Press Nationalism and
The New York Times Coverage
of the Bosnian War
W. Matt Meyer
631 C Old Humboldt Rd.
Jackson, TN 38305
(901) 664-7211 or (901) 425-9615
[log in to unmask]
Dr. Daniel Riffe
E. W. Scripps School of Journalism
Press Nationalism and
The New York Times Coverage
of The Bosnian War
This study uses content analysis to examine the New York Times by looking
at how changes in its coverage of the Bosnian War may have been linked to
changing American foreign policy. This study also uses extramedia sources to
examine links between the press, the policymakers and the public. The study
found a marked diffference between early and later coverage, with Times coverage
focusing more on American-centered story topics and actors later in the war.
Press Nationalism and the New York Times . . .
Press Nationalism and
The New York Times Coverage
of the Bosnian War
In his 1993 book, Witness to Genocide, a collection of first-hand accounts
of the death camps and atrocities in Bosnia-Herzegovina, Newsday foreign
correspondent Roy Gutman criticized both mainstream media and Western
politicians for inaction at the beginning of the Bosnian War. Gutman marveled
that he was first to report on the genocide being committed by Serbian
nationals: "Since when in the age of spy satellites does a reporter come up with
such a scoop?," Gutman asked in the book. His answer? Western officials wanted
to play down the tragedy to avoid entanglement in a military mess. For example,
in July 1992, President George Bush dismissed the Balkan situation as a "hiccup"
in world importance. In truth, though, the same month saw the beginning of
"ethnic cleansing" as Serbs separated Muslim men from their families and shipped
train cars full of these "war criminals" out of land the Serbs wanted.
Did the press follow the government's lead by failing to attend to the war?
This study examines New York Times coverage of the Bosnian War by looking at how
changes in coverage may have been linked to changing American foreign policy.
Previous studies have often claimed to show that American media report or
"frame" international news in terms of U.S. interests. Rather than acting as
independent adversary, watchdog or public surrogate, researchers say, the press
"follows the flag" when covering certain international events.
But in addition to the content analysis of media coverage, this study
attempts to examine the press-policy-public triangle by measuring policymaker
attention to the war and an index of public concern.
One of the first studies linking American foreign policy to this "press
nationalism" was Kriesberg's analysis of 1918-1946 New York Times coverage of
Soviet news. He found that increases in unfavorable editorials were related to
increases in unfavorable news stories about Soviet events reported in terms of
In 1964, Lynch and Effendi studied editorial treatment of India in the New
York Times, concluding that improved India-West relations were paralleled by
improved editorial treatment. Sahin's 1973 study of Turkish-American relations
also determined that Times assessments of Turkish policy changed as policy
changed from favorable to unfavorable.
Yu and Riffe explored how Mao Tse-tung and Chiang Kai-shek were depicted in
three U.S. news magazines. Their analysis suggested that, over time, the
magazines increasingly portrayed Mao as a noble statesman and actor on the
global stage, and portrayed Chiang as an "exiled" leader of an outcast nation D
depictions consistent with U.S. policy.
Ronald Reagan's China policy was the focus of Chang's 1984 analysis of
three elite American papers before and after the 1980 presidential election.
Reagan's pre-election, pro-Tawain China policy was treated unfavorably, but
after the election, his new-found, anti-Tawain policy was treated favorably.
This reversal is consistent with the long-standing policy of non-recognition of
While the studies reviewed above are representative explorations of
coverage-policy linkages, some recent studies in this area have compared
coverage to external or extramedia data on policy. For example, Ramaprasad and
Riffe used U.S. State Department Bulletins and 1973-1980 congressional
publications D "official" documents D to operationalize shifts in U.S. policy
toward India. An analysis across the four periods of relations D as defined by
State Department records D showed that the coverage did not follow policy
directly. However, when the first three periods were combined and compared to
the last one, coverage did mirror policy and become more positive.
Chang also used extramedia data in two studies of U.S.-China policy. The
first compared 1950-1984 New York Times and Washington Post references to China
and Tawain with references in The Public Papers of the Presidents of the United
States and The Weekly Compilation of Presidential Documents. As relations with
China improved over the 34-year time period, the use of condensational symbols
(terms such as "Communist China," "Red China," or "Nationalist China" that
carried ideological overtones) in reference to China in both government
documents and newspapers decreased and use of neutral symbols increased.
Using the same time period and sources, Chang later analyzed how
presidential statements affected news reports of Chinese-American relations from
1950 to 1984. He found that the more attention the government gave U.S.-China
policy, the more press coverage the policy received. When official relations
were favorable, the press coverage was favorable.
This study's first research objective was to describe coverage of the
Bosnian War in the New York Times. We examined a 45-month sample of Times
coverage, by looking at different content variables and trends in coverage over
While this longitudinal analysis is eminently suited for tracking gradual
changes in coverage, our second objective was to compare Times coverage
variables before and after what arguably represented a major American foreign
policy change in Bosnia. That "break point" in the data was the bombing in the
Sarajevo marketplace in February of 1994.
The third research objective was to compare Times coverage of the Bosnian
War to several extramedia indicators of government policy. We anticipated that
Times attention to the war in Bosnia would parallel government attention to the
war. (This study thus differed from others that focused on positive or negative
press coverage or on good or bad foreign relations.) Finally, we examined the
public side of the press-policy-public triangle by comparing Times coverage and
measures of government attention with public attention to the war.
Coverage of the Bosnian War in The New York Times from April 4, 1992, to
December 31, 1995, was analyzed. April 1992 marked the beginning of the siege on
Sarajevo, and December 1995 was when peacekeeping troops were sent to aid the
peace process in Bosnia.
The Times provides a fairly comprehensive D but admittedly unrepresentative
D example of press treatment of U.S. policy. It is influential among world media
and is acknowledged as a leading paper.
Using a random starting point, every fourth story of the 3,652 listed under
the topic "Bosnia" in the Times CD-ROM index was selected. Editorials, sports
stories and stories from the "Week in Review" section were omitted. Each story
was coded for date, length, page placement, whether it was staff or wire, where
it was filed from, topic, location of event (e.g. Washington or Bosnia), primary
actors and whether it was issue-oriented or event-oriented.
An actor (either a country or an international organization) was chosen as
a primary one if it was involved in the main conflict of the story. In other
words, actors were primary if the story could not have happened without them.
Topic was determined on the basis of headline, subheads and the first seven
to eight paragraphs of the story. Five topic categories were used:
y Military actions. Fighting, taking prisoners, cease fire
announcements, bombings, sniper fire.
y War's effects. Refugees, relief efforts, atrocities,
displacement of citizens, war crimes, ethnic cleansing, land mine
removal, U.S. troops coming to Bosnia to enforce peace, indigenous
people dealing with the Dayton treaty after it was signed, warring
parties moving from the front after Dayton treaty.
y Diplomacy. Embargoes, sanctions, World Court hearings,
international debate over roles in Bosnia; policy stories involving two
or more countries.
y Policy and politics. Policy actions in only one country were
placed here. Voting, power struggles among Yugoslavs, uprisings,
turmoils, governments discussing their role in Bosnia, how Bosnian
situation affects politics in the United States.
After training, coders reached agreement on 87% of the ten percent sample
(90 stories) of the expected population of 900 stories.
Presidential attention to the Bosnian War was measured by counting the
number of non-repeating index entries under the headings "Bosnia-Herzogovinia,"
"Serbia," or "Former Yugoslavia" in the index of The Papers of the Presidents
of the United States and The Weekly Compilation of Presidential Documents.
A second measure of government attention was operationalized as the
percentage of Bosnian mentions in State Department press briefings. "Foreign
Policy on CD-ROM" was used. The number of briefings in which Bosnia was
discussed per month were tallied and divided by the total number of press
briefings in a month, yielding a relative percentage of each month's briefings
that referred to Bosnia.
Finally, measures of issue salience for the public used results for the
Harris Poll's commonly asked question, "What are the two most important issues
facing the government today"; and the monthly news interest index gathered by
the Pew Research Center (formerly Times Mirror).
Findings and Discussion
This sample of Times coverage of the Bosnian War from April 1992 to
December 1995 yielded 862 articles. The average article was 881 words long and
27.6 % of all stories appeared on page one (See Table 1).
Table 1 shows a plurality of stories were filed from Bosnia (35.4 %) and
that a plurality of the events covered in the stories (45 %) happened in Bosnia.
This 10 % discrepancy may reflect the "listening post" journalism that occurs
when reporters cannot get to the scene of an event and file from another site.
One in five stories was about Bosnia-related events in Washington, D.C.
Diplomacy (36.8%) and policy (16.7 %) stories made up over half of all
stories about the Bosnian War in the Times. Were these the most important
aspects of the war, or the easier to report, courtesy of available sources?
Reporters and government sources have often formed close, symbiotic
relationships. The emphasis in the Times coverage on diplomacy may have been
a function of the close, working relationship between Times reporters and the
government sources, as well as the number of easily-accessible, press-friendly
Arguably, stories about war's effects could have occupied a larger portion
of the Times coverage. Some critics say more stories about the war's effects
would have prompted quicker Western action and a speedier end to the war.
Table 2 shows the dominance of particular actors, by topic. Bosnian Muslims
(40.5 %), Bosnian Serbs (38.1 %), the United States (34.5 %) and the United
Nations (29.8 %) appeared as primary actors most often. (The percentages, using
the total number of stories for each topic as a base, represent the percentage
of times an actor was coded as primary in a topic.) Comparatively short shrift
was given to some of the indigenous parties, with Serbians (12.6 %) and
Croatians (8.5 %) appearing less often than the Bosnian peoples or the Western
powers. This indicates the relatively minor role that the ideas and people of
Serbia and Croatia played in Times war coverage, despite the fact that Serbia's
desire to form a "Greater Serbia," as well as Serbia's war and subsequent treaty
with Croatia, spurred the events which led up to the siege of Sarajevo and the
fighting in Bosnia.
Bosnian Muslims (53.5 % of war's effects and 68.5 % of military action
stories) and the Bosnian Serbs (37.8 % of war's effects and 83.7 % of military
action) appeared mostly in fighting-oriented coverage. The fact that they were
infrequently in stories about diplomacy or politics (Muslims = 28.4 % in
diplomacy and 18.1 % in politics; Serbs = 26.8 % in diplomacy and 14.6 % in
politics) suggests their primary role in Times coverage as combatants, rather
than diplomats at the bargaining table. The data may reflect limited
correspondent access to the inner workings of Bosnian Muslim or Bosnian Serb
politics during the war.
By contrast, the U.S., the U.N. and Europe appeared in a high percentage of
the diplomacy stories D the United States in 44.8 %, the U.N. in 43.5 % and
Europeans in 29.7 %. This finding may indicate a Times coverage focus on the
West's diplomatic role. It also suggests that Western sources were more
accessible and more willing to talk to Times reporters about diplomacy than
Similarly, Table 2 shows that the United States appeared as a primary actor
in a majority of politics and policy stories (appearing in 58.3 %); recall that
the study included Bosnia stories originating in Washington. Reliance on
easily-accessed official sources in Washington and in U.S. diplomatic circles
served to limit the range of actors in these stories.
Table 3 examines the stability of content variables across the 43-month
period. The trend scores (rho) suggest that only three variables showed any
consistent, significant change over time. Issue-oriented stories, stories in
which NATO appeared as a primary actor and stories in which Croatians appeared
as primary actors all showed significant increasing trends in Times coverage.
The trend towards more issue-oriented stories (rho = .383, p = .009) may
indicate Times efforts to educate readers by providing more coverage of the
"issues behind the events." Perhaps, however, this increasing trend suggests
that the war had to develop its own "history" before these kinds of analytical
stories could emerge.
While trend analysis is useful for examining the gradual or overall
increase or decrease, coverage can also change abruptly, often in response to a
foreign policy change.
Such a policy shift followed the February 6, 1994, bombing of the Sarajevo
marketplace by Bosnian Serbs. This tragedy, which left 68 civilians dead, was
noted by critics as a turning point which clearly solidified Western policy in
Bosnia. After the bombing, intense NATO strikes on Bosnian Serbs began, the
West increasingly ignored the arms embargo against the Muslims and Western
diplomatic pressure intensified.
To examine the "effects" the bombing and the policy shift had on the
coverage, the content data were divided into two time periods D before (the
22-months from April 1992 to January 1994) and after the bombing (the 23 months
from February 1994 to December 1995). Table 4 shows the frequency of the content
variables in the two periods.
The results indicate a marked change in coverage. Following the bombing,
the increase in the prevalence of diplomacy stories (37.7 % before, 39.3 %
after) and military action stories (17.8 % before, 25.1 % after), and the
increased primary actor role of the United States (29.7 % before, 38.8 % after)
and NATO (4.6 % before, 11.8 % after) suggests that the bombing and subsequent
policy change led the Times to focus its coverage more on U.S.-related topics
This increased focus on American-centered topics and actors may have also
served to "crowd out" non-American topics (war's effects decreased from 26.7% in
the first period to 18.6% in the second) and actors: the Serbians (17.6% to 8%),
the Bosnian Croatians (5.8% to 2%), and the Bosnian Muslims (43.7% to 37.5%).
In short, the post-bombing focus on U.S. involvement occurred at the
expense of attention to other topics and actors. This is hardly surprising D the
Times is an American newspaper D but it did impact how the Bosnian story would
The study also examined how Times attention to the war related to
government and public attention to the war. Monthly frequencies of Times stories
were correlated (Pearson's r) with monthly measures of government and public
The correlation in Table 5 between monthly measures of Presidential
attention to Bosnia and frequencies of Times stories (r = .7601, p = .0001)
indicates a strong relationship. Times attention to the war was also highly
correlated with State Department attention to the war as operationalized by the
press briefing percentages (Pearson's r = .5273, p =.002) and to the Pew public
attention measures (r = .5499, p = .02).
Times attention to the war mirrored government attention to the war. The
coefficients do not indicate any directional relationships D who set whose
agenda D but they do suggest that the press and the government came to agree on
the importance of, or how much attention to pay to, the "hiccup" of the Bosnian
This study has limitations. First, the content analysis used only five
crude story topics. Second, generalizability suffers because this study examined
only the Times. However, the Times influences the American elite D government
leaders, business leaders and intellectuals D who contribute to foreign
policy and diplomatic decisions. How the Times covers an event is arguably
important to these leaders and shapes their view of the world.
Another potential limitation involved the operationalization of
presidential and State Department attention to the war. The measure of each was
dependent on the accuracy of the index of the collected public briefings and
speeches and whether mentions were initiated by the press or not. Information
was lost because the indexes show frequency of mention and not the length,
substance or importance of the mention. Still, this study asserts that these
references represent a measure of government attention worth further
More critical as a potential limitation is this study's conceptualization
of "press nationalism" in terms of attention level rather than overtly biased
content. Determination of press bias favoring "official viewpoints," however,
has been rare in press nationalism studies, in terms of overt labeling and
negative assertions. However, some studies have suggested that increased
press attention, due to greater American involvement in an issue or to some
other change in policy, may affect coverage. That is, as press attention to a
country increases, the extra attention results in an increased variety of story
topics, which D because reporters are often drawn to "bad news" D may yield more
negative portrayals. Press nationalism may not be evident in editorial
"slant" D language which is overtly pro or con D so much as in topics covered.
What this study asserts, however, is that press nationalism may also be
viewed simply as the increased emphasis in the Times coverage on American
involvement in the Bosnian war D to the detriment of the coverage of
non-American topics and actors D following a change of U.S. policy in Bosnia.
In this study, a shift in foreign policy affected portrayal of other people
and topics D simply because attention to U.S. involvement increased following
the February 1994 bombing. Did atrocities and war crimes D the subjects of war's
effects stories D stop occurring following this bombing? Of course not. This
study suggests that the shift towards more diplomacy and military action stories
was because these topics were clearly the ones which concerned the American
officials the most. This study suggests that after the bombing, the Bosnian War
was less of a story about people dying in the fields of Srebrenica, and more of
one about the Western diplomats around the table in Geneva.
The increase of American-oriented topics and actors in coverage effectively
"crowded out" other kinds of topics or actors D topics like war's effects and
actors like the indigenous people involved in the war, the Serbians and the
Bosnian Croatians. These changing amounts of U.S.-related story topics and
primary actors over time may be an indication D rather than overtly biased
statements or an increase in negative topic coverage D of the impact of policy
on press coverage of the war.
Future studies in press nationalism should continue to compare media
content to extramedia data when examining the relationship between press
coverage and policy. Extramedia indicators are usually able to provide an
objective record of events with which to compare press coverage. New uses for
existing reference documents should also be explored for their use in press
In addition, these studies should examine press coverage before and after
policy-defining events that may be identified in the extramedia sources. Instead
of concentrating on finding explicitly biased statements, press nationalism
studies should continue to look for other kinds of effects on coverage of
changes in American foreign policy.
 Roy Gutman, A Witness to Genocide. New York: Macmillan, 1993.
 Yang-Chou Yu and Daniel Riffe, "Chiang and Mao in U.S. News Magazines,"
Journalism Quarterly 66 (Winter 1989): p. 914.
 Martin Kriesberg, "Soviet News in the New York Times," Public Opinion
Quarterly 10 (1946): p. 540-564.
 Haluk Sahin, "Turkish Politics in New York Times: A Comparative Content
Analysis," Journalism Quarterly 50 (1973): p. 685-689.
 Yu and Riffe, p. 913-919.
 Tsan-Kuo Chang, "How Three Elite Papers Covered Reagan China Policy,"
Journalism Quarterly 61 (Summer 1984): p. 429-432.
 Jyotika Ramaprasad and Daniel Riffe, "Effect of U.S.-India Relations On New
York Times Coverage," Journalism Quarterly 64 (1987): p. 537-543, 663.
 Tsan-Kuo Chang, "The News and U.S.-China Policy: Symbols in Newspapers and
Documents," Journalism Quarterly 65 (1988): p. 320-327.
 Tsan-Kuo Chang, "The Impact of Presidential Statements on Press Editorials
Regarding U.S.-China Policy, 1950-1984," Communication Research Vol. 16., No. 4
(August, 1989): p. 486-509.
 Carol H. Weiss. "What America's Leaders Read." Public Opinion Quarterly 38
(1974): p. 1-22.
 W. Phillips Davison, "Diplomatic Reporting: Rules of the game," Journal of
Communication 25 (Autumn 1975): p.140.
 Gutman, p. viii.
 "The Road to Ruin." U.S. News and World Report, Dec. 12, 1994: p.64.
 George Mitchell, "What Scandal?" The New York Times. June 17, 1996: Sec.
A, p. 15.
 Weiss, p. 1-22.
 Yu and Riffe, p. 914.
 Ramaprasad and Riffe, p. 538.