celebrities came to visit the
siege of Sarajevo they invariably saw only the Muslim-held centre of the city, and
ignored the Serb-populated areas. Although Sarajevo was a supposedly demilita-
rized “safe area,” Bosnian government forces in the city were accused by UN
General Francis Briquemont of provoking the Bosnian Serbs “on a daily basis”
(quoted in Binder 1994–5: 73). This pattern was repeated elsewhere, such as at
Gorazde, where an April 1994 NATO airstrike against Bosnian Serbs was pre-
sented as a response to the Serbian assault on the town, in which thousands were
said to have been killed or wounded. It later transpired that the attack—deliber-
ately provoked by Bosnian Muslim forces from within the “safe area”—caused
“closer to 200 than 2,000 casualties,” mostly soldiers, and that the extent of the
fighting had been exaggerated by UN officials. General Sir Michael Rose revealed
that the visible destruction of the town had largely resulted, not from the Serbian
shelling, but from previous Bosnian Muslim attacks which had driven 12,500
Serbs from the area (Bogdanich and Lettmayer 2000).
Through a selective and one-sided style of journalism which Gowing (1997:
25–6) has called the “secret shame” of the media in Bosnia, the war was portrayed
as a simple battle of Good versus Evil. As senior BBC correspondent John Simp-
son (1998: 444–5) subsequently wrote: “a climate was created in which it was
very hard to understand what was really going on, because everything came to be
seen through the filter of the Holocaust.” The Serbs were depicted as Nazis com-
mitting genocide against innocent Bosnian Muslim civilians. Reporters even
discovered Nazi-style “death camps” in Bosnia in 1992—though typically they
noticed only the Serb camps, when according to the Red Cross in autumn 1992
the Bosnian Muslims actually had more camps (12 compared to eight), with
nearly as many detainees (1,061 compared to 1,203) (Johnstone 2002: 71). The
equation between Serbs and Nazis which reporters—and the PR firm employed
by the Bosnian government, Ruder Finn—attempted to create in the public mind
invoked moral absolutes in a way that resonated powerfully with contemporary
sensibilities. Mick Hume (1997: 18) suggests that the war thereby provided “a
twisted sort of therapy, through which foreign reporters [could] discover some
sense of purpose—first for themselves, and then for their audience back home,” as
journalists undertook a “moral mission on behalf of a demoralised society.” As
Ignatieff (2003: 42) admits: “The Western need for noble victims and happy end-
ings suggests that we are more interested in ourselves than we are in the places,
like Bosnia, that we take up as causes.”
183
PHILIP HAMMOND
184
H U M A N I Z I N G WA R
although Blair declared at the end of the war that “we intend to start building [the]
new Europe in Kosovo” (June 10), neither is multiculturalism much in evidence in
Kosovo itself, where most of the non-Albanian population have been violently
driven out. By October 2001, the OSCE reported that, since NATO’s victory in
June 1999, there had been more than 5,000 terrorist attacks, over 1,000 murders
and more than 1,000 abductions.4
As in Bosnia, amid all the talk of tolerance, the Serbs were demonized, some-
times in the crudest terms. In the Telegraph Patrick Bishop suggested that “‘Serb’ is
a synonym for ‘barbarian’” (March 26, 1999), while Steve Crawshaw wrote in the
New Statesman (May 31, 1999) that “millions of Serbs” were “liars on a grand scale”
who had “gone mad.” The Independent’s Marcus Tanner (May 11, 1999) asked: “Do
Albanians look like Serbs?”
No . . . The Serbs often have black or dark brown hair and are generally
darker and more heavily built than Albanians. Their appearance is fairly
typical of southern Slavs. By contrast, the Kosovars look Celtic to a
British eye. They have curly hair, which is often blonde or rust coloured,
and their skin tends to be very pale and covered in freckles. Their eyes are
often green or blue and their build is much more slender than that of the
Serbs. They have longer heads. It is not surprising that they look so differ-
ent as they belong to different races that have very rarely intermarried.
Tanner’s clumsy attempt at racial typology was clearly intended to make Kosovo
Albanians more appealing to white British readers while portraying the Serbs as
suspiciously thick-set and swarthy. Even more bluntly, the Sun (April 14, 1999)
described the Serbs as “animals,” who were “an affront to humanity,” and urged
that they be “shot like wild dogs.” As a study commissioned by the Holocaust
Educational Trust noted: “the utilization of anti-Slav stereotypes during the
Kosovo crisis arguably evoked the use of similar stereotypes . . . during the Nazi
era,” when the German media portrayed the Serbs as subhuman (Cica 2001).
Such imagery was also employed during the Bosnian war, when, for example, the
Independent’s leader-page cartoon depicted Bosnian Serbs as apes in combat gear
(May 29, 1992).
Despite the apparent differences between exponents of “tribal” explanations
and advocates of “ethical” intervention to uphold human rights, all sides in the
debate assume the superiority of the West—whether this is understood in civiliza-
tional or moral terms. As Ignatieff (2000: 213) notes:
185
PHILIP HAMMOND
Ignatieff warns against “an absolutist frame of mind which, in defining all human
rights violators as barbarians, legitimizes barbarism” (Ignatieff 2000: 213), but
seems to have adopted just such a mindset himself, arguing that “force and the
threat of it are usually the only language tyrants, human rights abusers and terror-
ists ever understand” (The New York Times, September 7, 2003).
186
H U M A N I Z I N G WA R
for ever-greater interference in other countries in the name of human rights. Falk
complains that:
the Bush administration has been doing its best to wreck world order as
it had been evolving, and . . . part of the wreckage is the abandonment
of legal restraints on the use of international force, the heart and soul of
the UN Charter.
(Nation, July 14, 2003)
Yet if anyone wrecked the UN system it was the advocates of “human rights inter-
vention” who put the West’s moral duty to intervene above the principle of
sovereign equality. Ignatieff argues that upholding sovereign equality means
“defending tyranny and terror” (The New York Times, September 7, 2003): this was
the complaint of ethical interventionists throughout Falk’s “golden age” of humani-
tarian intervention. It was what led many, including Falk himself, to approve the
Kosovo bombing as “illegal but moral.” The alternatives for the future, Ignatieff
suggests, are leaving the UN Charter as “an alibi for dictators and tyrants” and let-
ting the US go its own way, or reforming the UN so it will more readily use force “to
defend human rights” multilaterally (The New York Times, September 7, 2003.). Or
as Todd Gitlin (2003) puts it: “not empire, but human rights with guns.”
Between those who advocate military force to promote human rights, and
those who seek to legitimize a limitless war on terrorism with humanitarian
rhetoric, we face an uninviting choice. It may be easy to see through the cynical
justifications of the “war on terrorism,” but the underlying consensus in favor of
“moral” intervention still needs to be challenged.
Notes
1 Renaud Girard, interview with Diana Johnstone, January 25, 2000.
2 This and other similar German government documents are posted at emperors-
clothes.com/articles/german/Germany.html
3 UNMIK Office of Missing Persons and Forensics, press release, February 3, 2003
(www.unmikonline.org/press/2003/pressr/pr917.htm)
4 See www.osce.org/kosovo/documents/reports/minorities/
References
Bell, Martin (1998) “The journalism of attachment,” in Matthew Kieran (ed.) Media
Ethics, London: Routledge, pp. 15–22.
Binder, David (1994–5) “Anatomy of a massacre,” Foreign Policy, no. 97, Winter, pp. 70–8.
Bogdanich, George and Martin Lettmayer (2000) Yugoslavia: the Avoidable War, documen-
tary film, USA: Frontier Theatre and Film.
Chandler, David (2003) “Culture wars and international intervention: an “inside/out”
view of the decline of the national interest,” unpublished paper.
Cica, Natasha (2001) Truth, Myth or Genocide?, research paper, London: Holocaust Educa-
tional Trust.
187
PHILIP HAMMOND
Gitlin, Todd (2003) “Goodbye, new world order: keep the global ideal alive,” Mother Jones,
July 14 (www.motherjones.com/commentary/columns/2003/29/we_478_01.html)
Gowing, Nik (1994) Real-Time Television Coverage of Armed Conflicts and Diplomatic Crises,
Cambridge, MA: Harvard University.
Gowing, Nik (1997) Media Coverage: Help or Hindrance in Conflict Prevention? New York:
Carnegie Corporation.
Hansen, Lene (1998) Western Villains or Balkan Barbarism?, PhD dissertation, University of
Copenhagen.
Herman, Edward S. and David Peterson (2002) “Morality’s avenging angels: the new
humanitarian crusaders” in David Chandler (ed.) Rethinking Human Rights, Basing-
stoke: Palgrave.
Hume, Mick (1997) Whose War is it Anyway? London: Informinc.
Huntington, Samuel P. (1993) “The clash of civilizations?,” Foreign Affairs, vol. 72, no. 3,
Summer, pp. 22–48.
Ignatieff, Michael (2000) Virtual War, New York: Metropolitan Books.
Ignatieff, Michael (2003) Empire Lite, London: Vintage.
IICK (2000) The Kosovo Report, Oxford: Oxford University Press.
Johnstone, Diana (2000) “The French media and the Kosovo war” in Philip Hammond
and Edward S. Herman (eds) Degraded Capability: the Media and the Kosovo Crisis, Lon-
don: Pluto.
Johnstone, Diana (2002) Fools’ Crusade: Yugoslavia, NATO and Western Delusions, Lon-
don: Pluto.
Kaldor, Mary (2000) “Europe at the Millennium,” Politics, vol. 20, no. 2, May, pp. 55–62.
Kaplan, Robert D. (1994) “The coming anarchy,” The Atlantic Monthly, vol. 273, no. 2,
pp. 44–76.
Kavran, Olga (1991) “Cold War lives on in Yugoslavia reporting,” Extra!, November/
December (www.fair.org)
Keegan, John (1993) A History of Warfare, London: Pimlico.
Kenney, George (1999) “Rolling thunder: the rerun,” The Nation, June 14, pp. 48–51.
Klug, Francesca (1997) “Can human rights fill Britain’s morality gap?,” Political quarterly,
April–June, vol. 68, no. 2 (accessed electronically).
Little, Allan (2001) “The West did not do enough,” BBC Online, June 29 (http://news.
bbc.co.uk/hi/english/world/from_our_own_correspondent/newsid_1413000/1413764.sm)
Mason, Barnaby (2000) “Rambouillet talks ‘designed to fail,’” BBC Online, March 19
(http://news.bbc.co.uk/hi/english/world/europe/newsid_682000/682877.stm)
O’Neill, Brendan (2003) “Cross-border terrorism: a mess made by the West,” Spiked,
July 24 (www.spiked-online.com/Articles/00000006DE7B.htm)
Ricchiardi, Sherry (1996) “Over the line?,” American Journalism Review, September
(accessed electronically).
Rieff, David (1995) Slaughterhouse: Bosnia and the Failure of the West, London: Vintage.
Rieff, David (1999) “A new age of liberal imperialism?,” World Policy Journal, vol. 16,
no. 2, Summer, pp. 1–10.
Rieff, David (2002) A Bed for the Night: Humanitarianism in Crisis, London: Vintage.
Robertson, Lord George (2000) “Law, morality and the use of force,” May 16 (www.nato.
int/docu/speech/2000/s000516a.htm)
Simpson, John (1998) Strange Places, Questionable People, London: Macmillan.
UKFAC (2000) Fourth Report, May 23 (www.publications.parliament.uk/pa/cm199900/
cmselect/cmfaff/28/2802.htm)
188
H U M A N I Z I N G WA R
UNHCR (1999a) “UN inter-agency update on Kosovo humanitarian situation report 82,”
March 4 (www.reliefweb.int/w/rwb.nsf/0/29291b9f2df28d62c125672b004a07fa?Open-
Document)
UNHCR (1999b) “Kosovo crisis update,” March 30 (www.unhcr.ch/cgi-bin/texis/vtx/
home/+2wwBmDevEudwwwwMwwwwwwwhFqnN0bItFqnDni5AFqnN0bIcFqEQd5d
VdagGo5o5aupmwBnaTwGqrDzmxwwwwwww/opendoc.htm)
US Committee for Refugees (1999) “Fighting heats up Kosovo winter; fresh displacement,”
March (www.refugees.org/news/crisis/kosovo_u0399.htm)
189
10
PRISONERS OF NEWS VALUES?
Journalists, professionalism, and
identification in times of war
Howard Tumber
The battle for information and the contest over the winning of public opinion is
a feature common to all conflicts. Attempts by the US government and the mili-
tary to control and manage news during the invasion phase of the 2003 Iraq
conflict involved a number of different measures and procedures. Using familiar
techniques of censorship, misinformation, obfuscation, and psychological opera-
tions to varying degrees, the US was able to frustrate journalists and news
organizations in their search for information. But it is the process of embedding
journalists with military units that is the subject of discussion here and its impli-
cations for the future reporting of conflict.
The embedding of reporters within the military was a key event of the commu-
nications set-up of the 2003 conflict. Unlike the Falklands conflict in 1982 when
journalists were “embedded”1 with the British Task force almost by accident,
during the war in Iraq there was a deliberate plan set out by the US Department of
Defense in consultation with news organizations for journalists to be “situated”
within various parts of the military. The thinking behind this “innovation” had
been developing for some time. Ever since the Vietnam War, governments and the
military had experimented with different methods of “controlling” and “manag-
ing” the media. The information policy adopted by the British Government and
the military during the Falklands conflict was poorly organized and lacked plan-
ning. There was an absence of agreed procedure or criteria, no centralized system
of control, and no coordination between departments (see Morrison and Tumber
1988: 189–90). But whatever impromptu measures the British introduced at the
time, together with post hoc justifications, were clearly based on the myth of Viet-
nam, that somehow the media and television in particular were to blame for the
US losing the war. US defense officials, using the experience gained during the
Falklands conflict, have since tried various measures—pooling arrangements,
stationing reporters in military units—in different operations over the last 20
years. Embedding in various guises has been used before, both in World War II and
Vietnam, and in a limited way in more recent times in Haiti in 1994, Somalia in
190
P R I S O N E R S O F N E W S VA L U E S ?
1992–5, and Bosnia in 1995 (see Porch 2002). The protests by news organizations
following the lack of media access operating in Gulf War I and the restrictions
experienced in recent conflicts in Kosovo in 1999 and Afghanistan in 2001 led the
Pentagon to reassess the reporting rules for the current conflict (see Tumber and
Palmer 2004).
The embedding process in Gulf War II was planned well in advance of the
invasion of Iraq in March 2003. A number of briefings took place in Washington
between Pentagon officials and news organizations to discuss the process and
journalists began attending military training courses in November 2003.2
191
H O WA R D T U M B E R
The situation was that you were a propagandist; and that’s how it turned
out. So there wasn’t any need to put pressure on anyone to write gung-
ho copy because everyone was doing it without any stimulus from the
military. And that’s how most of the reporters felt. They were all very
patriotic and “positive” about the whole thing. So the military didn’t
have to lean on them.
(cited in Morrison and Tumber 1988: 98)
Among the journalists in the Falklands a shift took place in the values that ques-
tioned the virtue of remaining untouched and unmoved toward those they lived
with. There was simply no escaping the military’s embrace. As Mick Seamark,
then of the Daily Mirror, and another of the journalists accompanying the mili-
tary, commented shortly after he returned to the UK:
On this occasion you were always part of it. You couldn’t get off, you
couldn’t say at the end of one day: “Let’s go and have a drink some-
where, away from the story.” You were part of the story in that sense.
The journalists certainly began to feel that they belonged.
(cited in Morrison and Tumber 1988: 98)
This belonging began to change the journalists’ language. As David Norris, then
of the Daily Mail, relayed on his return from the Falklands: “I found I was refer-
ring to ‘us’ collectively when we were on shore” (cited in Morrison and Tumber
1988: 98). Even Gareth Parry, of the liberal Guardian, mentioned that, with the
approach of danger, his attitude changed from originally saying “British” to (after
a few weeks) “us” or “we.”
What in effect happened was that the process of involvement and identifica-
tion by the journalists had little to do with their individual private views and
feelings about the conflict or the attitudes of their news organizations. The
dynamics of the situation were so strong that they overwhelmed all this. As
Morrison and Tumber suggested:
What was happening to the journalists was that their professional need
to cover a story in a detached way was slowly being swamped by the very
real, human need to belong, to be safe. The comradeship and closeness
demonstrated by the troops, which the journalists so admired, were not
just the random product that any occupational association throws up,
but the response to having to work closely together especially during
military exercises and having to solve tasks as a group. At the same time
the enclosed world of a group of men (and all the media personnel were
men as were most of the military) living together means that the need for
emotional expression, fulfillment, and release, the talking over of wor-
ries, fears or any of a myriad small problems which beset individuals,
192
P R I S O N E R S O F N E W S VA L U E S ?
The evidence of this emotional embrace by the journalists was clear in the com-
ments made by the journalists after they returned to Britain. For example, Robert
McGowan, then of the Daily Express, reflected:
I think you’ll find that most of the journalists made lifelong friends with
some of the units. I certainly have. What is more, if the setting is very
dangerous, time spent in close proximity to others who are sharing the
same experiences submerges the individual’s personal characteristics.
The other person becomes “you”: he knows what you are going through.
There is a transparency to feelings and relationships not commonly
found in civilian settings.
(cited in Morrison and Tumber 1988: 100)
In the Falklands conflict, the difficulties the journalists faced in performing their
own professional duties obliged them to make allies of the military, driving them
further into cooperation. For the journalists there was no escape from the mili-
tary’s embrace. There was no possible chance of physically distancing themselves
or of gaining psychological respite by returning to a hotel to write their stories. As
Morrison and Tumber described:
The journalists not only merely observed their subjects, but lived their
lives and shared their experiences, and those experiences were of such
emotional intensity that the form of prose which journalists use to take
the reader into that experience—the “I was there” form—provided not
only a window for the reader, but also a door for partiality irrespective of
any desire to remain the detached professional outsider.
(Morrison and Tumber 1988: 95–6)
193
H O WA R D T U M B E R
medium than in print journalism. In addition these journalists were better able to
weather pressure toward partiality because, compared with the press, they were
more regularly in touch with their home offices (see Morrison and Tumber 1988:
102–3, 105).
For the journalists in the Falklands, events conspired to make them totally
involved and, although analytically impartiality and objectivity are not regarded
as the same, for practical purposes, given the dynamics of the situation, to lose
impartiality in those circumstances was journalistically to lose objectivity. “Only
by remaining completely impartial and unmoved by events would it have been
possible to hold a totally objective account” (Morrison and Tumber 1988: 104).
For these journalists the entire experience had brought them to understand the
nature and purpose of the campaign, the behavior of the military, and the con-
duct of the war, as participants rather than observers. The journalists had entered
a different, closed world and their values shifted with them.
Revisiting the Falklands some 20 years later, the comparisons with the experi-
ences of the journalists “embedded”4 with military forces during the 2003
invasion of Iraq are compelling.
Embedded in 2003
Keith Harrison, of the Wolverhampton (UK) Express and Star, reporting from
Umm Qasr, the southern port in Iraq, in 2003 admitted to the quickness of the
socialization process. His account provided a feel of his and other correspondents’
experiences:
It’s been less than a week, but already the experiences of the war have
had a startling effect on everyone taking part—and those reporting on
it. In our combat kit, we look and sound like soldiers, which is a tribute
to the Army’s embedding system, in which journalists are trained and
attached to military units for the duration of a campaign. We answer to
the Commanding Officer, we follow orders, we share the rations, we eat
where the soldiers eat and we sleep where they sleep. The Royal Logistic
Corps—where they go, we go. The military language that first seemed
like talk from another world is now our mother tongue. Terms like
“sitrep” (situation report) and “be advised” have not so much crept into
our language as carried out a military coup. Place is now location. Car is
always vehicle. Pardon has become “say again.” ETD, ETA, and IAD—
estimated time of arrival/departure and immediate action drill—are now
used almost constantly as we communicate with the soldiers and officers
of the RLC. We know that dobhi is laundry, gash is rubbish and “chogie”
is an affectionate term for the local workers. We say ablutions, not toi-
lets, and put up with flies, food and facilities that we would have sniffed
at just three short weeks ago. Sniffing at the toilets today would be
extremely unwise. The novelty of the American MRE (Meals Ready to
194
P R I S O N E R S O F N E W S VA L U E S ?
Eat or Meals Rejected by Everyone) has long since worn off, but at least
we’re now experts in heating the food with the chemical packs involved.
We’re becoming indoctrinated and recognise the sights and sounds of
army life instantly. We’ve endured no fewer than 30 air raids since war
began. Many have been false alarms—others have carried the chilling
threat: “Incoming ballistic missile! Take cover! Missile in air!” . . . Hours
earlier, when we left our US base, we were given lengthy and frankly dis-
concerting farewells from those staying behind in reserve. As our vehicle
was being prepared, TV pictures showed an Iraqi bunker being blown to
smithereens at close range by a US tank and I found myself cheering
along like a bloodthirsty Dallas cowboy.
(Harrison 2003)
It is hardly surprising then that the reports and language used by the journalists
embedded with the troops would echo these developing attachments.
Some of the “unilaterals”5 believed that it became complicated for the embed-
ded reporters to report uncompromisingly on the actions of their units. “Once
you bond with these guys, once reporters have come to like these guys, they’re
not going to report how horrible anybody can be in war, how they were laughing
as they shot people” (broadcast network correspondent cited in Cook 2003).
Jeff Gralnick, a journalist with experience in Vietnam warned of the dangers
facing the embedded reporters and likened them to the fate of hostages with their
captors:
But that is not, as we know, why you’ve opted to be embedded. Not for
combat. Not that. You’re all going over to report. Truth. Honesty. The
real story. But that is going to be difficult because once you get into a
unit, you are going to be co-opted. It is not a purposeful thing, it will just
happen. It’s a little like the Stockholm Syndrome.
You will fall in with a bunch of grunts, experience and share their
hardships and fears and then you will feel for them and care about them.
You will wind up loving them and hating their officers and commanders
and the administration that put them (and you) in harm’s way. Ernie
Pyle loved his grunts; Jack Laurance and Michael Herr loved theirs; and
I loved mine. And as we all know, love blinds and in blinding it will alter
the reporting you thought you were going to do. Trust me. It happens,
and it will happen no matter how much you guard against it.
Remember also, you are not being embedded because that sweet old
Pentagon wants to be nice. You are being embedded so you can be con-
trolled and in a way isolated.
Once you’re in the field, all those officers and commanders you now
hate, because you love your grunts, you will hate even more because they
will have total control over where you can go, what you can see and what
you can do. Vietnam was easier, we came and went—serial embedees—
195
H O WA R D T U M B E R
Gralnick also warned of the need to remember that the experience of reporting as
an embedded reporter was only one small part of the jigsaw:
One more thing to remember. War is a macro kind of thing. Units in the
field and their grunts are the micro parts of it. So if you are in the field
with a unit where bad things happen, you are seeing only what is around
you. Nothing else. You have no idea how the war is going, only how
your war is going, so never turn what you have in front of you into some-
thing that ends with cosmic conclusion about the war and policy
themselves. Many of us 35 years or so ago—hawk reporters or dove—did
just that and many of us regret it to this day.
(Gralnick 2003)
For some of the journalists the socialization process had begun some four months
earlier (in November 2003) when they attended one of a series of Pentagon-held
week-long training seminars for journalists at a Marine Corps base in Quantico,
Virginia.6 The seminar provoked the journalists into assessing their relationship
with the military and even made news itself. The issue that arose was how to
ensure “separation.”
Upon arrival at the camp, journalists received military issue equipment such as
backpacks, helmets, flak-jackets, and NBC (nuclear, biological, chemical) suits,
which they then used in training exercises (DeFoore 2003). According to Wash-
ington Times staff photographer Gerald Herbert “at first they enjoyed getting their
hands on the new ‘toys’” (cited in DeFoore 2003) but a few of the journalists
quickly realized the dangers of donning all the military gear.
DeFoore outlined the story:
196
P R I S O N E R S O F N E W S VA L U E S ?
Spencer Platt, Getty News staff photographer, an attendee at the training camp,
commented:
On the final night of boot camp the journalists learned they were about
to become the subjects in a massive photo-op organized by the military.
The thought of marching five miles in full gear with still and TV cam-
eras documenting their every move spooked many of the journalists
there. So before the big event, many decided to present themselves in
more of an independent light when the time came for their pictures to
be taken.
(DeFoore 2003)
Gerald Herbert, a Washington Times staff photographer and another attendee, was
concerned about the implications:
All of a sudden the media were trying to spin the media. That question
was nagging me all week long and came to a head that day: at what
point are we observing and at what point are we participating? Some of
the journalists used white tape and black markers to designate them-
selves as press, while others wore jeans and one guy even drew a peace
symbol on his shirt. In the back of your mind you’re wondering how
much is too cosy and when do you become your subject. It’s a very diffi-
cult line and it’s still something people are trying to sort out. How much
do you assimilate into [the military’s] mode and how much do you main-
tain a profile of visual separation?
(cited in DeFoore 2003)
197
H O WA R D T U M B E R
Army’s 3rd Infantry Division, had previously covered other conflicts. For him
“action is an attractive flame”:
combat was rarely more than moderate for us. It offered us the thrill of
close brushes with death, but it was the Iraqis who did all the dying
when A Company, “The Assassins,” rolled up in their tanks flying the
Jolly Roger flag.
(Crittenden 2003)
The following comments from Crittenden indicated both his closeness to the
troops and his praise for those reporters trying to remain independent:
198
P R I S O N E R S O F N E W S VA L U E S ?
“You gotta love this. This is like Patton in that German palace at the
end of World War II,” said Lt. Col. Philip de Camp, commander of the
4/64 Armor battalion, with a big toothy grin. He was sitting in one of
Saddam’s chairs, with one of Saddam’s gold-plated AK-47s in front of
him on one of Saddam’s big conference tables strewn with American
military maps and other articles of a warfighting battalion’s business.
De Camp handed me one of the battalion’s battle coins, with the
“Tuskers” elephant head emblem on it and the motto “We Pierce.”
“You earned it,” de Camp said.
(Crittenden 2003)
I’ll walk away with other souvenirs. The rare privilege of becoming close
to a good group of soldiers and riding with them into battle. The names
and faces of all those soldiers and everything we shared in a little more
than a month. The experience of riding with a conquering army into the
capital of my nation’s enemy, into that enemy’s own yard. The memory
of this strange tour of Mesopotamia, the Cradle of Civilization. Too bad
civilization grew up and left home. We missed all the old ruins but saw a
lot of new ones.
(Crittenden 2003)
For some of the reporters, like their predecessors in the Falklands, their experi-
ences living alongside the military were reflected in their copy. Chris Ayres of the
Times recognized his emotions and the impossibility of remaining detached. In
writing about Marines shooting Iraqis who failed to stop at a checkpoint, he
wrote:
In both the Falklands conflict and in Gulf War II the journalists’ adoption of mil-
itary jargon, however strenuously they sought to excise it from their vocabulary,
indicates how efficiently they were being assimilated.7 As war progressed in Iraq,
some correspondents became very involved.
The story of Sanjay Gupta is an interesting one in the context of the journalist
at war. Gupta, a CNN medical correspondent, received a brief to deal with chem-
ical weapons stories. But he was then invited to travel with a mobile military unit
known as the “Devil Docs” and, as the only trained neurosurgeon among a group
199
H O WA R D T U M B E R
of military doctors, there were times when he was pressed into medical action. He
worked on a number of patients, not only US military but also a two-year-old
Iraqi boy and an Iraqi soldier (Huff 2003; see also Bauder 2003).
Gupta outlined his position: “I knew I was going into a place where people
were going to be injured, I didn’t volunteer for this. I made it quite clear my mis-
sion was journalism not medicine” (cited in Huff 2003). But as he later said:
Today, one of the other doctors said they were flying someone in with a
significant head wound, “would you mind taking a look?” he said. Sure
enough it was a penetrating wound to the brain. I knew this person was
not going to survive a flight to Kuwait.
(cited in Huff 2003)
Compare Gupta’s position with the hypothetical one posed in a discussion held
in 1989 and raised again by Amitai Etzioni on his website during Gulf War II (see
Etzioni 2003). “In a future war involving US soldiers, what would a television
reporter do if he learned the enemy troops with which he was travelling were
about to launch a surprise attack on an American unit?” The question involved a
hypothetical conflict between the north Kosanese and the US supported south
Kosanese. Both Peter Jennings, the ABC anchor, and Mike Wallace, correspon-
dent of 60 Minutes, responded that getting ambush footage for the evening news
would take priority over warning the US troops.
At first, though, Jennings responded that “if he was with a north Kosanese unit
that came upon Americans, I think I personally would do what I could do to warn
the Americans.” Wallace countered that other reporters, including himself,
“would regard it simply as another story that they are there to cover. I’m a little
bit at a loss to understand why, because you (Jennings) are an American, you
would not have covered that story” (cited in Etzioni 2003). When Wallace was
asked “Don’t you have a higher duty as an American citizen to do all you can to
save the lives of soldiers rather than this journalistic ethic of reporting fact?” he
responded: “No, you don’t have a higher duty . . . you’re a reporter.” This argu-
ment convinced Jennings, who conceded, saying “I think he’s right too, I
chickened out” (Etzioni 2003).
Wallace was apparently mystified when Brent Sowcroft, the former National
Security Adviser, argued in relation to the hypothetical, that “you’re Americans
first, and you’re journalists second.”
In contrast to Wallace, Gupta’s position became that of a participant. In his
case he had his “other” professional values to fall back on. His Hippocratic oath
and medical professional values enabled him to participate with a “clear” con-
science. His decision was “automatic” and did not involve a deliberation or
“professional” dilemma. Other journalists did not have “an other” to fall back on
unless it is some wider moral and ethical cultural system.
200
P R I S O N E R S O F N E W S VA L U E S ?
Observing or participating?
The idea of objectivity can be regarded in two distinct ways. In the first instance,
objectivity is conceptualized as an impossible goal. “Objective reporting” is asso-
ciated with ways of gathering news (knowledge about places, people, and events)
and conveying them in a detached, impersonal way free of value judgments. Nev-
ertheless, the act of reporting itself places limitations (such as space, time, and
pertinence) on the ability to report the whole known truth. It follows, therefore,
that the necessity of selection and the hierarchical organization of a story, sug-
gests more of a subjective rather than objective outcome (Bovee 1999: 114–16,
121; Bourdieu 1996: 21). In addition, the structural environment of the institu-
tions of reporting is also restricted by economic and political factors that lead to a
subjective outcome (Schudson 1978). In the second instance, objectivity is used
as a strategic ritual, allowing for the defense of the profession (Tuchman 1972;
Bovee 1999: 123–4). The procedures of the verification of facts, the separating of
“facts” from “analysis,” the presenting of conflicting possibilities and supporting
evidence, the judicious use of quotation marks, the structuring of information in
an appropriate sequence and the criterion of common sense in assessing news
content, while enabling the claim to objectivity (which functions as a shield from
criticism), do not guarantee objectivity. Instead, they only allow an operational
view of objectivity (Tuchman 1972: 662–79; see also Tumber and Prentoulis
2003).
Objectivity is a prized status within journalism. The institutionalized and thus
professional phase of journalism started with the elevation of objectivity to the
dominant ideology within the profession (Trice 1993: 60). Objectivity became the
foundation for the social responsibility claims of the journalistic identity. In war
corresponding the principles of detachment are a key element in the social
construction or formation of identity. But it is here that problems emerge. Cor-
respondents face criticism in two ways. By following the tradition of detached
reporting, journalists are censured or condemned for their dispassionate stance
often in the form of accusations of a lack of patriotism and for eschewing the per-
ceived national interest. At the same time the “journalism of attachment,” the
human, emotional face of war corresponding, has been criticized for opening the
door to mistaken accounts of the conflicts, and for being “self-righteous” and “mor-
alizing” (Ward 1998; McLaughlin 2002: 166–8; Tumber and Prentoulis 2003).
The conditions of detachment are also important because they are a facilitator
of objectivity, and it is in this regard that governments and military practices
attempt to ensure that the “objective” coverage of stories is often restricted or
negated during international conflicts.
In the Falklands and Iraq a clear clash occurred between two competing senti-
ments. On the one hand the journalists carried the occupational ideology of
impartiality and objectivity while the military rucksacks on their backs symboli-
cally carried more than the single source of their provision: in effect, where did
their commitments lie—to traditions of journalistic practice or to those who
201
H O WA R D T U M B E R
could and did protect them, the military? (See Morrison and Tumber 1988;
Tumber and Palmer 2004.)
In Gulf War II the conditions for the journalists were different from those during
the Falklands and the nationalities of the correspondents were not restricted as
they had been during the Falklands. But elements of the situation were very similar.
To repeat the question posed by Gerald Herbert of the Washington Times; “at
what point are we observing and at what point are we participating?”
Are the acknowledgments of their predicaments by journalists during the Falk-
lands and Ayres in Gulf War II “admissions of irrevocably compromised
journalistic integrity, or evidence of a degree of self awareness that others respon-
sible for covering the conflict might do well to emulate?” (Mangan 2003).
In conflict situations like the Falklands and Gulf War II, journalists enter a
realm of professional uncertainty. The values that “normally” ensure a “protective
distance give way to affinity values born of proximity.” Journalists become emo-
tionally involved with the troops with whom they are located—and want “their”
side to win.8
Few people can stand aside in the face of anguish and suffering, especially when
those who need help have been companions, hosts, and mentors. In the Falklands
conflict, Max Hastings, who had already covered 11 wars, understood the ambi-
guities of the war reporter’s role; and resolved the tension between observer and
participant by openly deciding, at least on one occasion deliberately, to assist the
efforts of the Task Force by his writing. There was no self-delusion on Hastings’
part about what this meant in terms of standard journalistic practice (Morrison
and Tumber 1988: 113). Hastings, on his own admission, was partial in his
approach in that he made a conscious decision not to report anything detrimental
to the war effort, but such partiality did not exclude him from being objective.
Writing about the Falklands conflict, Morrison and Tumber suggested that a
reasonably well-defined role (in this case) within British journalism—the partici-
patory journalist—had evolved. Furthermore, in certain circumstances the
participant is likely to be more critical than their erstwhile “objective” colleague,
the observer.9 The journalist as observer works within a system of news values
placing and judging events providing for a degree of predictability in news selec-
tion. The journalist as observer, therefore, is more secure and is protected by the
accredited acceptance of understandings of the profession and news organizations
than the participant, who, while not operating entirely outside such values—
must still “make” the news—intercedes with their own individual personal
judgment of events related to emotions unconnected to journalism itself. An
extra dimension is added to the news gathering process which in extreme cases
leaves the participant vulnerable to challenge. The responsibility for the story or
story selection, therefore, is for the participant a greater individual act than it is
for the observer where it is dispersed among collective values. Thus the partici-
pant, having dismissed a story not on the grounds of newsworthiness, but from
personal values, sees the accepted story as their own property. Because of the
exclusivity of the story, the personalized decision, the direct emotional as opposed
202
P R I S O N E R S O F N E W S VA L U E S ?
This last point is crucial for the US Defense Department officials and military
who were instrumental in the planning and implementation of the embedding
process. If military success is achieved then the dangers of having to deal with
“uncontrollable” participant journalists are minimized. It is on this basis that
future planning and honing of the embedding process is ongoing by Pentagon
officials (see Tumber and Palmer 2004, ch. 4). But for news organizations and the
public the dangers of embedding are all too evident. The enmeshing and identifi-
cation of journalists within the military can lead to problems in the integrity of
the information supplied. For the observer, being a “prisoner of news values”
bound by the rules and procedures of an occupation, ensures a protective safety.
203
H O WA R D T U M B E R
The problem is to maintain those values in a social world (the military) where
the plausibility structure supporting those values remains sufficiently intact.
For the large news organizations operating in Gulf War II, unlike the situation
during the Falklands conflict, the potential problem of identification arising from
the embedding process was mitigated through the presence of their other “unilat-
eral” journalists working in the region and other reporters stationed at the media
center at Centcom headquarters in Doha, Qatar. This allowed for alternative sto-
ries and perspectives about the war in addition to those from correspondents
embedded with military units. The problems for the “unilateral” journalists were
the physical dangers they encountered not only from Iraqi attack but also from
so-called “friendly fire.”10 The portents for news coverage of future wars and con-
flicts are not good. Governments and military will plan more embedding formats
with all the associated risks for “objective” reporting, increase warnings to non-
embedded journalists about their safety, and continue the usual policies of
censorship, misinformation, and delay in confirmation of events. For the future,
the struggle for news organizations will be in resisting these increasing pressures
and, for journalists, a continuation of the current reassessment about the nature
of witnessing, truth, and objectivity.
Notes
1 Embedded was not a term used during the Falklands conflict. Because the organization
of the placing of reporters on board the Task Force ships was done quickly and haphaz-
ardly, little thought was given to the process, and hence its naming, by military and
defense officials.
2 For details see Tumber and Palmer 2004.
3 The study involved interviewing all the journalists who sailed with the task force to
cover the war (see Morrison and Tumber 1988).
4 Approximately 500+ journalists were “embedded” with the military during the 2003
war in Iraq. The United States Government allowed journalists to accompany forces
in the field, on ships, and at departure airfields.
5 The term given to those reporters not “embedded” with the military.
6 Over 50 members of various news organizations attended the course which included
staged hostile environment scenarios and instruction on chemical weapons protection
(DeFoore 2003, see also Tumber 2002).
7 In Gulf War II some journalists “escaped” from the “embedding process” and left their
units. In some cases it was a difficult decision because news organizations were pre-
vented from sending in replacements (the no substitute rule). Toward the end of the
war the numbers leaving their units increased considerably.
8 Hollywood often portrays the journalist as making an emotional choice. In the feature
film Under Fire, Nick Nolte playing the part of a photojournalist on assignment in
Nicaragua for an American news organization is asked by one of the Sandinistas,
“Whose side are you on?” “I don’t take sides, I take pictures” is his response. Later in
the film his “objective” stance changes to partisanship when he agrees to help the
“rebels” by faking a photograph purporting to show that the dead rebel leader is still
alive (Tumber 1997).
9 The “participant” role has been discussed further in the context of the “journalism of
attachment” see Tumber 1997 and 2002; Carruthers 2000, p. 240; McLaughlin 2002,
p. 166; Seib 2002, p. 68; Tumber and Prentoulis 2003, p. 225.
204
P R I S O N E R S O F N E W S VA L U E S ?
10 For details of incidents and subsequent investigations see News Safety Institute 2003;
Tumber and Palmer 2004, ch. 3.
References
Bauder, D. (2003) “CNN reporter performs surgery in Iraq,” www.EditorandPublisher.com
April 3.
Bourdieu, P. (1996) On Television and Journalism, London: Pluto.
Bovee, W.G. (1999) Discovering Journalism, Westport, CN, and London: Greenwood Press.
Carruthers, S.L. (2002) The Media at War, Basingstoke: Macmillan Press.
Cook, J. (2003) “Military, media meet off battlefield to debate war coverage,” www.chicago
tribune.com August 18.
Crittenden, J. (2003) “Embedded journal: leaving Iraq,” www.Pointer.org. April 14.
DeFoore, J. (2003) “Photo-news access vs. independence: thoughts on media boot camp,”
www.pdonline.com March 26.
Etzioni, A. (2003) “Reporters first, Americans second?” www.Amitai-notes.com March 21.
Gralnick, J. (2003) “Lessons to survive by,” www.Tvweek.com March 21.
Harrison, K. (2003) “War reporting: diary of a journalist with the army,” www.Timeson-
line.co.uk March 24.
Huff, R. (2003) “CNN’s own ‘Devil Doc’ scrubs up for duty again,” www.Nydailynews.com
April 9.
Mangan, L. (2003) “Them and us: the singular language of the embeds,” Media Guardian,
April 7, p. 5.
McLaughlin, G. (2002) The War Correspondent, London: Pluto.
Morrison, D.E. and Tumber, H. (1988) Journalists at War, London: Sage.
News Safety Institute (2003) “Dying to tell the story,” News Safety Institute.
Porch, D. (2002) “No bad stories,” www.Nwc.navt.mil/press/review/2002/winter/art5-w02
Seib, Phillip M. (2002) The Global Journalist: News and Consciousness in a World of Conflict,
Lanham, MD, and Oxford: Rowman and Littlefield.
Schudson, M. (1978) Discovering the News: a Social History of American Newspapers, New
York: Basic Books.
Trice, Harrison M. (1993) Occupational Subcultures in the Workplace, Ithaca and New York:
ILR Press.
Tuchman, G. (1972) “Objectivity as a strategic ritual: an examination of newsmen’s
notions of objectivity,” American Journal of Sociology, vol. 77, pp. 660–79.
Tumber, H. (1997) “Bystander journalism, or journalism of attachment,” Intermedia, vol.
25, no. 1, pp. 4–7.
Tumber, H. (2002) “Reporting under fire: the physical safety and emotional welfare of
journalists” in B. Zelizer and S. Allan (eds), Journalism after September 11, London:
Routledge.
Tumber, H. and Palmer, J. (2004) Media at War, London: Sage.
Tumber, H. and Prentoulis, M. (2003) “Journalists under fire: subcultures, objectivity and
emotional literacy,” in D.K. Thussu and D. Freedman (eds), War and the Media,
London: Sage.
Ward, Stephen J. (1998) “Answer to Martin Bell: objectivity and attachment in journal-
ism,” The Harvard International Journal of Press/Politics, vol. 3, no. 3, pp. 121–5.
Zinsmeister, K. (2003) “Different worlds,” Nationalreview.com April 7.
205
11
OUT OF SIGHT, OUT OF MIND?
The non-reporting of small wars
and insurgencies
Prasun Sonwalkar
206
O U T O F S I G H T, O U T O F M I N D ?
the nether world of “absence” of news in media content about most wars and con-
flicts within and across nations. There is much that can be mined by effecting a
shift from “presence” to “absence.”
The thesis of “symbolic annihilation” that was hitherto viewed in terms of the
marginalization of news about women (Tuchman 1972) and other minorities can
also be applied to the coverage of wars and conflicts that involve swathes of pop-
ulation in the developing world. There are countless wars and conflicts raging
across Africa, Asia, and elsewhere that figure only in small print in research sta-
tistics of think tanks or when citizens of Western countries fall victims there. The
selectivity evident in scholarship and media content, preferring wars and con-
flicts involving Western powers, has obvious implications in an increasingly
interconnected world, where the movement of humans, material, and money
across continents to finance terror is no longer a difficult proposition—as the
events of September 11 showed. The need to bring most—if not all—wars and
conflicts into the international public sphere has never been more important
than now. One can no longer be sure where or when a terrorist will strike, and
why. The distinction between “our” terrorists and “their” terrorists can scarcely be
maintained any longer.
At least in theoretical terms, the value of human lives lost in wars in Africa or
Asia and elsewhere cannot be lower than those lost on September 11. The
absence of news, or according low priority to the many “invisible” wars and con-
flicts, means that they do not figure in the public agenda, within national public
spheres, or at the international level. The media is a key factor in conflict resolu-
tion efforts, but one first needs to be aware of something before public opinion is
created and authorities compelled to “do something.” As Wolfsfeld observed:
“The very fact that policy makers and citizens are encouraged to think about
some challenges rather than others is, by itself, likely to affect the allocation of
public resources and how people relate to the political world” (1997: 13).
Selectivity is built into the processes and dynamics of news production. This
is evident at various levels: when the Western media routinely ignore wars and
conflicts in the developing world, and also within national settings when elite-
controlled media systems, equally routinely, ignore conflicts involving minorities.
Violence and terrorism have long been privileged as key news values, but they
remain hostage to the defining news value of cultural proximity. Wars and con-
flicts are not intrinsically newsworthy; they need to be culturally proximate
enough to become news, in international as well as national settings. Even a high
degree of violence as part of serious conflicts may not always qualify as news if it
does not involve the “right” perpetrators or victims in the “right” geographies of
culture. For every conflict that is well covered or analyzed, there are countless
that remain unreported, ignored, or marginalized. It is nonetheless true that con-
flicts in the developing world may figure intermittently on the international
media’s radar, and more often within national settings, but there still remain
scores of wars and conflicts that do not make it to the news columns or television
bulletins even within national settings.
207
P R A S U N S O N WA L K A R
Journalists are invariably drawn from the “national mainstream” and circularly
cater to this section of society and its value system. Thus, events and issues that
do not fall within the paradigm of interest to this section are unlikely to be con-
sidered by journalists as newsworthy. Even if wars and conflicts involving the
“other” meet the “most salient, operational news value of violence” (Hall 1981:
237), they may still not resonate in the media. Journalists breathe the same dom-
inant socio-cultural air of the “national mainstream,” and this cannot but
influence their work, notwithstanding the demands and claims of being profes-
sional, objective, or ethical in their routines.
The fact remains that, despite the real and stated commitments to multi-
culturalism in democratic societies, at the grassroots level, the existence of a
socio-cultural binary of “us” and “them” or “we” and “they” is very much a reality
and can scarcely be denied. This is true not only in Western societies but also
within multicultural countries such as India. But this socio-cultural binary is not
a given; it is created historically during processes of nation formation when
majorities and minorities are manufactured. The binary also shifts according to
focus, locale, and perspective (Shils 1975; Elliott 1986; Elias 1994), and deeply
influences news production. I argue that at its most fundamental level, news must
essentially be about us, and that even though the contours and constituents of
“us” usually remain amorphous, journalists always have a clear conception of
what will interest “us.” As Alistair Hetherington, former editor of the Guardian,
put it, the instinctual news value of most journalists simply is: “Does it interest
me?” (1985: 8).
In communications research, race has been the prime node through which
“othering” in news discourse is examined. Several studies have highlighted the
way reporting, particularly in tabloids, panders to racial prejudices and attitudes.
Racism has become so commonsensical that it has almost become banal, to use
Billig’s term (1995). As Hall argued, there is a racist “common sense” that per-
vades British society: “Since (like gender) race appears to be ‘given’ by Nature,
racism is one of the most profoundly ‘naturalized of existing ideologies’”
(1990: 9). The non-white sections of population come to be routinely seen as the
“they” of British society; a position that is so ingrained as to be unquestionable.
Announcing a special series on British Islam, soon after September 11, the
Guardian asked in a full-page insertion with a veil in the background: “How much
do we know about them?” The paper obviously did not feel the need to identify
who it considered to be the “we” and “them.” The question precluded the possi-
bility that British Muslims might also be its readers.
Hartley (1992) analyzed a similar situation with regard to the representation of
Aboriginals in Australia, and he suggested that journalistic strategies of inclusion
and exclusion had become so ingrained and naturalized that they had become part
of the “common sense” or what Shils (1975) calls the “central value system” of a
society. In India, a similar situation is evident in relation to “caste.” The domina-
tion of the upper caste or Brahminical order of things, or the events and issues
involving the Hindi-speaking people, is so pervasive as to make it commonsensical
208
O U T O F S I G H T, O U T O F M I N D ?
and natural.
Another node of “othering” is at the level of nation, which is also a selective
project. The discourse of nations and nationalism is premised on several assump-
tions about common myths and historical memories, a common, mass culture,
a historic territory, common legal rights, and duties of members, etc. The
“we”/“they” dynamic is central to nation formation and in the case of young
nations, to nation building. It assumes a version of hegemony by which one view
of society is made to appear as the “natural” order of things, beyond rational ques-
tioning, which may completely delegitimize or even obliterate alternative
versions. As Nag observed:
Nations have always been concerned about “us” against “them.” Nations
are obsessed with “self” and discriminate the “other.” The construction
of the national self has always been only vis-à-vis the “other.” The basis
of such construction is differentiation. The “self” consisted of people
who share common cultural characteristics and such commonalities
could be measured only by contrasting against those who do not.
(Nag 2001)
Nossek referred to the socio-cultural binary, but in a discussion about the cover-
age of foreign political violence:
But this binary may well be conceptually applied within national settings, where
journalists decide instantly whether events and issues, and also wars and conflicts,
are “ours” or “theirs.” Applying the binary to this Indian case study, I demonstrate
how prolonged insurgencies in the northeastern part of the country do not always
qualify as news while similar events and issues in Kashmir are routinely privileged
in the discourse of the English-language national press based in New Delhi. The
focus here is less on the better-known Kashmir conflict than on the weakly res-
onating insurgencies in the ethnic cauldron of India’s northeast.
I first present an overview of lesser known wars and conflicts across the globe
and then outline some basic characteristics of the Indian press and emphasize its
209
P R A S U N S O N WA L K A R
centrality in modern India. The subsequent section looks at key features of the
two conflict zones. Then interviews with senior Indian journalists reveal the
salience of the binary and the differences and disjunctures in their perceptions
toward the two conflict zones. Both bear similarities in terms of the nature and
intensity of violence, secessionist goals, foreign involvement, use of sophisticated
arms, degree of local alienation, challenge to the nation-state, etc. In other
words, Kashmir and the northeast demonstrate “comparable message potential”
(Entman 1991: 9) for news coverage. The account is also based on my experience
of covering the northeast conflicts and Kashmir for The Times of India and other
publications over a decade.
210
O U T O F S I G H T, O U T O F M I N D ?
nia, Kashmir, Tamil separatism in Sri Lanka or conflicts in the Middle East. How-
ever, they constitute but a fraction of the number of ongoing conflicts, given that
an overwhelming majority of the 189 members of the United Nations are non-
homogenous and internally diverse nation-states, and that many face conflict
situations in various forms.
The power geometry of international relations ensures that conflicts involving
major actors such as the United States, western European states, or NATO are
extensively covered by the news superpowers based in the West. But the situation
is different in relation to the large number of ongoing conflicts in the developing
world; many of them now involve challengers with an international reach. In
Britain, the only media focus on “invisible” conflicts is the Channel 4 series,
Unreported World, though it does not deal only with conflicts and routinely
records low viewership owing to its late night telecast.
The international media largely ignored the tinderbox situation in Afghanistan
when the Taliban took over in 1996, but provided blanket coverage only after Sep-
tember 11 happened. The Western media woke up to realities in Afghanistan only
when “we” were attacked. Until then, only a few journalists such as Robert Fisk
reported on the depredations of the fundamentalist regime in a country that had
already been ravaged by the Soviet occupation. By late 2003, in another example of
Western news superpowers pursuing the foreign policy of their governments, atten-
tion quickly shifted from Afghanistan to Iraq, amid pious declarations that this
time the West would not “let Afghanistan down.” Except for stray news,
Afghanistan had soon ceased to be a story in much of the Western media.
211
P R A S U N S O N WA L K A R
212
O U T O F S I G H T, O U T O F M I N D ?
the northeast when the Red Army overran the Indian state of Arunachal Pradesh
and entered Assam, before pulling back.
Kashmir and the northeast figure prominently in India’s official list of “prob-
lem areas” (UMHA 2003). Kashmir is part of the federal state that is officially
called “Jammu and Kashmir,” while the northeast region is a conglomeration of
seven states: Assam, Arunachal Pradesh, Mizoram, Manipur, Tripura, Nagaland,
and Meghalaya. The state of Jammu and Kashmir is India’s only Muslim-majority
province, and by the logic of British India’s partition in 1947 into Hindu-
majority India and Muslim-majority Pakistan, it should have become part of the
Islamic state of Pakistan. But Kashmir’s former Hindu ruler opted to join India in
the face of invasion by Pakistani troops immediately after independence, and
modern democratic India has since tenaciously held on to the picturesque valley
as the prime symbol of its commitment to secularism. Its presence in India is also
sought to serve as a symbol of assurance that a Muslim-majority province can
thrive in a Hindu-majority India, which anyway has more than twice as many
Muslims as in Pakistan.
The Kashmir conflict involves two of the largest religious communities in
South Asia: Hindu and Muslim, who have a mixed history of close interaction
and attrition over the years. Their issues dominate mainstream public discourse
in India. Kashmir has been the most contentious issue between India and Pak-
istan. As Bose (1999) observed,
Both sides hold on to their respective positions with equal tenacity. Among the
factors driving India’s position is the not unrealistic apprehension that
Kashmir’s exit from the Indian Union would set off powerful centrifugal
forces in other parts of the country. [The] stakes for both states involve
far more than territorial claims: the question of control of Kashmir goes
to the very basis of the state-building enterprise in South Asia.
(Ganguly 1996)
213
P R A S U N S O N WA L K A R
Source: Ajay Sahni, “Survey of conflicts and resolution in India’s northeast,” in Faultlines, vol. 12.
the grip of more than one form of conflict: secessionist insurgency, separatism
within India, struggles for local autonomy, inter-, and intra-tribal clashes, locals
versus “outsiders,” locals versus immigrant Muslims from Bangladesh, language
tussles, boundary clashes among the seven states, etc. The recent calendar of
casualties in the region runs into thousands (see Table 11.1).
The conflicts are played out within the overall context of the region’s histori-
cally tenuous links with mainland India. Its people bear a deep sense of alienation
from the rest of the country; the feeling is strong that “India,” “New Delhi,” and
the “Central Government” are solely interested in the region for its tea, oil, and
other natural resources. Until 1972, it was India’s External Affairs Ministry that
administered two of the region’s seven states, Nagaland and Arunachal Pradesh.
A key fact central to the “otherness” of the region is that British rule spread there
nearly a hundred years after the British had subjugated the rest of India which
had implications for its participation in the freedom struggle and subsequent inte-
gration in modern India.
Owing to prolonged unrest, the northeast is often referred to as “India’s Ulster”
or as a potential “East Timor” (Dutt 1999). At least 50,000 people have been
killed in insurgent violence since 1947, of whom 25,000 alone perished in vio-
lence linked to the Naga insurgency, which is as old as the Kashmir conflict.
Since World War II—when Kohima (capital of Nagaland) and Imphal (capital of
Manipur) were prominent theaters of battle between the Allied powers and the
Japanese—“the political scene in north-east India . . . has never been placid”
(Singh 1987: 22). Tribes were apprehensive of the vast Hindu–Muslim popula-
tion in the rest of India, and they had heard that Hindus revere the cow, that
they would ban cow slaughter and deprive them of their food: beef. The sense of
“otherness” was constructed at various levels, including religious, ethnic, and cul-
tural (including culinary).
In human geographical terms, the region is literally peripheral to the project of
modern India (see Figure 11.1). Less than one percent of the northeast’s boundary
is contiguous with mainland India, while the remaining 99 percent form India’s
international borders with China, Bangladesh, Myanmar, and Bhutan. Kashmir
214
O U T O F S I G H T, O U T O F M I N D ?
Figure 11.1 Kashmir and northeast conflict zones in India (courtesy of Indo-Asian News
Service)
and the northeast are located at opposite ends of the Himalayan range. The north-
east covers a land area of 255,037 square km (approximately the combined size of
the United Kingdom and Ireland). The northeast’s political representation is 24 in
the lower house of parliament (Lok Sabha: the House of the People) that has 545
members.
The “otherness” of the northeast is acknowledged in Indian public discourse.
The Constitution recognizes that the region is “special” and needs statutory
215
P R A S U N S O N WA L K A R
protection; a clutch of laws and regulations has been enacted for the purpose, and
new states were created to meet tribal aspirations. But as the Planning Commis-
sion, a federal government department, puts it: “The Northeast tends to be seen
as a distant outpost, some kind of land’s end” (1997: 2). Federal civil servants are
loathe to serve in the region.
On the other hand, Kashmir has an evocative resonance in the mists and
myths of early Hinduism; it is also widely known for its pacific brand of Islam:
Sufism. Unlike the northeast, Kashmir has inspired much poetry and Bollywood
themes over the years. The low status of the northeast is driven by an “absence of
cultural and psychological integration with the mainstream” (Datta 2000). In
mainland India, inhabitants from the northeast with Mongoloid features are
viewed as strangers or foreigners. As Horam (1988: 64) observed: “As soon as a
Naga crosses north-east India and ventures westwards, he is mistaken for a Thai,
Cambodian, Chinese or Japanese and treated as a foreigner . . . Even Indians fail
to recognize them as Indians.”
The region’s “otherness” invokes themes such as civilized (mainstream)–
barbaric (northeast), developed–undeveloped, cultured–uncultured, etc. The
pejorative attitude toward tribes is ingrained in the mainstream psyche domi-
nated by the Hindu–Muslim dynamic. The situation is not dissimilar to what
Hartman and Husband observed in the British context; that there are elements in
the British cultural tradition that are derogatory to non-whites (1981: 274). Most
conflicts in the northeast involve tribes who have close knowledge of the difficult
hilly terrain and are known for their fierce martial prowess.
Modern India’s determination to preserve territorial boundaries inherited at the
time of independence prompted the state to deploy the army, air force, and other
security forces in the northeast in large numbers. Owing to prolonged unrest, sev-
eral laws have been enacted, many dubbed “draconian” by human rights groups.
Violations of human rights during counter-insurgency operations often worsen
the ground situation but, as journalists admit during interviews, they are hardly
deemed worthy of reporting in the newsrooms in New Delhi. Issues and events in
Kashmir or other areas that involve or affect the Hindu–Muslim problematic con-
tinue to be privileged in news discourse.
216
O U T O F S I G H T, O U T O F M I N D ?
there was a strong feeling in newsrooms that “there is no interest among readers;
somehow the region, its peoples and issues do not seem to matter to us” (YV).
DD, a journalist and human rights activist working in Nagaland, said:
According to YV, the instinctual feeling about the northeast among journalists is
that “whatever happens there, somehow does not affect us and our readers.”
Kashmir, on the other hand, is routinely seen as “something that deeply affects
us” (VVPS). Journalists see tribes in the northeast as “primitive,” which carries
the implication that the non-tribal section of India’s population is “modern.”
DAT, a New Delhi-based special correspondent of The Times of India, was not
even aware if the paper had a correspondent in the northeast, because he did not
remember the last time the paper had any report from the area (it was later dis-
covered that the paper did have a correspondent there).
DPS said he often had to struggle to get his northeast-related stories published
in his paper. His seniors would ask: “Who is bothered, why do you bother?” He
said he once had to explain to his colleagues that “Kuki” was not a biscuit but
the name of a major tribe of Manipur, where hundreds of people had been killed
in clashes with Nagas. Indian culture, he said, was still largely considered Aryan
culture while the Mongoloids of the northeast were considered alien. He would
like to see the region covered extensively, like Kashmir, but he said he felt help-
less owing to “complete lack of interest” among his seniors and colleagues.
Three journalists explained how the low status of tribes and the northeast
region is part of the cultural air in which they operate:
Right from an early age, we are not made aware of our country’s ethnic
diversity. We only know of Bengalis in eastern India, nothing beyond.
Most of us will not be able to tell a Naga from a Mizo. The attitude
towards the tribes is that they are a necessary evil, so all we have to do is
to hold on to them and their territory by force. The feeling is that we are
superior—the concept of Aryan supremacy. The image is that people
there drink, eat all kinds of meat, they are amoral Christians who
believe in polygamy and who fight bitterly. In sum, it is a society we
don’t need to know much about. The Indian army is there; give them
two slaps and they will be quiet.
(NM)
217
P R A S U N S O N WA L K A R
Prejudices are very deep even among highly paid officers. The attitude is
that they are jungle people, and the same is the attitude among media
men. There is discrimination by rulers, ministers, bureaucrats, and intel-
ligence agencies. My colleague wanted to go to the northeast for his
honeymoon; he became a major target of ridicule in our newsroom. The
northeast holds no appeal; also, it is too distant. It is remote to me cul-
turally, I do not relate to their languages, culture, or race. They are all
chinks.
(VVPS)
The northeast’s low status is fed into the sub-conscious right from child-
hood in north India. In folktales, at least in eastern Uttar Pradesh where
I come from, it is portrayed as a region of women from where one never
comes back, or as a region that is not worth going to. On the other
hand, Kashmir has the opposite place. It is a place that everyone aspires
to visit before dying. So when something happens or goes wrong there,
the people are disturbed. The poetry of great [Hindi] poets such as
Dinkar or Pant resonates with Kashmir imagery. Kashmir is considered
the crown, the most important part of the body. The northeast too has
the same importance, but there is no awareness of this.
(RBR)
218
O U T O F S I G H T, O U T O F M I N D ?
used to cover it. Due to its complexity, the attitude of newspapers is: this
is a beehive, why should we put our hands in it, why should we bother?
We don’t know, we don’t cover.
In such an environment, on the rare occasions that news from the region is pub-
lished, it is invariably marked by errors: incorrect names of state capitals or names
of important tribal politicians/actors or incorrect mention of constitutional provi-
sions for the region. According to AJP, “among journalists, there is no stake in
the northeast. They often go wrong on facts; worse, nobody notices it.”
The Kashmir conflict, with its relatively simpler Hindu–Muslim–Pakistan
dimension, is considered “sexy” (DAT) to cover and ensures prominent exposure
to journalist output. The conflicts in the northeast are numerous and complex; it
is impossible to report them without basic knowledge about the kaleidoscope of
tribes, religions, languages, and the plethora of constitutional and legal provisions
for the region. News reporting demands that events are conveyed in a simple and
easy to understand manner, which may preclude lengthy explanations. As HC, a
veteran journalist who covered the northeast in the 1950s, remarked, journalists
like to take the easy route by ignoring complex events and issues.
ZA, former northeast correspondent on The Times of India, hinted at an Orien-
talist frame that governed coverage of the region:
The northeast has better chances of figuring on the radar of the national press if
news reports are seen as “juicy, sexy, or exotic” by the Indian mainstream. There
is clearly no reason to believe that the notion of Orientalism is confined only to
the West–Middle East dimension. As the interview responses indicate, this sense
of the “other” can also be constructed within the Orient.
The routine non-coverage or low coverage of the northeast in the national
press adds to the local sense of ennui in the region. AB, editor of an Assamese
daily, said:
219
P R A S U N S O N WA L K A R
when so many get killed. They do not play the role they should; they are
responsible for this chaotic situation here. The rulers and the intellectu-
als based in New Delhi have no time for us; nobody bothers. The Delhi
press also supports them. At least write some facts. At least reflect 50
percent of the situation here. Now not even five percent is reflected in
the Delhi press. There is no democratic reaction in other parts of the
country to events happening here, because no news from here is pub-
lished. They don’t know, so they don’t react, so no pressure is put on the
government. Assam is seen as distant, as a forest area, as if people here
are not civilised or cultured.
Conclusion
If acts of violence and terrorism were so irresistibly newsworthy, as the interlock-
ing discourses of violence/terrorism/conflict and the media suggest, the host of
insurgencies in northeast India should have resonated at least as much as the
single insurgency based in Kashmir, if not more. Violence and terrorism have
been long privileged in the production of news, while the trope of the margins
has been mostly confined to the politics of representation. Research has rarely
examined the non-coverage of events and issues involving the minorities or the
margins. This gap assumes more salience owing to the fact that the movement of
men, material, and money to finance terror knows no national borders, as events
in Iraq, Afghanistan, and Turkey show. If news discourse is seen as being by the
elite, for the elite, and of the elite, the exclusion of minorities’ life situations is
likely to be institutionalized in newsrooms.
As the example of northeast India shows, even when insurgent outfits resort to
220
O U T O F S I G H T, O U T O F M I N D ?
References
Allan, S. (2001) News Culture, Buckingham: Open University Press.
Appadurai, A. (1996) Modernity at Large: Cultural Dimensions of Globalisation, Minneapo-
lis, MN: University of Minnesota Press.
Billig, M. (1995) Banal Nationalism, London: Sage.
Bose, S. (1999) “Kashmir: sources of conflict, dimensions of peace,” Economic and Political
Weekly, March 27.
Brown, M.E. (2001) “Preface” in Nationalism and Ethnic Conflict, Michael E. Brown et al.
(eds), MIT Press: Cambridge, MA.
Chenoy, A.M. (2002) “Militia mentality,” The Times of India, September 17.
Datta, S. (2000) “Northeast turmoil: vital determinants,” Strategic Analysis, March, vol.
XXIII, no. 12, pp. 2,123–33.
Dutt, J.K. (1999) “An East Timor in India?,” The Statesman, November 12.
Elias, N. (1994) The Established and the Outsiders: a Sociological Enquiry into Community
Problems, London: Sage.
Elliott, W.A. (1986) Us and Them: a Study of Group Consciousness, Aberdeen: Aberdeen
University Press.
221
P R A S U N S O N WA L K A R
222
O U T O F S I G H T, O U T O F M I N D ?
Smith, A. (1980) The Geopolitics of Information: how Western Culture Dominates the World,
New York: OUP.
Sonwalkar, P. (2002) “‘Murdochization’ of the Indian press: from by-line to bottom-line,”
Media, Culture and Society, vol. 24, no. 6, pp. 821–34.
Tuchman, G. (1972) “Objectivity as strategic ritual: an examination of newsmen’s notions
of objectivity,” American Journal of Sociology, vol. 77, pp. 660–79.
UMHA (2003) website of the Union Ministry of Home Affairs, Government of India,
http://mha.nic.in/scen.htm accessed August 28, 2003.
Viswanath, K. and Karan, K. (2000) “India,” in Handbook of the Media in Asia, Shelton
A. Gunaratne (ed.), London: Sage.
Weimann, G. and Winn, C. (1993) The Theatre of Terror: Mass Media and International
Terrorism, London: Longman.
Windmiller, M. (1954) “Linguistic regionalism in India,” Pacific Affairs, vol. 27, no. 4,
December 1954, pp. 291–318.
Wolfsfeld, G. (1997) Media and Political Conflict: News From the Middle East, Cambridge:
Cambridge University Press.
223
12
THE BATTLEFIELD IS
THE MEDIA
War reporting and the formation of national
identity in Australia—from Belmont to Baghdad
Michael Bromley
The invasion of Iraq by the so-called “coalition of the willing,” led by the USA in
March and April 2003 was arguably the most reported military conflict in history.
More than 3,000 correspondents, ranging from backpack journalists working
alone with wireless digital equipment to television anchors and their production
crews, were assigned to cover events across the Middle East. Such intensity of
focus on limited military action which, unlike the “total war” experiences of most
of the twentieth century, involved non-combatant Western domestic populations
only indirectly, had been building since at least the Vietnam War, reaching a pre-
vious historical peak during the earlier Gulf conflict of 1991 (Tiffen 1992b: 44).
This reflected two apparently contradictory long-term general trends. On the one
hand, those holding power were made more accountable to the public, and the
capacity to call them to account has increased exponentially (Tiffen 1992a:
118–19). Part of this was exhibited in the extent to which public life was shaped
by the routines, rituals, demands, and expectations of journalism (Blumler and
Gurevitch 1996: 127–8). As a consequence, individual journalists were less easily
controlled; for example, the appointment of official war correspondents, particu-
larly during the World Wars of 1914–18 and 1939–45, and the censorship
inherent in the system, foundered on the work of journalists like Wilfred
Burchett, whose eye-witness account of the effects of the bombing of Hiroshima
in 1945 was unsanctioned by Pacific Command, and James Cameron, who
angered the British government by reporting the brutality of Republic of Korea
forces during the Korean War in the 1950s.
Parallel to this growing claim to a freedom to report, the ethos and means of
public communication diffused far more widely and deeply in society (Hartley
224
T H E B AT T L E F I E L D I S T H E M E D I A
2000: 41). Thus, the myths, images, and ideas of war were shaped for the past 150
years by both highly professionalized, technocratic journalists and the accounts
captured in letters, diaries, photographs, etc. of ordinary people (McCrum 2001).
On the other hand, the management of public communication (“spin”) gave rise
to the notion of “the public relations state” (Deacon and Golding 1994: 5–6).
States and the military tried especially to curtail journalists’ freedoms in and
around combat zones, and to neutralize the inquisitive, perhaps even mischie-
vous, roving reporter, exemplified by the logistical and psychological controls
exerted on British correspondents during the Falklands War in 1982 (Morrison
and Tumber 1988). A system of “pooling” enforced by the US in and around the
Gulf in 1991 contributed decisively to the ways in which combat was presented
in the media. Moreover, the professionalization of the military worked to elimi-
nate unofficial first-hand combatants’ accounts, although not always successfully
(McCrum 2001).2 As a result, how wars are reported became a highly contested
extension of the military action, and possibly even of greater importance than
much of the combat itself, presenting “an increasingly important second front”
(Tiffen 1992b: 44, 55).
Australia’s involvement in the 2003 invasion of Iraq resonates with many of
these broader issues. Australians (originating in white, European colonization
and settlement) fashioned a sense of national identity through participation in
war around the foundation of the State at the beginning of the twentieth century.
Reporting and recollecting war were crucial to this project. The deployment of
military personnel in 2003, therefore, had enormous potential symbolic signifi-
cance, and the main mechanism for realizing this potential was the way in which
Australia’s journalists recorded events for the public at home. This juxtaposition
of combat and the reporting of combat helped unravel the conundrum of Aus-
tralia’s minimum military contribution to the “coalition,” and how that was
massaged into a major international intervention.
225
MICHAEL BROMLEY
from the highly secretive, so-called special forces (Defence Ministry 2003;
Marcus 2003). At the outset, Howard insisted that Australia would send no
infantry or other types of “ground troops.” This reflected a more general shift to a
higher reliance on secretive special forces in the ADF, their numbers having dou-
bled in the five years to 2000 (Ferguson 2003), making Australia’s army more
specialist and less citizen-based. “In modern warfare,” Howard said in a radio
interview, “a lot more can be achieved with highly mobile technologically profi-
cient but small numbers of troops” (Radio 4BC 2003). At the same time,
Australian politics have also become more controlled and secretive, and increas-
ingly reliant on media management (Phillips 2003).6 As a consequence,
according to the former diplomat Alison Broinowski (2003: 10–11), Australians
were “confused” and “unclear” about the country’s involvement in the invasion
of Iraq. In September, 67 percent of 1,200 people surveyed for an Australian news-
paper poll believed that the government had misled them on the issue of Iraq’s
possession of WMD (a key argument in support of the invasion), and 36 percent
felt that the deception was deliberate (ABC Online 2003b). Furthermore, during
the actual fighting in Iraq, the Australian government and military maintained
an almost total silence on operational matters.
The role of journalists in interrogating Australia’s participation in this military
conflict might have been expected to assume a particular importance, then. As
The Australian’s media commentator pointed out
Many Australian journalists complained that they were prevented from doing
their jobs by military and civilian officials who imposed stringent restrictions on
reporting (Cohen 2003; Mottram 2003). An Australian parliamentary analysis
agreed that “Australians [were likely to] receive an overwhelming amount of
information from American sources . . . but little information about Australia’s
contribution to the conflict” (Miskin, Rayner, and Lalic 2003).
226
T H E B AT T L E F I E L D I S T H E M E D I A
A majority of Australian correspondents and their news crews were actually scat-
tered elsewhere in the Middle East—“distant from the action . . . milling about”
(Wright 2003). The most prolific sources of information on Australia’s involve-
ment in Iraq were said to be the military briefings taking place in the Australian
federal capital, Canberra. The popular Australian perspective on the invasion of
Iraq is likely to have been impacted, therefore, not just by the media per se
(Keeble 1997), but by the media operating at a considerable distance from any
direct experience of Australian participation (a dimension perhaps captured in
Nicholson’s cartoon—see Figure 12.1).
The invasion of Iraq appeared to many Australians to be no more than a “sym-
bolic war,” therefore (George Munster Forum 2003). In this way, it was fought in
the Australian media as much as, if not in some ways more than, on the ground.
As international tension mounted early in 2003, a former editor-in-chief of the
Sydney Morning Herald newspaper identified the media and journalism as equally
culpable with politics for widespread public cynicism and disbelief, which could
be countered only by journalism based on “moral courage” (Bowman 2003). Yet
the media seemed predisposed to only one view of the conflict. Roy Greenslade’s
(2003) revelation that all 175 newspapers owned by the global media proprietor
Rupert Murdoch, including seven of the 12 national and metro dailies in Aus-
tralia, espoused editorially his personal support for the invasion of Iraq, was
widely cited in Australia (Jackson 2003a).
227
MICHAEL BROMLEY
Figure 12.1 “So this is what war is really like!” Nicholson cartoon for The Australian,
March 20, 2003 (courtesy of the artist)
228
T H E B AT T L E F I E L D I S T H E M E D I A
Newspapers
In 2000 News Limited controlled 68 percent of major daily newspaper circula-
tions, nearly 60 percent of Saturday papers sold, more than three-quarters of
Sunday paper sales, almost half of the market in suburban weekly newspapers,
and nearly a quarter of the regional press. Fairfax newspapers accounted for 21.5
percent of the major daily sales, more than a quarter of Saturday circulations,
nearly a quarter of Sunday sales, 18 percent of the suburban press, and 15 percent
of regional papers. Only three other companies (APN News and Media and Rural
Press with 30 percent and 15 percent respectively of regional sales, and West
Australian Newspaper Holdings with about 10 percent of both the metro daily
and suburban markets) have significant press holdings, which in total amount to
sales of no more than about 1.35 million of about 13 million newspapers sold
(2–3 million national and metropolitan dailies, 620,000 regional dailies, 3.5 mil-
lion Sundays, and 6.4 million suburban titles). Although concentration of
ownership and control has been a feature of the Australian press since the second
decade of the twentieth century, the trend accelerated markedly after the mid-
1980s. Between 1987 and 2000 News Limited alone closed ten newspapers (Lewis
2001: 101–6).
Television
PBL runs Channel 9, Australia’s largest national television network, with more
than a 30 percent share of the free-to-air audience, and nearly a quarter of all
television audiences. PBL also operates three metropolitan and one regional tele-
vision licenses (Cassar 2003). Australian television is dominated by three
free-to-air commercial services which in mid-April 2003 accounted for 77.4 per-
cent of prime time audiences, and more than two-thirds of all viewing (source:
OzTAM in Jackson 2003c). Channel 7 (controlled by Kerry Stokes) and Chan-
nel 10 (the Canadian conglomerate CanWest) each has about a quarter of the
free-to-air audience and a fifth of all television audiences. PBL and News Limited
share equally a half stake in the main paid-for television supplier, Foxtel, and PBL
has a 33 percent share of News’ subscription service, Sky News (Given 2000:
43–5; Cassar 2003). In 2003, pay-TV had about 13 percent of the 24-hour view-
ing audience (Murphy 2003b). The ABC, based on the BBC model, and the
Special Broadcasting Service, a public service oriented multicultural and multi-
lingual broadcaster introduced in 1980, share about a fifth of prime-time
television audiences (16 percent and 4 percent respectively), and 17 percent of
the 24-hour audience. This situation has provoked some to ask whether public
service television in Australia is “an endangered species” (Jacka 2000).
Eighty percent of Australians get their news from television, and more from
the Channel 9 network than any other source (Flew 2002: 177). Editions of
National Nine News regularly head the ratings (with audiences of around 2 mil-
lion), and the 60 Minutes current affairs program is routinely in the top 10.
229
MICHAEL BROMLEY
Nevertheless, since the late 1980s there has been “a major and potentially cata-
strophic decline in network news and current affairs audiences” (Turner 2001:
56). These shows have been positioned as commercial “flagship formats” built on
prioritizing the credibility of personality presenters rather than investments in
news gathering, and on tabloidization (Turner 2000: 91–2). All the same, there is
evidence that crises (such as the invasion of Iraq) temporarily boost ABC news
viewing (Tiffen 1991: 10–11; Jackson 2003c).
Radio
The main owners control more than three-quarters of radio broadcasting over
250 commercial stations (Miller and Turner 2002: 145). Austereo, the largest
commercial radio broadcaster in Australia operating two national networks each
with five stations, is also the biggest outside the US (see www.austereo.com.au).
The British newspaper and broadcast company, DMGT, has 63 radio stations in
Australia (see www.dmgt.co.uk), while the Australian Radio Network (12 sta-
tions) is part of APN (see www.apn.com.au). All three have various partnership
arrangements with the US Clear Channel company. The conglomerates have
been heavily criticized for cutting costs, cutting jobs, and cutting news to the
extent of sacrificing “public interest objectives [such] as plurality, diversity and
competition [which are essential] in order to ensure a free, vigorous and indepen-
dent media sector” (IRB 1999). Music formats predominate, but “talkback” radio
has considerable currency: journalists and news producers rank it among the four
most important news and current affairs services—in part because it is viewed as
“a broader litmus test of community opinion” (Pearson and Brand 2001).
Attempts to dismantle the ABC’s radio news and current affairs service failed in
the late 1990s (Miller and Turner 2002: 144). In 2003 ABC operated a range of
about 50 national, regional, and local stations, including Radio National and
NewsRadio (Dunn 2001).
Magazines
The five largest magazine publishers control more than 80 percent of audited
sales; the top two alone nearly 70 percent. PBL, through its subsidiary Australian
Consolidated Press, has 45 titles with sales of more than 53 million—nearly 50
percent of the market. Stokes’ Pacific Publications (9 titles) accounts for 21.5
percent. The smallest of the “big five,” Murdoch Magazines (3 titles) has a 2.4
percent share (Jackson 2003e).
In sum, the mainstream Australian mediascape amounts to a carve-up, favor-
ing a small coterie of commercial operators. Griffen-Foley (2003) in particular
has traced a close, at times corrupt, relationship between Australian politicians
and the country’s powerful patrician commercial media. This situation has been
characterized by a general, and perhaps unsurprising, lack of journalistic indepen-
dence, or editorial vigor—the modest supposed exceptions being a brief, and
230
T H E B AT T L E F I E L D I S T H E M E D I A
ultimately largely unsuccessful, attempt between the late 1960s and the mid-
1970s to assert “professional standards of integrity” over the growing partisanship
and self-interestedness of the less than a handful of key media proprietors, and a
more vigorous pursuit of “the Fourth Estate role of the news media” between 1980
and the mid-1990s (see also Schultz 1998: 195–229). Of all the media, newspa-
pers seem to exercise the most influence on public views of politics with
television some way behind, and talkback radio a further distance off. Surveying
so-called Middle Australia, Pusey (2003: 127–32) concluded that media audi-
ences were more or less polarized between those who valued a critical,
independent Fourth Estate operating in the public realm, and those who turned
to the media for projections of their own social prejudices, resentments, and pri-
vate anxieties. Responding to suggestions that politicians had lied to justify the
invasion of Iraq, a British correspondent wrote that Australians preferred “to con-
centrate on their desire for good times” (Fickling 2003b).
231
MICHAEL BROMLEY
masking human suffering—at least to the extent that the Australian government
felt that media coverage in 2003 had dwelt too much on “the awfulness of the
war” (Dodson 2003b). One commentator observed that “nearly non-stop live
television of a kind never seen before” was delivering “footage . . . of the battle for
Iraq [which] is so pore-close, graphic and in-spite-of-yourself compelling it has
been dubbed ‘war porn’” (Jackson 2003b). Events in 2003, no less and possibly
more, than in 1991, were subjected to “saturation coverage”—from SMS text
news messages supplied to mobile phones by Channel 9 to the post-war recollec-
tions of The (Melbourne) Age newspaper’s journalistic team published as a colour
supplement. In the first four days of fighting, Channel 9 aired 56 hours of cover-
age, and the ABC 41 hours (Agenda 2003; Media Watch 2003c). Several
newspapers published special editions, the Sydney Daily Telegraph, for one, even
temporarily resurrecting its otherwise defunct afternoon edition (Meade and
Brook 2003).
Viewerships for television news reportedly rose immediately on the afternoon
of March 20 by between 22 percent (ABC) and 196 percent (SBS) over the pre-
vious week. Channels 9 and 7 posted 64 percent and 58 percent rises respectively,
although their prime time evening newscasts had more modest increases of 12–13
percent. SBS’s evening bulletin audiences rose by 15 percent and 30 percent and
those for the morning shows the following day (Sunrise on Channel 7 and Today
on Channel 9) by 30 percent combined. However, the stations began to return to
their normal schedules after the first week of fighting, and most of the gains were
lost, except by the ABC (Murphy 2003a). Tens of thousands of extra copies of
the major newspapers were printed during the opening days of the conflict, but it
was difficult to determine whether these captured additional readers: over the 12-
month audit period including March and April, readerships remained mostly
static (Jackson 2003d). By early April, 62 percent of a sample of the Australian
population surveyed felt that there had been too much media coverage, and only
3 percent that there had been too little (Dodson 2003b).
Debates about the quantity of journalism are almost inextricably tied, however,
to both assessments of the quality of journalism and to the technologies of news.
Tiffen (1992a: 139) makes the point that the earlier military action in Iraq repre-
sented something of a turning-point in these respects in Australia as the ADF
had not been in combat since the Vietnam War. Satellite communications, allied
to intense market competition, among other factors, made feasible in 1991 on-
demand “real time” broadcasting from multiple locations and sources. The same
combination of factors which enhanced journalists’ capacities to monitor govern-
ments also increased the cost advantages of relying on generic, pre-packaged
internationally syndicated feeds, however. The system of “pooling” enforced by
the US in 1991 angered many journalists, and excluded all those from Australia
(Tiffen 1992a: 117ff; Miskin, Rayner, and Lalic 2003). Similar anxieties were
expressed about the system of embedding journalists with US military units in
2003, and the role of the Doha media centre, which Australian correspondents
nicknamed Operation Mushrooms because they suspected that its real purpose
232
T H E B AT T L E F I E L D I S T H E M E D I A
was to keep them in the dark (Callinan 2003). Australian journalists working
outside closely accredited assignments (known colloquially as “unilaterals”) were
far more numerical—and more vulnerable in what was described as a conflict that
was of greater danger to journalists than it was to soldiers (Byrne 2003). The
camera operator Paul Moran, working for the ABC, was killed on March 23 in
northern Iraq in a bomb attack. News Limited’s correspondent Ian McPhedran
was later expelled from Baghdad by the Iraqis. McGeough reflected that the Pen-
tagon, too, did not “like the foreign press being in Baghdad. They spoke about us
quite contemptuously as ‘the independents’” (The Media Report 2003).
These factors appear to be constitutive of a longer-term trend evident in Aus-
tralian journalism, in which reporting from the field, especially overseas, has been
in decline in favor of informed analysis and comment, often originated domesti-
cally and produced locally (Tiffen 1992a: 119, 139). Notwithstanding general
continuing developments in these areas (notably, but not exclusively, digitiza-
tion), it may be conjectured that as far as Australian journalism was concerned,
the changes apparent in 2003, compared to 1991, were fewer, and less significant,
than those which Tiffen noted occurred between 1965 and 1991. Both the scope
and scale of journalism produced in March and April 2003 are far too great to
sample comprehensively with any degree of reliability at this temporal proximity.
Nevertheless, some aspects of the coverage may be gauged from a more or less
impressionistic snapshot analysis of the main newspapers, television, and radio.
Television not only made extensive use of satellite-based communications
(notably the videophone) for “live” reporting feeds, wholesale news exchanges,
and studio links (Jenkins 2003; Wilson 2003), but also broadcast direct and some-
times uninterrupted the output of a variety of American and British news
organizations, such as the BBC, ITN, Sky, PBS, NBC, and CNN, piggybacking
on the greater US and UK news investments. Australian news organizations also
made considerable use of the Internet, trawling the Web chiefly to check on what
other media around the world were publishing and broadcasting (Vermeer 2003).
The reporting of the relatively small number of Australian journalists in the field
was thus heavily supplemented by material emanating from the far larger corps of
American and British journalists and crews assigned to the story worldwide, but
captured and/or edited locally. At the same time, ABC’s Lateline program particu-
larly, but not exclusively, utilized satellite links for extended studio-based analyses
and commentaries. Media demand for expert domestic input was at times “almost
unquenchable” with the Strategic and Defence Studies Centre at the Australian
National University holding briefings daily for the first 16 days of the conflict and
more intermittently thereafter. “We were essentially shaping the news,” said the
organizer (Hull 2003). In this way, major international events were substantively
domesticated.
Even so, the demand for actuality of Australian involvement on the ground
was so great that it offered the Defence Ministry opportunities to “spin” mislead-
ing coverage. SBS, Channel 7 and Channel 10, as well as the Daily Telegraph,
reproduced Ministry images of Australian commandos “deep behind enemy lines”
233
MICHAEL BROMLEY
(Channel 7), which turned out to have been from a photo-shoot staged during
preparations three days before the invasion began (Media Watch 2003f). Such
duplicity was occasionally matched by the media themselves: Media Watch
(2003g) reported that the Channel 7 current affairs show Today Tonight edited
stock footage from Iraq with a report by correspondent David Richardson to pur-
port to show him having penetrated “into the unknown” when he had merely
walked ten metres into the demilitarized zone on the Kuwaiti border. Later both
the Daily Telegraph and the weekly news magazine Bulletin held to the story that
an American flag wrapped by US Marines around a statue of Saddam Hussein in
Baghdad had been recovered from the debris of the Pentagon following the ter-
rorist attack of 9/11, when it had been purchased for $9 US from the US Senate
gift shop (Media Watch 2003j, 2003k).
Tabloidization
Debates about both the processes and effects of tabloidization have been a feature
of and around the Australian media for about 20 years (see Turner 2001). A
sample of the presentation of the opening days of the conflict in a number of
newspapers provides a sense of how the Australian press negotiated the chal-
lenges of engaging their audiences. On March 21 (the first full day of the invasion
in Australian time), the tabloid-sized, but otherwise “serious,” Australian Financial
Review wrapped its usual Monday–Friday five-column, type-dominated front page
with a poster “lift-off” showing gun-carrying troops silhouetted against a fiery
orange and yellow sky. The main headline was simply “WAR.” The tabloid Daily
Telegraph, identifying for its readers the “MOMENT THE WAR BEGAN,” pub-
lished a similar poster front page of a photograph of a missile exploding over
Baghdad above the headline “GET SADDAM.” On March 23, the tabloid Bris-
bane Sunday Mail used a double-page poster wrap around the newspaper to carry a
picture of the presidential palace in Baghdad erupting in flames “as cruise missiles
find their target with deadly accuracy.” The headline was “AWESTRUCK.” The
following day, in a “SPECIAL WAR EDITION” the front page of the broadsheet
Australian was dominated by a half-page photograph of a British marine firing a
guided missile at an Iraqi position. The rest of the page was taken up by text, and
the main headline was a more sober “IRAQ FIGHTS BACK.” In Broinowski’s
view (2003: 111–12), however, the otherwise measured Australian approached
the invasion with a surprising degree of zealotry. Other News Limited papers
offered readers facilities for sending messages to “our troops”: the Brisbane
Courier-Mail’s service remained available on the paper’s website in early October
(www.couriermail.news.com.au). On the other hand, tabloidization also provided
occasional relief from the chorus: on the day the rest of the press reported the
attack on Baghdad, the main story in the Northern Territory News in Darwin was
“Topless woman attacks picnic” (Media Watch 2003h).
The use of “our” (usually in conjunction with “troops”) was commonplace in
television, especially on Channels 7 and 10. The ABC’s Geoff Thompson apolo-
234
T H E B AT T L E F I E L D I S T H E M E D I A
gized for using “we” while reporting on US forces (Meade and Brook 2003). The
executive producer of Channel 9’s Sunday program acknowledged that “News
judgments may occasionally have been skewed by the desire to support ‘our
troops’” (Rice 2003: 12). Giving the news a distinctive Australian slant was taken
to be a high priority in television—even at SBS. Iraq’s time zone meant that
many events were better scheduled for the lighter morning news shows, however.
Substantial numbers of Australians used pay-TV services, with BBC World,
CNBC, Sky News, CNN, and Fox News audience levels doubling or more
(Matthew 2003). Giving Australian journalism a “distinction” grade, Rice (2003:
12) nevertheless indicated the traps into which television coverage could—and
sometimes did—fall:
The way in which the invasion was handled by the ABC’s major networked radio
news program AM raised the most visible debate. The Communications Minister
Richard Alston (2003a) produced a 68-point document detailing complaints
against the program which included allegations of exaggeration, “gratuitous
barbs,” jumping to negative conclusions, mockery, dismissiveness, trivialization,
cynicism, put-downs, “putting the boot in,” taking sides, “immature and irrelevant
abuse,” “unreasonableness, scoffing, making furious attacks,” sarcasm, derision,
vitriol, contempt, and using “strong and pejorative” language—all directed against
the US and the coalition. The corporation’s head of news and current affairs had
referred to the military as “lying bastards” during an international conference
eight months previously (Alston 2003b). The Media, Entertainment, and Arts
Alliance, which includes the Australian Journalists’ Association, condemned
Alston for trying to influence news coverage (ABC Online 2003a). An internal
inquiry by the ABC, while admitting that in two instances AM reports had been
sarcastic and “excessive,” generally cleared the program of bias. “Overall, I believe
our coverage of the conflict was balanced and delivered in a professional manner
upholding the standards of objective journalism,” the ABC’s managing director
said (News.com.au 2003).
235
MICHAEL BROMLEY
the stories we have told each other about Australians at war are power-
ful and resonant. They endure in the national mind. And they were, in
the main, first told by war correspondents.
There is no permanent archive or exhibition in any Australian institu-
tion which does justice to the extraordinary record of this distinguished
band of Australians whose contribution to understanding our history
goes far beyond their dispatches from the front.
. . . Few other countries can boast news men and women of such
renown who have involved themselves in reporting war to the world and
have, at the same time, exerted such a singular influence on their own
country’s political, cultural and literary traditions.
(Rees 2002)
While this tradition can be traced back at least to the work of A.B. “Banjo”
Paterson, a notable devotee and promoter of the bush who reported on the Boer
War for two Australian newspapers (Gerster 1987: 17–18), two journalists stand
out in this regard. C.E.W. Bean, as Australia’s first official war correspondent,
236
T H E B AT T L E F I E L D I S T H E M E D I A
Conclusion
Since the 1990s politicians have contested the right to lay claim to “the Aus-
tralian legend with its working-class roots in egalitarianism, mateship, the fair go
237
MICHAEL BROMLEY
Notes
1 Figure published by the CEW Bean Foundation.
2 During the invasion of Iraq weblogs were used by some military personnel to post
public communication on the worldwide web, and more utilized email for private cor-
respondence. Because of the sheer volume of the material, this chapter does not
address either blogs or websites.
3 Accurate statistics on military deployments in and around Iraq in 2003 were notori-
ously difficult to come by. The Australian government always used the word “about” as
a qualifier. The figures used here are the most reliable available, but should still be
approached with caution. See the Defence Ministry (2003) Web pages devoted to
Operation Falconer.
4 Estimates varied from 70 Australian troops on the ground in Iraq to 800 stationed in
the region in August 2003 (see Banham 2003).
5 Australian news workers also continued to die. At the beginning of July the Aus-
tralian sound recordist for the US television network NBC, Jeremy Little, was
wounded in a grenade attack in Fallujah. He died in hospital in Germany just over a
week later (The Mercury, 2003).
6 Marr and Wilkinson (2003) trace the censorship, secrecy, and misinformation origi-
nating with the Australian government associated with the so-called Children
Overboard affair in 2001.
238
T H E B AT T L E F I E L D I S T H E M E D I A
References
ABC Online (2003a) “MEAA condemns Alston’s ‘intimidation’ of ABC” (nd), posted at
www.abc.net.au/news accessed October 6, 2003.
ABC Online (2003b) “Most Australians feel misled on Iraq: poll” (September 24), posted
at www.abc.net.au/news/ accessed September 24, 2003.
Agenda (2003) “Iraq: stories from the front line,” The Age (June), pp. 1–7.
Alston, R. (2003a) Minister for Communications, Information Technology and the Arts,
media release (May 28), posted as “Alston seeks urgent investigation into AM’s Iraq
coverage” at www.dcita.gov.au accessed October 6, 2003.
Alston, R. (2003b) ‘”Analysis of the AM program’s Iraq coverage” (May 30), posted as
“Alston attacks ABC coverage of Iraq war,” AustralianPolitics, at www.australianpoli-
tics.com.au/news accessed October 6, 2003.
ANZAC Commemoration Committee (2003) “Gallipoli,” posted at www.anzacday.org.au
accessed June 13, 2003.
Banham, C. (2003) “Security force lifts troop numbers to 1000,” Sydney Morning Herald
(May 2), posted at www.smh.com.au accessed August 22, 2003.
Blumler, J. and Gurevitch, M. (1996) “Media, change and social change: linkages and
junctures” in J. Curran and M. Gurevitch (eds) Mass Media and Society, 2nd edition,
London: Arnold, pp. 127–8.
Bowman, D. (2003) “The Fourth Estate: war and moral courage,” Australian Policy
Online (January 1), posted at www.apo.org accessed September 29, 2003.
Broinowski, A. (2003) Howard’s War, Melbourne: Scribe Books.
Byrne, C. (2003) “Watchdog condemns Iraq deaths,” The Guardian (May 2), posted at
www.media.guardian.co.uk accessed September 9, 2003.
Callinan, R. (2003) “Secret service,” The Australian Media (April 3–9), p. 4.
Cassar, E. (2003) “Government granting content control to media proprietors,” Guido-
media.com at www.guidomedia.com.au/erincassar.html accessed November 5, 2003.
Clark, A. (2002) untitled, “The first annual Dymphna Clark lecture,” Manning Clark
House (March 2), posted at www.manningclark.org/papers/DClecture-2002.html
accessed November 7, 2003.
Cohen, D. (2003) “Look back in wonder,” The Walkley Magazine, vol. 22, p. 40.
Crikey.com (2003) “Australian journalists in the war zone” (March 24), posted at www.
crikey.com.au/media accessed September 10, 2003.
Day, M. (2003) “Lies, damn lies and governments,” The Australian Media (June 5–11),
p. 6.
Deacon, D. and Golding, P. (1994) Taxation and Representation: the Media, Political Com-
munication and the Poll Tax, London: John Libbey.
Defence Ministry (2003) Disarmament of Iraq: Operation Falconer, posted at www.defence.
gov.au/opcatalyst accessed September 30, 2003.
Dodson, L. (2003a) “Uniting Australia, the PM’s greatest challenge,” The Age (March 21),
posted at www.theage.com.au accessed June 19, 2003.
Dodson, L. (2003b) “Howard doesn’t like what the media shows us,” The Age (April 4),
posted at www.theage.com.au/articles accessed September 1, 2003.
Dunn, A. (2001) “What have you done for us lately?: public service broadcasting and its
audiences” in M. Bromley (ed.), No News is Bad News: Radio, Television and the Public,
Harlow: Longman, pp. 210–18.
Ferguson, P. (2003) “Our busiest soldiers,” The Australian Business (June 21), p. 27.
239
MICHAEL BROMLEY
Fickling, D. (2003a) “Australia criticised for troops pullout,” The Guardian (April 18),
posted at www.guardian.co.uk accessed September 15, 2003.
Fickling, D. (2003b) “Spinning out of control,” The Guardian (June 16), posted at
www.guardian.co.uk accessed September 15, 2003.
Fitzgerald, R. (2002) The Federation Mirror: Queensland 1901–2001, St Lucia: University of
Queensland Press.
Flew, T. (2002) “Television and pay TV” in S. Cunningham and G. Turner (eds), The Media
and Communications in Australia, St Leonards NSW: Allen and Unwin, pp. 173–87.
Franklin, M. (2003) “Foreign correspondent’s view of Iraq,” posted at www.reportage.
uts.edu.au accessed September 29, 2003.
George Munster Forum (2002) “Did September 11 dull journalists’ critical faculties?’
Australian Centre for Independent Journalism, University of Technology, Sydney
(October 8), posted at http://journalism.uts.edu.au/acij/forums accessed October 23,
2002.
George Munster Forum (2003) “Reporting the symbolic war: intelligence sources and jour-
nalists,” Australian Centre for Independent Journalism, University of Technology,
Sydney (September 19). Broadcast in Big Ideas, ABC Radio National (September 28),
posted at www.abc.net.au/rn/bigidea/stories/s951897.htm accessed September 29, 2003.
Gerster, R. (1987) Big-noting: the Heroic Theme in Australian War Writing, Melbourne: Mel-
bourne University Press.
Given, J. (2000) “Commercial networks: still the ones?’ in G. Turner and S. Cunningham
(eds), The Australian TV Book, St Leonards, NSW: Allen and Unwin, pp. 35–51.
Greenslade, R. (2003) “Their master’s voice,” Media Guardian (February 17), posted at
www.media.guardian.co.uk accessed September 15, 2003.
Griffen-Foley, B. (2003) Party Games: Australian Politicians and the Media from War to Dis-
missal, Melbourne: The Text Publishing Company.
Hargreaves, I. (2003) Journalism: Truth or Dare?, Oxford: Oxford University Press.
Hartley, J. (1992) The Politics of Pictures: the Creation of the Public in the Age of Popular
Media, London: Routledge.
Hartley, J. (2000) “Communication democracy in a reactional society: the future of jour-
nalism studies,” Journalism: Theory, Practice and Criticism, vol. 1, no. 1, pp. 39–48.
Henderson, G. (2002) “Anzac Day: an unfinished story,” The Age (April 23), posted at
www.theage.com.au accessed October 8, 2003.
Henderson, G. (2003) “Crean runs afoul of new patriotism,” The Sydney Institute
(April 29), posted at www.thesydneyinstitute.com.au accessed October 9, 2003.
Hilmer, F. (2001) “The media industry in Australia: public policy choices,” A.N. Smith
Memorial Lecture, University of Melbourne (October 30). Excerpts posted at
www.fxj.com.au accessed November 4, 2002.
Hoar, P. (2000) “Live from Chunk Bair: new media/old media and Gallipoli mythology,”
posted at www.greatwar.org.nz accessed June 19, 2003.
Hull, C. (2003) “Thirst for information,” Quarterly Bulletin, vol. 4, no. 2, Research School
of Pacific and Asian Studies, Australian National University, posted at htpp//:
rspas.anu.edu/qb/articles accessed September 1, 2003.
IRB (1999) submission by the Australian Association of Independent Regional Broadcast-
ers (IRB) to the Productivity Commission Broadcasting Inquiry.
Irving, H. (2001) “Tragic choice for a national myth,” The Australian (April 27), posted at
www.theaustralian.news.com.au accessed June 19, 2003.
Jacka, E. (2000) “Public service TV: an endangered species?” in G. Turner and S. Cun-
240
T H E B AT T L E F I E L D I S T H E M E D I A
ningham (eds), The Australian TV Book, St Leonards, NSW: Allen and Unwin,
pp. 52–68.
Jackson, S. (2003a) “Pushing their opinions,” The Australian Media (February 27–
March 5), pp. 4–5.
Jackson, S. (2003b) ‘“War porn’ obscures the gruesome truth,” The Australian Media (April
3–9), p. 5.
Jackson, S. (2003c) “ABC news wins the viewing war,” The Australian Media (April
10–16), p. 12.
Jackson, S. (2003d) “News posts best readership gains,” The Australian Media
(August 28–September 3), p. 12.
Jackson, S. (2003e) “The top five magazine groups increase their dominance,” The Aus-
tralian Media (November 6–12), p. 10.
Jenkins, C. (2003) “Real-time view from the eye of the storm,” The Australian Media
(April 3–9), p. 6.
Keeble, R. (1997) Secret State, Silent Press, Luton: University of Luton Press.
Kelly, P. (2003) “How PM stole a legend,” The Australian (August 9), posted at
www.theaustralian.news.com.au accessed October 9, 2003.
Kirk, A. (2000) “John Howard to visit Gallipoli,” AM, ABC local radio (April 22), tran-
script posted at www.abc.net.au/am accessed October 9, 2003.
Lewis, K. (2001) “Pluralism in the Australian print media,” AsiaPacific Media Educator,
vol. 11, pp. 100–12.
McCrum, R. (2001) “War without witnesses,” The Observer (October 28), posted at
www.guardian.co.uk accessed November 5, 2003.
McGeough, P. (2003) In Baghdad: a Reporter’s War, Crows Nest, NSW: Allen and Unwin.
Marcus, J. (2003) “Analysis: Australia’s military role in Iraq,” BBC News (March 14),
posted at http://news.bbc.co.uk accessed September 30, 2003.
Marr, D. and Wilkinson, M. (2003) Dark Victory, Crows Nest, NSW: Allen and Unwin.
Matthew, P. (2003) “Of war and media,” posted at www.bandt.com.au/ articles/ f1/
0c0158f1.asp accessed October 5, 2003.
Meade, A. and Brook, S. (2003) ‘Deadly theatre of war,” The Australian Media (March 27–
April 2), pp. 4–5.
The Media Report (2001) “How well are we served by foreign correspondents?,” ABC Radio
National (November 1), posted in transcript at www.abc.net.au/rn/talks accessed Sep-
tember 6, 2002.
The Media Report (2002) “Frontline to the front page,” ABC Radio National (April 25),
posted in transcript at www.abc.net.au/rn/talks accessed October 2, 2003.
The Media Report (2003) “What happened to the war?”, ABC Radio National (May 15),
posted in transcript at www.abc.net.au/run/talks accessed June 19, 2003.
Media Watch (2003a) “Where are our troops?,” ABC Television (March 24), transcript
posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003b) “Embedded truth,” ABC Television (March 24), transcript posted at
www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003c) “A new national broadcaster?” ABC Television (March 24), tran-
script posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003d) “The moment of truth,” ABC Television (March 24), transcript
posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003e) “Pilots lite,” ABC Television (March 31), transcript posted at
www.abc.net.au/mediawatch accessed May 13, 2003.
241
MICHAEL BROMLEY
Media Watch (2003f) “4RAR poses,” ABC Television (March 31), transcript posted at
www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003g) “Sluggo in Baghdad tonight,” ABC Television (March 31), tran-
script posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003h) “NT News boycotts war,” ABC Television (March 31), transcript
posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003i) “Propaganda and the Tele,” ABC Television (April 7), transcript
posted at www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003j) “Saddam falls,” ABC Television (April 14), transcript posted at
www.abc.net.au/mediawatch accessed May 13, 2003.
Media Watch (2003k) “Blair is toppled,” ABC Television (May 5), transcript posted at
www.abc.net.au/mediawatch accessed May 13, 2003.
The Mercury (2003) “Wounded Australian newsman dies’ (July 7), posted at www.themer-
cury.news.com.au accessed September 9, 2003.
Miller, T. and Turner, G. (2002) “Radio” in S. Cunningham and G. Turner (eds), The
Media and Communications in Australia St Leonards, NSW: Allen and Unwin,
pp. 133–51.
Miskin, S., Rayner, L., and Lalic, M. (2003) Media Under Fire: Reporting Conflict in Iraq,
Current Issues Brief 21, Parliament of Australia, Department of the Parliamentary
Library, posted at www.aph.gov.au/library accessed May 13, 2003.
Morrison, D. and Tumber, H. (1988) Journalists at War, London: Sage.
Mottram, M. (2003) “Australia left in the dark, say reporters,” The Age (March 25), posted
at http://new.theage.com.au accessed September 10, 2003.
Murphy, K. (2003a) “Viewers turn on the news as bombs blast Baghdad,” The Australian
Media (April 3–9), p. 12.
Murphy, K. (2003b) “Hewitt helps Seven back into play,” The Australian Media (Septem-
ber 25–October 1), p. 12.
News.com.au (2003) “ABC denies bias in Iraq coverage’ (July 21), posted at www.news.
com.au accessed October 6, 2003.
Pearson, M. and Brand, J. (2001) Sources of News and Current Affairs, Sydney: Australian
Broadcasting Authority, vol 1, pp. 5, 6.
Phillips, M. (2003) “Ministers’ advice bill hits $20m,” Courier-Mail (September 23), p. 2.
Pilger, J. (2002) “Journalism in Australia has a courageous history, but Murdochism has
turned it into a disgrace,” New Statesman (February 21), posted at http://pilger.cartl-
ton.com accessed November 18, 2002.
Pilger, J. (2003) untitled talk at John Pilger Forum, Australian Centre for Independent
Journalism, University of Technology Sydney (March 28), transcript posted at
http://acij.uts.edu.au/seminars/2003pilger.html accessed September 10, 2003.
Pusey, M. (2003) The Experience of Middle Australia: the Dark Side of Economic Reform,
Cambridge: Cambridge University Press.
Putnis, P. (1996) “Producing overseas news for Australian television,” Australian Journal of
Communication, vol. 23, no. 3, pp. 1–22.
Radio 4BC (2003), John Howard interview with John Miller and Ross Davie (June 12),
transcript posted at www.pm.gov.au/news/interviews accessed September 11, 2003.
Rampton, S. and Stauber, J. (2003) Weapons of Mass Deception: the Uses of Propaganda in
Bush’s War on Iraq, New York: Jeremy P. Tarcher/Penguin.
Rees, J. (2002) “Foundation origins,” CEW Bean Foundation, posted at www.npc.org.au/
CEWbean/foundation_origins.htm accessed September 10, 2003.
242
T H E B AT T L E F I E L D I S T H E M E D I A
Reynolds, H. (1999) Why Weren’t We Told?: a Personal Search for the Truth about Our His-
tory, Ringwood, Victoria: Viking.
Rice, S. (2003) “The charge of the lite brigade,” The Walkley Magazine, vol. 21, pp. 10–12.
Schultz, J. (1998) Reviving the Fourth Estate: Democracy, Accountability and the Media, Cam-
bridge: Cambridge University Press.
Sekuless, P. (1999) A Handful of Hacks, St Leonards, NSW: Allen and Unwin.
Simms, M. (2002). “The media and the 2001 election: Afghans, asylum seekers and
anthrax” in J. Warnhust and M. Simms (eds), 2002: The Centenary Election, Brisbane:
University of Queensland Press.
Spearritt, P. (2003) “Unfriendly fire: Australia and the war on Iraq,” Brisbane Institute
(April 2), posted at www.brisinst.org.au accessed September 1, 2003.
Stewart, C. (2001) “The legend that must not die,” The Weekend Australian (April 26),
posted at www.theaustralian.news.co.au accessed October 8, 2003.
Tiffen, R. (1991) “The second casualty: the ABC and the Gulf War,” Current Affairs Bul-
letin, vol. 67, no. 11 (April), pp. 10–18.
Tiffen, R. (1992a) “News coverage” in M. Goot and R. Tiffen (eds), Australia’s Gulf War,
Carlton, Victoria: Melbourne University Press, pp. 114–39.
Tiffen, R. (1992b) “Marching to whose drum?: media battles in the Gulf War,” Australian
Journal of International Affairs, vol. 46, no. 1 (May), pp. 44–60.
Turner, G. (2000) “Television news and current affairs: ‘Welcome to Frontline’” in
G. Turner and S. Cunningham (eds), The Australian TV Book, St Leonards, NSW:
Allen and Unwin, pp. 89–102.
Turner, G. (2001) “Sold out: recent shifts in television news and current affairs in Aus-
tralia,” in M. Bromley (ed.), No News is Bad News: Radio, Television and the Public,
Harlow: Longman, pp. 46–58.
Uechtritz, M. (2003a) “On balance, ABC news coverage is not biased,” The Australian
Media (June 5–11), p. 7.
Uechtritz, M. (2003b) “Complaints, Minister?,” The Walkley Magazine, vol. 22, p. 5.
Van Druten, R. (2003) “Neo-fire fighter: a new profile for the international news corre-
spondent,” BA (Hons) thesis, Brisbane: Queensland University of Technology.
Vermeer, T. (2003) untitled talk, Australian Centre for Independent Journalism, Univer-
sity of Technology Sydney (April 4), transcript posted at http://acij.uts.edu.au/
seminars/2003vermeer.html accessed October 5, 2003.
Veterans Support and Advocacy Service (2003), posted at www.ausvets.com.au/tradi-
tion.htm accessed June 12, 2003.
Vidal, G. (nd) “War, seen through photographs darkly: visual traces and interpretive possi-
bilities in the photographic representation of Gallipoli,” posted at www.arts.monash
.edu.au/history/events/genidwar/papers/callister.html accessed May 14, 2003.
Ward, R. (1958) The Australian Legend, Melbourne: Oxford University Press.
Wilson, P. (2003) “Ingenuity that delivers pictures on demand,” The Australian Media
(April 3–9), p. 6.
Wright, T. (2003) “Fog of war,” The Bulletin, vol. 121, no. 13) posted at http://bulletin.
ninemsn.co.au/bulletin accessed September 10, 2003.
Zin, C. and Campbell, D. (1999) “Queen of Australia looks set to survive historical vote,”
The Guardian (November 6), posted at www.guardian.co.uk accessed October 9, 2003.
243
Part 3
Stephen D. Reese
War has become an increasingly common tool of US national policy. Rather than
Congressionally declared states of war periodically punctuating otherwise harmo-
nious periods of peace, military conflict has become a condition of modern life.
Now the National Security Strategy of the United States has formalized the case
for pre-emptive unilateral military action, a policy of great significance for inter-
national relations. This plan, advocated for years by neo-conservatives who
ascended to key positions in the latest Bush administration, was put into practice
most recently in Afghanistan and Iraq, and even contemplated with respect to
Iran and Syria. Advocating overwhelming US world military superiority to pre-
vent the emergence of rivals follows in line with other administration steps away
from multilateral international agreements on arms control, the environment,
and other issues. This unilateralism of military force is rationalized by its archi-
tects as “power that can be trusted” (e.g., Armstrong 2002). Although anti-war
voices have been at work, American military action has taken place largely
against a backdrop of public support, or at least acquiescence. To understand how
this support is developed and sustained we look to the relationships among the
military, state, and media.
At a basic operational level, many analysts have looked at specific media cen-
sorship, public relations, and other manipulatory actions taken by military and
administration officials in shaping media coverage. At a broader systemic level,
others have considered how the ideological leadership of the media serves the
interest of the US “empire.” It is helpful, however, to combine these insights to
examine from a sociological perspective specifically how news organizations enter
into routinized relationships with military and other newsmakers, and how news
of conflict is placed into particular frames of reference, which serve to anchor war
in familiar cultural terms. These “routinized frames” are revealed through the
recurring combination of visual and verbal elements within media coverage,
showing what organizing principles are at work in the decisions of news managers
and news sources. In this way, we may better see how the media perform their
jobs in communicating news of war, national policy, and public debate over it.
247
STEPHEN D. REESE
US media performance in this arena becomes ever more crucial given the coun-
try’s lone super-power status, and the administration’s decision to act alone if
need be in wielding military force. Of course, the attacks of 9/11 in New York and
Washington launched a new “war on terrorism,” a loaded and elastic frame used
to help justify and fast-track the new unilateralist foreign policy. Patriotic post
9/11 television news graphics provided short-hand labels describing how “Amer-
ica strikes back” quickly mutating into “America’s new war,” with that “war”
invoked as a main justification for the 2003 American invasion of Iraq. That is
why implying the dubious connection of Saddam Hussein to those attacks was so
strategically important in justifying military action against him. To the extent
they can help examine such claims and facilitate reasoned discussion of these
policies, the American news media have major implications for the rest of the
world.
Chapter purpose
In this chapter I return to the Desert Shield/Desert Storm operation in the Persian
Gulf War of 1990/1991 (Gulf War I) and consider how specific frames within news
coverage at the local level create an implicitly pro-policy position, delegitimating,
and marginalizing dissent.1 The structured routines of newswork give rise to cer-
tain predictable ways of making sense of military conflict, particularly in the public
debate that follows. We need to understand coverage of front-line conflict, but a
broader “war at home” takes place away from the scene of actual combat as the
government tries to build support for the policy behind it and policy opponents
attempt to mount their challenges. These two processes are carried out in large
part via the media. Although most analysis of the news media in wartime has
focused on the front-line war, these actions are connected to and color the cover-
age of the domestic front. Thus it is important to consider how these two wars are
organized for public consumption by the news media, how one feeds the other, and
how that coverage works to advance or prevent a healthy public debate.
I focus here specifically on a local television news station as a lens into how
coverage of the conflict in 1991, even far away from the front-line, created a no-
win situation for the anti-war position. In Gulf I, the local community was an
important site of public debate, including rallies for and against the war and the
ubiquitous yellow ribbons. Although the national debate and network level
media drew much of the scholarly analysis, people found support for and gave
voice to their opinions in local schools, churches, and locally organized political
speech. Furthermore, local television showed the commercial imperative of audi-
ence appeal writ large, which highlighted the processing by which news converts
military action into an audience-friendly story line. Of course, this happened at
the national level too, but within a single community the news organization’s
decision-making in connection to specific events, relationships with sources, and
the resulting coverage can be easily explored. Entertainment values too make this
analysis even more relevant. In the recent 2003 war in Iraq, stories such as the
248
M I L I TA R I Z E D J O U R N A L I S M
249
STEPHEN D. REESE
The strength of military logic in the broader conflicts of recent years, particu-
larly post 9/11, is undeniable. Since Gulf War I, the more formalized conflict of
armies from that war has been supplanted by asymmetric warfare, yet with the
same premises and approaches applied to this more fluid conflict in which the
“enemy” does not agree to play by the rules of traditional combat. The “war on
terrorism” frame for this new condition carries in its terminology a traditional
Defense Department solution, which, although it may be partly accurate, over-
shadows other interventions in this jointly sociological, economic, political, and
religious issue. (The “war on drugs” worked similarly, privileging military and law
enforcement solutions to an issue that was also a public health matter. The two
became linked when government officials argued that using illegal drugs helped
provide revenue for terrorists.) Military logic becomes mapped onto every other
discussion, becoming the dominant organizing principle and short-circuiting
debate. The success of military action as a policy response may, as a result, be said
to be determined through criteria of the military’s own choosing.
The dominance of this military logic frame was aptly illustrated in President
Bush’s famous photo-op jet landing on the aircraft carrier Abraham Lincoln last
May 2003 off the coast of San Diego. Changing out of his flight suit, he addressed
the cameras with a banner behind him, “Mission accomplished.” The clear mes-
sage was that in the military completing its major combat operations, the
president’s national mission had been effectively completed too—one being con-
flated with the other in this mediated imagery and symbolism. Though the
military branch was an instrument of national political policy, the president wear-
ing a flight suit visually overrode this distinction.4 The power of this “war on
terrorism” and its associated “axis of evil” was further illustrated by the fact that
the majority of Americans were reported to hold Saddam Hussein responsible for
the 9/11 World Trade Center attack, even though no evidence supported such a
link. He was also implicitly linked by his inclusion in the “axis,” against which
the war on terrorism was arrayed. As mentioned earlier, this framing cast a pre-
emptive strike national policy into a self-defense context, making it more
intuitively palatable to most Americans.
The routines of newswork draw attention to the structured ways that journal-
ists enter into relationships to obtain desirable goals (Shoemaker and Reese
1996). In front-line coverage these routines are often clearly delineated; military
officials desire positive accounts of their activities, to “get their story out,” and to
simply accommodate the demands of the many news organizations seeking access
to the story. Systems are developed to meet those needs. Journalists, of course,
want the most exciting material possible that will be of interest to their organiza-
tions and audiences. The Vietnam-era memory of these relationships, particularly
among military officers of that generation, is adversarial, with journalists “not on
the team.” The more typical modern characterization of this relationship is sym-
biotic, and a perceived anti-military attitude risks a journalist being excluded
from interview opportunities and other desired access. These routine structures
impose their own logic, working against alternative frameworks of interpretation.
250
M I L I TA R I Z E D J O U R N A L I S M
In the case of war reporting, they contribute to what Kellner (1992) called the
“militarization of consciousness.” Law enforcement and military institutions are
particularly important agents of social control in society and act as “primary
definers,” on which journalists have come to rely heavily for sources or news. Hall
et al. (1978) argue that the media stand in structured subordination to these insti-
tutions which in the case of the military, is able to exert great definitional
power—not only on its own realm but also in a way that carries over into others.
The post-Vietnam image of the military emerged with damage but was gradu-
ally rehabilitated both in the political and wider cultural spheres (Baritz 1985).
President Reagan’s policies emphasized a greater ideological justification for using
the military and a willingness to deploy it in tune-up conflicts like Grenada. In
popular culture, movies such as Rambo and Missing in Action carried a revisionist
version of Vietnam history, advancing the notion that the military was under-
mined by spineless politicians and forced to fight with “one hand tied behind its
back.” As the Rambo character said at the end of First Blood, “I did what I had to
do to win, but somebody wouldn’t let us win!” President Nixon had been able to
distract attention from the unpopular southeast Asia policy by focusing public
attention on the prisoners of war issue, a matter in which there was much more
fervent and exploitative strategic government gamesmanship than evidence
(apart from Chuck Norris movies) for POWs still behind enemy lines (Franklin
1992). With the advent of Gulf War I, national officials were able to draw on this
restored image of American forces to engage public support. The potency of this
focus was seen in its power to invert one post-Vietnam principle of military
policy. Before Gulf War I, officers like Colin Powell advocated building public
support before any large-scale commitment of troops; paradoxically, however, the
US administration showed that by committing the troops they could engage the
public. Once significant forces had been deployed in Saudi Arabia and a January
deadline set (“showdown”) for Hussein to leave Kuwait, the “support the troops”
motif exploited a powerful cultural value, which found its way into news framing.
They, the troops, engaged support precisely because they were there, effectively
obliterating any challenge to the policy that got them there.
251
STEPHEN D. REESE
by KVUE-TV, which at the time was the ratings leader in Austin, Texas. Inter-
views were conducted with station producers and reporters, and raw footage of
several community public demonstrations was also examined. Anchor introduc-
tions to stories were emphasized because these lead-ins displayed the most
obvious encapsulation of the frame by compressing the essence of the story into a
few attention-getting words.
Conflict
In 1991, a variety of public protests and demonstrations were going on through-
out the community. Once the January 15, 1991, deadline for Saddam Hussein set
by President George Bush (Bush senior) arrived, public opposition to the con-
flict was framed to domesticate its focus. The language in news reports clearly
worked to downwardly adjust perceptions of anti-war protest strength. Indeed,
anti-war protest was probably more vocal in this relatively liberal city, making
the framing job perhaps more clearly necessary. (My italics are added below for
emphasis.)
January 16
Reporter: (on anti-war protest at the University of Texas) . . . Protestors out-
numbered those supporting the war by 2 to 1, but supporters say that’s only
because the anti-war folks are more vocal.
January 17
Anchor: . . . Even though anti-war protesters outnumbered Bush supporters two to
one, conservatives say they are tired of staying silent.
January 17
Reporter: Anti-war protesters have demonstrated almost continuously since yester-
day evening and conservatives felt it was time to defend themselves.
January 16
Anchor: There are many, many Austin residents who support President Bush’s
decision to bomb Iraq and they say they want to be heard. They plan a
candle-light vigil in Waterloo Park tonight.
January 20
Anchor: . . . In the beginning pro-war forces were relatively quiet, now they are
gaining in momentum . . . [after shots of rally, in conclusion] . . . Later the pro-
252
M I L I TA R I Z E D J O U R N A L I S M
war group was confronted by those opposed to the US presence in the Per-
sian Gulf region.
January 17
Anchor: A stark contrast tonight to the overwhelming crowds that have gathered
recently to protest the war. Tonight about 30 people stood by in City Park
. . . a quiet candlelight vigil to support President Bush. They were small in
number but their feelings were just as strong . . . Those in attendance had to
dodge rain showers, but that didn’t dampen their spirits.
Other reports presented a less positive view of dissenters, calling attention to dis-
ruptions, minimizing their strength, and challenging their symbolism.
January 17
Anchor: At the State Capitol today, anti-war protesters were anything but peaceful.
January 19
Anchor: Police and war protesters had estimated that a peace rally at the State
Capitol this afternoon would reach some 20,000. Instead 1,500 to 2,300
showed up, far short of the anticipated crowd.
January 26
Anchor: . . . Anti-war protesters carried flag-draped caskets symbolizing war dead
. . . But [notes the anchor] so far, US military officials say one American ser-
viceman has been killed in combat.
This attempt by the anchor to “correct” the symbolism of the protesters, which
presumably referenced deaths on both sides, implicitly restricted war dead to
American casualties.
Station officials reported that audience complaints often made them sensitive
to airing footage of protest, arguing that it allowed them to “consider all sides.”
One producer said, “The people who supported the troops were a kind of silent
majority.” Thus, in this case the opposing sides became the anti-war position, on
the one hand, and the “support the troops” position, on the other—not an anti-
policy and pro-policy side. Examining the linguistic composition of these frames
shows how strongly intertwined the “pro-troops” position and the related stance
of “get behind the president” became in coverage.
January 23
Anchor: 150 demonstrators supporting the war effort demonstrated at the Univer-
sity of Texas and listened to people speak about patriotism. As a counterpoint,
253
STEPHEN D. REESE
these five protesters at the State Capitol are all who are on hand for a war
protest that began on the 15th.
In a story from an area public school, the reporter even overtly made a point to
separate policy and troops before implicitly joining them again.
January 16
Reporter: [about school kids’ reaction] . . . In the meantime the students are fol-
lowing through on their commitment to support not the war itself but rather
the Americans in the Middle East fighting for peace.
This distinction between troops and policy is further eroded, with local officials
adding their voice of support—again implicitly joining the two.
January 23
Anchor: Austin County Commissioners came out in support of American men
and women serving in the Middle East and against the actions of Saddam.
January 23
Reporter: . . . There are others who say they don’t necessarily want to fight in a
war either but will do whatever it takes to protect their country’s interests.
January 17
Anchor: . . . The peace protests are hard for families whose loved ones are in the
Persian Gulf. One military wife says she can handle the stress and anxiety of
knowing her husband is in the thick of things, but it’s harder when she’s con-
fronted by scenes of angry protesters demonstrating against the war.
Woman: There are lots of families hanging on to every word that the news is
putting out and I think it’s really destructive to them.
January 17
Anchor: . . . The anti-war sentiment is unsettling for families whose loved ones
are involved in Operation Desert Storm and for those who back President
Bush’s decision to go to war.
Later reporting further served to reinforce this clear pitting of the anti-policy
stance against the pro-troops/patriotism position in a binary opposition.
254
M I L I TA R I Z E D J O U R N A L I S M
January 19
Anchor [over shots of protesters]: People converge on the State Capitol shouting
their pleas for peace while a patriotic group of small town residents sing their
support for American soldiers at war.
January 24
Anchor [over video of veterans rally]: The US must show 100 percent support for
our troops in the Middle East. That’s the message from veterans who say they
are upset over the number of anti-war protests. They say it sends a bad mes-
sage to the troops in the Middle East, that we don’t support them.
January 24
Wife of serviceman: It’s time for all Americans to unite behind the young men and
women who are serving their country.
Anchor: . . . Many are upset over the number of anti-war protests and say they
should stop.
January 19
Reporter [over pictures of flag-waving rally in adjoining town]: They are the
images of Americana . . . The pictures of heartfelt pride and support for sol-
diers in the Middle East. The war in the Middle East has revived patriotism
here.
January 20
Anchor: As the battlefield gets more intense, more Americans are working to
show their support for the troops who are under attack in Saudi Arabia.
255
STEPHEN D. REESE
Reporter: On one side of the UT campus, several hundred people who are opposed
to the war carried on a protest that began last night.
Anti-war speaker: During the war in Vietnam we lost over 58,000 young Ameri-
can lives.
Pro-war speaker: The legacy of Vietnam will die with this conflict.
Reporter: A few feet away supporters of the President held their own rally.
Same person: Because Iraq is not Vietnam.
Reporter: It was smaller but feelings ran just as strong.
Chanting males: USA, USA.
Student: How many troops do they have compared to ours?
Reporter: With two groups so close together there was inevitably conflict.
Students: [unintelligible argument]
Student: The sheep can preach the virtues of vegetarianism until hell freezes over,
but the wolf isn’t listening. You’ve got to deal with people in a language
they’re capable of understanding and Saddam Hussein only understands
violence.
The reporter moved from this bi-polar pairing of positions to reach and conclude
with a consensual, but ultimately “pro-troops,” and therefore “pro-policy,”
theme.
Reporter: Some who came here were motivated by a deeper feeling, a sense of
commitment. [Said over shots of anthem and flag to a woman who has a
brother in the Gulf.]
Woman: When your family’s over there all you know is to support them.
Reporter: Students raised during a time of peace are now debating their genera-
tion’s war [over shots of signs, peace signs]. Some of the slogans have
changed, some haven’t. But the emotions raised by patriotism and violence
[Saddam Hussein’s?] run just as strong.
Thus, again the anti-war position was pitted against the high ground of those
with a “deeper commitment.” Indeed, opposing the war was tantamount to
opposing the woman interviewed and her family. But this framing was part of a
routine package that made it possible for news organizations to handle protest
stories easily and with a minimum of audience complaint. Reporters were not
expected to have expertise in the policy issues. They were able to present the
“form” of balance as an easily followed format, which would then yield a consen-
sual “patriotic” middle ground.
Consensus
The “support the troops” concept became a crucial element in the conflict and
consensus frames and a way to manage public dissent over government policy.
Particularly with regard to local television, Kaniss (1991) argued that, given the
256
M I L I TA R I Z E D J O U R N A L I S M
nature of the large and fragmented audiences, stations are driven to find unifying
symbols and themes, such as sports franchises, which help to promote a sense of
community solidarity.6 Thus, the “support the troops” message was tailor-made for
news coverage seeking to restore a community threatened by the divisive disputes
over war policy. Frames derive their power in large part because they are internal-
ized “organizing principles” that news workers can apply routinely. Interviews
with station reporters showed how this occurred for the pro-troops element. As
one admitted:
Look, almost everyone had strong feelings about the war . . . not like
they were “pro-war” but that everyone backed the troops. They wanted
the troops not to get hurt over there. No one wanted them hurt. I have
to admit maybe I was too close to the story. I had relatives—close rela-
tives—over there fighting.
“The troops” became the nation’s home-town team, indeed the consensual glue
used by reporters to symbolically hold the community together, especially when
trying to frame expressions of conflicting public opinion.
January 22
Anchor: . . . Although both sides of the war issue are still battling back and forth,
one thing seems to hold the factions together: support for the men and
women in Saudi Arabia . . . [Referring later to flag sales] Although everyone
may not choose to show their support in the same way, at least for some,
the support for the troops is there no matter what the belief about the war
itself.
January 22
Anchor: . . . People may be divided about how they feel about US involvement in
the Middle East, but one feeling seems to be shared by everyone: support for
the troops who are over there now.
News routines show that not all stories require balanced voices. According to
Hallin (1986), those stories that deal with subjects within either the sphere of
deviance or sphere of consensus are by their nature not ones that require even
treatment. Those, however, within the “sphere of legitimate controversy” do.
Thus, as a consensual story, “support the troops” stories came to no longer require
balance, as in this report on efforts at a local school.
January 23
Anchor: . . . Those who support the American forces in the Persian Gulf War are
trying to make themselves more visible . . . Among other things, the students
257
STEPHEN D. REESE
signed a huge Happy Valentine’s card to be sent to the troops and passed out
yellow ribbons.
Of course, there was no shortage of stories from the community and surrounding
areas that served to exemplify traditional values: placement of yellow ribbons, ral-
lies, flag-flying, and veterans meetings. Even the veterans themselves were
processed through the consensus frame to eliminate any troubling qualms about
war in general. As one veteran was quoted as saying:
January 16
“War produces dead bodies. Let’s hope this one’s over quickly. War is
hell . . . it’s just you can’t describe it.”
The reporter, given this threatening notion of war’s consequences, quickly reas-
sured the audience:
Despite the knowledge of how horrible war can be, for every ounce of fear
among members of this group, there’s still a ton of patriotism . . . These men
have been there . . . They know first hand the turmoil, the desperation of
war . . . But all are very proud tonight and holding their heads up high.
With military success, the pro-troops element soon morphed into the “heroes of
Desert Storm,” a label that continued linking the troops to the policy. A
reporter’s January 18 story glorified local Bergstrom Air Force Base reconnais-
sance pilots as the “unsung heroes of the war.” The characterization suggests one
who embarks on a worthy undertaking, so it is difficult to celebrate the heroes
without also endorsing the mission on which they were sent. This theme was a
valuable resource for routine story construction by providing an easily con-
structed story-line, drama, and meaning to the conflict.
In this chapter I have largely centered my attention on how war and its public
debate were handled within a specific geographical community. The routinized
structure of media/military relationships rooted conflict in frames of reference
that held audience appeal and accessible cultural meaning. Coverage of this con-
flict as seen in the first Gulf War was closely related to coverage of dissent at
home within an overall military logic, finding particular expression in support for
“the troops.” During the interval between this Gulf conflict and that which fol-
lowed, the local community was superseded in many ways by global public
communities, which had implications for public support for military action.
258
M I L I TA R I Z E D J O U R N A L I S M
audiences. With greater competition among US cable news networks, and the
more forceful patriotic voice of Fox News, national news became more closely
aligned with the commercial imperatives of local television. Indeed, the trauma
of 9/11 drove news to appeal to the same sense of one community in the name of
national solidarity that was typical of local news. This had an impact on the
extent to which conflicts were easily framed within military logic and dissent was
marginalized. The local/national, “vertical” frames of reference came into
increasing tension with more globalized, multi-level “horizontal” orientations of
world news gathering. Global communities, if not supplanting local ones, cer-
tainly added an important layer to the public sphere. The current Bush
administration took an active role in framing national policy very explicitly in
the shape of the “war on terrorism” and the “axis of evil.” These perspectives
became more pointed and publicly resonant than the vague, negatively conno-
tated sense of a “new world order,” employed by the earlier Bush administration
in Gulf War I. But they were also more open to contestation. How might we com-
pare the potential for framing dissent, as we reflect on differences between Gulf
War I and Gulf War II? Although it is difficult to visualize a public sphere pro-
jected globally, some suggestive anecdotal outlines emerge in the way that world
publics react and interact through the media.
In many respects, a globalized public opinion came of age following the attacks
of 9/11 and the subsequent US efforts to engage militarily in Afghanistan and
Iraq. Public protests around the world on February 15, 2003, were a particular
watershed event, which created a boundary-spanning anti-war movement acting
in its various locations but in a simultaneous global arena, supported by transmis-
sion of global news and other communication (from email to CNN, weblogs,
news sites, etc.). The national media continued to cater to the parochial views of
their officials and mass audience, and local media were still limited to covering
locally based public protest actions. Global elites, however, increasingly took into
account world opinion, driven by alternative sources of information to any spe-
cific locally based channels. Dissent was not so easily marginalized in this more
diffused media environment.
The new ability of citizens to mount a globally coordinated expression of
opposition produced corresponding political consequences. Thus, compared to
the first Gulf War of 1990 and 1991, the US administration had much greater
difficulty operating free of constraint in implementing what amounted to a pre-
emptive strike national security policy in the 2003 Iraq invasion (it is, of course,
true that the international community was more unified behind Gulf War I). An
anti-war public in many countries made it politically treacherous for national
leaders to support the Bush administration. Media and public opinion, particu-
larly as seen recently in Europe, were less apt to follow government policy.
Forging multilateral agreement for a unilateral policy came with greater difficulty
in a world with global communication supporting different dimensions of public
opinion, and where the purported rationale for policy was subjected to world
scrutiny, helping expose disconnects between surface discourse and underlying
259
STEPHEN D. REESE
260
M I L I TA R I Z E D J O U R N A L I S M
about their coverage (while obliging them to monitor more closely the work of
non-US journalists—particularly, for example, from countries such as Abu
Dhabi). Ted Koppel of ABC News Nightline was unable to resist seeking an
embedded position (an enhanced one attached to the division commander) and
prefacing one report with ominous heroic imagery from Shakespeare (“Unleash
the dogs of war!”). An embedded reporter from the Atlanta Journal-Constitution
exhibited this professional ease with the heroic theme by presenting a photo-
graphic retrospective of his work following his assignment, accompanied by an
arrangement of Samuel Barber’s sublime “Adagio for Strings” (also used in the
movie Platoon).7 Indeed, I would argue that the unilateralist policy that got
the troops to Iraq made this perspective even more necessary, for it was a policy,
after all, predicated on the assumption that US military superiority would be
used wisely and was “power that can be trusted.” Thus, perhaps it was not surpris-
ing that American journalists internalized this assumption and that non-US
journalists were more likely to at least make a distinction between taking the
perspective of the unit and the side of “the American war machine.”8 Thus,
although the embedding program worked to reinforce a military logic and a
heroic frame, within the still emerging global norms of newsgathering there was
evidence of a fault line between a nationalistic unilateralism and a multilateral
world.
Thus, embedding was a form of control that created a strong dependency rela-
tionship between journalists and their units (not only for getting the story but
for protection in a dangerous place). Even the training supervised by the military
for aspiring embeds underscored the premise that “we know what we’re doing,
and you don’t.” Nevertheless, on the ground of military conflict, it became more
difficult to manage information in an environment more fluid and porous than
just 12 years prior, with satellite phones and other technologies making commu-
nication easier and quicker. In some ways, this more fluid communication field
made controlling the “story line” more crucial, with the Pentagon and the US
administration seeking foremost to frame the story as “mission accomplished.”
The availability of satellite phones, for example, made it possible for many more
journalists to instantly transmit their first-hand observations to editors any-
where in the world. Peter Arnett was alone in transmitting via satellite phone
from Baghdad in 1991; reporting from the same city in 2003 he noted far greater
competition, with 200 to 300 such phones in the city and a dozen video uplinks
and video phones (Blumenthal and Rutenberg 2003). In the first Gulf War,
restricted coverage led many viewers to give little consideration to civilian suf-
fering, while during that later war reporters had greater access to the impact of
the conflict.
In the current Iraq conflict of Gulf War II, even the powerful “support the
troops” component within the frames of dissent cuts both ways, that is, with
dual and contrary consequences. On the one hand, the media dynamic remained
similar today, with news organizations clearly “on board” with the policy. A
recent report from NBC Nightly News, for example, documented the hospital
261
STEPHEN D. REESE
262
M I L I TA R I Z E D J O U R N A L I S M
closely for alternative points of view concerning the war in Iraq. According to
Croad (2003), “Much of the US media’s reaction to France and Germany’s
intransigence on the Iraqi war issue has verged on the xenophobic, even in the
so-called ‘respectable’ press.” As a result, she observed that the feedback to these
European websites suggested that people no longer rely only on their own
national media, exercising instead their need for information on a global scale.
Web-based autonomous media emerged such as Indymedia.org, a collective of
independent media organizations and journalists, that provided a critique of war
coverage in the mainstream press, reframed issues away from military strength to
diplomatic relationships and, as it reported, promoted “global citizenship.” So,
information globalization means that citizens have access to the policy record in
a way never possible before, and other countries have access to well-informed
points of view around the world. Thus, greater transparency has developed con-
cerning US policy objectives, even if not from the government itself, making it
harder for national leaders to “go it alone” with the expectation that the world
public will fall in line.
As a globalized public sphere becomes more complex and interconnected, it
will become important to theorize the implications for public support for military
conflict. Local news organizations during Gulf War I effectively structured sup-
port for the policy as they applied a military logic to local debates. As these
debates over military conflict become globalized and denationalized, beyond the
scope of any single local community, there remains the hope that these policies
can be debated clearly through a more multilateral cultural media lens.
Notes
1 A fuller analysis is contained in a previous article (Reese and Buckalew 1995).
2 Lynch was an American soldier captured and taken to an Iraqi hospital, where by all
accounts she was treated humanely before being retrieved in what was hailed at the
time as a heroic special rescue operation. The Pentagon has denied staging the rescue
as a media event, but it provided to the media its video footage of the operation, acted
no doubt as the source of many details in news accounts attributed to anonymous
sources, and failed to later correct erroneous details that didn’t conform to the story
line (Lynch was captured without resistance, although early reports had her emptying
her gun at the enemy before being overcome).
3 I understand that in the Arab world this most recent Iraq war is Gulf War III, with the
Iraq–Iran war being the first.
4 News stories in November 2003 at this writing discuss the White House’s attempts to
distance the president from the “mission accomplished” banner, a jarring symbolic
memory given the ongoing presence of US forces in Iraq. The president has blamed
the navy for posting the banner, but the administration’s skill in framing visual back-
drops for his speeches and controlling every other aspect of media interaction casts
doubt on this innocence. What appeared to be a classic presidential photo opportunity
and a golden opportunity for campaign advertising now ironically may be just that, for
the opposition that is.
5 In the original study, another frame, “control,” was also explored. It emerged from the
tight relationship between local news organizations and law enforcement, making it
easy to slip into a “police work” perspective and cast public dissent as a threat to
263
STEPHEN D. REESE
social control. Dissent, as a result, was often treated as a matter of police work, keep-
ing unruly crowds in check and focusing on procedures in place to manage public
gatherings.
6 This interlock with sports continues on a national level, as seen, for example, in the
National Football League promoting its “Intrepid fallen heroes fund” meant to support
the families of military personnel “who have given their lives in the current operations
in defense of our country” (www.nfl.com/heroesfund, November 10, 2003). My point
is certainly not to diminish the loss of these individuals but to suggest how deeply
ingrained the troops are in the national psyche, reinforced in this case by initiatives
supported by commercial enterprises. The frequent analogies of sports to war and vice
versa is another lengthy subject.
7 My observations from this section are drawn from various comments at recent profes-
sionals’ meetings. Insights into the foreign press are from the Newsworld International
meeting for news professional in Dublin (October 20–3, 2003). The Koppel and
Atlanta details are from a symposium on war reporting at the University of Texas at
Austin (November 4–5, 2003).
8 The latter view was expressed by BBC correspondent David Loyn at the same News-
world meeting referenced above (October 21, 2003).
References
Armstrong, D. (2002) “Dick Cheney’s song of America,” Harper’s Magazine, October,
vol. 305.
Baritz, L. (1985) Backfire: American Culture and the Vietnam War, New York: Ballentine.
Blumenthal, Ralph and Rutenberg, Jim (February 18, 2003) “Journalists are assigned to
accompany US troops,” The New York Times, p. A12.
Byrne, C. (October 22, 2003) “BBC was ‘distrustful’ of embedding,” The Guardian.
Croad, Elizabeth (February 27, 2003) “US public turns to Europe for news,” dot.journalism,
www.journalism.co.uk/news/story576.html.
el-Nawawy, M. and Iskandar, A. (2002) Al-Jazeera: how the Free Arab News Network
Scooped the World and Changed the Middle East, Cambridge, MA: Westview.
Franklin, H. (1992) M.I.A: Mythmaking in America, Brooklyn: L. Hill Books.
Gerbner, G. (1992) Persian Gulf War: the Movie in H. Mowlana, G. Gerbner, and
H. Schiller (eds), Triumph of the Image: Persian Gulf War in Global Perspective, Boulder,
CO: Westview, pp. 243–65.
Graubard, S. (1992) Mr Bush’s War, New York: Hill and Wang.
Hall, S., Critcher, C., Jefferson, T., Clarke, J., and Roberts, B. (1978) Policing the Crisis:
Mugging, the State, and Law and Order, London: Macmillan.
Hallin, D. (1986) The “Uncensored” War: the Media and Vietnam, Berkeley, CA: University
of California Press.
Kaniss, P. (1991) Making Local News. Chicago, IL: University of Chicago Press.
Kellner, D. (1992) The Persian Gulf TV War, Boulder, CO: Westview.
Media Guardian (February 17, 2003) “Spin caught in a web trap.” Media.Guardian. co.uk/
iraqandthemedia/story/0,12823,897330,00.html
Reese, S. (2001) “Framing public life: a bridging model for media research” in S. Reese,
O. Gandy, and A. Grant (eds), Framing Public Life: Perspectives on Media and Our
Understanding of the Social World, Mahwah, NJ: Erlbaum.
Reese, S. and Buckalew, B. (1995) “The militarism of local television: the routine framing
of the Persian Gulf War,” Critical Studies in Mass Communication, vol. 12, pp. 40–59.
264
M I L I TA R I Z E D J O U R N A L I S M
Shoemaker P.J. and Reese S.D. (1996) Mediating the Message: Theories of Influences on Mass
Media Content, New York: Longman.
Smith, J. (1992) George Bush’s War, New York: Holt.
265
14
WAR O R PEACE?
Legitimation, dissent, and rhetorical closure in
press coverage of the Iraq war build-up
The global space within which much news and media comment are produced and
circulate has never been clearer than in the contentious build-up to the recent
UK/USA war in Iraq. As disputes within and between national governments over
the very definition of the issues at stake intensified, the global circulation of critical
perspectives on the expected war was striking and cut across the divisions between
official government positions. Whatever the local tendencies toward closure of the
issues from a specific national perspective (and as the war began in the UK, those
tendencies intensified), in order to understand the conflict fully, it is essential to
comprehend the global character of dissent and opposition. The global nature of
elite media and political discourse was matched by the globalization of opposition
to a UK/USA invasion of Iraq. On February 15 over 8 million people marched in
five continents to express their dissent (although the large majority of them
marched through the streets of major cities in western and southern Europe). The
analysis of both media discourse and popular dissent as a consequence demands a
cosmopolitan approach (Beck 2000). In this chapter we will focus on press dis-
course in the UK but we see this very much as a contribution to a broader
cosmopolitan project that does not, however, overlook national specificities.
There are good reasons to focus on the UK beyond the limitations of the
authors and their circumstances. Not only was the UK America’s closest ally,
diplomatically and militarily, but the UK government was renowned for its public
relations, having won two landslide elections in 1997 and 2001 with apparent
ease, leaving the major opposition party in disarray. Moreover, it was unusual that
a supposedly left-of-centre government, unlike other European social democratic
parties, should support a neo-conservative US Republican executive and that
consequently the two major UK political parties were united in their support for
the USA. Despite this, only 38 percent of the British population surveyed in an
opinion poll supported a “unilateral” war (a war without UN sanction) against
Iraq immediately before the outbreak of war. (After the advent of war there was a
dramatic shift in favor of military action.) The low point in terms of support for
266
WA R O R P E A C E ?
the war was between mid-January and mid-February 2003 when opinion polls
revealed that only 30 percent and 29 percent of representative samples of the
British population supported war (The Guardian, ICM). On February 15 an
unprecedented one and a half million people marched through the streets of
London to voice their dissent. The numbers took most people by surprise. Only a
few days before the march newspapers were predicting 500,000 demonstrators but
it was clear that a momentum was developing. A march organizer commented
that week: “it’s a new movement, out of anyone’s control. It’s like a tidal wave.
The people organizing it are not in control. It has its own momentum” (Burgin,
The Guardian, December 2, 2003: 6).
We will analyze the reporting of the conflict by seven national newspapers
during a key week of this low point in support for war. The first day of analysis
coincides with the publication of the Blix Report on January 27. George Bush
delivered his second State of the Union speech to Congress on January 28. Tony
Blair travelled to Camp David for “a council of war” at the end of the week (at
which time journalists were presented with copies of a new intelligence dossier
quickly dubbed the “dodgy dossier” by most of the British media because of its
extensive plagiarism of dated academic work downloaded from the Internet and
passed off as based on new intelligence sources). It was thus a key week for news-
papers to take their position with regard to the possibility of war.
The degree of dissent from the pro-war position of the UK government and
official opposition poses an interesting but welcome problem for critical media
researchers. It has become the received wisdom among critical media scholars
that the mainstream media generally act as handmaidens to the public relations
state in the manufacture of consent. Whether or not this describes accurately the
normal relationship of media and state, it is clear that, during the early months of
2003 at a time of geopolitical crisis, relations between some sections of the mass
media and state were and indeed remain severely strained. The degree of media
dissent may have also helped to legitimate and mobilize popular dissent, although
we also argue that the narrow terms on which some of that dissent was drawn
may, in the longer-term, have contributed to the fragility of the anti-war majority.
The relationship between media dissent and popular dissent is, of course, com-
plex and multi-causal, and requires, ideally, an holistic approach, both to media
(texts, production, and consumption) and to broader social and cultural change,
beyond that which we can attempt here.
267
NICK COULDRY AND JOHN DOWNEY
key question for understanding how social change occurs in complex, mediated
societies and is essential for assessing the prospects of democratization. Echoing
the work of Alberto Melucci (1996) and others, Habermas argues that the great
issues of the last decades—feminism, ecology, nuclear disarmament, and global
poverty—have all been raised initially by new social movements and subcultures
who through effective dramatization (for example, by non-violent symbolic acts of
civil disobedience) of their concerns have persuaded the mass media to place the
issues on the “public agenda.” Of course, while opposition to the war was surprising
and momentous, it prevented neither the UK’s participation in the war nor a
sudden shift in public opinion in favor of war in March and April 2003.
While Habermas’ account (1996) possesses a certain plausibility, it needs to be
supplemented by considering how these groups may penetrate the confines of the
public sphere. Habermas seeks to explain this largely in terms of the mass media’s
self-understanding in liberal democratic societies (rightly or wrongly) as objective
observers of society. However, the ability of counter-publicity groups to make
their voices heard in the mass media depends not only on this self-understanding
but also on the existence of crisis in the public sphere, manifested through medi-
ated disagreement and controversy within economic, political, and cultural elites.
The destabilization of the public sphere is both a top-down (centre–periphery)
and a bottom-up (periphery–centre) process whose dimensions may be mutually
reinforcing. It follows that, to understand both the generation and outcomes of
crisis, one must grasp the dynamic relationship within and between elite and pop-
ular discourses, and between actors in the mass media public sphere and in the
counter-public sphere (Downey and Fenton 2003). Indeed, this is an essential
and overlooked task if we wish to understand social change in global modernity
(Fenton and Downey 2003).
We are interested primarily in three broad processes: the construction of con-
sensus (and dissent), the construction of authority (specifically authority to
represent the reality of what is happening in the world), and the naturalization of
facts or frameworks of interpretation (Potter 1996).
Taking these in turn, the build-up to a major international war is, obviously, a
time when many actors are intensely concerned with the representation, or con-
struction, of consensus around that war; what was immediately striking, however,
from the early days of the Iraq war build-up, was the degree to which consensus
against the war was also being constructed not just by media, but also by elements
within the military, diplomatic, political, and cultural elites. This was why we
chose the representation of consensus and dissent as our principal focus from the
outset. Consensus is however never just consensus; it is used, rhetorically, as a
warrant of truth (Potter 1996: 117). Hence the importance of the second theme:
the construction of particular actors as “entitled to know particular sorts of things
[so that] . . . their reports or descriptions may thus be given special credence”
(Potter 1996: 114), against which there is the equally important construction of
other actors as having a “stake” in this or that statement which disqualifies them
as credible sources (Potter 1996: 124–5). The construction of consensus and
268
WA R O R P E A C E ?
authority occur within a third and wider construction, more difficult to detect:
what Potter calls “constructing out-thereness,” that is, the construction of certain
claims “as not being constructed’ (Potter 1996: 151, added emphasis).
This is a complex process: certain major explicit claims are presented as simply
factual (and therefore beyond contestation) on the basis of other claims that are
left implicit (but whose obviousness is assumed). The selection of background
and foreground “facts” is obviously crucial to what forms part of the apparently
natural “surface” of events and what does not. During the Iraq war build-up the
relative exclusion of certain issues from the frame of possible discussion (for
example, perspectives which challenged the relevance and justification of the US
timetable toward war) was important if other claims and statements (specifically
US and UK claims about what was happening) were to appear as “just” facts. This
complex process of light and shade is what Steve Woolgar has called “ontological
gerrymandering” (quoted in Potter 1996: 183–4). There was a lot of it around in
the early months of 2003.1
Our analysis focuses on press articles from the six days beginning January 27,
2003.2 Seven newspapers were chosen (the four broadsheet dailies—Daily Tele-
graph, Times, Guardian, Independent—and the top three tabloid dailies in terms of
circulation—Sun, Daily Mirror, Daily Mail) to represent broadsheet and tabloid
opinion in the UK. All war-related articles were analyzed (the initial selection
used the Lexis-Nexis database and contained 955 articles), from which articles
(news items, but also editorial and “independent” comment columns) were chosen
for a more detailed discourse analysis on the basis of being broadly representative
either of the discourse positions and/or rhetorical strategies of newspapers. A full
list of the latter articles is contained in Appendix 1. Our analysis does not, there-
fore, pretend to be an exhaustive study of the full range of comment present (or
absent) during this period (this would have required a much more extensive study
that would have also considered images as well as written texts), but rather an
indicative analysis of certain key discourse positions that seem to us significant in
the broader construction of the crisis. Different discourse positions could be found
within the same newspaper during the Iraq crisis. This is indicative in itself of both
crisis and flux in the mass media public sphere concerning the then impending
invasion of Iraq.
269
NICK COULDRY AND JOHN DOWNEY
same day as a letter signed by eight European leaders in support of the US’s
stance, even holds out the prospect of constructing a united European–US posi-
tion. The chief stumbling block to this, of course, was the Franco-German
position that the UN inspectors should be given more time to complete their
work and that war should be contemplated only as a last resort. The predicational
strategy of the Times is illuminating. Jacques Chirac is accused of “posturing.”
This implies that his present opposition is not sincere, calculated in order to
bring about certain effects that would be to the advantage of France, and that
France’s opposition will be reversed once suitable accommodations could be
found. Schroeder is accused of “strategic pacifism.” Given that Germany under
Schroeder took part in the Kosovo conflict (its first military engagement since the
end of World War II), it is somewhat curious to accuse Schroeder of pacifism. Of
course, the charge of pacifism means that one can both explain and dismiss Ger-
many’s opposition by reference to this principle without having to justify the
present conflict by reference to the principle of a “just war.” The Times not merely
misrepresents the German position but also attempts to dispel pacifism’s positive
connotations by suggesting Schroeder’s was not a principled pacifism but adopted
for strategic reasons. Without claiming the Franco-German position was some-
how interest-free, our point is the Times’ contrast between the “universal”
interests represented by the UK/USA position (international security) and the
“particular” interests ascribed to the Franco-German position.
This editorial position had been developed in a comment article by Daniel
Finkelstein on January 28. Finkelstein supports war, with or without the UN’s res-
olutions. Finkelstein adopts a Kantian-sounding moral vocabulary that gives the
impression of possessing some intellectual authority. We have, according to
Finkelstein, a moral duty or obligation to maintain international security and this
demands that we should support the invasion of Iraq whether it has the sanction
of the UN or not. Whereas the UK/USA is presented as obeying a Kantian cate-
gorical imperative and as acting selflessly, the UN as an institution is brought into
question: “the Security Council is not a panel of disinterested philosophers. Its
decisions all too often are based on national prejudice, imperial adventurism, the
vanity of individuals, and the murderous impulses of dictators.” This juxta-
position of the UK/USA and the Security Council is contradictory and rather
ironic bearing in mind that the UK and USA are two of its five permanent mem-
bers and thus are presumably as interested as other members. The article’s clear
strategy is to remove the argument from matters of fact (whether or not Iraq pos-
sesses WMD and poses an imminent threat to the world) and, assuming that
“fact” as widely recognized, to convert the argument to one about morality. The
moral case for war is wrapped in a pseudo-Kantian vocabulary and presented as a
contrast between the dutiful and selfless UK/USA (going to war to protect the
universal good of international security) and the war’s immoral opponents.
The editorial of the Daily Telegraph “Why Britain should fight” on the day of the
publication of the Blix Report (January 27: p. 21) admitted that three-quarters of
the British public were opposed to war and argued this was because anti-war cam-
270
WA R O R P E A C E ?
paigners were presenting the better argument; Tony Blair by pursuing a “narrow
legal” case for war against Iraq (i.e. via UN resolutions) had left the majority of the
public confused as to “what they are fighting for.” The only way to overturn the
anti-war consensus, the editorial argues, is to invoke the national interest irrespec-
tive of the reports of the UN inspectors; not only is the regime of Saddam a
military threat to the UK but also “let us not be shy of saying that it is in no one’s
interest for the [sic] some of the world’s key oil supplies to be in the hands of an
unstable dictator.” Ultimately, then, the war is about “who is the boss.” An Anglo-
American hegemony would also be good for Iraq, the region, and the world.
The Daily Telegraph is here arguing for a new era of imperialism based on liberal
representative democracy and free trade under the auspices of the benign powers
of the UK/USA. The account that it provides is strikingly similar to the radical
critique of the war aims. Of course, what is different is the evaluation of the out-
come. Prima facie, the Daily Telegraph’s assumption that, once anti-war
campaigners’ diagnosis of the war rationale is admitted, the majority of public
opinion will switch from being anti- to pro-war, is paradoxical; the paradox disap-
pears, however, if one assumes a natural consensus in favor of that rationale, once
directly stated.
While the Daily Telegraph’s assessment may indeed have been close to the un-
official government reasons for going to war, the open espousal of such a position
hardly helped Blair who at this stage was relying on winning UN Security Coun-
cil support for a war to win over public opinion and, therefore, emphasizing
the supposed threat of Saddam rather than the benefits of “regime change.” Not
only therefore was there no consensus for war but also no consensus among the
war’s supporters about how to wage the rhetorical battle for public opinion.
Indeed the clear anti-war consensus meant that assorted supporters of the UK
government felt at liberty to advocate various rhetorical repair jobs, thus adding
to the sense of confusion concerning the war’s justification and the impression
that the official justification was a screen to cover imperial ambitions. (In this
context, the contradiction between the Daily Telegraph’s claim that the UN
inspectors were irrelevant on January 27 and its editorial (January 28: 21) the day
after the Blix report’s publication stating the “case for war [was] still strong” seems
less surprising.)
The Daily Telegraph’s discourse position was consistently adopted across genres
(news reports, comment columns, and editorials). Even the devastating and sur-
real comment article by comedian Armando Iannucci, that offered an immanent
critique of the UK/USA attitude toward the authority of the UN and the notion
of a pre-emptive self-defense, may be seen as consistent with the newspaper’s
stress on realpolitik (the overwhelming importance of projecting Anglo-American
power in oil rich regions of the world).
Whereas the Daily Telegraph clearly supported a war against Iraq if not entirely
for the reasons used by the UK Government, the Daily Mail came out against the
war in editorials on January 27 and 28 (p. 10 on both occasions), stating that the
UK and USA had failed to provide evidence that Iraq was an “imminent threat”
271
NICK COULDRY AND JOHN DOWNEY
and consequently the war was not justified. The Daily Mail’s doubts went beyond
the evidential, however. Rather than sharing the Daily Telegraph’s judgment that
a post-Saddam Anglo-American empire would be good for UK and global inter-
ests, the Daily Mail, after noting the great likelihood of conflict, comments: “at
what cost to the Middle East, world oil supplies, the war against terrorism, the
Western alliance and the public’s trust in the prime minister remains to be seen”
(January 27: 10). It is not that the Daily Mail is against an Anglo-American
empire, just that it believes that this enterprise is likely to backfire. The Daily
Telegraph and Daily Mail agree on the criteria by which the world should be
judged but have radically different projections of the consequences of war.
Whereas the Daily Telegraph was consistent in its discourse position, the Daily
Mail adopted a number of positions across different genres. Its columnists, for
example, ranged from the skeptical (in line with the editor: for example Peter
McKay January 27: 13; Keith Waterhouse January 27: 14) to that of Melanie
Phillips (January 27: 10) who bolted a “clash of civilizations” thesis between the
“West” and Islam onto a “decline of European civilisation” argument, reaching
general conclusions about Islam from the activities of neofundamentalist groups
and distinguishing liberal Europe (unwilling to defend itself, therefore likely to be
crushed by the “Islamist tiger”) from the USA (strongly nationalist, deeply reli-
gious, and prepared to fight). The meaning of Phillips’ argument is clear: Europe’s
survival is dependent upon becoming more like the USA and rejecting liberalism
in all shapes and forms.
Despite an editorial line that was skeptical of the UK/USA position, Daily Mail
news journalists accepted the UK/USA framing of the conflict. Thus coverage
written by David Hughes of the Blix report claimed the report exposed Iraq’s
“charade” (adopting uncritically the phrase of Jack Straw), so that “the count-
down to war quickened last night” (January 28: 4–5). The illogical idea of a
countdown quickening (rather than, say, being continued or interrupted) is a
strategy of intensification taken from the UK/USA. The elision of the actors (the
people setting up the “countdown”) serves to make conflict appear an unavoid-
able, natural process rather than a humanly constructed, and thus entirely
mutable, series of events. The same journalist employs the same strategies a day
later when writing of the “looming conflict” (January 29: 15) as though the con-
flict had a life of its own, beyond human control.
The Independent adopted a consistently anti-war position across genres with
Robert Fisk spearheading its coverage and analysis of the conflict. Fisk used the
occasion of the day of the Blix Report to launch a broadside against UK/USA
“deceptions” (January 27: 5). The first “deception” is that Saddam is a dictator
who poses an imminent threat to the region and the world in the manner of
Hitler’s Germany in the 1930s, making the anti-war position one of “appease-
ment.” Not only was this Saddam/ Hitler elision intended to bring the conflict
closer to home, thereby making the threat appear more real, but it also borrowed
World War II’s legitimacy for the present conflict while intimating a successful
conclusion. This was a key strategy of the UK/USA because it provided the moral
272
WA R O R P E A C E ?
justification for war and was an argument designed to appeal to liberals, leftists,
and pacificists by questioning the morality of their moral opposition to the pre-
sent conflict. The difficulty in disrupting this analogy lay, Fisk argued, in the
obviously brutal character of Saddam’s regime. The easier task was to disrupt the
idea that Saddam posed the same threat to the world after a crushing military
defeat in the 1991 Gulf War and 12 years of sanctions and containment as Hitler
did in the 1930s after the German annexations and invasions of Czechoslovakia
and Poland. That is, of course, why the issue of “weapons of mass destruction” was
crucial. Fisk asks whether “we are prepared to pay the price of so promiscuous a
war” and points to the likelihood that thousands of Iraqis will die and that the
UK/USA will be seen as an occupying power that will strengthen support for
neofundamentalist groups. The second deception, Fisk argued, was that the war
was not about oil. While the UK/USA insisted that the war was exclusively about
WMD, the Iraqi regime and protesters insisted that the war was about the imper-
ial control of Iraq’s oil and that the issue of WMD was a rhetorical fig leaf to
cover naked ambition and self-interest. This was the central argument of more
radical anti-war protesters for whom the UN “weapons inspections” were a public
relations charade. This radical position seemed to win widespread support in
everyday life in the weeks leading up to the war. Even many war supporters did
not believe the UK/USA official version of the war.
Fisk presents the UK/USA as relatively isolated: “The only other nation push-
ing for war—save for the ever-grateful Kuwait—is Israel.” This serves of course to
emphasize the lack of consensus in favor of war internationally and to damn the
UK/USA through association with an already occupying power. Domestically the
populations of the UK/USA, despite “being told to go to war by their newspapers
and television stations and politicians,” are becoming increasingly skeptical of
the claims of their governments. Indeed, the “popular” consensus in Britain is
anti-war. What Fisk does not explain, however, is how this might be so: is this
popular anti-war consensus generated from the periphery? Does Fisk overstate the
elite consensus? Or do both play a role in the generation of popular dissent? In
any case, it is this “popular” consensus for which Fisk claimed to speak.
A striking feature of press coverage in this period, notwithstanding this signifi-
cant dissent about the ends and means of war, was the de facto consensus
constructed around the time-frame of the UK/USA war build-up. The dominant
news-frame almost everywhere was the momentum building toward war around
the UK/USA diplomatic agenda. Turning to the three remaining papers in our
sample, this was virtually the only perspective referred to in the Sun and it also
dominated the news coverage in the Guardian; only in the Daily Mirror did other
perspectives contribute to news reports, and then always within a context deter-
mined by the UK/USA official agenda.
Since the Sun has been the most belligerent UK newspaper, its construction of
national and international consensus for its position was hardly surprising. This
was expressed not only in terms of UK “hearts and minds” (January 30: 9) but also
in terms of a broad coalition of “the West v the Rest” (headline January 30: 9)
273
NICK COULDRY AND JOHN DOWNEY
and even “the world” being on course for war (January 28: 8). There were other
more disturbing aspects to this construction of consensus: the denigration of
Islam and asylum-seekers set up a situation where Muslims were seen as the
“enemy within.” Where dissent from this “consensus” was mentioned, it was
always in disparaging terms: unspecified “anti-war campaigners” (January 28: 8),
“rebel lefties” and the “loopy left” (January 30: 9).
Since the Guardian was the newspaper whose editorial opposition to war was
most predictable from its general discourse, its reproduction of the momentum for
war in its lead news items is more surprising: for example, the suggestion of a diplo-
matic “consensus” after the Blix report that Iraq was not cooperating (January
28: 1); in addition, the isolation of France implied by its comment (after the Blair/
Bush summit on January 31) that, while the UK and US were seeking to convince
“the international community,” Blair felt increasing “frustration with the French”
(who, however, would have “the squeeze” put on them) (February 2: 1).3 The sig-
nificance of these suggestions in Guardian news reports emerges more clearly when
we look later at their close reliance on UK and US diplomatic agendas; for now, we
should just note that it was at odds with the Guardian’s clear editorial position
(January 28, January 30) against the war.
In contrast, the Mirror followed its editorials’ anti-war position into its news
articles, interpreting diplomatic reactions after the Blix report as a consensus
against war that left the US isolated (January 28: 4) and mocking Blair’s Janu-
ary 29 House of Commons performance as “My war against the world” (headline,
January 30: 2). Here there was an overlap with the editorials, if not the news cov-
erage, of the Guardian which argued (January 28: 21) that the UK/USA reaction
to the Blix report “will not be how most of the world views” that report and (Jan-
uary 30: 23) insisted that Blair should overturn his existing pro-war, pro-Bush
policy and instead “speak for this nation.” The idea that, far from war tapping
into a national consensus, war went directly against the national consensus
(noted already in Robert Fisk’s writing for The Independent) was developed later in
the week by the Daily Mirror (January 31: 6), reporting its commissioned YouGov
poll that showed 75 percent currently against the war and, more strikingly, only 2
percent believing that the war would make them safer from terrorist attack. The
resulting image of Blair as the isolated leader battling against the tides of popular
opinion remained, however, ambiguous, as we note below.
To sum up, the Times, Daily Telegraph and Sun (the two biggest circulation
broadsheet papers and the biggest tabloid) supported the UK/USA war at this
stage but the last two used arguments for war (for example, control over oil, the
West versus the rest) that were antithetical to the official UK/USA position.
These arguments presumably only helped to confirm popular doubts about the
truthfulness of the official line. Only the Sun claimed an international and
domestic consensus existed in favor of war. The other newspapers clearly recog-
nized and commented upon both the international and domestic absence of such
a consensus, even if in more subtle ways they generally reinforced, rather than
challenged, the event-frame assumed by the UK/USA position. The four news-
274
WA R O R P E A C E ?
papers that took anti-war editorial stances did so for contrasting reasons. While
the Daily Mail was simply concerned about whether the national interest would
be served by war, the Independent, Guardian, and Daily Mirror raised moral objec-
tions concerning the consequences of war.
Media elites, then, were split not only in terms of which action would further
national interests but also of which actions were moral. The confusion between
the two in anti-war positions is one explanation why, once the war started, some
newspapers and some of the public swung in behind the UK/USA position. It was
not that they supported the war but that once the war which appeared irre-
versible had started, apparently consensual appeals to “the nation” (for example,
the call to support “our” armed forces) trumped prior doubts concerning whether
the war was, in fact, in the national interest.4
275
NICK COULDRY AND JOHN DOWNEY
fact (“Saddam is using guerrilla tactics to sabotage the hunt for his doomsday
weapons, it emerged last night,” January 27: 2); in the Guardian (January 27: 1)
the story was the fact of the intelligence briefing itself and its diplomatic signifi-
cance, although there was little reference in its main news report to alternative
interpretations of the claims in the briefing. Only in the Mirror was there substan-
tial skepticism, with its sub-headline “War in weeks as Blair gives ‘evidence’ for
attack” (January 27: 4).
Similar differences were played out in news treatment of diplomatic sources.
While the Sun presented UK and US diplomatic sources without any suggestion
of distance from them, the Guardian on occasion indicated skepticism (for exam-
ple in noting (January 28: 1) the difference between the UK Foreign Secretary’s
“bellicose” interpretation of the Blix report and the UK ambassador to the
United Nations’ more cautious interpretation). On other occasions, however, it is
striking how close the Guardian’s news reports stayed to the interpretation that
the UK and US administrations were encouraging; its front page February 1
report on the Blair/ Bush Washington summit read more like a Whitehall press
release (“Mr Blair impressed on the Americans . . . Mr Blair secured support [from
the US],” and so on). The implication—one that UK diplomats no doubt encour-
aged—was that the summit was about diplomacy (Blair restraining Bush from war)
even though, as the same report made clear, Blair had already secured Bush’s sup-
port on the need for a second UN resolution by phone on the evening before the
summit. Why then the time and expense of Blair’s transatlantic visit? The reason,
already anticipated in media comment earlier that week, emerged clearly in the
Sun’s news report, but was fudged in the Guardian: “The President and the PM
thrashed out final details for an onslaught beginning in mid-March—as exclu-
sively revealed in yesterday’s Sun.” So much for diplomats’ claim (reported by the
Guardian without demur) that the summit was a “council of diplomacy”! Only
the Mirror kept a more consistent distance from official UK/USA sources.
A quite different issue of authority concerned Blair’s own standing as prime
minister. Some personalization of the war build-up is hardly surprising. The per-
sonalization, however, that really mattered for the British public’s perception of
the issues at stake concerned Blair himself. A theme, more dominant in press
coverage nearer to the outbreak of war, was the presentation of Blair as the lone
leader, bravely opposing the skepticism of his people at considerable personal
cost. It is worth noting the assumptions about the credibility of Blair’s self-
representation as a man of “ideals” upon which this depended. Possibly the
strongest attack on Blair’s policy during the week we analyzed came in a Guardian
editorial (January 30: 23), which argued that his policy would have results
directly at odds with his ideals (of “global justice” and so on). Even this criticism
already conceded that the prime minister was motivated by “ideals,” rather than,
say, by a calculation of Britain’s strategic interests; yet this was a reading of Blair’s
actions and motives on which he later played himself when under maximum
pressure just before war started. We see how, behind the surface of dissent from
the British government’s position, there were significant limits to that dissent.
276
WA R O R P E A C E ?
277
NICK COULDRY AND JOHN DOWNEY
writing for the Daily Mirror, Independent, and Guardian who directly raised oil
resources as a key reason lying behind the UK/USA drive to war.
The naturalization of the UK/USA perspective took the form, first, of constant
references to “time running out” for peace. This was, of course, the stated
UK/USA position, but it became naturalized when, for example, UN inspectors
were described as having “earned themselves” a few further weeks, whereas the
UK had “nudge[d] back Bush’s decision to go to war” (The Guardian, January
28: 1); or when the UK’s release of “intelligence” information just before Blix’s
report was described as if it were a disinterested speeding-up of the weapons
inspections: “Britain is aiming to prevent the process from dragging on indefi-
nitely, by handing over and publicizing sensitive intelligence which allegedly
shows that Iraq is flouting the UN” (January 1: 1). The word “allegedly” hardly
counters the naturalizing force of that apparently neutral phrase “the process.”
“The process” is not the UN inspections as such, but those inspections as inter-
preted by the US and UK (as “delay” to their underlying war timetable; otherwise
how could a few months’ inspections be seen as “dragging on indefinitely”?). The
Guardian’s editorial (January 28: 23) made a concerted effort to dislodge this nat-
uralization, by arguing that it was the inspections process that was “natural” and
should be left undisturbed: “if the Bush administration and its admirers wish to
curtail or cancel this UN process, after a mere two months or so, it is up to them
to explain why. They have not done so to date . . . ” Yet this fundamental point
failed to influence the Guardian’s own news reports later in the week, most impor-
tantly in its reports of the Blair/ Bush summit (see above).
It is worth noting what metaphors came to dominate in newspapers’ coverage
of the war build-up. We might have expected the metaphor of war as something
to fear (Mirror, January 27), although in fact it was surprisingly rare; we might
also have expected the Sun’s celebration of Blair confronting his critics in the
House of Commons on January 29 as a man of action: “the Prime Minister raised
the stakes . . . [he] was stung into action . . . under fire from all sides . . . it was the
first time he had lined up Stalinist tyrant Kim Jong II for a possible military
strike.” With an image of a watchful rifle-carrying UK soldier above the article,
and a comic-book picture of the globe with members of the “axis of evil” named
within jagged balloons (like mini-explosions), it was almost as if the war had
begun (Sun, January 30: 8). Less expected, however, was the way that this “Boy’s
Own” picture of Blair—as the isolated, embattled, but brave quasi-military
leader—circulated beyond the pages of the Sun and into articles elsewhere that
prima facie were strongly critical of the Prime Minister (for example, the colum-
nist Jackie Ashley’s article: The Guardian, January 30: 21).
Such an idea of Blair as the embattled leader was however double-edged, as
became clear in the article published in the Mirror the next day (January 31: 6)
by Ashley’s fellow Guardian columnist, Jonathan Freedland, headed “A leader
who has left behind his people.” This article analyzed the devastating findings of
that day’s YouGov poll (referred to above) and concluded that Blair was isolated
from his people as never before. However:
278
WA R O R P E A C E ?
Not surprisingly, this was a line used to great effect by Blair’s supporters (and even
some of his formal Conservative opponents) later in the war build-up. The article
concluded:
How will historians look back at this solo stance by Tony Blair? That
depends on the outcome of the coming war. But they will either say this
was his defining act of great statesmanship—or the decision that ulti-
mately led to his downfall.
Naturalized here are a number of assumptions: first, that the war was inevitable
(remember this “critical” piece was written a full two weeks before the largest of
the global anti-war protests on February 15); second, that Blair’s position was dic-
tated solely by a sense of what is right (otherwise, how can the mere success of the
war be grounds for attributing his stance to “statesmanship,” rather than, say,
lucky miscalculation?); third, that Blair will survive, if he does, because his policy
proves a success, rather than because his opponents fail to oppose him (much
closer to the truth, as we write); and finally, and most obviously, that if the war is
a “success” on UK/USA terms, it will be impossible to interpret otherwise than to
Blair’s credit, which precisely reproduces the UK/USA framework for interpret-
ing the build-up to war as “inevitable.”
If the image of Blair, the isolated leader, rose to prominence, it is worth asking
what other themes (less favorable to the UK/USA position) received less promi-
nence in UK press coverage. A minimal list would be:
279
NICK COULDRY AND JOHN DOWNEY
Yet it must also be noted that, in contrast to this naturalization of the UK/USA
framework for interpreting the coming war, the Mail, Guardian, Independent, and
Mirror began to give the anti-war movement greater prominence, especially in
the period directly after our sample week. The Mirror was the most campaigning
anti-war paper, urging its readers to sign a petition (by February 15 it claimed to
have collected 195,000 signatures) and also sponsoring the “Stop the war coali-
tion” march on February 15. On the morning of the march the Mirror headline
read “The world against the war” with the contents devoted to details of the glob-
alization of the anti-war protests and accounts of the preparations for the London
deomonstration. The Guardian and Independent ran stories about first-time “ordi-
nary” protesters. The Mail provided a map for its readers wishing to join the
march. The anti-war movement, if only for a while, cut across ethnic, class, reli-
gious, and political boundaries, as newspapers helped to construct as well as
simply reflect the diverse character of the movement.
Conclusion
How are we to make overall sense of this complex picture? The fundamental
point is that elite media and political discourse concerning the waging of a war
against Iraq in the UK was deeply divided. Opinion was divided concerning
whether war was in the national interest and/or right morally. Generally speaking
right-wing papers (Times, Telegraph, Sun, Mail) either supported or opposed the
war along lines of perceived national interest whereas liberal and left-of-centre
newspapers (Mirror, Independent, Guardian) employed arguments questioning the
morality of the UK/USA position. Divisions within elite media discourse and the
consequent legitimation of dissent helped to establish the preconditions for a suc-
cessful mobilization of one and a half million people on the streets of central
London in winter. To understand this mobilization fully, one must acknowledge
not only the legitimacy crisis in the public sphere but also the creative disobedi-
ence of counter-public spheres and alternative media in encouraging such a
display of public opposition.
Of course, public opposition to the war did not prevent it taking place (for
reasons which we have also explored) and, when the war started, media represen-
tations and public opinion shifted to being pro-war. In the longer term, as the
memory of “liberation” fades in the face of the realities of occupation, critical
media voices are returning and popular disaffection is growing. In the post-war
situation, there is no naturalized “timetable” on which the UK/USA position can
rely to close down popular dissent. On the contrary, the situation in Iraq, the UK,
and the US is open-ended and uncertain. It remains to be seen what conse-
quences the long-term legacy of dissent from the war at all levels will have for
national and international politics.
280
WA R O R P E A C E ?
Notes
1 Nor of course, in the UK, was it unique to this period: compare Fairclough’s analysis of
New Labour language in relation to “the international community” during the Kosovo
war (Fairclough 2000, p. 152–3).
2 Sunday papers were excluded since the principal war-related events (Blix report, State
of Union address, Blair–Bush summit) all occurred during Monday to Friday.
3 The same article’s sub-headline was “Blair gains extra time to win over waverers.”
4 There were other reasons for this shift: the limitations inherent in the framework of
mediated dissent (see next two sections) and probably a general fatalism (cf Croteau
1995, p. 115).
References
Beck, Ulrich (2000) “The cosmopolitan perspective: sociology of the second age of moder-
nity,” British Journal of Sociology, vol. 51, no. 1, pp. 79–105.
Chiapello, Eve and Fairclough, Norman (2002) “Understanding the new management
ideology: a transdisciplinary contribution from critical discourse analysis and new soci-
ology of capitalism,” Discourse and Society, vol. 13, no. 2, pp. 185–208.
Croteau, David (1995) Politics and the Class Divide, Philadelphia, PA: Temple University
Press.
Downey, John and Fenton, Natalie (2003) “New media, counter publicity and the public
sphere,” New Media and Society, vol. 5, no. 2, pp. 185–202.
Fairclough, Norman (1995) Media Discourse, London: Arnold.
Fairclough, Norman (2000) New Labour, New Language, London: Routledge.
Fairclough, Norman and Wodak, Ruth (1997) “Critical discourse analysis” in T. Van Dijk
(ed.), Discourse as Social Interactions, London: Sage, pp. 258–84.
Fenton, Natalie and Downey, John (2003) “Counter public spheres and global modernity,”
Javnost, vol. X, no. 1, pp. 15–31.
Habermas, Jurgen (1996) Between Facts and Norms, Polity: Cambridge.
Hall, S., Critcher, C., Jefferson, T., Clarke, J., and Roberts, B., (1978) Policing the Crisis:
Mugging the State, and Law and Order, London: Macmillan.
Melucci, Alberto (1996) Challenging Codes, Cambridge: Cambridge University Press.
Potter, Jonathan (1996) Representing Reality, London: Sage.
Wodak, Ruth and Meyer, Michael (eds) (2001) Methods of Critical Discourse Analysis, Sage:
London.
Appendix 1
List of 2003 articles chosen for detailed analysis:
January 27
The Daily Telegraph, editorial, p. 21
Daily Mail, editorial, p. 10
Daily Mail, comment, Peter McKay, p. 13
Daily Mail, comment, Keith Waterhouse, p. 14
Daily Mail, comment, Melanie Phillips, p. 10
281
NICK COULDRY AND JOHN DOWNEY
January 28
The Daily Telegraph, editorial, p. 21
Daily Mail, editorial, p. 10
Daily Mail, news, David Hughes, pp. 4–5
The Times, comment, Daniel Finkelstein
The Guardian, lead story, Borger/White/Macaskill, p. 1
The Sun, Kavanagh/Flinn article, pp. 8–9
The Mirror, Wallace article, pp. 4–5
The Guardian, editorial, p. 21
The Sun, editorial, p. 8
Mirror, editorial, p. 6
January 29
Daily Mail, news, David Hughes, p. 15
January 30
The Times, editorial
The Guardian, lead comment, Jackie Ashley, p. 21
The Sun, lead comment, Kavanagh, p. 9
The Mirror, lead story, Hardy, plus Routledge comment, p. 2
The Guardian, editorial, p. 23
The Sun, editorial, p. 8
January 31
The Guardian, lead story, Wintour/Campbell, p. 1
The Sun, lead story, Pascoe-Watson, p. 2
The Mirror, lead comment, Jonathan Freedland, p. 6
The Mirror, pp. 1–9
February 1
The Guardian, lead story, Wintour/Borger, p. 1
The Sun, lead story, Pascoe-Watson, p. 2
The Mirror, news article, Roberts, pp. 4–5
282
15
HOW BRITISH TELEVISION
NEWS REPRESENTED THE CASE
FO R THE WAR IN IRAQ 1
Introduction
There has been much interest in whether there has been a fundamental transfor-
mation in governments’ attempts to manage news coverage of international
conflict in the post-Cold war era (Thussu and Freedman 2003). The effects of
globalization and technological development on the production, distribution,
and consumption of the media have been seen as crucial: the proliferation of new
sources of information (Allan 2002); the emergence of transnational, 24-hour
rolling news channels (Thussu 2003); the uncertain future of state-owned and
public broadcasters in the context of deregulation, privatization, and the growth
of transnational entertainment-based, mainly US-owned media conglomerates;
and the development of transnational media services such as Al-Jazeera serving
diasporic communities that transcend national boundaries (Miladi 2003). The
events of September 11, 2001 and the adoption by the US government of policies
based on pre-emptive military intervention have given urgency to this discussion
(Zelizer and Allan 2002; Thussu and Freedman 2003).
Two key aspects of the British news media’s role in covering the war in Iraq in
the UK cut across most of these themes. The first was an attack led by the British
government on the role of particular news organizations in undermining its case
for the war: its targets being Al-Jazeera and other Arabic media, and, most signif-
icantly, the BBC.2 The second was a debate about the implications of the
Pentagon-led policy of embedded reporting. Purportedly an exercise in facilitat-
ing access for news organizations to the frontline (enabled by the development of
portable communications technology), the policy had attracted criticism for
encouraging a sense of identification among reporters with the military units on
whom they were dependent for everyday survival.
Contributions to these two debates have often focused on particular episodes.
The most notable was BBC Radio 4 Today correspondent Andrew Gilligan’s report
that members of the British security services were unhappy that the government’s
283
JUSTIN LEWIS AND ROD BROOKES
dossier on Iraqi WMD had been “sexed up” by 10 Downing Street. In many quar-
ters, this single episode was held up as symptomatic of the BBC institutional
anti-war bias. In this article we take a broader view in order to test competing
claims over different broadcasters’ coverage of the case for the war. Our analysis is
based on an extensive content analysis of weeknight bulletins by the most-
watched UK television news providers during the war itself.
In the six months leading up to the beginning of the war, opinion polls sug-
gested that public opinion was divided. Those in support of a war against Iraq to
remove Saddam varied between 30 percent and 40 percent throughout those six
months.3 Support for the war with Iraq rose significantly on two conditions: first, if
it could be shown that Iraq possessed WMD, or second, if the war had the approval
of the United Nations. In an ICM poll for the Guardian, for example (September
20–2, 2002), only 37 percent approved of a military attack to remove Saddam
Hussein, as opposed to 46 percent against. However the number approving if the
government provided evidence that Saddam possessed WMD rose to 65 percent.
Conversely, polls indicate that even with evidence that Saddam possessed WMD
about a quarter of the population were still against a war. The polls therefore indi-
cated that there was a significant group who could be won over to either side
(Kellner 2003).
Three arguments were central to the case for war: the assertion that Iraq pos-
sessed—and might use—WMD; the brutality of the Saddam Hussein regime; and
the notion that an invasion was in support of (and supported by) the Iraqi people
themselves.
In the ensuing battle for public opinion, a significant shift occurred during the
war itself, when polls suggested support for the war—without the two conditions
having been met—rose to high enough levels for the government to be able to
claim majority (though by no means overwhelming) support. Research suggests
that this was partly a function of a residual feeling of patriotism—once war began,
some people felt it was important to put aside doubts and support the mission of
the British armed forces.4 But we were also interested to see how the govern-
ment’s case for war was presented during the conflict.
Methodology
We carried out a content analysis of the four main British television news sources
during the Iraq war in order to examine the patterns of coverage and to assess the
degree to which the government’s case for war was validated or undermined by
that coverage. We chose BBC News at Six; ITV Evening News at 6:30 p.m.; Chan-
nel 4 News at 7 p.m. and Sky News at Ten. The early evening news programs on
BBC and ITV were chosen because they are generally the most popular, and
because they were more consistent in terms of time slot and length throughout
the war than the late evening bulletins.
Each news bulletin was broken down into discrete units of analysis—something
increasingly difficult to do in contemporary styles of broadcasting, especially when
284
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
one news event dominates the news bulletin, and various reports flow into one
another. In this study a “report” involved an identifiable, authored segment of the
news program: a news anchor’s introduction—if it contained substantive informa-
tion in its own right—and a correspondent’s report would typically be coded as
two discrete reports.
The sample consisted of 1,534 reports, all of which were coded by type/
authorship. This allowed us to look at the overall shape of news coverage in
order to see which part of the news operations were important in relaying infor-
mation to viewers. The main categories of authorship used in our analysis were
as follows:
• Embedded reporter
• Baghdad reporter
• Qatar reporter
• unilateral reporter
• available footage
• studio analysis
• interview with reporter
• interview with expert(s)
• anchor.
285
JUSTIN LEWIS AND ROD BROOKES
footage.” This category could include footage from embedded reporters and uni-
laterals from the same news organization or others through pools, news agencies,
news exchanges as well as footage from the Pentagon and the Ministry of Defence
(MoD).
Stories filed by embedded reporters, Baghdad-based unilaterals, roving uni-
laterals and Doha-based correspondents, were coded thus when they were clearly
marked as such. Two-ways between anchors and correspondents in the region
were coded under the category of reporter they were with, on the basis that it was
the correspondent who provided most of the information during the exchange.
Some reporters—notably political correspondents—delivered much of their
reports during two-ways with anchors, and these were categorized as “interviews
with reporters.”
The first stage of analysis gave us a broad picture of how the war was covered.
The second stage of analysis focused on the three main themes central to the case
for war:
We explored the presence and treatment of these themes during the war, in order
to see the extent to which broadcasters embraced or rejected the government’s
case.
286
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
that the Iraqi people would welcome liberation, and that civilian casualties could
be kept to a minimum.5 We therefore coded every reference to the condition and
attitude of the Iraqi people in order to see whether they were represented as sup-
porting and celebrating the invasion, or as suspicious, angry, or hostile. We also
coded every reference to news of Iraqi casualties (although it would be a mistake
to assume that news of civilian casualties was necessarily a problem for the gov-
ernment’s case, as the numbers reported in these cases were usually quite small,
which could be interpreted as supporting the idea that casualties were minimal).
287
JUSTIN LEWIS AND ROD BROOKES
288
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
289
JUSTIN LEWIS AND ROD BROOKES
Table 15.2 Number of stories including one or more reference to themes relevant to the
government’s case10
a little less likely to concern itself with the state of the Iraqi people, while ITN
was more likely than any other channel to run stories on the themes of the Iraqi
people and the depravity of the Iraqi regime.
We then looked at who was used as the source of claims about these issues. This
includes all sources shown directly (i.e. through coverage of press conferences, in
interviews, etc.) as well as where such claims are attributed to a particular source
verbally by anchors or correspondents or visually through graphics. Obviously
any content analysis is limited in its ability to account for journalists’ use of
sources in that it cannot account for unattributed use of sources, unofficial brief-
ings, or off-screen use of official briefings. Nevertheless such an analysis indicates
which sources were given the opportunity to present their version of events as
authoritative.
We identified 185 instances of where on-screen sources were used to back up
claims related to the case for the war. Given that there were 549 reports that
included one or more reference to these claims, it is striking how sparing was the
use of identifiable sources overall. We are left with the conclusion that the claims
made in the vast majority of the stories were not backed up by an attributable
source. Overall, Channel 4 was the most likely to attribute claims: of the four
broadcasters in our sample, Channel 4 accounted for 40 percent of our total
number of sources identified.
Table 15.3 shows that of the sources used, most were, fairly predictably, official
US/UK government or military (48 percent), followed by official Iraqi sources
(30 percent). This almost certainly underestimates the importance of British or
US military sources, however, as many claims that came from military sources
were not attributed. So, for example, when we analyzed the television coverage of
four stories that came from military sources—all of which turned out to be
unfounded11—we found that nearly half the claims made were unattributed
(Lewis et al. 2003).
What is notable from Table 15.3 is that, despite allegations that the BBC’s cov-
erage was anti-war, the BBC made more use of British/US government and
military sources than any other, outnumbering the use of Iraqi sources by more
than 2 to 1. Channel 4’s coverage, by contrast, struck the closest balance between
US/UK and Iraqi sources.
Other types of sources were used sparingly. Iraqi citizens were not often used
290
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
as sources—indeed, in our sample the BBC did not use Iraqi citizens as sources
at all. Thus while a number of claims were made about the mood of the Iraqi
people, these tended to be made on their behalf. Other types of source were also
used—spokespersons for non-governmental organizations (NGOs) such as the
Red Cross—but these amounted to only 12 percent of occasions, with Channel 4
accounting for half of these.
These data confirm the extent to which the coverage focused on the war itself,
the majority of sources identified being government or military sources. Those
who might have shed light on the validity of this case as the war progressed—
whether NGOs, weapons inspectors, academics, or experts on and in the Arab
world—played very little part in the story told by television news.
291
JUSTIN LEWIS AND ROD BROOKES
Table 15.4 Number of stories including one or more references to Iraqi WMD capability
Implying capability 87 85 91 96 91
Doubting capability 33 20 19 4 15
292
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
Table 15.5 Number of stories containing one or more reference to the theme of the state
of the Iraqi people
more likely to support the government’s case than undermine it. And since much
of the pro-invasion imagery happened toward the end of the war—with a great
deal of coverage given to the “celebrations” and “joy” of Iraqis tearing down stat-
ues of Saddam Hussein in Baghdad and Kirkuk—it could be argued that the
lasting impression was of a people happy to be free of Saddam Hussein, and less of
a people disgruntled about being invaded and occupied.
293
JUSTIN LEWIS AND ROD BROOKES
Table 15.6 Number of stories including one or more references to the theme of Iraqi
regime depravity
Embeds or in-beds?
We cross-tabulated the types of news reports against the number of stories
including references that supported the government’s case for war, in order to
explore a number of key questions. Did the practice of “embedded reporters”
result in news coverage that was more sympathetic to the government’s case for
war than types of reports that used different newsgathering methods? Did the
deployment of correspondents under Iraqi government supervision in Baghdad
lead to stories about the public mood of Iraqi citizens as more hostile than the
government’s case suggested, or to more stories about Iraqi citizen injuries or
deaths?
We tabulated the stories including references supporting the government’s case
on the existence of WMD according to type of news report, and compared the
results with the overall distribution of our sample according to type of news report
(Table 15.7). So, for example, reports including references supporting the govern-
ment’s case on WMD were most likely to come from anchors (24 percent) but,
given that anchor reports dominate our overall sample (48 percent) this suggests
that anchors were proportionately less likely to refer to such claims. This is
indicative of the fact that the WMD story tended not to make headline news
(which, given none were found, is hardly surprising).
Perhaps more significant is that reports from embedded correspondents were
also less likely to contain references supporting the government’s case on WMD.
By contrast, reports supporting the government’s case on WMD were more likely
to come from correspondents based at the official military briefings in Qatar.
The fact that embedded reporters were less likely than most other forms of
reportage to imply the presence of WMD would seem to go against the idea that
embedded reporters were more likely to reproduce government and military
claims. Indeed, had broadcasters been more reliant upon military briefings, it
seems likely that assertions about Iraqi WMD would have been more widely
repeated. This suggests that, while embedded reporters will inevitably be present-
ing a limited view of war, their reports are more likely to be independent than
information coming from military briefings.
More significant, in our view, is the breakdown of representations of the Iraqi
294
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
Table 15.7 Stories including one or more reference to the theme of Iraqi WMD capability,
tabulated against type of news report
Embedded reporter 6 5 1 5 9
Baghdad reporter 5 4 0 0 6
Qatar reporter 10 9 1 5 4
Unilateral 1 1 0 0 1
Available footage 23 20 6 32 15
Studio analysis 10 9 4 2 4
Interview with reporter 11 10 0 0 6
Interview with expert(s) 9 8 3 16 3
Anchor 27 23 4 21 48
Other 14 12 0 0 5
people. Table 15.8 indicates that, whereas the ratio between stories portraying
the Iraqi people as welcoming liberation as opposed to antagonistic to invasion in
our total sample was nearly 2 to 1 (151 to 85) in favor of the government’s case,
reports by embeds were much more balanced. Embedded reporters were, in other
words, notably more likely than news reports in general to portray the Iraqi
people as antagonistic to or suspicious of US/UK military action. Put another
way, out of the 85 stories which included references representing the Iraqi people
as unenthusiastic about or hostile toward the invasion, 32 percent were based on
embedded reports (remembering that embedded reports constituted only 9 per-
cent of our total sample).
More predictably, reports from correspondents based in Baghdad were also as
likely to include references to the Iraqi people opposing US/UK intervention as
they were to welcoming it (by 19 to 16). This type of report accounted for nearly
a quarter of all the reports including references suggesting that the Iraqi people
opposed US/UK military intervention.
By contrast, in reports by anchors the ratio between stories including refer-
ences to the Iraqi people welcoming compared to opposing US/UK military
intervention was almost 7 to 1 (34 stories as opposed to 5). It would seem that the
account of the war given by television news anchors, highly trusted by the public
according to surveys, was overwhelmingly more likely to give the impression that
the Iraqi people welcomed US/UK military intervention.
While we would not want to draw too much from this, the picture presented
here is an interesting one. We have found no evidence that embedded reporters
were more likely to cover issues critical to the government/military case
sympathetically. On the contrary, they tended to be more balanced than other
kinds of reporters. The distinction implied by Table 15.8 is not between
embedded reporters and those based in Baghdad—who both gave a similarly
balanced account, but between reporters on the ground and scripts written for
295
JUSTIN LEWIS AND ROD BROOKES
Table 15.8 Number of stories with references to the state of the Iraqi people, against type
of news report
Embedded reporter 29 27 29 17
Baghdad reporter 16 19 21 25
Qatar reporter 3 0 4 2
Unilateral 0 2 2 3
Available footage 54 25 31 47
Studio analysis 2 2 3 1
Interview with reporter 8 1 8 2
Interview with expert(s) 2 3 1 2
Anchor 34 5 56 32
Other 1 1 4 3
Total 151 85 160 131
anchors in London. In short, while reporters in the region told a fairly nuanced
story about the reaction of the Iraqi people, the editorial line in London echoed
the government’s position. If there was bias here, in other words, it was in spite of
rather than because of the embedded reporters.
296
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
WMD 53 4 0 0 7 64
Iraqi people 20 51 6 7 13 97
Regime depravity 10 0 4 5 2 21
Total 83 55 10 12 22 182
297
JUSTIN LEWIS AND ROD BROOKES
an average for all four broadcasters of 22 percent. Channel 4 ran more stories
including reports of Iraqi casualties: 55 percent of stories on the Iraqi people fea-
tured reports of casualties as opposed to an average of 42 percent. Embedded
correspondents for Channel 4 were more likely to report Iraqi opposition to
US/UK intervention and Iraqi citizen casualties than the other channels. Finally,
Channel 4 was least likely to use official UK/US government or military sources
than other broadcasters: 42 percent of all its sources as compared to the average
for all four broadcasters of 46 percent.
It should still be noted that, while our analysis seems to indicate that Channel
4 news was relatively more critical of the government’s case for the war than the
other news broadcasters, overall all broadcasters were significantly more sympa-
thetic to the government’s case than they were critical.
Our analysis suggests that the most pro-war broadcaster, by these measures, was
Sky, which was most likely to portray Iraqis as welcoming the invasion and was
most likely to suggest the presence rather than the absence of WMD. Again, we
should not exaggerate this, as Sky and Channel 4 represent opposite ends of a
fairly small spectrum.
Conclusion
While British broadcasters clearly did not submit to the kind of cheerleading that
characterized much of the US network coverage, our research suggests that the
wartime coverage was generally fairly sympathetic to the government’s case. This
manifested itself in various ways, notably: the focus on the progress of war to the
exclusion of other issues, the tendency to portray the Iraqi people as liberated
rather than invaded, the failure to question the claim that Iraq possessed WMD,
and the focus on the brutality or decadence of the regime without putting this
evidence in a broader historical and geopolitical context.
This is not to say that all the coverage was sympathetic—but we can see how
the overall weight of the coverage might have encouraged some hitherto uncon-
vinced people to support the war. Opinion polls during the war indicated support
increased from 54 percent at the beginning to 63 percent in the days following
the destruction of Saddam’s statue in Baghdad.16
As for concerns that embedded reporting would result in more sympathetic
coverage toward the case for the war, we found, on one level, no evidence to sup-
port this. Rather, reports filed by embedded correspondents were a major source of
information suggesting Iraqi citizens had mixed feelings about US/UK military
intervention. Whereas reports by Baghdad-based unilateral correspondents
accounted for a significant proportion of the coverage, the tiny amount of reports
filed by roving unilateral correspondents suggests that the Pentagon’s attempt to
replace independent reporting with embedded reporting was largely successful.
What the embedded reporters did do, however, was help to make the main nar-
rative a simple story of the progress of a war. It is in this context that the debate
about embedded reporting—and media coverage of war generally—needs to
298
B R I T I S H T V N E W S A N D T H E C A S E F O R T H E WA R I N I R A Q
move away from simple notions of censorship toward a more complex under-
standing of the media–military relationship. The Pentagon’s embed strategy was
ingenious because it increased rather than limited access to information. By giving
broadcasters access to highly newsworthy action footage from the front line, they
were encouraging a focus on the actions of US and British troops, who would be
seen fighting a short and successful war. The story was thus all about winning and
losing, rather than a consideration of the context in which the war was fought.
Notes
1 We are indebted to our research assistant Kirsten Brander for her work on this part of
the project.
2 In a widely reported interview with the Australian Broadcasting Corporation (G.
Jones, “Blair calls for claims to rebut ‘negative’ Arab media,” Daily Telegraph, April 1,
2003). Downing Street Director of Communications Alastair Campbell expressed
dismay at the reporting of allegations of atrocities committed by UK/US forces by
Arabic television and newspapers, and announced that as a result, Downing Street was
setting up an “Islamic media unit” to rebut claims allegations raised in the Arab
media. In the context of this attack on the Arab media, Campbell also attacked the
Western media for allowing dictatorships to exploit what they see as the weaknesses of
democratic media systems, by which he implied that media organizations were allow-
ing dictatorships to exploit their freedom to report claims critical of the conduct of the
war. The BBC had already been attacked by a number of columnists working for news-
papers widely recognized as editorially antagonistic to the BBC: the Sun’s Richard
Littlejohn had accused the BBC of treating “statements by coalition spokesmen . . .
with scepticism bordering on cynicism, while any old drivel put out by Baghdad is
taken as gospel” (R. Littlejohn, opinion, Sun, March 28, 2003).
3 ICM conducted a regular poll for The Guardian asking the question “would you approve
or disapprove of a military attack on Iraq to remove Saddam Hussein?” between August
23–5, 2002, and April 11–13, 2003, at variable intervals (www.icmresearch.co.uk).
4 Our own survey of opinion before, during, and after the war suggested that much of
the shift to a pro-war stance was based on a desire to support the troops—see Lewis et
al. 2003.
5 The reporting of civilian casualties became a major issue in debates around US tele-
vision news coverage of the war in Afghanistan. US television news channels tended
to minimize the reporting of such figures—partly because of lack of access—and some
channels (CNN) issued guidelines indicating that reporting of civilian casualties
should be accompanied by a reminder of how many US citizens died on 9/11.
6 A search of the Lexis-Nexis Executive database for the term “embedded reporters,”
conducted on August 29, 2003, revealed only five articles before January 1, 2003, as
compared to 914 articles for the first quarter of 2003; 1,457 articles for the second.
7 Percentages have been rounded up, and thus add up to 101 percent.
8 Many journalists found it hard not to get carried along with this shift, hence the
moment of victory was interpreted by many as a vindication of the war (such as
Andrew Marr on the BBC), even though few advancing the anti-war case did so on
the basis that the US-led forces would lose the war.
9 This percentage is lower than the figure derived from adding the three categories in
Table 15.2, as some reports contained references to more than one theme.
10 The percentages here refer to the proportion of reports in which these themes were
covered as a percentage of all reports broadcast during the war.
11 These involved unfounded claims that Iraq had fired Scud missiles at Kuwait; of a pop-
299
JUSTIN LEWIS AND ROD BROOKES
ular uprising against Saddam Hussein in Basra, a large tank column leaving Basra, and
premature claims of the capture of Umm Qasr.
12 These add to more than 100 percent, as some reports contained both claims and
counter claims about WMD.
13 “What Baghdad really thinks,” poll conducted for the Spectator/Channel 4 News, July
8–10, 2003, www.YouGov.com
14 These two things are, of course, quite separate, although they tended to get lumped
together in much of the coverage.
15 The tone of condemnation adopted by many journalists at this lavish lifestyle was, in
some ways, curious, given the wealth of Britain’s royal family in a country that has, by
European standards, a conspicuously large disparity between rich and poor.
16 ICM Guardian polls March 21–3 and April 11–13, 2003; YouGov polls taken at the
same time are very similar: 53 percent (March 20) rising to 66 percent (April 10)
agreed with the statement that the US and UK were right to take military action
(Kellner 2003, p. 13).
References
Allan, S. (2002) “Reweaving the Internet: online news of September 11” in B. Zelizer and
S. Allan (eds), Journalism after September 11, London: Routledge.
Bennett, W.L. and D.L. Paletz (1994) Taken by Storm: the Media, Public Opinion and Us
Foreign Policy in the Gulf War, Chicago, IL: University of Chicago.
Franks, T. (2003) “Not war reporting—just reporting,” British Journalism Review, vol. 14,
no. 2, pp. 15–19.
Glasgow University Media Group (1985) War and Peace News, Milton Keynes: Open Uni-
versity Press.
Hallin, D. (1986) The Uncensored War—the Media and Vietnam, Berkeley, CA: University
of California.
Kellner, D. (1992) The Persian Gulf TV War, Boulder, CO: Westview.
Kellner, P. (2003) Iraq—the Public and the War: a Report on YouGov Opinion Surveys before,
during and after the Conflict, www.yougov.co.uk
Lewis, J., Brookes, R., Mosdell, N., and Threadgold, T. (2003) “Embeds or in-beds?”
BBC/Cardiff School of Journalism.
Miladi, N. (2003) “Mapping the Al-Jazeera phenomenon” in D.K. Thussu and D. Freed-
man (eds), War and the Media, London: Sage.
Thussu, D.K. (2003) “Live TV and bloodless deaths: war, infotainment and 24/7 news” in
D.K. Thussu and D. Freedman (eds), War and the Media, London: Sage.
Zelizer, B. and S. Allan (eds) (2002) Journalism after September 11, London: Routledge.
300
16
EUROPEAN NEWS AGENCIES
AND THEIR SOURCES IN THE
IRAQ WAR COVERAGE
Terhi Rantanen
301
T E R H I R A N TA N E N
302
EUROPEAN NEWS AGENCIES
role as sources of news for the media. This, in turn, resulted briefly in an increas-
ing scholarly interest in news agencies and their critical evaluation (the
UNESCO New World Information and Communication Order debate in the
1970s and early 1980s)—this has now almost completely disappeared.
Much of the earlier research took the form of quantitative content analysis,
untheorized or under-theorized (Boyd-Barrett and Rantanen 1998: 2–3). Most
studies were based on a mechanistic model of a communication chain leading
back to news agencies as primary sources of news. Metaphors of “flows” or
“chains” were frequently employed to describe a one-way dissemination of news
from powerful international agencies to national agencies, and from national agen-
cies to national and local media. The “flow” metaphor was not foreign to early
models of communication, which depicted linear processes of message transmis-
sion from senders to receivers without much reciprocity. As Galtung and Ruge
(1965: 64) wrote in their classic study “The structure of foreign news,” in which
they analyzed the presentation of the crises in Congo, Cuba, and Cyprus in four
Norwegian newspapers,
we shall treat the news media as non-personal invisible entities and not
distinguish between the journalist in the field in the news-sending coun-
try, the local press agency bureau, the district bureau, the central bureau
of the press agency, the district bureau on the receiving end, the local
bureau in the news receiving country, the news editor in the receiving
paper, the layout man [sic] and what not—to indicate a chain with seven
or eight steps in it [emphasis added].
Much has changed both theoretically and empirically in the almost 30 years since
Galtung and Ruge published their study. The division between wholesalers and
retailers of news is no longer entirely valid; increasingly traditional wholesalers
(news agencies) find themselves in competition with retailers such as CNN and
the BBC, or with their own media clients who either directly subscribe to global
news sources or even compete themselves with national agencies. News agencies
have also themselves become retailers by delivering news on the Internet. As
Gurevitch et al. (1991: 197) argue, the institutional arrangements for transmitting
and exchanging news materials spawned by the availability of new technology
have transformed the global structure of news dissemination around the world
toward a greater decentralization of news. According to Gurevitch et al., we have
also entered a relatively new stage, the globalization of news (see also Boyd-Barrett
and Rantanen 1998).
The globalization of news challenges earlier seven- or eight-step chain models.
We now live in a largely convergent media environment in which simple chain
models leading to a single source are no longer valid. It is now time to start re-
theorizing the global news system as a complex network of different media where
every medium both receives and transmits news. We need to bear in mind, how-
ever, that there are still some clusters which are significantly larger and more
303
T E R H I R A N TA N E N
304
EUROPEAN NEWS AGENCIES
even operate globally. The largest of these is AFP (France) with a turnover of
€195 million. It is followed, by dpa (Germany) at €99.8 million, ANSA (Italy) at
€91.6 million, the PA (UK) at €87 million and EFE (Spain) at €70.6 million. At
the other extreme is one of the smallest European agencies, ANA (Greece) at
€6.6 million. The average turnover for these agencies is €16.5 million.
Two very strong national agencies, the German dpa and the Spanish EFE,
increasingly operate as global agencies. So far, although many national agencies
face severe financial difficulties, news is primarily still exchanged between global
and national players. In most European countries there is one national news
agency, which serves most of the media in that country, although it often finds
itself in competition with its own clients or with the global media. The situation
is much more complicated than it used to be, but national news agencies are still
important sources, especially of foreign news for small and medium-sized national
and local media which cannot afford to maintain their own foreign correspon-
dents or other direct sources. With the collapse of Communism and thus of the
position of the Soviet TASS, there is no longer a clear difference between, on the
one hand, western and, on the other hand, central and eastern European media
in terms of the sources they use. All central and eastern European media now
increasingly use Western sources.
Hence news agencies maintain their importance, even if they are no longer the
only sources of foreign news, especially in times of crisis such as international
conflicts when the costs of sending and maintaining one’s own correspondents
can become very high. We do not have accurate data on the number of news sto-
ries coming from different sources during the Iraq war. The study by Lewis et al.
(2003) of British television news states that 48 percent of reports were delivered
by anchors in the studio, while nearly one in five (19 percent) were edited pack-
ages or studio-based analysis. A high percentage of this news probably came from
agency sources. The number of news stories in other media, especially local
media, coming from agency sources is even higher. As a rule of thumb, the
smaller the medium, the higher the percentage of foreign news which comes from
agencies.
It seems more than reasonable to study how European news agencies covered
the beginning of the war in Iraq in the first part of 2003. This is an especially
interesting object of study since, although the conflict is outside Europe, the two
countries which invaded Iraq, the USA and the UK, are also the home countries
of the world’s news duopoly. Their governments and military forces also waged a
news war, which was probably more carefully planned than any previous news
war. The second European player in the world news duopoly is AFP in France,
where the government took a very different stance to the war than the UK and
US governments, and opposed their military actions. In general public opinion in
the EU was very divided, with huge anti-war demonstrations taking place around
Europe. So, what sources did the European agencies turn to in this situation?
The results presented here are based on a questionnaire, which I sent to the
members of the European Alliance of News Agencies (EANA) in July 2003. The
305
T E R H I R A N TA N E N
French AFP, the German dpa, and the Spanish EFE are all members of EANA,
along with small national agencies which operate mainly inside their own coun-
tries. The British Reuters, however, is not a member. Nineteen out of 28 member
agencies returned questionnaires. They were AA (Turkey), AFP (France), ANA
(Greece), ANP (Netherlands), ANSA (Italy), APA (Austria), C̆TK (Czech
Republic), dpa (Germany), EFE (Spain), HINA (Croatia), Itar-TASS (Russia),
LUSA (Portugal), MTI (Hungary), NTB (Norway), Ritzau (Denmark),
Rompress (Romania), STT-FNB (Finland), SDA-ATS (Switzerland), and STA
(Slovenia). The respondents were mostly chief editors or foreign news editors.
Golding and Elliot (1979: 92) divide the production cycle of newsmaking into
four stages: planning; gathering; selection; and production. The questionnaire
focused mostly on the first two stages, since I was mainly interested in their use of
sources. This is different from earlier research, which saw news agencies solely as
the source of news, without realizing that they, like any other media, are depen-
dent both on each other and on other media, i.e. they use each other as sources.
Furthermore, I did not study the content of news, i.e. what sources they actually
used, but instead tried to find out what they thought about the different sources
they subscribed to. My main research questions were:
306
EUROPEAN NEWS AGENCIES
pare their coverage. Approximately half of the agencies started their planning as
early as the autumn of 2002, and the other half in January or February 2003.
Many agencies sent correspondents to Iraq and neighboring countries early in
2003. There were only a couple of agencies which made no advance preparations.
These were smaller agencies, often with a small corps of correspondents.
307
T E R H I R A N TA N E N
zones. The Finnish STT sent five of its journalists to a one-day training course at
the Finnish Defense Forces International Center, where they learned what to do
if wounded or taken hostage.
308
EUROPEAN NEWS AGENCIES
One ground-breaking aspect of the Iraq war was the emergence of Al-
Jazeera and other Arabic-language television stations as major providers
of battlefield news. The journalistic community should encourage the
development of a pluralistic, indigenous media, coupling it with training
programs to foster a more rigorous and objective approach to the news.
309
T E R H I R A N TA N E N
make a concerted effort to be balanced and report from both sides of the conflict,
including political, economic, and social consequences as well as human interest
angles.
Both US and Iraqi officials were mentioned most often as the least reliable
sources. As one editor put it,
The briefings in Qatar were virtually useless, filled with slogans more
than information, as well as disinformation and a lack of explanation.
Also absent were the background briefings offered in Riyadh on various
aspects of the war, as well as around-the-clock access to the US or
British spokesmen’s office. The Washington briefings on Iraq, featuring
Secretary of Defense Donald Rumsfeld and Chairman of the Joint
Chiefs of Staff General Richard Myers, provided more quotable copy but
were still full of vagaries and propaganda.
310
EUROPEAN NEWS AGENCIES
This result is very different from those of Lewis et al. (2003) in their study of how
the British media covered the war. They found no evidence to support the idea
that embeds were necessarily “in bed” with the military or the US/British govern-
ment in the tone of their reporting. They wrote:
Indeed, our evidence suggests that the embeds provided a much more
balanced account of events than some non-embedded reporters—espe-
cially studio based anchors, whose scripts could be seen, on some issues,
to be inadvertently tilted toward certain pro-war assumptions.
The agencies also faced another problem, which they were fully prepared for. The
presence of so many journalists on the ground, whether embedded or not, also
made the Iraq war one of the most dangerous for the media, with more than a
dozen reporters killed over a relatively short period. This raises the question of
journalists’ security in war zones and their need for training and protection.
Conclusion
There is no question that news agencies in Europe share a professional journalis-
tic culture which goes beyond national boundaries. They subscribe mainly to the
same sources and, even more importantly, also trust the professionalism and
integrity of these sources. This shared professional culture does not disappear
when war breaks out, even if the government of the country in Europe where one
311
T E R H I R A N TA N E N
agency is located is involved in the war. In the Iraq war, sources outside Europe
were seen as less reliable and this applied to both the US and Arab media.
It is somewhat surprising that US sources were seen as so biased, since US and
European media have traditionally shared to a great extent the same journalistic
ideology of objective, reliable, and unpartisan news. The fact that the US Associ-
ated Press had lost its position in western Europe before the war and been
replaced there by CNN indicates that a structural change is taking place.
Whether this means that CNN has managed to become a global source able to
serve the diversified interests of its clients around the world better than other US
media remains, now that the war has ended, to be seen. But CNN has already
succeeded in becoming part of the professional journalistic culture in Europe.
It is less surprising that Arabic media are also considered less reliable. They are
after all new players with less experience of operating in the global market. They
do not share the same Western journalistic culture as other agencies. However,
the success of Al-Jazeera cannot be underestimated. It has become a global actor
noted and quoted around the world.
The high trust in agencies’ own correspondents gives support to what Gure-
vitch et al. (1991: 206) noted already 16 years ago, i.e. how the global is being
domesticated. Own correspondents were seen as the most reliable, almost covering
the war “with our eyes.” The trust in them illustrates a situation where there was
so much news available that it was difficult to know which source to trust. But
very few agencies actually had their own correspondents. What is the next
“domesticated” source if you do not have it available in your own country? In the
coverage of the war in Iraq, the next trustworthy source was a European source. It
did not even matter if the source was located in the UK, a country that was allied
with the USA against Iraq, as long as it was a European source.
The trust in European sources can be explained in at least in two different ways.
First, one could say that because of the political unification of Europe in the frame-
work of the European Union a new European identity is emerging and that
identity is reflected particularly in conflict situations. Second, in more narrow and
more professional terms, the fact that the agencies kept on trusting Reuters, some-
thing they have done in western Europe for a long time, reveals that they did not
see Reuters losing its professional integrity even if the UK was involved in the war.
The evidence is still scarce but indicates that a structural change may take
place. Tunstall wrote about the global duopoly that was based on the power of US
and British media. However, what we witnessed in the war coverage was a clear
drift between western European and US sources where the former were favored
over the latter. Of course, this situation is different from normal day-to-day cov-
erage when conflicting interests do not become visible and remain hidden. In this
sense, war coverage brings the issues of impartiality again to the surface and under
public scrutiny. Most European news agencies were very aware of their responsi-
bility before the public.
Whether the drift between US and western European sources is permanent
remains to be seen. But it challenges our previous theorization of the nature of
312
EUROPEAN NEWS AGENCIES
Western media: we can no longer take for granted that there is a holy alliance
between the US and western European media, and that they are the same. We
should concentrate more on the differences between the two rather than the sim-
ilarities. This is even more important when not only the alliance of the US and
Western European agencies has become fragile, but also a new global player, Al-
Jazeera, has emerged. It still has a different status compared to Western sources
and is being viewed critically. Its status is partly based on the conflict between the
Western and Arab worlds, to use very crude categories, but its rapid rise shows the
niche it has filled. Again, we do not know how permanent the Al-Jazeera phe-
nomenon will be after the day the Iraq war is over. Whether Al-Jazeera becomes
today’s equivalent of the Cold War Soviet TASS—a global source with a recog-
nized status, but clearly in a different category compared to its Western
counterparts—is also something that can only be seen in the daily routines of
global news transmission in years to come.
The war against Iraq, like wars in general, not only reveals the existing media
structures, but also questions them. As one editor put it, “too many interests col-
lided under the surface of the war, so the spectrum of coverage was far more
comprehensive than in the case of an average war, and in a way indicated the
birth of a new world order.” The old world order of news is still in place, but if one
looks carefully at it, one can see fractures that may cause more permanent
changes in the future.
The author would like to thank the members of the European Alliance of News
Agencies (EANA) for their help.
Notes
1 news.bbc.co.uk/1/hi/world/middle_east/737483.stm, October 3, 2003
2 www.azcentral.com/specials/special50/articles/1016apme-conversation17-ON.html,
16/11/2003
3 www.ojr.org/ojr/workplace/1061417078.php November 16, 2003
4 Libération, September 18, 2003
References
Boyd-Barrett, O. (1998) “‘Global’ news agencies,” in O. Boyd-Barrett and T. Rantanen
(eds), The Globalization of News, London: Sage, pp. 19–34.
Boyd-Barrett, O. and Rantanen, T. (1998) “The globalization of news,” in O. Boyd-Barrett
and T. Rantanen (eds), The Globalization of News, London: Sage, pp. 1–14.
Boyd-Barrett, O. and Rantanen, T. (2000) “European national news agencies—the end of
an era or a new beginning?” in Journalism Theory, Practice and Criticism, vol. 1, no. 1,
pp. 86–105.
Cassey, J. (2003) “The news coming out of Reuters makes for pretty grim reading: agents of
its own misfortune,” The Guardian, February 10.
el-Naway, M. and Iskander, A. (2002) Al-Jazeera: how the Free Arab News Network Scooped
the World and Changed the Middle East, Cambridge, MA: Westview Press.
313
T E R H I R A N TA N E N
Galtung, J. and Ruge, M.H. (1965) “The structure of foreign news”, in Journal of Peace
Research, no. 1, pp. 64–90.
Golding, P. and Elliot, P. (1979) Making the News, London: Longman.
Gurevitch, M., Levy, M.R. and Roeh, I. (1991) “The global newsroom: convergence and
diversities in the globalization of television news”, in P. Dahlgren and C. Sparks (eds),
Communication and Citizenship, London: Routledge, pp. 195–216.
International Press Institute (1953) The Flow of News, Zurich: IPI.
Kayser, J. (1953) One Week’s News: Comparative Study of 17 Major Dailies for a Seven Day
Period, Paris: UNESCO.
Lewis, J., Brookes, R., Mosdell, N., and Threadgold, T. (2003) “The role of embedded
reporting during the 2003 Iraq war: summary report,” a report commissioned by the
BBC: The Cardiff School of Journalism.
Miladi, N. (2003) “Mapping the Al-Jazeera phenomenon,” in D.K. Thussu and D. Freed-
man (eds), War and the Media: Reporting Conflict 27/7, London: Sage, pp. 149–60.
Rantanen, T. and Boyd-Barrett, O. (2004) “Global and national news agencies: the un-
stable nexus”, in A.S. de Beer and J. Merrill (eds), Global Journalism: Topical Issues and
Media Systems, Boston, MA: Allyn and Bacon/Longman.
Schramm, W. (1960) One Day in the World’s Press: Fourteen Great Newspapers on a Day of
Crisis, Stanford, CA: Stanford University Press.
Sigal, L.V. (1986/1999) “Reporters and officials: the organization and politics of news-
making”, in H. Tumber (ed.) News: a Reader, Oxford: Oxford University Press,
pp. 212–23.
Sreberny-Mohamamadi, A. with K. Nordenstreng, R. Stevenson, and F. Ugboajah (1985)
Foreign News in the Media: International reporting in 29 countries, Reports and Papers on
Mass Communication, no. 93, Paris: UNESCO.
Teh, K. (2002) “Towards a regional perspective in media and globalisation theories: the
rise of regional news,” a comparative study on Al-Jazeera and Channel NewsAsia,
unpublished honours thesis, University of Cambridge: Fitzwilliam College.
Tunstall, J. (1999) The Anglo-American Media Connection, Oxford: Oxford University
Press.
314
17
AL-JAZEERA AND WAR
COVERAGE IN IRAQ
The media’s quest for contextual objectivity
Across the world, the word Al-Jazeera has become synonymous with war. Since
the station’s emergence on the global media scene in 1996, the Qatar-based satel-
lite station has been the go-to channel for conflict coverage from the Middle
East. As Al-Jazeera reaffirms its position as a top transnational news organization,
its coverage of wars in Afghanistan and Iraq has catapulted it into a prominent
position and stirred much international controversy, placing its approach to jour-
nalism in the spotlight. Despite having had reporters arrested, licenses rescinded,
bureaus closed, and offices bombed, Al-Jazeera’s audience figures continue to soar.
Why does Al-Jazeera’s brand of journalism draw so much attention, debate, and
controversy? No evaluation of Al-Jazeera and its operations is coherent or com-
plete without a close look at the media landscape from which the network
emerged.
This chapter is a discussion of Al-Jazeera’s journalistic model and the network’s
coverage of the US-led war in Iraq as a case study for thinking about the journal-
ism of conflict. Since most historical landmarks in the development of Arab
media relate to conflict coverage, we briefly describe three major stages in the
development of journalism in the region and show how they articulate the dilem-
mas facing journalists, suggesting that the impact of foreign programming and the
Arab world’s quest for press freedoms have culminated with Al-Jazeera and other
satellite news channels. Two contesting forces in war coverage of the region—
editorial decision-making and network responsibility to audiences—help explain
how Al-Jazeera works, and we describe this tension as contextual objectivity and
illustrate how it is employed generally by Al-Jazeera and specifically in the case of
its coverage of the war in Iraq.
315
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
316
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
Overall, the history of foreign radio programming to the Arab world seems to
have gone full circle. Early messages from Italian, British, German, and American
stations sought to win Arab popular sentiments. Following World War II, the
credibility of a foreign radio broadcast was built on its ability to deliver fair and
balanced news to a region that depended on external feeds for uncensored jour-
nalism. Once again, the events of September 11, the war in Iraq, the prospects of
a protracted conflict between the Palestinians and Israelis, and the threat of an
expanded war in the Arab world have redirected foreign media messages back to
the mandated agendized “diplomatic” discourses of yesteryear. It is this sidelining
of news balance which threatens the perceived credibility of foreign broadcasters,
both radio and television, in the Arab world. This first stage of broadcasting
is therefore characterized by its marked emphasis on “diplomacy” rather than
objectivity.
A second stage signaled the earliest major indigenous experiment in Arab
broadcasting, which occurred in Egypt. This venture, which would eventually
establish the country as a leader of the Arabist movement and home of the most
developed media, film, and theatre industry, was facilitated by Egypt’s efforts in
being first among the Arab nations to build a high-powered medium-wave and
short-wave transmitter for radio broadcasting (Boyd 1999). Built both to reach
the indigenous population within the country’s borders and to project and dis-
pense the message of pan-Arab nationalism throughout the region, the venture,
called “Voice of the Arabs,” allowed Arab anti-colonial revolutionaries to use the
Egyptian radio service to broadcast dissenting messages to their home states.
Situated as the prescribed vehicle for pan-Arabism in the 1950s and 1960s,
“Voice of the Arabs” became immensely popular throughout the Arab world.
Nasser’s speeches and the late Egyptian singer Um Kalthoum captured and
mesmerized audiences across the region through their broadcasts, consolidating
Egypt as a leader in the Arab world, both as a political and media power (el-
Nawawy and Iskandar 2003). However, “Voice of the Arabs” soon fell from grace
in the Arab world, when, during the first few days of the 1967 war, its news
bulletins painted a misleading picture of the war, suggesting that Arab armies
were successful and neglecting to report casualties and damage in their ranks.
The fact that “Voice of the Arabs” had kept a general optimism throughout
the war was seen by many Arabs as a betrayal. Though the station was trans-
formed after the war, it was too late to salvage its reputation with its listeners
(Boyd 1999). Nonetheless, some characteristics of the nationalistic “Voice of the
Arabs” model are still emulated throughout the region. At its side, however,
an elaborate printed press in some Arab countries developed a tradition conso-
nant with principles of free discourse. Lebanon and Egypt had for long allowed
newspapers with leanings across the political spectrum to publish, and various
political publications allowed public discourse to flourish among the literate
(Boyd 1999; el-Nawawy and Iskandar 2003). This trend, however, was not con-
sistent throughout the Arab world, with countries applying varying degrees of
pressure on opposition press. Furthermore, while some nations in the Arab world
317
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
had loosened the belt on the print press, until the advent of satellite television,
none had allowed this phenomenon to translate to the broadcast media which
were virtually all extensions of government.
Until the late 1990s, most Arab countries had a Ministry of Information—
formally a regulatory body—which was primarily responsible for monitoring and
censoring all mass communication in the nation. That is primarily why foreign
broadcasters had initially built such a large audience in the region. The rise of
Arab satellite television stations would soon displace Arabic language foreign
programming.
A third stage of development began with a chain reaction which followed
access to CNN in the Arab world during the Gulf War of 1991 and led to the
transformation of the Arab media landscape in the years that followed. It her-
alded the birth of investigative reporting and war correspondence in the Arab
world. The Arab satellite networks’ coverage of the war was nothing short of a
revolution in the region and amounted to a phoenix rising out of the ashes of pre-
vious experiments such as “Voice of the Arabs.” The first Arabic station to
emulate the CNN formula was the London-based Saudi-funded Middle East
Broadcasting Center (MBC). A mixed content station, MBC, offered news,
entertainment, sports, and specialty programs. In the 1990s, a new satellite sta-
tion was launched virtually every year. Over time, there would develop two types
of satellite television stations in the region, those that were an extension of the
local government’s programming, and those considered “offshore”—with little or
no “clear” relationship to the government of a particular nation. Most of those
listed here are considered “offshore” satellite networks. Other major contenders
to MBC that tried to carve a niche for themselves in this domain included the
Saudi-owned Arab Radio and Television (ART) and Orbit. While these stations
were not directly and overtly connected to particular governments, they were not
subject to the work of censors, at least not formally. The Lebanese Broadcasting
Corporation (LBC) and Abu Dhabi television were also among the newcomers
that made their mark on regional broadcasting and continue to be influential
today. However, it was not until Al-Jazeera came on the scene in 1996 that the
Arab world had its own 24-hour all-news station with an extensive international
bureau system.
A product of the 1996 editorial fallout between the BBC and the Saudi
investors in the BBC Arabic Television News Service, Al-Jazeera was the brain-
child of Qatar’s comparatively progressive emir, Sheikh Hamad bin Khalifa
Al-Thani (el-Nawawy and Iskandar 2003). While Saudi investors had pulled the
plug on the BBC service after a row over the airing of a documentary on execu-
tions in the Saudi Kingdom in 1996, Qatar’s emir saw this as a unique opportunity
to hire a group of critically minded, independent, and well trained journalists to
better represent the interests of the region. The emir and Al-Jazeera’s founders
recruited the staff of the fallen BBC service and relocated them to Doha to com-
prise the start-up crew for the first 24-hour Arabic television news channel
(el-Nawawy and Iskandar 2003). The differences between the BBC’s Saudi fun-
318
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
ders and Qatar’s emir embody the contrasting principles of their respective net-
works.
Today, Al-Jazeera’s headquarters in Doha are filled with professionals from
most Arab countries, comprising a demographic microcosm of the 22-nation
Arab world. The diverse pan-Arab workforce of 350 journalists in its main office
and 35 international bureaus reflects every political leaning, religious affiliation,
and educational background in the Arab world. However, as far as funding, Al-
Jazeera is exclusively Qatari. The emir provided a start-up fee of $140 million to
launch the network and continues to make regular loans (almost $100 million
annually) to help sustain the station’s operations (el-Nawawy and Iskandar
2003).
While some have called into question Al-Jazeera’s financial dependence on the
Qatar government (el-Nawawy and Iskandar 2003), it seems that much of the
network’s editorial policy is independent of direct or indirect governmental con-
trol as far as international affairs are concerned. Alternately, we have been
critical of Al-Jazeera’s negligence toward domestic news in Qatar and have
brought forward questions regarding political connections between the network
and the Qatari government. This, however, seems to have a marginal impact on
coverage of international news (el-Nawawy and Iskandar 2003).
It was not long before Al-Jazeera raised eyebrows with its aggressive investiga-
tive reporting style, demonstrating dissent against regional governments, hosting
volatile political debate shows, and its journalistic professionalism. The station’s
news style and political debates have been imitated to varying degrees by a grow-
ing number of other channels in the region. The latest members of the now-highly
competitive media terrain are Al-Arabiya and the reinvented Abu Dhabi tele-
vision and Dubai television.
In addition to rattling governments in the region, Al-Jazeera’s coverage of wars
in Afghanistan and Iraq triggered a global debate about conflict in the media and
the nature of “objectivity.” The network’s immense success, with an estimated
audience of 50 million viewers worldwide, has drawn attention to its brand of
journalism.
While most Arab news organizations imitate Al-Jazeera’s approach and style,
the network’s global reputation and reach have made it the regional victor. Like
news organizations everywhere, these stations all have one thing in common—
the pursuit of journalistic integrity. This is notably due to a well articulated
journalistic philosophy and a unique way of employing it. The ways in which this
philosophy is conceptualized are exemplified by its application during coverage of
the recent war in Iraq.
319
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
whether they are CNN, Fox News Channel, or Al-Jazeera. Networks’ decisions to
include or exclude items from their news bulletins arouse lively and often volatile
discussion in the print media and the public sphere. Such discussions ultimately
call into question and reflect on journalistic standards, invoking the institutional
ideal—“objectivity.”
The pursuit of objectivity in journalistic reporting has been a cornerstone of
the ideals of news coverage. Every journalism textbook and news reporting course
enshrines and standardizes objectivity as the prime responsibility of a keen
reporter. Although it seems to have its roots as far back as the 1830s, the notion
of “objectivity” became popular following World War I among news organizations
as they developed an appreciation for a scientific validation of truths (Schudson
1978).
Mass communication scholars and media critics have also grappled with this
concept for decades, articulating its variable contexts and applying multiple philo-
sophical interpretations to it. More recently, the aftermath of September 11 and
the war in Iraq have triggered a renewed debate on media’s responsibilities, culmi-
nating with the conceptualization of objectivity as a media practice of information
collection, processing, dissemination, and as an overarching attitude (el-Nawawy
and Iskandar 2002; Zelizer and Allan 2002; Tumber and Prentoulis 2003).
The term objectivity itself, when used within the journalistic context, signifies
the adoption of a position of detachment, rather than neutrality, toward the sub-
ject of reporting. It also suggests the absence of subjectivity, personalized
involvement, and judgment. This ideal or mirage is one that is particular to the
media practitioner and the institution in which she operates. By virtue of her
occupational responsibilities as a reporter—and reflected in her training—she is
trained to avoid leanings in matters of dispute.
The notion of contextualization offers a corrective to some of the limitations
inherent in the notion of objectivity. Contrary to the standard of “objectivity” to
which journalists and news organizations aspire, media audiences are opinion-
ated, partial, and highly invested in the news content. Audiences are expected to
espouse certain opinions and to express a degree of partisanship. While journal-
ists are trained to excavate the “truths” that lie beyond the “context,” audiences
are invariably and inevitably contextualized. Contextualization demonstrates a
situational position, a way by which collectivism among participants within the
same “context”—whether cultural, religious, political, or economic—is realized
and engaged. It is precisely this contextualization that aggravates and complicates
the pursuit of “objective” coverage within the news media setting.
Contextualization further confuses attempts at even-handedness and efforts to
cover all sides of a story. Particularly in times of war, it is the context within
which a reporter operates that makes communication with the “enemy” unac-
ceptable. Context may be seen as the reason why dissenting voices in the US
during the build-up for war on Iraq were perceived and represented negatively by
many policy-makers, the American media, and subsequently by the public.
Accordingly, the ratings for networks that demonstrated a predominantly pro-
320
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
government and pro-war stance, such as the Fox News Channel (the network
that offered the most support for military action in Iraq), skyrocketed throughout
and during the war—suggesting a growing political “context” in the US. One
instance that demonstrated this was a visit by two US senators to Iraq before the
war to evaluate the humanitarian crisis caused by United Nations sanctions . The
reaction to the visit was swift and intense. Many television commentators and
politicians saw this visit as an act of “national betrayal” and defection. Evidently,
if politicians were heavily criticized for communicating with the “enemy,” then
how would the media fare if they did the same?
That is precisely what Al-Jazeera did—communicate with the “enemy.” In our
discussion of the network’s role in the world of post-September 11 journalism, we
offered the concept of “contextual objectivity,” in an attempt to articulate and
capture the eclectic discourse and epistemological tensions between the relativism
of message receivers and the empirical positivist attempts of message-builders (el-
Nawawy and Iskandar 2002, 2003). Should the media lead the masses or be led by
them? “[T]he theory of contextual objectivity—the necessity of television and
media to present stories in a fashion that is . . . impartial yet sensitive to local sensi-
bilities—is at work” (el-Nawawy and Iskandar 2003: 54).
Contextual objectivity, the perpetual tension between the decontextualized
messages of the news deliverer and the nuanced and colored perceptions of the
receiver of news messages, can be witnessed on virtually every news bulletin of
war on every media outlet in the world today, not the least CNN and Al-Jazeera.
It permeates every story, and has become increasingly emblematic of the struggle
for the construction of mediated messages.
We used the Minotaur as a metaphor to describe the contestation forged by the
new media to accomplish both seemingly-contradictory duties (el-Nawawy and
Iskandar 2002). Like the Minotaur, the Cretan mythological character that bore
the head of a bull and the body of a man, contextual objectivity reflects the ten-
sions of the instinctive and rational, the relativist and the positivist. The
Minotaur’s hybrid identity and the contradictory co-existence of human and beast
within the same entity is precisely what makes it a compelling analogy for contex-
tual objectivity. At the root of both the Minotaur’s existence and of contextual
objectivity is an internal tension and turmoil which facilitates a consensual balance
between the two natures, allowing for their cohabitation in the same creature. Both
the Minotaur and contextual objectivity represent a balance between two forces, a
balance that can tilt either way. On the scale’s pivot is the ideal intermediate
between context and objectivity—best articulated as fairness and balance. Like the
Minotaur who struggles to maintain equilibrium between his beastly and human
qualities, the media try to strike the balance between audience appeal and “objec-
tive” coverage.
The media have a dual role as both informants and a mirror for society, hence,
they are held to a rather stringent policy by their viewers. They must meet their
dual duties of being balanced (objectivity) while reflecting the views of
their public constituency (contextualization). For instance, American television
321
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
coverage, under no overt government influence, may reflect the views of main-
stream American audiences in the aftermath of the September 11 attacks while,
at the same time, creating public opinion.
While most networks engage in contextual objectivity, consciously or other-
wise, in their coverage of war, Al-Jazeera is perhaps the first network to articulate
this approach as a network philosophy. The channel’s mottos, repeated frequently
during program intermissions, are operationalizations of contextual objectivity:
“The opinion and the other opinion,” “Freedom, objectivity, accuracy,” and “With
all the colours of a rainbow.” These slogans demonstrate explicitly the contesting
dimensions of “contextual objectivity”—seemingly suggesting that “truth” is the
culmination of multiple conglomerated subjectivities. This implies that neither
objectivity nor context should be the sole priorities—thereby dismissing the
exclusivity of either. Instead, onus is placed on the gray area in-between the two—
where fairness and balance are situated. As Al-Jazeera attempts to harness the
advantages of both context and objectivity, they represent an experiment that
redefines modern journalism. Their attempts at delivering news and commentary
that juxtaposes multiple opinions and realities into a single mosaic are a testament
to the network’s pursuit of contextual objectivity. However, attempts to strike
equilibrium between the tensions of context and objectivity have left Al-Jazeera
celebrated by admirers and battered by critics (see el-Nawawy and Iskandar 2003
for examples).
Since its conception, reactions to the concept of contextual objectivity within
scholarly circles have been generally positive, while the popular press and media
appeared more ambivalent about its ability to explain differences in coverage
between networks. For example, a review article in the September 7, 2002 issue
of The Economist characterized the concept of contextual objectivity as a symp-
tom of the “struggle to defend the network [Al-Jazeera] from its detractors,”
arguing that the concept, which suggests that networks offer news with a particu-
lar worldview, was a “dubious” notion, “at best a muddle, at worst, an evasion”
(“Island in the sun,” 2002). By contrast, an article published in the Washington
Diplomat in June 2002 included an interview with Robin Wright, the chief diplo-
matic correspondent at the Los Angeles Times, who suggested that covering the
Israeli–Palestinian conflict generally emerged from a “set opinion or acceptance
of a certain moral value” on the part of reporters, a view in consonance with the
notion of contextual objectivity (Beyerle 2002).
When the concept of contextual objectivity is applied to Al-Jazeera, it is obvi-
ous that the network faces two major dilemmas: making the news comprehensive
and placing the stories within a meaningful historical account. However, the
inclusion of context and analysis almost inevitably leads to the encroachment of
opinion. For instance, while the network labors to bring forth multiple perspec-
tives on the Palestinian–Israeli conflict, the Arab public’s opposition to Israel’s
occupation of Palestinian territories is a prevailing context that permeates Al-
Jazeera’s coverage (el-Nawawy and Iskandar 2003: 209).
Much of the controversy and criticism Al-Jazeera has garnered since its incep-
322
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
323
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
decision to launch this war, it also aired documentaries showing the tough living
conditions inside Iraqi prisons and the brutality of Saddam’s regime. Moreover,
despite the fact that the overwhelming majority of Arabs were opposed to the
war, Al-Jazeera correspondents went out of their way to interview members of the
Iraqi opposition who lived overseas and who supported the war.
In some respects, Al-Jazeera correspondents had an edge over their Western
counterparts, owing to the fact that they spoke Arabic and were more familiar
with the Iraqi culture. This enabled them to interview average Iraqi citizens on
the street and to give their audience an overall sense of the general mood on the
Iraqi streets. The following is a report from Al-Kahki (Al-Jazeera’s only embed-
ded correspondent), obtained through a direct feed to Al-Jazeera studios in
Qatar from Umm Qasr, where he was stationed with the coalition troops on
March 27, 2003. Al-Kahki was reporting on the situation inside Umm Qasr,
which was surrounded by coalition troops, but was still controlled by the Iraqis.
Al-Kahki said:
The feeling I got from the people in Umm Qasr is that they were “con-
fused” and weren’t sure exactly what their legal status was. They didn’t
know what type of land they were standing on; whether it was con-
trolled by the British/American troops or still under the control of the
Iraqi regime. All they said they wanted was living peacefully with food,
water, and milk for their children.
Responding to a question from the Doha studio anchor with regard to the general
security and the Iraqi resistance in Umm Qasr, Al-Khaki said:
Regarding the issue of resistance, we cannot predict it, since from time
to time we see some pockets of resistance despite the fact that some
British troops here have been searching for these pockets everywhere.
There is a big void, however, in security and safety in this area. The resi-
dents told us about some looting in the warehouses of Umm Qasr.
Moreover, the power outage is worsening the situation. People are suf-
fering, and they are living in darkness with hardly any infrastructure.
It was that kind of live reporting on the ground from places that were still inac-
cessible to Western correspondents that offered audiences a street-level view of
the war’s impact on Iraqis that made Al-Jazeera a number-one choice for Arab
viewers during the Iraq war. Despite hackers’ attacks, its battered website contin-
ued to top the charts for the most sought-after keyword on the Internet (Schatz
2003). Moreover, Al-Jazeera’s subscriptions skyrocketed to four million addi-
tional subscribers in Europe during the war (Cozens 2003). Its effort to cover
wide-ranging angles of the conflict throughout the war— including the Arab
street, Iraqi civilians, embedded reporters with coalition forces, Iraqi press confer-
ences, and US Central Command briefings—also earned Al-Jazeera the respect
324
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
325
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
(They were allowed to return after Al-Jazeera decided to completely halt all its
reporting operations in Iraq.) Al-Jazeera played to the general feelings of Arabs
who wanted to see an end to the Iraqi civilians’ suffering and who were strongly
opposed to a war they deemed unjust.
As disgusting as these gory images were, editors at the station argued that not
showing them would have been a denial of the reality witnessed by Arab
reporters. Had Al-Jazeera decided not to show these images, it would have risked
losing its audience to other Arab networks, such as the brand new Saudi satellite
channel, Al-Arabiya, which also showed close-up images of dead and wounded
Iraqis (el-Nawawy 2003a).
Al-Jazeera also focused on US losses in the Iraq war. For example, during the
early days of the war, Al-Jazeera beamed pictures of Iraqi farmers cheering the
downing of a US plane over the Iraqi city of Basra. The network also showed live
pictures of hundreds of Iraqis gathered on the banks of the Tigris River in Bagh-
dad, setting fire to cane fields and shooting into the river to flush out US pilots
who witnesses said had parachuted from the sky.
Most Arabs who followed these losses on television were pleasantly surprised,
perhaps even exhilarated, to see the much weaker Iraqis resist what they consid-
ered an “occupying superpower.” In a way, there is a desire in the Arab world to
support the underdog, and Arab networks, including Al-Jazeera, fed that desire
during the war. That was one main reason why US officials appeared dissatisfied
with Al-Jazeera’s war coverage.
However, it was Al-Jazeera’s broadcast on March 23 of video footage (taken
from Iraqi television) of American prisoners of war and dead American soldiers
that stirred much anger and outrage against Al-Jazeera in the United States. The
footage showed dead US soldiers, easily identifiable by the faces. One of them was
a young man lying diagonally across the screen, his head in the lower-right hand
corner. His eyes were closed, and blood was gushing beneath his head and soaking
his T-shirt. The American POWs, who were shown in the same video, were visi-
bly exhausted and somewhat confused. They were interviewed by an unseen
speaker, presumably from the Iraqi government or Iraqi state television. The
POWs were asked their names and their hometowns.
In a move that was considered a punishment for Al-Jazeera for airing these
images, the New York Stock Exchange and Nasdaq ejected Al-Jazeera reporters.
(The Al-Jazeera reporters have since been readmitted to the New York Stock
Exchange.) Moreover, hackers attacked Al-Jazeera’s new English-language web-
site, replacing it with a red, white, and blue US map and the slogan “Let freedom
ring.”
While some of Al-Jazeera’s coverage may be seen as “contextually” objection-
able to an American audience, it was appropriate for non-American viewers.
Regardless of judgment, the treatment the station received in the US was unwar-
ranted for several reasons. First, some European and American networks showed
still pictures from the same footage that was broadcast on Al-Jazeera but were not
singled out. Second, it was a dramatic demonstration of how US officials found
326
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
For these reasons, the reaction by which what may be the only indepen-
dent and uncensored television network in the Arab world was punished
because its style of war coverage was seen as unsuitable to certain people
seems to have been counterproductive. The question remains whether
other nations similarly have the right to ban US media outlets because
they disagree with their coverage?
(el-Nawawy 2003b)
On one evening during the second week of the war, Al-Jazeera broadcast a phone
interview with Mohammed Saeed al-Sahhaf, the former Iraqi Information Minis-
ter. Al-Sahhaf became a media star and a cult figure in the West for insulting the
Americans in charming and flowery Arabic, feeding Arab television with mis-
information about Iraq’s military power, and declaring Baghdad “safe and secure,”
even when the coalition troops were on the capital’s doorsteps. Following the
phone interview with the Iraqi official, Al-Jazeera aired a speech by President
George W. Bush, and then US Secretary of State Colin Powell came on to give
an exclusive interview to the Al-Jazeera anchor, Adnan Al-Sharif, a former
employee of Jordanian television and now Al-Jazeera’s interim managing director.
Powell told Arab audiences that the US wanted this war to end as soon as possi-
ble and that American forces had been doing their best to minimize Iraqi civilian
casualties. Realizing Al-Jazeera’s great popularity among Arab audiences, US offi-
cials in Washington have been lining up to get on the network.
327
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
Khanfar: During the past few weeks, there has been an increase in the attacks
against the American soldiers. Are you worried that these attacks
might include more cities in Iraq?
Bremer: I certainly hope the attacks do not spread. Until now, these attacks are
concentrated in the north and the west of Baghdad, and they seem to
be individual attempts by Ba’athists from the former regime. We are
doing our best to capture all the militants so that we can live in peace.
Khanfar: But sir, a lot of people are now unemployed and you might find them
joining the resistance groups.
Bremer: Obviously, senior members of the Ba’ath Party are unhappy since they
will not have any role in the new government. This is the most impor-
tant decision I took for this country. People on the streets thank me
for that decision. As for the middle and lower members of the Ba’ath,
328
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
Such open and frank interviews with US officials were much needed at that
point, given a deterioration of the American image in the Arab world since the
start of the Iraq war. For American officials to appear on the most popular Arab
network and talk face-to-face to members of the Arab audience about issues of
concern to them was a vital step in making clear the American point of view.
One of Al-Jazeera’s more notable achievements was engaging its audience in
an open dialogue, where people from various countries exchanged ideas in the
hope of fostering mutual understanding in a constructive and respectful environ-
ment. One popular Al-Jazeera program—“Open Dialogue”—allowed for this type
of exchange. A recent episode of the live monthly program (all Al-Jazeera’s pro-
grams are aired live, which does not give room for any kind of censorship or
previewing editing) invited students and their faculties from Egypt, Lebanon, and
Iraq to engage in a dialogue via direct satellite link about what the Arabs could
do to improve the situation in Iraq after the war. The episode was hosted and
moderated by Al-Jazeera’s veteran correspondent Ghassan bin Jeddo (July 5,
2003).
Although most comments began with students sending regards and wishes for
peace, the discussion heated up and included accusations and pleas from all sides.
Some members of the Iraqi side accused the Iraqis living abroad of being passive
and not helping their country. “Where are the Iraqi professors who are living
abroad? Why don’t they come back and help their country?” said one Iraqi arts
professor in Baghdad. A professor in Beirut said, addressing his words to the Arab
leaders, “Please don’t turn Iraq into a ‘deal’ by distributing its wealth among you
while the Iraqi people are starving to death.” Then a student from Cairo said,
“There are lots of reasons that led Iraq through this dark tunnel, but the number
one reason are the Arab countries; when I say that I mean the Arab leaders, not
the people.”
The Iraqi students in the Baghdad studio disagreed among themselves on the
way they viewed the resistance movements against the US troops. One Iraqi stu-
dent said, “I would like to tell everyone in Fallouja [an Iraqi Sunni-dominated
city which had witnessed repeated Iraqi attacks against the American troops]
please, don’t drive the Iraqis into another massacre; we have been through
enough already.” An Iraqi political science professor interrupted him and said:
“The fact that Al-Fallouja is being accused of resistance is an honor for us and for
all Iraqis. Any country facing an invasion has the right to resist.”
329
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
This debate and others like it showed that Al-Jazeera aired commentary and
multiple perspectives, and in doing so it broke ground by venturing into the realm
of open discussion rarely attempted by other broadcasters in the Arab world.
In the process of presenting pluralistic and open debates, Al-Jazeera also
assigned episodes of its key programs to assess the Iraq war coverage by Arab news
networks, including, of course, Al-Jazeera itself—an indication that Al-Jazeera
welcomed critical analysis of its own coverage. One such episode was aired from a
program titled “Behind the Events,” a daily program that started airing after the
war (May 12, 2003). The episode was hosted by Faisal Al-Kasim, one of the most
popular television hosts in the Arab world, who some people call “the Arab Larry
King.” Al-Kasim’s celebrity status came from hosting the popular program “The
Opposite Direction,” which is the flagship talk show on Al-Jazeera.
Four guests were invited to the “Behind the Events” episode that evening:
Jawad Maraka, former president of Qatari television, Asa’ad Abu Khalil, a politi-
cal science professor at the University of California, Mowafak Harb, director of
the American-sponsored Radio Sawa, and Mustafa Bakry, editor-in-chief of a
daily Egyptian opposition newspaper. The episode focused on the media revolu-
tion caused by the transnational satellite networks in the Arab world and how
these networks had affected coverage of the Iraq war. The following is a transcript
of part of the show:
Al-Kasim: Now we see Arab news channels are becoming more independent
and free while the Western networks have been cheering for the Iraq
war; even some American viewers complained that they weren’t get-
ting the full picture through their media. What do you think of that
phenomenon?
Harb: The Arab media have come a long way over the past few years. They
have advanced a lot in terms of technology, ways of obtaining the
news, and even in competition with the Western networks. Having
said that, I still think that most of the Arab news channels distort
reality on purpose because they get their commands from the Arab
governments.
Abu Khalil: We cannot look at the Arab media as one entity. There is a difference
between print and broadcast and a difference between the media of
one country and that of another. This difference was apparent in the
Iraq war coverage, where we saw more diversity of opinions in the
French and the British media; however, the American media were
biased and one-sided although they pretended to be giving a bal-
anced picture. Now, with the American networks, there is hardly any
room for the opposite opinion. For example, when Ashleigh Banfield
[MSNBC correspondent] tried to criticize the war, she was labeled as
a “traitor” and received threat notes. On another note, I believe that
Al-Jazeera and Abu Dhabi television did a better job in covering this
war than the American media; however, I am against the fact that
330
A L - J A Z E E R A A N D WA R C O V E R A G E I N I R A Q
Pluralistic dialogues such as this one engaged the viewers and helped them
become critical users of their own media. This is especially important in the Arab
world, where the viewers had never been exposed to a program discussing the role
of the Arab media in covering a major conflict and assessing the strengths and
weaknesses of the Arab networks in a free manner without any inhibitions or red
lines. In processing what they receive from the media, the onus falls on the audi-
ences to use a critical eye and to be aware of the different factors influencing and
biasing the media. In the meantime, media personnel have to be self-critical and
have to hold themselves accountable for any clearly partisan or incomplete
reporting.
Al-Jazeera’s coverage of the Iraq war was not devoid of context. For example,
its reporters used the term “martyrs” to refer to the Iraqi civilians killed during the
military operations (el-Nawawy 2003a). These leanings often seem inevitable as
networks cannot be devoid of perspective, a perspective that produces and
reflects context. However, this context’s overemphasis ushers in a certain degree
of slant. Our monitoring and analysis of the Al-Jazeera coverage suggests that the
network labored to present multiple sides of the Iraq conflict, despite strong criti-
cism from and measures taken by both Iraqi and US officials, and that these
multiple sides were presented systematically within a larger contextual frame that
made sense within the region, if not to the world at large.
To this day, Al-Jazeera continues to be the target of criticism, prosecution, and
punishment from many governments around the world, in what may be an inad-
vertent reaffirmation of the network’s success at employing, implementing, and
engaging contextual objectivity.
331
A D E L I S K A N D A R A N D M O H A M M E D E L - N AWAW Y
Much like the mythical Minotaur, who was thought to be responsible for cli-
matic changes and seasonal unrest, in its short and notorious history, Al-Jazeera
has been no less than a hurricane in the media landscape. On the contrary, unlike
the Minotaur, who in the end was slain, there is much reason to believe that Al-
Jazeera and its news formula—contextual objectivity—will outlive the criticism
and perhaps the network itself. Instead, the Iraq war and its coverage by Al-
Jazeera, have helped create an urgent necessity for a meaningful analysis and
re-evaluation of war coverage to include an assessment of contextual objectivity
as a barometer for fairness and balance in reporting around the world.
References
Beyerle, S. (2002, June) “Al-Jazeera says it tries to cover both sides in the Israeli–
Palestinian conflict,” Washington Diplomat.
Boyd, D. (1999) Broadcasting in the Arab World: a Survey of the Electronic Media in the
Middle East, 3rd edition, Ames: Iowa State University Press.
Cozens, C. (2003, March 26) “4m in Europe sign up for Al-Jazeera,” The Guardian.
el-Nawawy, M. (2003a, April 8) “Whose truth is being reported?” The Christian Science
Monitor, vol. 95, no. 92, p. 9.
el-Nawawy, M. (2003b, April 9) “Al-Jazeera shouldn’t be taking US fire,” The Star Ledger,
p. 17.
el-Nawawy, M. and Iskandar, A. (2002) “The Minotaur of ‘contextual objectivity’: war
coverage and the pursuit of accuracy and appeal,” Transnational Broadcasting Studies
Journal, 9 (fall/winter).
el-Nawawy, M. and Iskandar, A. (2003) Al-Jazeera: the Story of the Network that is Rattling
Governments and Redefining Modern Journalism, Boulder, CO: Westview.
Faine, J. (2003, November 10) “The US media could learn from Al-Jazeera,” The Age.
Hasso, F.S. (2003, April 17) “Who covered the war best? Try Al-Jazeera,” Newsday.
Hourani, A. (1988) “Foreword” in P. Partner, Arab Voices: the BBC Arabic Service
1938–1988, pp. vi–vii, London: BBC External Services.
“Island in the sun; Arab media,” (2002, September 5) The Economist.
Partner, P. (1988) Arab Voices: the BBC Arabic Service 1938–1988, London: BBC External
Services.
Schatz, A. (2003, April 1) “All that Jazeera,” The Lycos 50 Daily Report, retrieved on
November 28, 2003 from http://50.lycos.com/040103.asp
Schudson, M. (1978) Discovering the News: a Social History of American Newspapers, New
York: Basic Books.
Steele, B. (2003, March 26) “Wall Street attack on Al-Jazeera unwise and unfair,” Poynter-
Online.
Suellentrop, C. (2003, April 3) “Al-Jazeera: it’s just as fair as CNN,” Slate.
Tahboub, D.T. (2003, October 4) “The war on Al-Jazeera,” The Guardian.
Tumber, H. and Prentoulis, M. (2003) “Journalists under fire: subcultures, objectivity
and emotional literacy” in D. Thussu and D. Freedman (eds), War and the Media,
pp. 215–30, London: Sage.
Zelizer, B. and Allan, S. (eds) (2002) Journalism after September 11, London and New York:
Routledge.
332
18
BIG MEDIA AND LITTLE MEDIA
The journalistic informal sector
during the invasion of Iraq
Patricia Aufderheide
The brief moment in which the US armed forces with allies destroyed the regime
of Saddam Hussein in Iraq in spring and early summer 2003 offered a snapshot
view of the US media’s performance of the toughest job in democracy: managing
public discussion of controversial issues in a way that included significant minor-
ity voices. It was not a lovely picture, but it was a provocative one.
Electronic big media were bigger than ever, both domestically and internation-
ally, and along with the big moguls—Murdoch, Berlusconi— there were big and
growing corporations such as Disney, Vivendi, and Viacom, and big media merg-
ers, notably that of AOL and Time Warner. At the same time, electronic little
media flourished, a myriad grassroots attempt to fuel public opinion with informa-
tion and attitude—often living, however briefly, in the virtual realm of the
Internet. The vast burgeoning of the World Wide Web over the previous decade
had transformed the expectations of a generation about their ability both to
express their opinions and to reach others. The 1999 anti-globalization demon-
strations in Seattle, WA, that pre-empted a meeting of the World Trade
Organization (Kidd 2002), which used the Internet to launch do-it-yourself news
services and triggered the rise of “indymedia” centers globally, made the possi-
bilities vividly evident within the US. Other global organizations including
OneWorld (oneworld.net) had already combined the strength of grassroots exper-
tise with the power of distributed networking.
The invasion of Iraq—that is, the brief period of military conquest ending offi-
cially on May 1 when President George W. Bush declared on the flight deck of the
USS Abraham Lincoln that “the United States and our allies have prevailed”—
displayed the spaces in which both big and little media flourished. In particular, it
revealed the antic vigor of a journalistic “informal sector.” Borrowed from devel-
opment economics, the term referred to the unofficial, untaxed, unpoliced, and
often flourishing underground economy.
Big media had been on a roll in the US since the Telecommunications Act of
1996 had greatly increased concentration of ownership in radio and moderately
333
P AT R I C I A A U F D E R H E I D E
Big media
Consider four examples, none of them necessarily a demonstration of tendentious
ideological behavior on the part of news outlets. Rather, they all demonstrated the
consequences of a media entity’s size on coverage and the voicing of controversial
views. The first example concerned using the commercial airwaves as a platform to
rally pro-war demonstrators. As the invasion with Iraq drew close, pro-war demon-
strations suddenly started springing up all over the country. Dozens of “Rally for
America!” events were held. They were organized by a particular conservative
radio talk-show host, Glenn Beck, who positioned himself as someone standing up
against liberal media—by which he apparently mostly meant television and which
he decried for running footage of anti-war rallies (although most US protesting
334
BIG MEDIA AND LITTLE MEDIA
groups argued that coverage was meager and slighted the numbers). Supporting
the “75 to 78 percent in favor of military action,”—he told a group estimated at
20,000 in a Clearwater, Florida, rally to “Begin your day and end your day on your
knees . . . Pray for our troops and pray for our president” (Gregoire 2003).
Beck’s show, along with those of other conservatives such as Rush Limbaugh,
was carried by the gigantic national radio company Clear Channel. Clear Channel
used to be a small radio company until the 1996 Telecommunications Act rolled
back ownership restrictions on radio. Now it is by far the largest company owning
local radio stations in the US, and it lowers costs by sharing programming among
its more than 1,200 stations. Glenn Beck was a national voice with an ability to
reach into local events planning. Though Clear Channel denied any role in
Glenn Beck’s call to action, his efforts clearly didn’t hurt. Clear Channel, which
was in fiscal straits, stood to benefit from a federal government decision to further
deregulate radio—eventually made in its favor, although later challenged in Con-
gress (Krugman 2003; Schwartz and Fabrikant 2003).
A second case also concerned Clear Channel which, besides being the biggest
radio station owner in the US, was also the largest concert promoter. This inci-
dent concerned cross-ownership and control, and it was relayed via actor and
movie producer Tim Robbins, who spoke about the incident in Washington, DC,
in April 2003, at a press conference he called at the National Press Club. Com-
plaining about the way he and his wife Susan Sarandon had been treated by the
National Baseball Hall of Fame, when their invitation to celebrate the 15th
anniversary of Bull Durham was rescinded because of their dissident views on the
invasion, Robbins discussed his isolation within the entertainment community,
but also noted that people were secretly cheering him on.
“A famous middle-aged rock-and-roller called me last week to thank me for
speaking out against the war,” he said, “only to go on to tell me that he could not
speak himself because he fears repercussions from Clear Channel. ‘They promote
our concert appearances,’ he said. ‘They own most of the stations that play our
music. I can’t come out against this war’” (Robbins 2003).
Clear Channel corporate spokespeople denied intimidating and suppressing
speech in this instance. Of course, as an individual under US law, the corporation
does have full-fledged First Amendment rights, and because it is so big, its First
Amendment rights have a very big footprint. But the rocker did not say that
Clear Channel had denied him speech, like the National Baseball Hall of Fame
had done to Tim Robbins. He just said he was afraid. He could not take a com-
mercial risk, not with a company that had the unique power to pull the plug on
his concert-driven career.
A third example also demonstrated the power of size to diminish points of view.
Cumulus Media, a company that owns 262 stations, refused to carry the music of
the country music band Dixie Chicks because the group criticized Bush and the
war while touring Europe. The British left-wing newspaper, the Guardian, was the
first to publish the on-stage comment of Dixie Chicks singer Natalie Maines in
London: “We’re ashamed that the president of the United States is from Texas”
335
P AT R I C I A A U F D E R H E I D E
(Campbell). This remark reverberated across the ocean. Once Cumulus dropped
the Dixie Chicks, almost every country music radio station group, including all of
those supported by Clear Channel, dropped them during the invasion. Their
record sales fell 75 percent the following week, and they suffered a near-blackout
of their new album during the invasion. (The sales rebounded after the invasion,
though.) The choice to drop the Dixie Chicks was probably a canny market deci-
sion since country music listeners skew to the conservative. In fact, Cumulus
Media spokespeople argued that the company was just being sensitive to its listen-
ers’ viewpoints. Because of consolidation, though, a few decision-makers had a
blanketing effect, at least during the invasion (Bishop and Florida 2003; Segal
2003).
A fourth example shows the power of commerce to affect the news judgment of
local television stations. Television stations have long been understandably sensi-
tive to ratings because their income from advertisers depends on the number of
viewers they get. Local news is an area where local stations keep the money.
Frank Magid, one of the most respected and trusted media research firms in US
television and known particularly for its consultation with local television sta-
tions, told local stations the week after the invasion started that covering
anti-war protests could, according to a survey the firm had conducted, affect rat-
ings. The firm’s survey of 6,400 viewers showed that anti-war protest information
was the topic that tested the lowest, or the least of interest, and was most likely
therefore to lead viewers to change channels. Polls by the reputed independent
firm Gallup and others showed at that point that 70 percent of Americans were
for the war. Covering the opinions of the 30 percent could affect ratings, which
would affect profits. It seems that many stations got the message. Anchors were
uniformly supportive of the invasion, and station websites were heavy on the
“support your troops” weblinks (Farhi 2003).
Big media’s downplaying of dissent within the US also generated a counter-
effect. It rekindled the never-extinguished coals of customer suspicion and
disgruntlement. The bigger big media get, the more easily their customers move
from their default stance of cynicism and mistrust to anger and rejection. That is
one good part of the reason why big media executives are so very sensitive to
majority customer mood. And the more demographically-targeted big media
get—a function of increased specialization and multichannelization—the less
leeway big media executives have to deviate from a perceived predilection of its
demographic.
336
BIG MEDIA AND LITTLE MEDIA
Aggregating
Infomediaries abounded, which were established Internet sites that culled both
mainstream reporting and material from “alternative” international or specialized
presses, and also often encouraged “open publishing” (post your own work or com-
ments on others’), in order to compose a news agenda different from the prestige
dailies. These sites acted as semi-open gates for news creators and news seekers.
Indymedia, itself a child of 1999 anti-globalization protests, put the emphasis on
open publishing, although many sites linked on Indymedia were also infomediaries
(Anonymous 2002; Kidd 2002). Some, such as the Information Clearing House
(“News you won’t find on CNN or Fox Moooo’s”), were labors of love, fed inter-
mittently with donations from site visitors (www.informationclearinghouse.info/).
Others, such as oneworld.net and opendemocracy.net, were foundation-funded.
These latter had different missions. OneWorld hosted a virtual community of non-
profit organizations supporting social justice and human rights. Open Democracy
provided a forum for diverse and conflicting opinions on public affairs. Info-
mediary sites deliberately blurred the line between news providers and news
creators. At the same time, they provided a degree of moderating (indymedia.org
much less than others), which contributed to reader confidence in the veracity of
the material.
Do-it-yourself
We noticed a spontaneous creation of new media, often by activists testifying to a
reality that they found obscured in mainstream media and sometimes by people
demanding a voice in a time of crisis. Some of it was email letters, or petitions, or
a forwarded emailed testimony. We called this “electronic samizdat,” under-
ground duplication of information sneaked among friends in the tradition of
337
P AT R I C I A A U F D E R H E I D E
Figure 18.1 “War beyond the box”—this website encapsulated the Center for Social
Media’s scan of the informal media landscape (courtesy of the American
University School of Communication)
338
BIG MEDIA AND LITTLE MEDIA
339
P AT R I C I A A U F D E R H E I D E
340
BIG MEDIA AND LITTLE MEDIA
Figure 18.2 Bloggers’ links turned the “Weapons of Mass Destruction 404” parody web-
site into an international favorite (courtesy of Anthony Cox)
These jokes, like the testimonies, slideshows, and websites, fueled an under-
ground current of resistance. Strongly marked by a sense of community, the
journalistic informal sector was organized by networks of friends and communities
of belief and values. This had implications both for the growth and use of infor-
mation within that sector, and also for its potential participation in the public
opinion-shaping spheres now dominated by professional journalism.
341
P AT R I C I A A U F D E R H E I D E
342
BIG MEDIA AND LITTLE MEDIA
Too often during the Iraq invasion that meant choosing between branded info-
tainment and an uncharted information environment. Big media continued,
during the invasion, to be absolutely critical in establishing what we think is our
common reality, because they acted as big filters. The efforts of some sites in the
informal sector, such as OneWorld, to moderate information, to host exchanges
of viewpoints, and to encourage citizen input within a civil framework are all
interesting and encouraging attempts to address that problem.
The invasion of Iraq, in which the US administration so carefully managed
mainstream media, created unprecedented media activity among people who
insisted upon expression of their views, and who had been developing their digi-
tal skills at their workstations, in class, and while producing their family
newsletters. On the basis of this activity, viral networks grew and some sites
became more established sources of information, investigation, and commentary.
Months after the invasion, most of the sites we found and featured on “War
beyond the box” were still active, and many were still reporting.
The informal sector’s burgeoning activity during the Iraq invasion has prompted
a range of responses. Among what Wired magazine editors early dubbed the
“digerati,” some have been inspired to address at least one of the weaknesses in the
sector: its dispersion and lack of indexing. The Media Venture Collective,2 a net-
work of public-interest-oriented digerati, launched an experimental media project
related to the 2004 US presidential election. The central element of the project was
a repository for affirmatively public domain material (licensed under the terms of
the Creative Commons3) at the servers of the Internet Archive, a public domain
project backed by the dot.com plutocrat Brewster Kahle. It anticipated building
relationships with providers of raw material, producers using the material, and dis-
tributors—all of which were already in place and used during the invasion of
Iraq—to develop an open media network on the model shown in Figure 18.3.
Such a nonpartisan, affirmatively public domain repository for audio-visual
material, proposed by project designer Brad DeGraf deliberately as an experiment
to test the waters, offers an interesting approach to the challenges of the journalis-
tic informal sector today. It could create clear lines of antecedence and
origination, without losing easy access. It could give grassroots-created materials a
home, rather than depending on friends, family, Google, and accident to find
voices beyond big media. It could offer advantages to both traditional journalists
and those arising from the informal sector, in that this kind of central space could
permit spontaneity and flexible use of material by commercial and noncommercial
users, from both traditional and nontraditional sources and journalists. Function-
ing as a kind of public library of ephemeral, Web-based, digital information of all
kinds, such a service would need some kind of public backing, investment or
endorsement, rather than depending on the largesse of dot.com winners. It would
need to become a spot on the public media landscape.
Other models to aggregate and share information, already extant during the
Iraq invasion, will assuredly also develop. The left-culture indymedia sites, for
instance, continue to proliferate worldwide and to draw on the youthful energies
343
P AT R I C I A A U F D E R H E I D E
Figure 18.3 Open media architecture—the nonprofit Media Venture Collective proposed
a nonpartisan repository for informal media linked to events such as the 2004
US election (courtesy of Media Venture Collective)
of many volunteers, at the same time suffering from bouts of incivility and
attempts at sabotage by opposed ideologues. Blogs are becoming a staple journal-
istic hybrid, both personal and public, with some becoming celebrity sites.
Web-based public platforms such as Open Democracy and OneWorld are build-
ing in the learning from each experiment in Web-based public discourse, as they
develop their interactive, grassroots journalism. Meanwhile, mainstream media,
344
BIG MEDIA AND LITTLE MEDIA
especially in their electronic forms, have been made newly aware of the power of
interactivity and the public appetite for it.
The gap between big and little media in the Iraq invasion was enormous, and if
the energy released by the informal sector’s generating of information and opinion
was great, so was the confusion. The contradictions were boldly on display. Minor-
ity voices in the US—generally, those opposed to the war—had an unprecedented
range of expression open to them at unpredecented low costs. At the same time,
these very possibilities were capable of reinforcing paranoia, mistrust, and a sense
of embattlement. These new options could spread misinformation as quickly as
they could enlightenment, and often did.
The journalistic informal sector continued to grow on the sites we identified
on “War beyond the box,” as the post-invasion scenario turned into a grinding
conflict. It will continue to develop, particularly in times of crisis where distrust
of mainstream media turns into action. We may yet get to see the gap between
big media and little media become a spectrum, or even a structured network of
relationships.
For the effusions of informal journalistic sector to develop into a truly public
and participatory sphere of civic engagement as well as discrete communities of
belief, sustained public support and scholarly analysis are both important. Public
support should entail a revision of the current regulatory permission to media
concentration, given the power of highly concentrated media outlets to set agen-
das and intimidate. It should also acknowledge that ad hoc information sharing
in times of crisis indicates the desperate desire of ordinary citizens to contribute
to public decision-making. Such desires should be honored with public resources
to encourage citizen participation. Among many examples of government
fostering of civic informational engagement is the Canadian e-commons project
(http://ecommons.net/). Public funding for affirmatively public domain re-
sources, repositories, and for indexing and archiving systems are valuable invest-
ments in civic discourse. This informal sector also merits closer academic
study—case studies, cultural production studies, textual analysis and reception
studies. The powerful claims that journalists have made in the US over the last
50 years for professional journalistic standards and behaviors can become critical
principles brought to practices developing today in the informal sector.
Notes
1 Participants included principal investigator Professor Patricia Aufderheide, Project
Manager Agnes Varnum, and students Lisa Chan, Aaron Johnson, Navin Kul-
shreshtha, and Catherine Taylor. Professor B.J. Altschul acted as adviser.
2 The Media Venture Collective (www.mediaventure.org) is, as its website maintains, a
“grass roots, non-profit venture fund . . . focused on channeling citizen donations into
strategic investments that facilitate media democracy, and thereby maximizing their
social change impact.”
3 Creative Commons (http://creativecommons.org/) offers licenses that affirmatively
assign the default copyright that authors are granted under US law, in graduated terms,
ranging from highly conditional use only by certain kinds of parties to unconditional
345
P AT R I C I A A U F D E R H E I D E
public domain use. Creative Commons, which exists to facilitate the range of materi-
als available for creators, works with many partners; for instance, the Open Elections
Network expects to use Creative Commons licensing.
References
Aufderheide, P. (1999) Communications Policy and the Public Interest, New York: Guilford
Press.
Bishop, B. and Florida, R. (March 23, 2003) “O, Give me a home where the like-minded
roam,” Washington Post, Outlook edition, p. B5.
Campbell, D. (April 25, 2003) “Dixie sluts fight on with naked defiance,” The Guardian
(London), p. 1.
Cox, A. (July 10, 2003) “The war on the web,” The Guardian (London), www.guardian.
co.uk/online/story/0,3605,994676,00.html, accessed November 12, 2003.
Farhi, P. (March 28, 2003) “For broadcast media, patriotism pays: consultants tell radio,
TV clients that protest coverage drives off viewers,” Washington Post, Style edition,
p. C1.
Gregoire, N. (April 6, 2003) “Radio host, Gov. Bush cheered in Clearwater,” Tampa Tri-
bune, p. 1.
Hazen, Don (February 11, 2003) “Moving on: a new kind of peace activism,” Alternet,
available www.alternet.org/story.html?StoryID=15163 (accessed October 21, 2003).
Kidd, D. (2002) “Indymedia.org: the development of the communications commons,”
Democratic Communiqué, vol. 18, pp. 65–86.
Krugman, P. (March 25, 2003) “Channels of influence,” The New York Times, available
www.nytimes.com/2003/03/25/opinion/ (accessed April 4, 2003).
Robbins, T. (April 16, 2003) “A chill wind is blowing in this nation . . .,” transcript of the
speech given by actor Tim Robbins to the National Press Club in Washington, DC on
April 15, 2003, CommonDreams.org, available www.commondreams. org/views
03/0416-01.htm (accessed October 28, 2003).
Schwartz, J. and Fabrikant, G. (March 31, 2003) “War puts radio giant on the defensive,”
The New York Times, available www.nytimes.com/2003/03/31/business/media (accessed
April 4, 2003).
Segal, D. (April 25, 2003) “Dixie chicks bare their, uh, souls: band counters critics of anti-
war remarks,” Washington Post, Style edition, p. C1.
Shenk, D. (1997) Data Smog: Surviving the Information Glut, 1st edition. San Francisco,
CA: Harper Edge.
Shepard, B.H. and Hayduk, R. (2002) From ACT UP to the WTO: Urban Protest and Com-
munity Building in the Era of Globalization, London: Verso.
US Federal Communications Commission. (FCC) (June 2, 2003) “FCC sets limits
on media concentration,” press release available at www.fcc.gov/ Daily_Releases/
Daily_Business/2003/db0602/DOC-235047A1.pdf (accessed October 21, 2003).
346
19
THE CULTURE OF DISTANCE
Online reporting of the Iraq war
Stuart Allan
Writing in the London Review of Books at the time of the British military conflict
with Argentina over the disputed sovereignty of the Falkland/Malvinas Islands,
Raymond Williams (1982) sought to discern how this “unnecessary war,” to use
his apt turn of phrase, was being reported. He argued that underlying the typical
sorts of questions that arise when television news reports are examined, such as
“issues of control and independence; of the quality of reporting; of access and bal-
ance in discussion,” was a deeper problematic. In order to describe it, he coined
the phrase “the culture of distance.”
The central technical claim of television, Williams pointed out, is its capacity
to represent distant events. While reminding us that the televisual picture of
the world is a selective one, he argued that “what is much more significant is
the revealed distance between the technology of television, as professionally
understood, managed, and interpreted, and the political and cultural space
within which it actually operates” (1982: 14). Across this distance—via the
conventions of “familiar connections”—the tragic devastation endemic to war-
fare is recurrently taken-up and re-inflected by television news into an “antisep-
tic” representation of reality. Not surprisingly for someone who had experienced
conflict first hand, namely as a tank commander in World War II, this problem of
distance was particularly troubling, not least in moral terms. Commenting on the
daily news reporting of the build-up to the seemingly inevitable outbreak of war,
he wrote:
After several days of it, feeling the rhythm soaking in, I happened to pass
a bonfire of rags and oil in the village and suddenly, in an overwhelming
347
S T U A RT A L L A N
It started with an attack by two A-10 jets that danced in the air like
acrobats, tipping on one wing, sliding down the sky to turn on another,
and spraying burning phosphorus to mislead heat-seeking missiles before
turning their cannons on a government ministry and plastering it with
depleted uranium shells. The day ended in blood-streaked hospital corri-
dors and with three foreign correspondents dead and five wounded.
(The Independent, April 9, 2003)
348
T H E C U LT U R E O F D I S TA N C E
Fisk, well known for his incisive writing style, was clearly aiming to ensure that
his eyewitness description of the unfolding events in Baghdad resonated with his
readers. His report appeared beneath the day’s lead story, “The US advance, street
by street,” on the front page. In contrast with it, however, Fisk’s report provided
the kind of personal insight that ordinarily falls outside of the conventionalized
strictures of ostensibly objective, hard news reporting. Indeed, his commitment to
sustaining a reporter-centered narrative—“The A-10s passed my bedroom
window, so close I could see the cockpit Perspex, with their trail of stars dripping
from their wingtips, a magical, dangerous performance fit for any air show, how-
ever infernal its intent”—presumably would have been particularly valued by
many of The Independent’s readers, even though the events in question had tran-
spired the day before.
The issue of immediacy is important here. It is altogether likely, of course, that
most of these same readers would have been aware of much more recent develop-
ments in Baghdad than those described in their newspaper, courtesy of the
electronic media. Television, as one would expect, led the way in reporting the
battle for control of the city as it unfolded on April 9. While fighting continued in
some areas, images of a statue of Saddam Hussein in Firdous Square being pulled
from its plinth by a US armored personnel carrier (the bronze head promptly, if
only momentarily, draped by a US flag) featured in television newscasts around the
globe. In visual terms, the fallen statue promptly assumed a charged symbolic
status in some news reports, frequently being held to represent the collapse of the
central regime’s authority, the power of Saddam broken and removed. This tele-
vision coverage, it would be later claimed by some commentators, effectively
demonstrated the immediacy of real-time reporting and its benefits. No war, they
argued, had been better recorded, the sheer volume of words and images offering
an unprecedented degree of detail in near-instantaneous time. Citing factors such
as improvements in news technologies, as well as the use of reports from “embed-
ded” correspondents, they insisted that many of the criticisms first leveled at
24-hour news in the 1991 Gulf War had been laid to rest.
Meanwhile, observations of a different sort were being made time and again
on different Internet websites in response to the day’s television reporting (well
before newspaper reports covering the events in Firdous Square had gone to
press). Across chat rooms, bulletin boards, discussion forums, weblogs, and the
like, Internet users gave voice to their points of view about what these events
meant to them. Some rejoiced, while others, in sharp contrast, demanded to
know what was really happening on the ground in Baghdad. For some, the top-
pling of the statue appeared to have been almost choreographed for the benefit
of the cameras. Awkward questions were posed on various sites about whether it
was a spontaneous act (or one organized with US television schedules in mind?),
the composition of the “crowd” of onlookers (was their number made to appear
more substantial by the camera angles chosen?) and the extent to which these
“jubilant” people were actually “celebrating,” among other concerns. Many of
those writing online posts vented their anger at what they regarded to be the
349
S T U A RT A L L A N
pro-war, even jingoistic stance of mainstream news reports, singling out Fox
News for particular criticism owing to its perceived over-reliance on official “pro-
paganda” and “spin.” Regardless of differing political perspectives, however,
many users simply wanted much more by way of context, critique, and explana-
tion than television news was providing that day, and were too impatient to wait
for the next day’s newspapers.1
Accordingly, news sites—whether “official” ones associated with an estab-
lished news organization, or “unofficial” ones such as personal blogs—were prov-
ing to be indispensable resources. From the moment news of the first attacks
launching so-called “Operation Iraqi Freedom” on March 19, 2003 was reported,
Internet traffic to online news services surged dramatically. More people than
ever, according to companies monitoring Internet traffic such as Hitwise, Nielsen
Net Ratings, and the like, were surfing the Internet for news and information.2
For many Internet commentators, the US-led attack on Iraq represented the
“coming of age” of the Internet as a news medium. Regularly singled out for
attention was the role of high-speed, broadband Internet access, not least its
capacity to enable news sites to offer users live video and audio reports, multime-
dia slideshows, animated graphics, interactive maps, and so forth. The rapid rise
in the number of users availing themselves of the technology—over 70 million
people in the US at the time—meant that providers could further enhance exist-
ing types of digital reportage accordingly (Kirkpatrick 2003). Moreover, other
commentators pointed to the ways in which online news was consolidating its
position as a primary news source. Of significance here, for example, was the
extent to which users, especially office workers unable to watch television in the
workplace, were relying on the Internet for up-to-the-minute news of breaking
developments. Research conducted during the first six days of the war by the Pew
Internet and American Life Project (2003) indicated that 56 percent of online
users in the US had turned to news sites for reports about the conflict. “More
than half the people who are online are getting their news online—that’s never
happened before,” Lee Rainie, the project’s director, maintained. “It’s another
milestone moment for online news” (cited in Weaver 2003).
Moreover, this same opinion survey by Pew sought to determine what US citi-
zens thought about the conflict, how they were acquiring their news about it, and
what sort of impact developments were having on them.3 Briefly, its findings sug-
gested that 77 percent of the country’s 116 million adult Internet users had been
online in connection with the war in Iraq. Their main reasons for turning to the
Internet, as one might expect, included searching for information about the war,
seeking alternative opinions about the conflict, sending, and receiving emails
about pertinent events, and expressing their views and offering prayers. Evidently
a relatively small percentage of users were making use of email to mobilize others
in efforts to build collective support for their views about the conflict—while 10
percent of users received email from organizations against the war, 7 percent of
them received email from pro-war organizations. Approximately one in seven (or
14 percent) of users said that they had been online more frequently than usual
350
T H E C U LT U R E O F D I S TA N C E
351
S T U A RT A L L A N
was presumably this very gap in perception which motivated many users to look
abroad in the first place.
“In the Internet age,” as US journalist Elizabeth Putnam (2003) pointed out,
“overseas newspapers are a few keystrokes away, making them more available
and attractive to people who want to understand why there seems to be so much
anti-American sentiment around the world.” Nowhere were these sorts of
tensions more apparent than with regard to news sites in the Arab world. From
the vantage point of most US and UK users, however, no site in the region
would attract more intense interest during the Iraq war than Al-Jazeera (www.
aljazeera.net).
Al-Jazeera online
Often described as the “CNN of the Arab World,” Al-Jazeera (which means “an
island” in Arabic) is arguably the region’s most influential news organization.
Launched in the Qatari capital, Doha, in 1996, the 24-hour satellite television
network attracts an audience currently estimated to be about 35 million regular
viewers, making it the most widely watched Arab news channel. Available free of
charge throughout much of the Arab world, it is typically a pay-television channel
in Europe and North America. Although backed financially by the government of
Qatar, Al-Jazeera’s journalists consistently maintain that their editorial freedom is
not compromised as a result. That said, the network’s status as an independent
voice in the Arab world, encapsulated in its slogan “The opinion and the other
opinion,” is frequently called into question by its many critics. For some, the net-
work’s commitment to providing news coverage from an Arab perspective means
that it is ideologically compromised, and as such biased against the US and Israel.
Other critics, in contrast, have denounced Al-Jazeera for being a Zionist tool,
while still others insist that it is little more than a front for the Central Intelli-
gence Agency (CIA). In any case, above dispute is the fact that its news coverage
has recurrently placed a considerable strain on Qatar’s relations with other coun-
tries in the region, including Bahrain, Jordan, and Saudi Arabia where the
network’s offices have been closed on different occasions.
No stranger to controversy, Al-Jazeera came to prominence across the global
mediascape in the aftermath of the dreadful events of September 11, 2001, owing
to its decision to broadcast taped messages attributed to Osama bin Laden (see also
Zelizer and Allan 2002; el-Nawawy and Iskandar 2003; Thussu and Freedman
2003). News organizations around the world paid considerable sums to air edited
excerpts, much to the consternation of US officials—not least National Security
Adviser Condoleezza Rice, for example, who demanded of television network
executives that they “exercise judgment” (i.e. censorship) in re-broadcasting the
messages. Interestingly, most of the considerable traffic to the network’s site
(www.aljazeera.net) at the time was from the US, despite the fact that its content
was entirely in Arabic. During the subsequent “war on terror” in Afghanistan,
attention was once again directed at Al-Jazeera’s role in making available reports
352
T H E C U LT U R E O F D I S TA N C E
of the conflict that challenged the preferred definitions of reality set down by mili-
tary officials. For this reason alone, further controversy erupted in November 2001
when a US “smart” bomb destroyed the network’s Kabul offices. Intense specula-
tion ensued that the offices had been deliberately destroyed. For example, Nik
Gowing, a presenter on BBC World, stated afterwards that Al-Jazeera’s only crime
was “bearing witness” to events that the US officials would prefer it did not see. In
demanding that the Pentagon be called to account, he pointed out that when the
presence of journalists is “inconvenient” they risk becoming “legitimate targets” in
the eyes of the military—a charge promptly denied, as one would expect, by a Pen-
tagon spokesperson (see Wells 2001).5
Following the start of “Operation Iraqi Freedom,” subscriber numbers surged
dramatically in response to the intense demand for alternative insights into
the conflict. The number of subscribers to the channel in Europe, it was claimed
at the time, effectively doubled once the war was underway. The depth of its
reporting was recurrently singled out for praise—or condemnation—depending
on conflicting perceptions of the relative legitimacy of the war. In addition to
reporting from Central Command in Qatar, four of Al-Jazeera’s reporters were
“embedded” with the US and British military forces. In the main, however, the
network ensured that most of its journalists roamed more freely. Together they
covered the breadth of Iraq, including areas where Western journalists did not
venture. The Al-Jazeera television crews remained in Baghdad throughout the
conflict, as well as in other major battlegrounds such as Basra, Mosul, and in Kur-
dish-controlled northern Iraq. Not surprisingly, a very different kind of coverage
ensued. Tarik Kafala (2003), a BBC News Online reporter, identified a case in
point. “When Western journalists outside Basra were speculating about an upris-
ing on the basis of coalition briefings,” he observed, “Al-Jazeera’s correspondent
inside the city was reporting first hand that ‘the streets are very calm and there
are no indications of violence or riots.’” This type of disjuncture between the net-
work’s reporting and that of its Western rivals attracted considerable comment.
US Secretary of State Colin Powell, for example, criticized the coverage, con-
tending that it “magnifies the minor successes of the [Iraqi] regime and tends to
portray our efforts in a negative light” (cited in Delio 2003). For others, however,
it was the very extent to which Al-Jazeera’s reporting called into question the
more “sanitized” representations of the conflict that made its presence so impor-
tant—both on their television screens and, increasingly, on their personal
computers (see Gubash 2003).
Prior to the launch of Al-Jazeera’s website, Arabic speakers were typically most
interested in CNN.com (www.arabic.cnn.com) when looking for news online.
Since the September 11 attacks, however, the page views for the Arabic-language
site operated by Al-Jazeera reportedly grew from about 700,000 a day to 3 million,
with more than 40 percent of visitors logging-on from the US (Ostrom 2003).
Indeed, at the outbreak of hostilities in Iraq, aljazeera.net was widely recognized
as receiving the most “hits” of any Arabic site in the world. Of critical signifi-
cance here was its commitment to pushing back the boundaries of Western
353
S T U A RT A L L A N
At no time was this difference in news values cast in sharper relief than on
March 23, the night Al-Jazeera broadcast footage of US casualties, as well as Iraqi
television’s interviews with five US prisoners of war. Al-Jazeera’s decision to air
the interviews was promptly denounced by US Defense Secretary, Donald Rums-
feld, who alleged that it was a violation of the Geneva Convention protecting
prisoners of war. In reply, the network’s London bureau chief, Yosri Fouda, argued
that Western news reports were being constrained to the extent that they failed
to provide accurate coverage. Regarding the Geneva Convention, he insisted
that a double standard was being invoked. “We and other broadcasters were not
criticized for showing pictures of Iraqi dead and captured,” he stated, “or those
famous pictures from Guantanamo Bay” (cited in Kafala 2003).
The more heated the ensuing furore became, of course, the more news head-
lines it generated around the world. The very images deemed by Western news
organizations to be too disturbing to screen were being actively sought out by vast
numbers of people via online news sites. According to figures compiled by popu-
lar search engines, such as Google, Lycos, and AltaVista, the term “Al-Jazeera”
was quickly becoming one of the most searched-for-topics on the Web. Figures for
the week in question indicated that the term “Al-Jazeera” (and variant spellings)
was the term that showed the greatest increase on Google, while Lycos reported
that it was the top search term, with three times more searches than “sex” (a
perennial favourite with Web surfers). For Karl Gregory of AltaVista, the popu-
larity of Al-Jazeera’s online sites was clear evidence of “people branching out
354
T H E C U LT U R E O F D I S TA N C E
beyond their normal sources of news” (BBC News Online, April 1, 2003). The
decision taken at Al-Jazeera to broadcast the images, as well as to display them
online, was justified by its spokesperson, Jihad Ballout, as being consistent with
its journalistic ethos of reporting the war as it was being fought on the ground. In
his words: “We didn’t make the pictures—the pictures are there. It’s a facet of the
war. Our duty is to show the war from all angles” (cited in Whitaker 2003). In the
opinion of others, however, the network had become a mouthpiece for Iraqi pro-
paganda. Citing the images, some military officials began ignoring questions from
Al-Jazeera’s reporters at briefings. At the same time, two of the network’s finan-
cial reporters were evicted from the floor of the New York Stock Exchange
(Nasdaq would follow suit, citing “Al-Jazeera’s recent conduct during the war” as
the reason), their press credentials revoked. It was in cyberspace, however, that
the backlash registered most decisively as various pro-war individuals and groups
made clear their intent to make Al-Jazeera a target of retaliation.
News sites of all descriptions are always vulnerable to attack from hackers—
typically involving little more than webpage defacements and graffiti—but those
directed at Al-Jazeera’s sites were remarkably vicious. The “electronic onslaught,”
as aptly characterized by one Internet commentator, began on March 25, the
same day the English-language site, www.english.aljazeera.net, was launched. Two
days later, hackers “crashed” both sites, effectively forcing them offline by a
“denial of service” or DOS attack. This type of attack aims to close down a tar-
geted site by overwhelming the associated server with so much meaningless data
that it can no longer handle legitimate traffic. Few sites have sufficient resources,
such as the necessary bandwidth, to withstand millions of simultaneous page
impressions. Such was certainly the case with both Al-Jazeera sites. The English-
language site was disabled virtually from the outset, while its Arabic-language
counterpart struggled—with only limited success—to hold up against the storm.
Efforts to restore the sites, which reportedly included re-aligning them with
servers in France, encountered fierce resistance by repeated hack attacks. “We
come up for five or ten minutes,” stated Salah Al-Seddiqi, IT manager at Al-
Jazeera, “and then the attacks bring us down again” (cited in Roberts 2003).
Later the same day, even though security protocols had been reinforced for the
sites, matters went from bad to worse. Evidently, a pro-war hacker was able to
access the servers at Network Solutions Inc., a domain name registration service
based in Dulles, Virginia, that operates a database linking addresses (in this case,
www.aljazeera.net) with the identification numbers of the servers responsible for
maintaining its Web pages. This meant that Al-Jazeera’s domain was effectively
“hijacked” by the hacker, such that users were pointed to an altogether different
site instead. Specifically, traffic was redirected to a pro-war webpage featuring a
US flag, together with the messages “Let freedom ring” and “God bless our
troops,” signed by a self-proclaimed “Patriot.” It was quickly determined that this
latter site belonged to an Internet provider based in Salt Lake City, Utah, albeit
without their knowledge. Hackers calling themselves the “Freedom Cyber Force
Militia” had claimed responsibility for the attack, but in any case the registration
355
S T U A RT A L L A N
Blogs at war
Much of what passes for journalism in the US, in Mediachannel.org editor Danny
Schechter’s (2003) assessment, “is seen as nothing but propaganda by people in
other countries and by an increasing number of Americans, who are turning to
international Web sites to find the kind of news they can no longer get here.” In
addition to international sites such as Al-Jazeera’s, however, an altogether differ-
356
T H E C U LT U R E O F D I S TA N C E
ent type of site has similarly attracted a remarkable degree of attention during the
Iraq conflict. Specifically, news-oriented weblogs achieved widespread public
salience, being heralded as a new interactive form of participatory reporting,
commentary, and analysis of breaking news. Indeed, by the time of the formal
declaration of “Operation Iraqi Freedom” on March 19, 2003, the term “blog”
was rapidly being appropriated into the everyday language of journalism.
Weblogs, or blogs for short, may be characterized as diaries or journals written
by individuals with net access who are in possession of the necessary software
publishing tools (e.g. those provided by sites such as Blogger.com) to establish an
online presence. Most bloggers pull together their resources from a diverse array
of other sites, thereby situating a given news event within a larger context, and
illuminating multiple dimensions of its elements. The apparent facts or claims
being collected are usually time-stamped and placed in reverse-chronological
order as the blog is updated, making it easier for readers to follow its ongoing nar-
rative. Customarily the sources of the blogger’s information are acknowledged
explicitly and the accompanying hyperlink enables the user to negotiate a net-
work of cross-references from one blog to the next, or from other types of sites
altogether. In principle, the facts or claims presented in any one blog can be sub-
jected to the relentless double-checking of users, some of whom may be even
better informed about the events in question than the initial blogger. Any
attempt by a blogger to present a partisan assertion as an impartial statement of
fact is likely to be promptly recognized as such by other users.7
Many news bloggers—a small minority compared to the number of ordinary
netizens involved overall—consider themselves to be “personal” journalists,
intent on transgressing the border between “professional” and “amateur” report-
ing. By acting as “unofficial” news sources on the Web, these blogs link together
information and opinion which supplements—or, in the eyes of some advocates,
supplants—the coverage provided by “official” news outlets. The potential of
blogs in this regard was widely recognized during the tragic events of Septem-
ber 11, 2001. In the early hours after the attacks, most of the major news sites in
the US, as well as others such as the BBC’s site in London, were so besieged by
user demand that they were largely inaccessible (see Allan 2002). As one site
after the next refused to load properly, users turned elsewhere for news of breaking
developments. Hundreds of refashioned websites began to appear over the course
of the day, making publicly available eyewitness accounts, personal photographs
and in some cases video-footage of the unfolding disasters. Of particular impor-
tance here was the crucial role played by blogs in making these forms of
“amateur,” “guerrilla,” or “DIY” (do it yourself) journalism available. “Most of the
amateur content,” Kahney (2001) remarked at the time, “would be inaccessible,
or at least hard to find, if not for many of the Web’s outstanding weblogs, which
function as ‘portals’ to personal content.” Managers of these blogs spent the day
rapidly linking together any and all items of “personal journalism” from “amateur
newsies” onto their respective sites. In so doing, they rendered problematic the
familiar criteria defining what counted as news—as well as who qualified to be a
357
S T U A RT A L L A N
358
T H E C U LT U R E O F D I S TA N C E
Someone reporting live from the battlefield for CNN can’t come close” (cited in
Allemang 2003). Commenting on this type of “horizontal” communication,
Glenn Harlan Reynolds (2003) of InstaPundit.com noted wryly that “the term
‘correspondent’ is reverting to its original meaning of ‘one who corresponds,’
rather than the more recent one of ‘well-paid microphone-holder with good
hair.’”
While it is difficult to generalize, most warbloggers posting from Iraq seemed
motivated to share their eyewitness experiences of the conflict so as to counter-
balance mainstream news media coverage. The work of CNN correspondent
Kevin Sites was a case in point. In addition to filing his television reports, Sites
wrote “behind the scenes” features for CNN.com, all the while maintaining a
multimedia blog. Published on his own site, Sites’ blog provided his personal
commentary about the events he was witnessing from one day to the next, along
with various photographs and audio reports that he prepared. Perhaps in light of
the media attention Sites’ blog received, however, CNN asked him to suspend it
on Friday, March 21, 2003. A spokesperson for the network stated at the time
that covering war “is a full-time job and we’ve asked Kevin to concentrate only
on that for the time being” (cited in Kurtz 2003). Sites agreed to stop blogging,
later explaining that “CNN was signing my checks at the time and sent me to
Iraq. Although I felt the blog was a separate and independent journalistic enter-
prise, they did not” (www.kevinsites.net). Reactions from other bloggers were
swift. CNN’s response, according to Steven Levy (2003) of Newsweek, “was seen
in the Blogosphere as one more sign that the media dinosaurs are determined to
stamp out this subversive new form of reporting.”8
In contrast, MSNBC’s support for blogging meant that three warblogs were
focused on war coverage at the height of the conflict. “Weblogs are journalism,”
argued Joan Connell, one of the site’s executive producers. “They can be used to
great effect in reporting an unfolding story and keeping readers informed” (cited
in Mernit 2003). Nevertheless, while she does not share CNN’s stance that blogs
lack a sufficiently “structured approach to presenting the news,” she does believe
that there is a necessary role for an editor in the process. In her words: “Unlike
many Weblogs, whose posts go from the mind of the writer straight into the
‘blogosphere,’ MSNBC’s weblogs are edited. Our editors scrutinize our weblogs for
accuracy, fairness, and balance, just as they would any news story” (cited in
Mernit 2003). Not all bloggers on the front lines were associated with a major
news organization, however. Many worked as a “sojo” or “solo journalist,” writing
and editing their own copy for both online and print or broadcast media. Being
almost constantly on the move meant relying on mobile technologies, such as a
notebook computer and digital camera, or even a videophone and mini-satellite
dish. Still, for these bloggers, their relative freedom of movement enabled them
to pursue the stories which mattered most to them—and the readers of their war-
blog. Herein lay the popularity of the warblogs among users, which in the opinion
of journalist Bryony Gordon (2003) was hardly surprising: “if a television
reporter’s movements aren’t subject to Iraqi restrictions, then his [or her] report is
359
S T U A RT A L L A N
likely to be monitored by the Allied Forces. Devoid of such regulations, the Inter-
net is thriving.”
Freelancer Christopher Allbritton had announced his intention to be the
Web’s first independent war correspondent in the months leading up to the inva-
sion. His blog, titled “Back to Iraq. 2.0” (www.back-to-iraq.com), called upon
readers to help contribute to the financial support necessary to fund his travel and
expenses in Iraqi Kurdistan. “It’s a marketplace of ideas,” he maintained, “and
those who are awarded credibility by their readers will prosper” (cited in Warner
2003). Support was such that his expenses were met by some 320 donors, allow-
ing him to file daily stories from the country using a borrowed notebook computer
and a rented satellite phone. As his blog’s daily readership grew to upwards of
25,000, he became accustomed to receiving emails which posed questions and
suggested story leads, while others provided useful links to online materials. “My
reporting created a connection between the readers and me,” Allbritton (2003)
later observed, “and they trusted me to bring them an unfettered view of what I
was seeing and hearing.” This involvement on the part of his readers in shaping
his reporting worked to improve its quality, in his view, each one of them effec-
tively serving as an editor. “One of the great things about the blogosphere,” he
maintained, “is that there’s built-in fact-checking.” Given that so many people
will “swarm” over posts, “generally the truth of the matter will come out” (cited
in Glaser 2003).9
Precisely what counts as truth in a war zone, of course, is very much in the eye
of the beholder. Above dispute, in the view of many commentators, was that
some of the best eyewitness reporting being conducted was that attributed to the
warblog of “Salam Pax” (a playful pseudonym derived from the Arabic and Latin
words for peace), a 29 year-old architect living in middle-class suburban Baghdad.
Indeed, of the various English language warblogs posted by Iraqis, none attracted
a greater following than Salam’s “Where is Raed?” (dear_raed.blogspot.com),
which had begun to appear in September 2002. His motivation for blogging was
later explained as a desire to keep in touch with his friend Raed, who had moved
to study in Jordan. In the months leading up to the initial “decapitation attack,”
to use his turn of phrase, the blog contained material ranging from personal—and
frequently humorous—descriptions of everyday life, to angry criticisms of the
events around him. It was to his astonishment, however, that he discovered that
the international blogging community had attracted such intense attention to his
site. As word about “Where is Raed?” spread via other blogs, email, online discus-
sion groups, and mainstream news media accounts, it began to regularly top the
lists of popular blogs as the conflict unfolded. For Salam, this attention brought
with it the danger that he would be identified—a risk likely to lead to his arrest,
possibly followed by a death sentence. At the same time, speculation over the
identity of the Baghdad Blogger—and whether or not “Dear Raed” was actually
authentic—was intensifying. Some critics claimed that it was an elaborate hoax,
others insisted it was the work of Iraqi officials, while still others maintained that
a sinister CIA disinformation campaign was behind it. Salam responded to skep-
360
T H E C U LT U R E O F D I S TA N C E
tics on March 21, writing: “please stop sending emails asking if I were for real,
don’t belive [sic] it? then don’t read it.” Moreover, he added, “I am not anybody’s
propaganda ploy, well except my own” (cited in BBC News Online, March 25,
2003).
Enraged by both Saddam Hussein’s Ba’athist dictatorship and George W. Bush’s
motivations for the invasion, Salam documented life on the ground in Baghdad
before and after the bombs began to drop. This was “embedded” reporting of a very
different order, effectively demonstrating the potential of blogging as an alterna-
tive means of war reporting. His warblog entry for March 23, 8:30 p.m., was
typically vivid:
Today’s (and last night’s) shock attacks didn’t come from airplanes but
rather from the airwaves. The images Al-Jazeera are broadcasting are
beyond any description. . . . This war is starting to show its ugly face to
the world. . . . People (and I bet “allied forces”) were expecting things to
be much easier. There are no waving masses of people welcoming the
Americans, nor are they surrendering by the thousands. People are
doing what all of us are doing—sitting in their homes hoping that a
bomb doesn’t fall on them and keeping their doors shut.
Salam Pax, dear_raed.blogspot.com
Salam’s posts offered readers a stronger sense of immediacy, an emotional feel for
life on the ground, than more traditional news sites. For John Allemang (2003),
writing in The Globe and Mail, “what makes his diary so affecting is the way it
achieves an easy intimacy that eludes the one-size-fits-all coverage of Baghdad’s
besieged residents.” As Salam himself would later reflect, “I was telling everybody
who was reading the web log where the bombs fell, what happened . . . what the
streets looked like.” While acknowledging that the risks involved meant that he
considered his actions to be somewhat “foolish” in retrospect, nevertheless he
added: “it felt for me important. It is just somebody should be telling this because
journalists weren’t” (cited in Church 2003).
Multiple truths
Any bold declaration that online journalism will abolish once and for all the
“culture of distance” will invite a more considered response, once it is situated in
relation to the sorts of developments discussed above. As has been made appar-
ent, however, these emergent forms of journalism have the capacity to bring to
bear alternative perspectives, contexts, and ideological diversity to war reporting,
providing users with the means to connect with distant voices otherwise being
marginalized, if not silenced altogether, from across the globe. In the words of US
journalist Paul Andrews (2003), “media coverage of the war that most Americans
saw was so jingoistic and administration-friendly as to proscribe any sense of
impartiality or balance,” hence the importance of the insights provided by the
361
S T U A RT A L L A N
likes of Salam Pax. This “pseudonymous blogger’s reports from Iraq,” Andrews
believed, “took on more credibility than established media institutions.” This
point is echoed by Toby Dodge (2003), who argued that Salam managed to post
far more perceptive dispatches than those written by “the crowds of well-
resourced international journalists sitting in the air-conditioned comfort of five
star hotels.” Communicating to the world using a personal computer with unreli-
able Internet access, he reported “the traumas and more importantly the opinions
of Iraqis as they faced the uncertainty of violent regime change.”
To close, this chapter has taken as its focus some of the ways in which online
reporting opens up alternative spaces for acts of witnessing. Warblogs—together
with sites such as those of belonging to Al-Jazeera—have been shown to possess
the potential to throw into sharp relief the narrow ideological parameters within
which mainstream news media typically operate. Journalists’ routine, everyday
choices about what to report—how best to do it, and why—necessarily implicate
them in a discursive politics of mediation. The very multi-vocality at the heart of
their narrativization of reality renders problematic any one claim to truth, and in
so doing reveals that witnessing is socially situated, perspectival, and thus politi-
cized. Before online reporting can become interactively dialogical in any
meaningful sense of the term, however, it will have to counter the forms of social
exclusion endemic to the culture of distance. A first step in this direction, as this
chapter has sought to demonstrate, is to recognize that the culture of distance is,
simultaneously, a culture of othering. At stake, in my view, is the need to decon-
struct journalism’s “us and them” dichotomies precisely as they are taken up and
re-inflected in news accounts where the structural interests of “people like us” are
counterpoised against the suffering of strangers. To recast the imperatives of
“here” and “there,” and thereby resist the familiar pull of the culture of distance,
it is the corresponding gap between knowledge and action that will have to be
overcome.
Notes
1 To gain a quick sense of these sorts of interventions across the webscape, simply type
the words “Saddam,” “statue,” and “blog” into a search engine, such as Google
(www.google.com). At the time of writing, some 7,830 hits were generated by this
combination, thereby providing a flavor of the nature of the ensuing discussion and
debate.
2 In Britain that day, the level of traffic to The Guardian newspaper’s website soared by
nearly 30 percent to around 4.5 million impressions. According to Hitwise research,
The Guardian’s site was the leading online newspaper service with a 7.26 percent share
of the market, followed by FT.com (5.17 percent), the Sun (3.05 percent), The Times
(2.86 percent), the Telegraph (2.24 percent) and the Independent (1.51 percent). Of
the non-print sites, the British Broadcasting Corporation’s stand-alone news site was
ranked highest with a 4.69 percent share. Evidently traffic to this BBC site was up by
30 to 40 percent for the day, a level of demand which appeared to have caused the ser-
vice to repeatedly “crash” in the early hours (see Timms 2003). Over the course of the
days to follow, people going online during office hours appeared to be largely responsi-
ble for the surge in traffic to news sites. Many were seeking out alternative news
362
T H E C U LT U R E O F D I S TA N C E
sources, as well as wanting particular types of perspectives about the factors under-
pinning the conflict. “These figures show the desire of British surfers to get a real range
of informed opinion on the war,” argued Tom Ewing, a Nielsen Net Ratings analyst.
“This shows where the Internet comes into its own when fast-moving news stories are
involved” (cited by BBC Online, April 15, 2003).
In the US, Yahoo.com reported that in the first hour following President George
W. Bush’s announcement that the conflict had started, traffic levels to its site were
three times higher. The volume of traffic to its news section jumped 600 percent the
next day (Thursday, March 20) and again the day after. The sites associated with dif-
ferent television networks proved particularly popular. On the Thursday, CNN.com
evidently secured the highest figures for all news sites with 9 million visitors, followed
by MSNBC with 6.8 million (about half of the visitors for both sites were accessing
them from their workplaces). Other news sites witnessing a significant rise in demand
that day included Foxnews.com (77 percent increase), Washingtonpost.com (29 per-
cent increase) and USAToday.com (17 percent). “Without a doubt,” stated Daniel E.
Hess of ComScore, “people are glued to their Web browsers for virtually minute-by-
minute updates of the war as it unfolds” (cited in Walker 2003, see also Richtel 2003).
3 The evidential basis for the study’s findings was derived from a daily tracking survey,
carried out via telephone interviews among a random sample of 1,600 adults between
March 20 and 25, 2003 (999 of whom were Internet users). “For results based on the
total sample,” the Pew report states, “one can say with 95 percent confidence that the
error attributable to sampling and other random effects is plus or minus 3 percentage
points. For results based on Internet users (n = 999), the margin of sampling error is
plus or minus 4 percentage points. In addition to sampling error, question wording and
practical difficulties in conducting telephone surveys may introduce some error or bias
into the findings of opinion polls” (Pew 2003: 10).
4 Additional results worthy of attention here include the study’s finding that Internet
users in the US were likely to support their country’s war effort by a 3 to 1 margin.
Some 74 percent of users surveyed were found to be backing the war effort in the early
days of the campaign, compared with 22 percent who were opposed to it. Still, it
appears that in contrast with those who support the war, its opponents “are more polit-
ically active online, more anxious to discuss the war, and more likely to seek out a
variety of sources of information about the war” (2003, p. 8).
5 Among the dead foreign correspondents mentioned in Robert Fisk’s report (discussed
above) was an Al-Jazeera reporter, killed by the US air attack on the network’s office
in Baghdad. “Despite two separate assurances from the American government that Al-
Jazeera’s base of operations would not be targeted,” he wrote, “it was destroyed.”
6 Further details regarding who was behind the pro-war hacking attacks against Al-
Jazeera have begun to emerge in the months since these events transpired. In June
2003, John William Racine II, a website designer from Norco, California, pleaded
guilty to felony charges revolving around the hijacking of the Al-Jazeera site. Evi-
dently he had contacted the Federal Bureau of Investigation (FBI) himself on
March 26 in order to confess to the “scheme to defraud,” which also included inter-
cepting some 300 email messages sent to Aljazeera.net (US District Court, California,
case no. CR 03-557; filed June 9, 2003).
7 Interestingly, Fisk has seen his surname turned into a verb—“Fisking”—by some blog-
gers as a form of shorthand to describe the critical practice of deconstructing a
published news item on a point-by-point basis. A somewhat skeptical Brendan O’Neill
(2003) comments: “Fisk is now a kind of mythical figure, that strange British journalist
who dares to say the unthinkable—a view which, it has to be said, is often out of pro-
portion to any biting insight on Fisk’s part.” Evidently it is fair to say that, for some,
Fisking is a way to challenge mainstream journalism’s hegemony while, for others, it is
little more than an opportunity to engage in a politically partisan rant.
363
S T U A RT A L L A N
References
Allan, S. (2002) “Reweaving the Internet: online news of September 11,” in B. Zelizer and
S. Allan (eds), Journalism after September 11, London and New York: Routledge,
119–40.
Allbritton, C. (2003) “Blogging from Iraq,” Nieman Reports, Fall, pp. 82–5.
Allemang, J. (2003) “Where everybody is a war reporter,” The Globe and Mail, www.globe
andmail.com, March 29.
Andrews, P. (2003) “Is blogging journalism?,” Nieman Reports, Fall, pp. 63–4.
Bodi, F. (2003) “Al-Jazeera tells the truth about war,” The Guardian, March 28.
Carlson, P. (2003) “In the line of fire,” The Washington Post, April 3.
Church, R. (2003) “Interview with Salam Pax,” CNN International, transcript
no. 100302cb.k18, October 3.
Delio, M. (2003) “US tries email to charm Iraqis,” Wired News, February 13.
Dodge, T. (2003) “An Iraqi in cyberspace,” The Times Literary Supplement, October 24.
el-Nawawy, M. and Iskandar, A. (2003) Al-Jazeera: the Story of the Network that is Rattling
Governments and Redefining Modern Journalism, Boulder, CO: Westview.
Gallagher, D.F. (2002) “A rift among bloggers,” The New York Times, www.nytimes.com,
June 10.
Glaser, M. (2003) “Reading between the lines in Iraqi blogs and newspapers,” Online Jour-
nalism Review, www.ojr.org, November 7.
Gordon, B. (2003) “The Internet is having a field day—war ‘blogs’ are everywhere,” The
Telegraph, www.telegraph.co.uk, April 2.
Gubash, C. (2003) “New Arab TV channels show clout,” MSNBC.com, March 31.
364
T H E C U LT U R E O F D I S TA N C E
365
INDEX
ABC News (Australia) 226, 227, 229, 230, Al-Seddiqi, Salah 355
231, 232, 233, 235, 237 AltaVista 354
ABC News (US) 7, 80, 87, 88, 103, 109, Alternet 337
140, 142, 143, 144, 152, 200, 261, 356 Amanpour, Christiane 8, 9, 70, 182
Abu Dhabi television 81, 261, 307, 318, Amnesty International 142
319, 330 Anderson, Steve 7
Acheson, Dean 37 Andrews, Paul 361
Adams, Eddie 119, 121, 126 anthrax 48, 64
Adie, Kate 3, 5, 11 anti-war protests 10, 28, 31, 54, 103, 104,
advertising (influence on reporting) 12, 26, 108, 142, 143, 145, 159, 180, 247, 248,
38, 93, 101, 124, 125, 334, 336, 339 252, 253, 254, 255, 256, 259, 262,
Afghanistan 27, 28, 29, 30, 33, 34, 35, 36, 266–82, 299, 301, 305, 334, 336, 337,
45, 51, 65, 68, 69, 77, 78, 80, 87, 105, 340, 342
107, 120, 122, 123, 124, 129, 131, 153, AOL Time Warner 333
174, 175, 184, 191, 206, 211, 220, 247, Argentina 191, 347
259, 260, 299, 315, 319, 325, 352, 358 Arnett, Peter 159, 261
Africa 16, 63, 67, 120, 155–73, 207 A-shafi, Suleiman 88
Agence France Presse (AFP) 18, 44, 304, Associated Press (AP) 18, 68, 176, 304,
305, 306, 307, 308, 309 308, 309, 312, 339
agenda-setting 6, 11, 12, 15, 26, 30, 33, 54, asylum seekers 48, 166, 184, 262, 274; see
69, 71, 78, 96, 97, 99, 100, 102, 105, also refugees
158, 161, 162, 168, 212, 216, 231, 268, Atlanta Journal Constitution 67, 261
302, 337, 342, 345 Atomic Weapons Authority 275
Al-Arabiya 81, 326 Atta, Muhammad 90, 91
Albright, Madeleine 164, 176 Auchmutey, Jim 67
Al-Jazeera 8, 18, 19, 51, 65, 80, 81, 83, 84, Australia 17, 31, 208, 224–43, 299, 364
85, 86, 93, 102, 104, 121, 260, 283, 287, Austria 306, 307, 308
302, 308, 309, 312, 313, 315–32, 348, “axis of evil” 64, 77, 250, 259, 278
352–6, 361, 362, 363 Ayoub, Tareq 325
Al-Kahki, Amr 323, 324 Ayres, Chris 199, 202
Al-Kasim, Faisal 320, 330, 331 Ayyoub, Tayek 51
Allbritton, Christopher 339, 360, 364 Aziz, Tariq 8
Allemang, John 361
Allen, Dean 358 Baghdad Blogger; see Salam Pax
Alouni, Tayseer 84, 86, 93 Bahrain 50, 352
al-Qaeda 9, 45, 62, 64, 65, 66, 69, 70, 71, Baker, James A. 149
77, 86, 90, 93, 107, 110, 155 Balkans 13, 16, 120, 126, 174–89
al-Sahhaf, Mohammed Saeed 44, 327 Ballout, Jihad 355
366
INDEX
Bangladesh 214 Bush (Snr), George H. 15, 27, 33, 63, 72,
Barr, Cameron 66 142, 149, 150, 151, 153, 184, 252
Basque separatism 210 Bush, George W. 34, 37, 43, 47, 60, 61, 62,
Basra 7, 8, 105, 123, 225, 300, 326, 353 63, 64, 65, 69, 71, 73, 74, 138, 139, 145,
BBC World 10, 235, 304, 354 148, 150, 151, 156, 157, 162, 174, 180,
Beck, Glenn 334, 335, 342 186, 250, 252, 253, 254, 267, 274, 276,
Belgium 27, 141, 152, 168 277, 278, 287, 301, 325, 327, 333, 334,
Bell, Martin 11–12, 159, 182, 184 335, 340, 361, 363
Benn, Tony 277 Bush administration (George W.) 8, 14, 15,
Benton, Ross 54 60, 64, 67, 69, 71, 73, 106, 136, 137,
Berlin Wall 63, 129, 181 139, 140, 141, 142, 144, 145, 146, 147,
Berlusconi, Silvio 333 151, 152, 153, 156, 158, 62, 174, 187,
Bernstein, Carl 38 247, 259, 278, 358
bias (accusations of) 9, 68, 102, 228, 231,
235, 251, 284, 286, 302, 309, 311, 312, Caldicott, Helen 36
330, 331, 352, 358 Cambodia 34, 125, 216
bin Laden, Osama 48, 64, 80, 81, 83, 84, Cameron, James 224
85, 86, 87, 88, 91, 93, 94, 107, 156, 352 Campaign for Peace and Democracy 103
biological weapons (threat of) 8, 28, 72, Campbell, Alistair 105, 299
146, 153, 196, 286 Canadian Broadcasting Corporation
Bishop, Patrick 184, 191 (CBC) 4, 20
Black Hawk Down 156 Capa, Robert 124, 127
Blair, Tony 46, 47, 54, 55, 105, 106, 107, Carter, Jimmy 33, 35, 69
110, 178, 180, 184, 185, 267, 271, 274, cartoons 55, 185, 227, 228
225, 276, 277, 278, 279, 281, 287, 293 Castro, Luis 51
Bleifuss, Joel 141, 152 Catalonia 206, 210
Blitzer, Wolf 85, 86, 94 CBS News 7, 85, 86, 88, 103, 143, 144,
Blix, Hans (report of) 18, 267, 270, 271, 152
272, 274, 275, 276, 278, 281 celebrity 14, 51, 78, 80, 82, 86, 90, 91, 93,
blog; see weblog 183, 277, 330
Bodi, Faisal 354 censorship 8, 9, 13, 30, 50, 98, 120, 137,
Boer War 25, 236, 237 145, 152, 190, 204, 224, 238, 247, 299,
Borger, Julian 53, 282 317, 318, 327, 329, 352, 356
Bosnia 35, 72, 126, 175, 181–6, 191 Center for Social Media 19, 336, 338
Boston Globe 117, 129 Central Intelligence Agency (CIA) 31, 32,
Boston Herald 197 33, 38, 39, 46, 52, 144, 153, 177, 352,
Bourke-White, Margaret 119 360
Bremer, Paul 328–9 chemical weapons (threat of) 8, 28, 29, 48,
Briganti, Irena 9 52, 53, 69, 72, 144, 145, 146, 149, 153,
British Broadcasting Corporation (BBC) 3, 196, 199, 204, 286
8, 10, 11, 12, 18, 20, 34, 47, 159, 176, Chile 32
177, 178, 179, 182, 229, 233, 235, 260, China 37, 59, 126, 212, 214, 216, 279
264, 283–300, 303, 304, 308, 316, 318, Chirac, Jacques 270, 275
353, 357, 362, 363 Chomsky, Noam 25, 34, 38, 39, 53, 97, 98,
Brokaw, Tom 120, 144; see also NBC News 109, 143
Brown, Tina 8–9, 70 Christian Science Monitor 36, 66
Browne, Malcolm 119, 127 citizenship 5, 20, 28, 136, 207, 259, 263,
Bull Durham 335 336, 343, 345
Buoen, Roger 68 Civil War (US) 25, 119
Burkeman, Oliver 6 Clark, David 107, 110
Burns, John 3, 70, 71 Clarke, Wesley 176
Burrows, Larry 119 “clash of civilizations” 56, 181, 272
Burundi 120, 126 Clear Channel 230, 335, 336
367
INDEX
Clinton, Bill 63, 69, 72, 105, 158, 160, embedded reporting 5, 6, 9, 13, 16, 17, 18,
182, 184, 342 19, 28, 30, 31, 49–50, 70–1, 77, 91, 105,
CNN 6–12, 13, 16, 65, 70, 80, 81, 83, 84, 136, 166, 167, 190, 191, 193, 194–200,
85, 86, 91, 99, 100, 102, 104, 106, 108, 203, 204, 206, 226, 232, 249, 260, 261,
136, 148, 152, 153, 157–63, 164, 166, 262, 283, 285, 286, 287, 289, 294–6,
182, 199, 233, 235, 259, 260, 299, 302, 297, 298, 299, 307, 311, 323, 324, 349,
303, 304, 308, 309, 310, 312, 318, 320, 353, 354, 361
321, 337, 353, 359, 363 Ethiopia 63, 165, 210
Cold War 15, 16, 32, 36, 45, 63, 72, 74, 69, “ethnic cleansing” 16, 34, 177, 178, 179
97, 98, 99, 106, 107, 108, 109, 157, 158, Express and Star (Wolverhampton) 194
159, 175, 180, 181, 184, 283, 313
Colombia 33 Fairness and Accuracy in Reporting
“compassion fatigue” 161 (FAIR) 143, 145, 152
Connell, Joan 359 Falklands/Malvinas conflict 16, 28, 30,
Conservative Party (UK) 159, 279 104, 120, 122, 152, 190, 191–4, 199,
Cook, Robin 177, 184, 291 201, 202, 203, 204, 225, 347
Couso, Jose 51 Federal Bureau of Investigation (FBI) 32,
Croatia 180, 181, 183, 306, 307, 308 363
C-Span (US) 148 Federal Communications Commission
Cuba 36, 303; see also Guantanamo Bay (FCC) 334
Czech Republic 306, 308 Finkelstein, Daniel 270
Czechoslovakia 126, 273 Finland 36, 152, 306, 307, 308
Firdous Square 43, 101, 117, 298, 325, 349
Daily Express 48, 51, 193, 194 First Blood 251
Daily Mail 192, 269, 271, 272, 275, 281, Fisk, Robert 51, 117, 211, 262, 272, 273,
282 274, 282, 348, 349, 363, 364
Daily Star 52, 194 flak (criticisms of reporting) 38, 97, 98, 99,
Daily Telegraph 44, 52, 93, 184, 185, 232, 102, 109
233, 237, 269, 270, 271, 272, 274, 280, Fletcher, Kim 51
281, 282, 299, 362 Foot, Paul 51
Democratic Party (US) 34, 106, 146, 148, Fouda, Yosri 354
342 Fox News 6, 7, 8, 9, 10, 11, 50, 93, 235,
Democratic Republic of Congo 27, 116, 259, 260, 309, 320, 321, 350, 363
155, 164, 168, 303 framing 13, 17, 18, 29, 31, 52, 53, 55, 63,
Denmark 306, 307 64, 72–4, 77, 85, 88, 89, 97, 99, 104,
Detroit Free Press 127 105, 106, 107, 117, 126, 130, 136, 139,
digital technology 119, 224, 233, 334, 340, 143, 144, 152, 161, 163, 183, 219,
343, 350, 359 247–65, 268, 269, 272, 273, 274,
diplomacy 16, 29, 32, 46, 63, 71, 72, 106, 277–80, 301
138, 141, 146, 149, 151, 162, 175, 176, France 176, 260, 263, 270, 274, 275, 279,
177, 180, 182, 226, 263, 266, 268, 273, 301, 302, 304, 305, 307, 316, 330, 355
274, 276, 277, 316, 317 Freedland, Jonathan 180, 278, 279, 282
disinformation 9, 30, 47, 48, 70, 136, freelance journalism 336, 360, 364
137–41, 143, 144, 147, 310, 360 Friedman, Tom 73
Dixie Chicks 335–6 “friendly fire” 13, 44, 54, 146, 204
Dodge, Toby 362
Dyke, Greg 10; see also BBC Gellhorn, Martha 20
genocide 16, 61, 69, 72, 155, 157, 160, 161,
East Timor 116, 126, 206, 214, 163, 164, 165, 166, 168, 169, 179, 183
Economist, The (UK) 126, 322 Germany 12, 119, 124, 127, 144, 152, 176,
Egypt 11, 93, 94, 307, 317, 329, 330 178, 179, 181, 185, 187, 199, 238, 260,
email 19, 48, 50, 238, 259, 262, 337, 339, 263, 270, 272, 273, 277, 279, 301, 305,
342, 350, 360, 361, 363; see also Internet 306, 307, 308, 316, 317
368
INDEX
369
INDEX
370
INDEX
NBC News 7, 103, 143, 144, 152, 196, 233, 88, 89, 92, 93, 126, 127, 260, 316, 317,
238, 261, 364; see also MSNBC 322, 327
Netherlands 306 Panama 28, 30, 33, 36, 45, 152
New York Daily News 339 “parachute journalism” 165
New York Stock Exchange 102, 326, 355 Parry, Gareth 192
The New York Times 3, 7, 62, 65, 67, 69, 70, patriotism (impact of on news coverage) 4,
73, 80, 85, 86, 88, 90, 92, 93, 120, 121, 5, 7, 8, 9, 10, 13, 17, 18, 26, 28, 50, 51,
123, 126, 129, 142, 152, 158, 174, 186, 55, 68, 74, 78, 85, 91, 92, 97, 98, 115,
187, 219 125, 144, 145, 149, 192, 201, 249, 251,
New Zealand 236, 238 253, 254, 255, 256, 258, 259, 284, 358
news agencies 18, 286, 301–14 PBS (US) 139, 150, 233
news photographs 15, 44, 54, 55, 88, 89, Pearl, Daniel 88, 92, 196
115–35, 140, 142, 156, 191, 196, 197, Pearl Harbor 35, 64, 68, 127; see also World
204, 225, 234, 235, 236, 261, 287, 307, War II
338, 357, 359 Pentagon 6, 15, 29, 30, 331, 35, 36, 53, 65,
news values 16, 85, 193, 202, 203, 207, 70, 71, 136, 137, 139, 140, 141, 143,
208, 303, 354 144, 145, 146, 151, 153, 180, 191, 195,
Newsweek 88, 90, 91, 94, 126, 130, 140, 196, 203, 233, 234, 261, 263, 283, 286,
152, 358, 359 287, 288, 298, 299, 310, 325, 351, 353,
Nicaragua 33, 63, 204 358
Nightline (ABC) 261 Persian Gulf War (1991) 6, 15, 17, 25, 28,
9/11 14, 60, 64, 65, 66, 68, 73, 74, 155, 174, 30, 35, 36, 37, 43, 44, 47, 49, 53, 54, 55,
175, 234, 248, 250, 259, 358; see also 97, 104, 105, 119, 120, 125, 130,
September 11, 2001; World Trade 136–54, 157, 191, 224, 225, 247–65,
Center 273, 279, 287, 288, 302, 318, 340, 349
Nixon, Richard 34, 108, 251 “personal journalism” 357
Normandy 348 Pew Research Center 350, 351, 363
Norris, David 192 Philadelphia Inquirer 121, 123, 127
North Korea 26, 45, 66, 200; see also Philippines 36
Korean War Phillips, Melanie 226, 272, 282
Northern Ireland 206, 210 photojournalism: see news photographs
Norway 306 Pilger, John 31, 32, 34, 39, 44, 46, 51, 228,
nuclear technology 26, 28, 35, 36, 47, 48, 231
55, 64, 72, 149, 196, 275 Platoon 261
Nunberg, Geoff 67 Platt, Spencer 197
Poland 273
O’Sullivan, John L. 61, 62, 74 polls 47, 108, 143, 146, 148, 150, 153, 226,
objectivity 6, 7, 13, 16, 19, 49, 92, 159, 191, 266, 267, 274, 278, 279, 284, 298, 299,
193, 194, 201–4, 221, 255, 315–27, 300, 363; see also public opinion
332 pool system 28, 30, 49, 104, 136, 137, 142,
Observer, see The Observer 145, 148, 152, 190, 225, 232, 285, 286
Official Secrets Act 46 Portugal 306, 307
Omaar, Rageh 11 Powell, Colin 48, 174, 249, 251, 327, 340,
Omar, Mullah 85, 90 353
Oneworld 333, 337, 340, 343, 344 Powell, Michael 334
Open Democracy 337, 344 press briefings 30, 101, 104, 105, 141, 162,
Orwell, George 49, 53 191, 227, 233, 276, 288, 289, 290, 294,
ownership of news organizations 17, 38, 98, 310, 323, 324, 353, 355
108, 211, 228, 229, 231, 304, 333, 334, Preston, Peter 6
335, 364 Prime Time Live (ABC) 140
prisoners of war (POWs) 116, 145, 146,
Pakistan 93, 196, 210, 212, 213, 219 147, 148, 251, 287, 293, 326, 327
Palestinians 64, 65, 66, 68, 69, 77, 79, 84, professionalism 15, 16, 17, 18, 26, 30, 78,
371
INDEX
81, 82, 83, 84, 85, 86, 87, 91, 118, 121, Robertson, George 177, 178
159, 165, 190, 192, 193, 200, 201, 202, Rokke, Doug 35
203, 206, 208, 209, 221, 225, 231, 235, rolling news 6–12, 13, 165, 283, 287; see
260, 261, 302, 309, 311, 312, 319, 341, also 24-hour news and individual
345, 348, 357 networks
propaganda 8, 13, 16, 25, 31, 34, 35, 38–9, Romania 306
50, 52, 53, 54, 82, 85, 102, 136, 137, Rosenthal, A.M. 67
139, 141, 142, 143, 144–6, 147, 148, Rubin, James 176, 178, 180
160, 165, 174, 176, 179, 180, 206, 231, Rumsfeld, Donald 29, 48, 66, 310, 354
249, 262, 287, 293, 294, 310, 325, 350, Rushdie, Salman 67
355, 356, 361; see also spin Russia 25, 34, 35, 67, 126, 279, 301, 302,
Protsyuk, Tara 51 306, 307, 308
psychological warfare 30, 44, 190, 206, 225 Rutenberg, Jim 7, 261
public opinion 6, 15, 18, 19, 96, 99, 103, Rwanda 16, 27, 69, 72, 120, 126, 155–7,
137, 143, 148, 150, 151, 153, 180, 190, 160–9
207, 252, 257, 259, 268, 271, 280, 284,
301, 305, 316, 322, 333, 334, 341, 342; Said, Edward 56, 143
see also polls St Petersburg Times 140
public relations (PR) 6, 18, 97, 104, 136, Salam Pax 19, 339, 347, 348, 360–2
141, 174, 225, 231, 247, 266, 267, 273 sanitization of reality 28, 104, 144, 231,
public sphere 18, 100, 101, 102, 103, 131, 260, 347, 353, 358
207, 259, 262, 263, 267, 268, 269, 280, satellite technology 11, 19, 55, 81, 96, 100,
320 102, 136, 140, 141, 148, 151, 160, 232,
Purdum, Todd 66, 69 233, 261, 302, 315, 318, 326, 329, 330,
334, 340, 352, 359, 360
Qatar 50, 105 204, 227, 260, 285, 288, 294, Saudi Arabia 93, 137, 138, 139, 140, 141,
295, 296, 302, 307, 308, 310, 315, 318, 143, 144, 145, 147, 148, 152, 251, 255,
319, 323, 324, 325, 330, 352, 353, 356 257, 318, 326, 352
Saving Private Ryan 55
radio 100, 137, 151, 165, 226 230–1, 233, Schechter, Danny 30, 31, 356
235, 283, 316, 317, 318, 330, 333, 334, Schmemann, Serge 65
335, 340–1, 342, 348 Schroeder, Gerhard 270
Radio Free Europe 316 Schwarzkopf, Norman 141, 142, 150, 249,
Radio Sawa 316, 330 279
radioactivity 35, 36 Seamark, Mick 192
Rambo 250 Security Council 18, 27, 36, 191, 269, 270,
Rather, Dan 120, 144; see also CBS News 271, 275, 277; see also United Nations
Reagan, Ronald 33, 39, 47, 140, 142, 149, September 11, 2001 14, 30, 37, 45, 48, 61,
151, 153, 184, 251 63, 65, 66, 67, 69, 73, 77, 78, 80, 81, 83,
Red Brigades 79 85, 86, 91, 92, 93, 106, 120, 124, 127,
Red Cross 162, 174, 178, 183, 291 140, 155, 207, 208, 211, 283, 302, 316,
Reeves, Cris 44 317, 320, 321, 322, 351, 352, 253, 357,
refugees 123, 129, 142, 155, 161, 164, 166, 358; see also 9/11; Twin Towers; World
177, 178, 179, 184 Trade Center
Republican Party 9, 27, 34, 148, 174, 266, Serbia 34, 45, 51, 53, 106, 168, 176, 177,
334 178, 179, 180, 181, 183, 340
Reuters 18, 51, 67, 68, 128, 162, 304, 306, Shnev, Ismail Abu 88
307, 308, 309, 312 Short, Claire 277
Reynolds, Glenn Harlan 358, 359 Sierra Leone 155, 164, 169
Rice, Condoleezza 69, 85, 352 Silva, Victor 51
Ridge, Tom 71–2 Simpson, John 8, 9, 183; see also BBC
Ritter, Scott 29 Sirota, David J. 7
Robbins, Tim 335 Sites, Kevin 359, 364
372
INDEX
60 Minutes 80, 200, 229; see also CBS News The Times (UK) 50, 54, 143, 147, 269, 270,
Sky News 7, 8, 10, 12, 18, 44, 81, 109, 229, 274, 280, 282, 362
233, 235, 284, 287, 288, 289, 290, 291, The Times of India 210, 212, 217, 219
292, 293, 294, 297, 298 think tanks 31, 37, 207, 328, 358
Slovenia 180, 181, 306, 307, 308 Tianenmen Square 126
Smith, W. Eugene 127, 128 Time magazine 126, 129, 152
Snow, Jon 11 torture 142, 145, 146, 287, 293, 294
Solana, Javier 177 trauma 77, 86, 259, 362
Somalia 16, 45, 72, 106, 108, 116, 125, Tucker, Joanne 356
129, 130, 155, 156, 157, 158, 160, 161, Turkey 220, 236, 306, 307, 219
162, 163, 164, 184, 191 20/20 (ABC) 142
Sorenson, Erik 7, 93 24-hour news 6–12, 96, 99–101, 104, 108,
Soviet Union 31, 35, 38, 45, 63, 126, 129, 119, 158, 229, 260, 283, 302, 349, 352;
140, 144, 181, 211, 305, 313 see also rolling news; individual networks
Sowcroft, Brent 200 Twin Towers; see World Trade Center
Spain 51, 124, 210, 305, 306, 307 Tyler, Patrick 138, 139
Spanish-American War 36, 119, 122
Spanish Civil War 25, 124 Uganda 27, 161
spectacle (war as) 14, 15, 43, 45, 47, 49, 55, Umm Qasr 8, 105, 194, 300, 324
90, 136, 149, 150, 161, 186, 348 unilateral reporting 13, 50, 195, 204, 233,
“spin” 8, 14, 60, 71, 98, 104, 158, 198, 225, 262, 285–8, 295, 296, 298; see also
233, 350; see also public relations embedded reporting
Stalin, Joseph 117 United Nations (UN) 28, 29, 52, 60, 66,
stereotypes 63, 165, 185 88, 96, 103, 142, 143, 156, 159, 160,
Stockholm Syndrome 195 161, 169, 175, 177, 178, 179, 183, 187,
Stop the War Coalition 103, 280 211, 266, 270, 271, 273, 275, 276, 277,
Straw, Jack 12, 272 278, 284, 301, 303, 321, 340; see also
Sudan 116, 155, 161, 162 Security Council
Sun (UK) 48, 52, 53, 54, 55, 179, 185, 269, United Press International (UPI) 71, 196
273, 274, 275, 276, 278, 280, 282, 299, US Central Command 50, 105, 227, 288,
362 323, 324, 325, 353, 354
Sunday Times (UK) 177, 179, 184 US News and World Report 152
Swain, Jon 11 USA Patriot Act 340
Switzerland 36, 306 USA Today 9, 71, 152
Sydney Morning Herald 227, 228, 231 USS Abraham Lincoln 71, 250, 261, 263,
Syria 35, 45, 247, 307 333
Uzbekistan 73
tabloidization 228, 230, 234–5
Taliban 33, 51, 65, 73, 93, 107, 211; see also Viacom 333; see also ownership
Afghanistan videophone 233, 359; see also satellite
Tanner, Marcus 185 technology
Tanzania 164 Vietnam 25, 31, 34, 47, 50, 53, 63, 72, 73,
Tears of the Sun 166 79, 97, 98, 107, 108, 118, 119, 121, 122,
The Australian 31, 226, 228 125, 125, 126, 127, 129, 143, 150, 157,
The Chicago Tribune 123, 124 159, 160, 190, 195, 224, 232, 250, 251,
The Guardian 9, 39, 46, 50, 51, 53, 54, 55, 256, 348
69, 70, 90, 100, 174, 183, 184, 192, 208, Village Voice 151
262, 267, 269, 273, 274, 275, 276, 278, Vivendi 333; see also ownership
280, 282, 284, 299, 335, 362 Voice of America 316
The Hindu 212
The Nation 151 Waas, Murray 149, 153
The Observer 46, 53, 54, 120, 191 Walker, William 176
The Statesman 212 Wall Street Journal 152, 196
373
INDEX
Wallace, Mike 200; see also CBS News Wilson, Woodrow 36, 67, 106
Walt Disney Co. 304, 333; see also wire services 39, 68, 138; see also news
ownership agencies
Walton, Jim 9 Wolff, Michael 50
“War beyond the box” 337–45 Woodruff, Judy 139
“war on drugs” 25, 32, 33, 250 Woodward, Bob 139, 140–1
“war on terrorism” 14, 15, 16, 32, 45, 49, World Trade Center 64, 65, 68, 77, 82,
59–74, 107, 156, 174, 175, 186, 187, 127, 156, 250, 351; see also 9/11;
248, 249, 250, 259, 260, 352, 358 September 11, 2001
warblogs 19, 339, 348, 358–61, 362; see also World Trade Organization 333
weblogs World War I 12, 25, 50, 122, 124, 141, 224,
Washington Diplomat 322 236, 320
Washington Post 68, 71, 83, 93, 97, 110, World War II 12, 25, 27, 34, 38, 43, 47, 55,
138, 139, 152, 279 63, 64, 66, 119, 124, 127, 190, 199, 214,
Washington Times 196, 197, 202 224, 237, 270, 272, 317, 347, 348; see
weapons of mass destruction (WMD) 8, 9, also Holocaust
18, 29, 43, 48, 49, 64, 70, 71, 72, 83, 96, Wright, Dean 348
104, 105, 107, 174, 180, 226, 269, 270, Wright, Robin 74, 322
273, 284, 286, 289. 290, 291, 292, 294,
295, 296, 297, 298, 300, 304, 341 Yahoo.com 356, 363
weblogs 19, 151, 238, 259, 339, 340, 341, Year Zero 262
342, 344, 347, 348, 349, 350, 356–62, Yugoslavia 16, 25, 35, 161, 175, 187
363, 364; see also Internet; warblogs
Wells, Matt 11, 353 Z Magazine 151
Williams, Brian 262 Zaire; see Democratic Republic of Congo
Wilson, Joseph 138, 139 Zelnick, Bob 139
374