Journal of Language and Social Psychology-2014-Montiel-0261927X14542435

542435
research-article2014
JLSXXX10.1177/0261927X14542435Journal of Language and Social PsychologyJournal of Language and Social PsychologyMontiel et al.
Article
Journal of Language and Social Psychology
120
2014 SAGE Publications
DOI: 10.1177/0261927X14542435
jls.sagepub.com
Nationalism in Local Media

During International Conflict:
Text Mining Domestic News
Reports of the ChinaPhilippines
Maritime Dispute
Cristina Jayme Montiel1, Alma Maria O. Salvador1,

Daisy C. See1, and Marlene M. De Leon1
Abstract
Using the lens of positioning theory and a mixed methods strategy, we investigate
domestic medias production of nationalism during an international conflict. Philippine
and Chinese news accounts of the Scarborough Shoal conflict are text mined by
RapidMiner software. With almost 100% accuracy, mathematical models successfully
classify sets-of-words that belong to each domestic newspaper. Principal components
analyses show that Philippine Daily Inquirer reports revolve around words related to
local fishing and livelihoods. News accounts in Chinas Peoples Daily uphold national
sovereignty. A complementary qualitative analysis of the news reports gives our
mathematically derived themes more meaning. Philippine maritime claims cite recent
Philippine history and international law. Chinas story references lost sovereignty
and ancient rather than recent history. Positioning analysis demonstrates how each
newspaper claims its own country owns Scarborough Shoal and is the victim of the
other country. Our findings show how two contradictory and nationalistic versions
of an international conflict may be linguistically constructed by domestic media. Our
research also highlights the illuminative power of positioning theory in the field of
international relations and media analysis.
Keywords
nationalism, media discourse, positioning theory, text mining, China, Philippines,
maritime conflict
1Ateneo
de Manila University, Manila, Philippines
Corresponding Author:
Cristina Jayme Montiel, Department of Psychology, Ateneo de Manila University, PO Box 154, Manila
1099, Philippines.
Email: cmontiel@ateneo.edu
Downloaded from jls.sagepub.com by guest on May 30, 2015
On April 10, 2012, a Chinese vessel reportedly blocked the Philippine Navys
attempted arrest of Chinese fishers in the Scarborough Shoal. For 2 months, neither
side moved from the stand-off in a strategically located 55 square kilometer circumferential area of small islands, triggering another maritime dispute on the Pacific Ocean.
Although tensions deescalated by September 2012, the threat of international conflict
continues until this writing, with naval and fishing vessels from both China and the
Philippines sporadically occupying the disputed oceanic area. As the two countries
clash over contested maritime territory, both governments hurl accusations and counteraccusations against each other. Understandably, claims by both sides fail to coincide. Our research looks at the subjective meanings of the Scarborough Shoal conflict,
from the lens of China and Philippine domestic media.
To develop our theoretical frame, we first describe the emergence of politicized
national identities during international conflicts. We then shift our argument to unpack
domestic medias process of constructing a nationalistic meaning of an international
conflict through discursive positioning.
International Conflicts and Politicized National Identities

International conflicts such as the ChinaPhilippines maritime dispute can be considered large-scale intergroup clashes. An intergroup conflict involves a collision between
parties that see their respective group interests as incompatible with the other (Pruitt &
Kim, 2004). The Scarborough Shoal conflict covers a particular type of intergroup
conflict that is political in nature, where the collectives are entire states that hold
incompatible interests over one maritime claim.
During intergroup conflicts, parties relate to each other not as individuals but
rather as members of particular groups. Intergroup conflicts make prominent social
identities of the parties involved in the conflict (Louis, 2008). Furthermore, international conflicts evoke shared politicized identities that are national rather than ethnic, gender, or class based. Simon and Klandermans (2001) claimed that politicized
identities arise when people engage as self-conscious group members in a power
struggle on behalf of their group knowing that it is the more inclusive societal context in which this struggle has to be fought (p. 319). For example, the Scarborough
Shoal crisis stirs up identities associated with being Chinese or Filipino, and belonging to the nation state of China or the Philippines. In relation to the Scarborough
Shoal conflict, people who identify with their respective nation states likewise position other Chinese and Filipinos as either belonging to ones in-group or out-group
as the case may be.
Politicized national identities emerge in an international conflict as local citizens
turn self-conscious of belonging to their respective countries during the power struggle between two nation states. But how do massive populations acquire relatively
homogeneous and nationalistic understandings of an international conflict? In the
large-scale communication process to win over ones domestic in-group against the
other out-group country, domestic media stands at center stage.
Montiel et al.
The Discursive Production of Nationalism Through Local

Media
Media holds the power to craft shared meanings among vast populations (Pitts &
Nussbaum, 2006). Its social potency emanates from a capacity to encode social reality
and circulate these symbolic representations (Lunt & Livingstone, 2001) to vast audiences, at a high speed. Understanding media discourse entails viewing its text not just
as simple talk in private conversations but as public and politicized communication
that plays a key role in national life (Lunt & Livingstone, 2001).
During international conflicts, Mediaespecially mass mediaare the vehicle
for political processes. They are the transmitter between politics and citizens (Hass,
2009, p. 78). For example, studies in the past decade revealed that both Israel and
Palestine media report their own peoples as victims and the Other as terrorists in the
chronic IsraelPalestine conflict (Hass, 2009).
Politicized national identities in local media involve not only in-group association
with ones fellow country members, nor out-group classifications of members of the
rival country, but also an awareness of third-party countries who may be drawn into
the regional conflict on either side of the international power struggle. Hence, when
politicized news communication is about international conflict, Chinese media target
local Chinese audiences while Philippine media address its news to Filipinos. However,
beyond politicizing the local newspaper audience to favor ones nation against the
other, both China and Philippine media outlets may likewise shape their reporting
styles (Bell, 1984) toward any third-party audience (Simon & Klandermans, 2001),
like, for example, the United States, that may be involved in a regional conflict as an
international superpower.
To understand the phenomenon of intergroup communication during international
conflict, one needs to integrate various layers of analysis (Singelis, 1996). Our model
for intergroup communication traverses analytical levels. We examine how a politicized maritime conflict like the Scarborough Shoal dispute was communicated to local
populations of both clashing countries through news media in China and the
Philippines. Hence, we investigate local meanings of an interstate conflictual relationship that unfolded on the global arena by analyzing intrastate news discourses on the
maritime dispute.
Figure 1 illustrates the social psychological position of media in a politicized international conflict such as the Scarborough Shoal maritime dispute. It shows two nation
states and includes three analytical layerstheir respective governments, domestic
media, and the public-at-large.
Among the various analytical layers involved in an international conflict, the most
visible and observable conflict is between nation states, like the governments of China
and the Philippines. This macro view is often taken by studies in political science and
international relations (Snyder, 2002; Waltz, 1988). Psychologists, on the other hand,
may examine aggregates of individual minds among the public-at-large (Hermann &
Kegley, 1995). Our research, however, covers domestic media and views news reports
in the context of both the nation state and the public-at-large.
China
Government
China
News Reports
China
Public
At-Large
Philippine
Government
Philippine
News Reports
Philippine
Public
At-Large
Figure 1. Varying analytical layers of intergroup communication during international conflict:

The Scarborough Shoal case.
Media is strategically positioned to shape the public mind not only because it communicates rapidly and multiplicatively but more so because audiences tend to believe
that information emanating from media sources are social facts rather than subjectively constructed versions of reality (Billig, 1995; Lunt & Livingstone, 2001). News
reports are expected to be carriers of objective, unbiased, social truths. Because media
consumers view news accounts as social facts, there is a suspension of any critical
questioning about the subjective construction of such news. We point out, however,
that news reports are a kind of discourse, and are therefore subjectively constructed. In
order to understand the subjective nature of news reports, one needs to consider that
discursive productions are situated and action-oriented (Potter & Edwards, 2001).
Viewing news accounts as situated discourse implies that news text is not linguistically isolated but rather enmeshed with particular interactions and contexts. Here lies
the fusion of language and social psychology in media discourse, as language acquires
meaning in relation to the human interactions that contextualize the communicative
function of language (Hornsey, Gallois, & Duck, 2008). Furthermore, during social
conflicts, one needs to consider intergroup rather than personalized interactions, as
medias text is analyzed vis--vis antagonistic intergroup relations between ethnic clusters (Stewart, Pitts, & Osborne, 2010), countries, and regions (Chan, 2014). At the
height of international conflicts like the Scarborough Shoal dispute, domestic media
discourse is produced amid an intensifying contentious relationship between two
national governments quarreling over maritime claims, with both governments likewise
trying to position themselves favorably to third-party audiences in the global arena.
The embeddedness of media discourse in an international conflict is closely linked
to the second feature of discursive production, the action-orientation of news reports.
During international conflicts, the reading audience of local newspapers is the domestic public-at-large, whose support is much needed when nation states go to war. The
role of domestic media is crucial when the actual military-like confrontations with the
Montiel et al.
other nation are hundreds of miles away from where populations live. Because the
Scarborough Shoal is literally in the middle of an ocean, domestic media plays a vital
role in the production of this constructed international conflict in the minds of the
Chinese and Filipino public-at-large. As nation states clash, we posit that the actionorientation of local news accounts veers toward the production of nationalism in the
public mind of ones local population.
Local medias construction of nationalism during international crises need not utilize flagrantly obvious symbols of national loyalty like flag waving, anthem chanting,
or calls for mass obedience to political leaders. Such communicative strategies would
manifest nationalistic bias and counter the perceived credibility of mass media as
truth-bearing. To conserve medias power to construct symbolic representations in the
public sphere, news reports need to retain as much apparent objectivity as it can muster. However, loyalty to ones country can be constructed by media in more subtle
ways (Li, 2009; Tsaliki, 1995), resulting in what is now commonly known as banal
nationalism (Billig, 1995).
Linguistic Positioning of a Countrys Right Over

Contested Territory
How does one detect nationalistic discursive patterns in news reports if markings of
this action-orientation cannot be observed from texts that directly connote national
loyalty? We propose another way of seeing nationalism in news reports, through a
positioning lens. In describing language and media as a new field in social psychology,
Lunt and Livingstone (2001) emphasized how media can position people. We extend
this personalized idea of medias positioning power and point out that during international conflicts media can likewise position two contending nations rather than individual persons.
Briefly, a position refers to a discursively produced set of rights and duties associated with interlocutors (Harre & Moghaddam, 2003). Such rights and duties can be
deduced from storylines. A storyline is a constructed narrative that provides meaning
to a set of sequential events (Slocum-Bradley, 2010). When discovered, a storyline
produces meaningful consistency among a set of apparently unrelated utterances. For
example, in news reports about the Scarborough Shoal conflict, the identification of an
underlying storyline coheres bits of news published at different stages of the maritime
clash. Such storylines likewise position the two disputing governments to each other
by discursively allocating rights and duties of China and the Philippines over the contested maritime territory.
With its ability to unpack psychological landscapes of large-scale relations, positioning analysis has been employed in the field of social power and political processes
(Moghaddam & Harre, 2010; Moghaddam, Harre, & Lee, 2008). In relation to political processes, a positioning lens swerves away from causal queries like Why are they
doing that? (Lee, 2010, p. 202) and interrogates issues like what meanings are
assigned to phenomena and how meanings unfold in social scenarios.
We posit that as domestic media present their news stories about an international
conflict, these can likewise produce nationalism through linguistic positioning. Local
news stories position one country in a positive light and the rival country in a negative
way. More specifically, in a quarrel over contested maritime territory, China news
accounts would be linguistically constructed in such a way that China is given the right
to own Scarborough Shoal and the duty to defend the contested area. On the other
hand, Philippine news reports would be constructed with inverse positioning storylines, giving the Philippines the right to claim the Shoal and the duty to defend it
against China.
To investigate the discursive production of nationalism through domestic media,
we compare news accounts of the Scarborough Shoal conflict in leading newspapers
from China and the Philippines. Our research asks three questions: (a) Is there a significant difference in news reports about the Scarborough conflict in China and
Philippine newspapers? (b) If so, what does the Scarborough Shoal conflict mean in
both countries? (c) How are China and the Philippines positioned relative to each other
in their respective peace-and-conflict storylines?
Data Sources
We collected online news reports about Scarborough Shoal from two mainstream
domestic newspapers: The Philippine Daily Inquirer and Chinas Peoples Daily
(). Both sampled newspapers are leading media outlets in their respective
countries. The Philippine Daily Inquirer tops other domestic newspapers in terms of
readership with survey results indicating that there were 866,000 readers for the
Inquirer versus 621,000 readers for the Bulletin, and 577,000 readers for Star
(Philippine Daily Inquirer, 2012). On the other hand, Peoples Daily is one of Chinas
most prominent official media outlets and is therefore an appropriate data source for
analyzing media discourse (Ng, Ye, & Lee, 2011). Its articles are written and edited by
the Chinese Governments Central Committees Central Publicity Department (Starr,
2010). The Peoples Daily has a market share of 47.6% in the newspaper market
(Beijing Journalism Association, 1986). All articles in the Chinese version of the website are included in the English versions and are translated by the Central Publicity
Department of the Chinese Government. We used the English version of Peoples Daily.
Our Chinese-speaking team member compared the Chinese and English versions of the
Peoples Daily and affirmed meaningful equivalence between both languages.
We selected news reports by employing key word searches on the online archives
of the targeted Philippine and Chinese newspapers. Our search words included
Scarborough Shoal, South China Sea, territorial disputes; and place-specific terms
such as Panatag Shoal for the Philippine Daily Inquirer and Huangyan Islands for the
Peoples Daily. Our search list snowballed as we uncovered more terms linked to other
related articles online. The number of news reports peaked in April and May of 2012,
then dwindled considerably in the subsequent months. Hence, we collected only the
news reports in April and May.
Montiel et al.
The initial number of AprilMay articles from the Philippines reached 187; China
had 81. By random selection, we further reduced our number of articles to equalize the
sample sizes from both countries and to approximate a total data pool of 100 news
reports. Our final data pool consisted of 55 articles from the Philippine Daily Inquirer
(total words = 29,200; word range per article = 180-1,236; M = 531; SD = 249) and 57
reports from Chinas Peoples Daily (total words = 22,310; word range per article =
57-1,370; M = 391; SD = 241).
Overview of Data Analysis Procedures

We used mixed methods to answer our different research questions. Study 1 applied
text mining and used a quantitative lens. Utilizing the software RapidMiner, we
employed machine learning algorithms to test whether China and Philippine newspapers reported the Scarborough conflict in significantly different ways. We likewise
investigated principal components associated with the different meanings of the
Scarborough conflict.
Study 2 utilized a qualitative approach to study domestic meanings of the international conflict. In this second study, discourse analysis involved detecting storylines
embedded in the news reports. Finally, positioning inquiries were carried out. We
combined the findings of Study 1 and Study 2 to explain how domestic newspapers
claimed contrasting sets of rights and duties associated with their respective countries.
Table 1 summarizes our strategy for data analysis.
We present Studies 1 and 2 sequentially. Each Study report contains both methodological details and highlights of the research findings. After explaining each Study,
we combine the findings of Studies 1 and 2 to describe the discursive positions claimed
by each domestic newspaper. We then profile how China and Philippine news reports
evoke polemic sets of political rights and duties that favorably position their own government in the dispute over Scarborough Shoal.
Study 1
Lexical Analysis of PhilippineChina News Reports: A Quantitative Text
Mining Approach
In text mining, raw text is preprocessed by translating words into one common language, followed by filtering for most commonly used words such as articles (a, the),
linking verbs (is, was), and others. The end result of preprocessing is a set of nouns,
adjectives, adverbs, and verbs stored in what is commonly coined as a bag of words.
To create classification models that were capable of automatically differentiating
Chinese articles from Philippine articles, we first fed two separate bags-of-words from
Philippine and Chinese newspapers into standard learning algorithms as training data
for text mining. We then used these mathematical models to test for differences
between the two newspapers.
Table 1. Overall Strategy for Data Analysis.

Research question
Study
1. Is there a significant difference

in news reports about the
Scarborough conflict in China and
Philippine newspapers?
2. What does the Scarborough Shoal
conflict mean in both countries?
Study 1
Study 1
Study 2
3. How are China and the Philippines

positioned relative to each other in
their respective peace-and-conflict
storylines?
Study 1 and
Study 2
Analytical strategy
Text mining using RapidMiner:
Performance measures of models
for classifying Philippine and
Chinese news reports
Text mining using RapidMiner:
Eigenvalue scores of Philippine and
Chinese news reports
Discourse analysis: Detecting newsreports structure; Storylines
Positioning analysis: Identifying China
and Philippine rights and duties in
the Scarborough Shoal conflict
Testing for Significant Differences Between China and Philippines News Reports. First, we
addressed the query: Is there a significant difference in news reports about the Scarborough conflict in China and Philippine newspapers? We wanted to see whether our
mathematical models could successfully separate sets-of-words that belonged to Chinese and Philippine news reports when we fed the text mining program the entire
combined set of Chinese and Philippine news reports. This procedure is intuitively
similar to discriminant analysis, where the question asked is one of accurate separation
of a pooled data set into two predefined categories. In a broad sense, our procedure did
to words what discriminant analysis does to numbers.
We used machine learning algorithms to determine which features or sets of combined words would give the best performance in accurately and precisely distinguishing Chinese news reports from Philippine news reports. We utilized three commonly
used machine learning algorithms for text mining, namely, Nave Bayes, Support
Vector Machines, and k-Nearest Neighbor (Banea, Mihalcea, & Wiebe, 2010; Banea,
Mihalcea, Wiebe, & Hassan, 2008; Kanayama & Nasukawa, 2006; Kim, Li, & Lee,
2010; Yao, Wu, Liu, & Zheng, 2006). For a technical description of these machine
learning algorithms, see the appendix.
Our RapidMiner findings confirm a highly significant difference between
Scarborough Shoal news reports from China and the Philippines. With almost 100%
precision and accuracy, programmed mathematical models successfully classified
sets-of-words that belonged to each of the two domestic newspapers. Table 2 shows
the performance measures for the three classification models. All three models have
high performance results in distinguishing between the Philippine and Chinese articles, with Nave Bayes having the highest score for each performance measure.
Labeling Themes: Principal Components of Philippines and China News Reports. We
employed a second text mining strategy to answer the question: What does
Montiel et al.
Table 2. Performance Measures of Models for Classifying Philippine and Chinese Articles.
Measure
Kappa
Accuracy
Precision
Recall
F-Measure
Nave Bayes
Support Vector Machines
k-Nearest Neighbor
100.00%
100.00%
100.00%
100.00%
100.00%
98.20%
99.17%
98.75%
100.00%
99.33%
89.20%
94.62%
94.23%
96.00%
94.54%
the Scarborough Shoal conflict mean in both countries? We separated China from
Philippine news reports, then extracted principal components from our two separate
sets-of-words. This procedure is intuitively similar to what principal components analysis does to quantified variables. After running this second procedure, we defined the
pivotal themes or components contained in each set of domestic news reports.
We produced two separate sets of eigenvectors associated with the news from
the Philippine Daily Inquirer and Peoples Daily. Table 3 presents our results. To
read this matrix, one would go down the columns for the principal components
associated with each domestic newspaper, and across rows to find the sets of words
associated with each component. Due to lack of space we present only the top 10
words that loaded on each component. The matrix cells contain the loadings of each
word on their respective principal components. To extract the central themes associated with the Scarborough Shoal news reports in both countries, we labeled each
principal component by considering the set of words that loaded on each
eigenvector.
Our findings confirm two different sets of themes reported by Philippine and China
news articles. Philippine news about the Scarborough Shoal conflict revolves around
words related to local fishing and livelihoods. For example, at the height of the maritime crisis, a Philippine Daily Inquirer (PDI) article on April 13, 2012, quotes a
Philippine Coast Guard officers statement that Fishing boats continue to go to the
Scarborough Shoal. Since there is no fishing ban, they are free to fish in the area. It
(the shoal) is ours, so why would we prohibit them? A week later on April 21, 2012,
PDI reports that The Bureau of Fisheries and Aquatic Resources has advised Filipino
fishermen against going to the area in a bid to ensure their safety while preventing tension from escalating in the disputed territory.
On the other hand, Chinas Peoples Daily holds national sovereignty as its first
principal component. We provide some excerpts to elucidate how concerns for sovereignty were depicted in Chinas news reports about the maritime conflict. On April 28,
2012, Chinese news claimed, The recent incident at Huangyan Island was triggered
by a Philippine warships violent harassment of Chinese fishermen, and concerns
Chinas sovereignty, according to Liu. Similarly, a month later, on May 16, 2012, the
Peoples Daily reported how The Chinese embassy reiterated Chinas sovereignty
over Huangyan Island, and urged the Philippine side to stop immediately their illegal
activities and leave the area.
10
0.21 secretary_albert_
rosario
0.20 foreign_secretary
0.16 filipino
0.16 vessel
vessel
brp
rosario
vessels
rosario
president
fishermen
albert
albert_rosario
secretary_albert
0.41
0.25
0.25
0.25
0.24
0.22
fishermen
fishing
guard
coast_guard
president
coast
Terms
EV
score
armed
forces
armed_forces
rosario
defense
fishermen
Terms
0.14 states
0.15 military
0.14 brp
Terms
fishing
manila
boats
fishing_boats
fishery
military
0.16 liu_weimin
0.16 hong
0.18 china_sovereignty 0.17 dispute

0.17 dispute
0.16 liu
0.17 relations
0.46
0.41
0.23
0.20
0.18
0.18
Terms
hong
embassy
military
diplomatic
tensions
side
Terms
0.14
0.17
0.17
0.17
0.45
0.31
0.30
0.23
0.22
0.19
EV
score
PC 3 = Military/
diplomatic dispute
0.19 huangyan_
island_south
0.19 island_south
0.18 island_south_
china
0.17 philippine_side
0.39
0.33
0.31
0.27
0.21
0.21
EV
score
PC 2 = Local fishing
and livelihoods
(Chinas) ThePeoples Daily
EV
score
PC 1 = National
sovereignty
liu
manila
military
hong
spokesman_liu
spokesman_liu_
weimin
0.20 sovereignty
0.37
0.36
0.35
0.29
0.28
0.24
EV
score
PC 3 = Military and
external defense
0.15 united_states
0.55
0.33
0.25
0.15
0.15
0.15
EV
score
PC 2 = Foreign
affairs, pol-diplomatic
procedures
Terms
PC 1 = Local
fishing and
livelihoods
Philippine Daily Inquirer
Table 3. Principal Component Analysis Eigenvector (EV) Scores of 10 Highest-Loaded Terms From Philippine and Chinese Articles.
11
Montiel et al.
News
Report
Causal
Explanations
History
Context
(Previous
Events)
Actual
Episodes
Circumstances
Events
Figure 2. Schematic of the news structure used to classify passages of each news report.
Although the first principal components of China and Philippine newspapers are
sharply different, the secondary principal components suggest overlaps between the
two news outlets. Chinas Peoples Daily acknowledges fishing livelihoods as its second principal component, and both China and Philippine newspapers consider a combination of diplomatic-military dispute resolution strategies in their subsequent
principal components. However, a deeper analysis of subjective meaning through
positioning theorya methodological strategy to be discussed at a later part of this
articleshows that even apparently common themes like fishing-for-livelihood position the two countries in different ways. Chinese newspapers discursively position
their local fishers with the right to harvest fish in the contested shoal, whereas
Philippine newspapers claim that the right to fish belongs to Filipino fishers.
Study 2
Storylines in Philippine and China News Reports: A Qualitative Study
To expound on our principal components findings, and to give our mathematically
derived themes more meaning, we implemented a qualitative analysis of news reports
addressing the same question: What does the Scarborough Shoal conflict mean in both
countries? We looked for storylines embedded in various segments of a news
structure.
First, we read each of the 112 news reports and classified the different parts of
every article into structural segments of a news report, following an analytical strategy
adapted from Van Dijks (1988) study of news structures. Figure 2 illustrates our classificatory scheme for the contents of each of the news reports. We selected parts of the
news that provided a causal explanation of the Scarborough Shoal conflict, and further
classified causal accounts into whether these were about history or previous events in
the context. We also chose parts of the news that covered actual episodes and categorized these segments into the episodes circumstances and actual events. Our
12
qualitative data pool consisted of news segments coded according to the topics of history, context, circumstances, or events.
We wrote one summarizing storyline for each part of the news structure. This reading and rereading was done by an expanded group during a series of workshops participated in by members of the research team, and psychology graduate and
undergraduate students. In these workshops, eight research groups were assigned to
specific news segments. Their task was to read and summarize all the news passages
contained in one coding category. We obtained eight summary storylines corresponding to the Philippines and Chinese news versions of the Scarborough conflict, organized around the four topics of history, context, circumstances, and events.
To show how storylines inform and elucidate the meaning of each principal component, we juxtaposed the qualitatively derived storylines with the principal components
computed from text mining. Table 4 combines both the storylines and principal component findings. It presents segments of the storylines that enhance ones understanding of the difference between news reports from the Philippines and those from China.
The Philippine story about Scarborough Shoal comes from reporting about the episodes circumstance and events, and recent Philippine history. It is a story about
Chinas bullying Filipino fisher folk, Philippines appeal for international mediation,
and seeking U.S. help to defend the Philippines against Chinas aggression. Chinas
news reports contain more references to past contexts and ancient rather than recent
history. The news reports talk about longstanding territorial sovereignty, harassment of
Chinese fishermen by Filipinos, ancient historical rights, and peaceful dispute
resolution.
Domestic Media Positioning Favoring National Identities in the ChinaPhilippines Scarborough Shoal Conflict. Local media in each of the antagonistic countries report two relatively different pictures of the same international conflict. Based on the Philippine
Daily Inquirer, the Scarborough Shoal conflict is about how
Filipino fishermen fishing in their rightfully owned Scarborough Shoal are bullied by
China. China continues to be aggressive in the South China Sea. The Philippines attempts
to resolve this conflict diplomatically through international mediation bodies. The US has
the duty to defend the Philippines because of a US assistance pact.
The Peoples Daily () claims that the Scarborough Shoal conflict is about
how
Ancient historical records show China holds territorial sovereignty over Huangyan
Island. Chinese fishermen have been fishing in the area over a very long historical time,
but now they are harassed by the Philippine Navy. China has to defend their fishermen,
but will do so through peaceful dispute resolutions.
Salient and incompatible discursive positions are produced by each countrys newspaper reports. Each domestic media outlet constructs news reports that favor their
13
Montiel et al.
Table 4. RapidMiners Principal Components and the Location of Their Respective
Storylines in the News.
Philippine Daily Inquirer
Storylines in the news
structure
Circumstances
Principal
components
PC 1: Local fishing
and livelihoods
Peoples Daily ()
Principal
components
PC 1: National
sovereignty
Filipino fishermen
own shoal
Chinas bullying of
Filipino fishermen
Events
PC 2: Foreign affairs, PC 2: Local fishing

pol-diplomatic
and livelihoods
procedures
Philippines attempt
to lift issue to
international
mediation blocked
by China
Philippine history
PC 3: Military and
external defense
PC 3: Military/
diplomatic dispute
Recent
Storylines in the news

structure
Context
Chinas territorial
sovereignty over
Huangyan Island
Harassment of
Chinese fishermen
by Filipinos
China history
Ancient China
records of
Huangyan Island;
longtime fishing
grounds
Circumstances
Prevent/defend
Chinese fishermens
arrest by Philippines
Events
Philippine Navy
harassment
Peaceful dispute
resolution
Chinas aggression
in South China Sea
U.S. assistance and
defense pact
respective national identities in the Scarborough Shoal conflict. Table 5 summarizes

the contradictory discursive positions taken by the Philippine Daily Inquirer and the
Peoples Daily. The Philippines main story revolves around the loss of fishing livelihoods whereas Chinas primary storyline emphasizes a threatened national sovereignty. Each newspaper claims that their own country is the victim of the other
country.
We emphasize, however, that the discursive incompatibility is not about a difference in primary and secondary themes in news stories, because although primary
themes are different, secondary themes overlap. On a deeper level of discursive analysis using positioning theory, one can detect the subtle but powerful production of
14
Table 5. Incompatible Discursive Positioning of National Rights and Duties by China and
Philippine News Reports.
Incompatible discursive positions
Who has the right to own
Scarborough Shoal?
Where does this right come
from?
Who is the victim?
Who/What is the central theme
of victimization?
Who is the aggressor?
Who has the duty to defend the
victim from the aggressor?
Philippine news
China news
Philippines
China
International laws
Ancient history
Philippines
Filipino local fishing and
livelihood
China
Philippines and the United
States
China
Chinas sovereignty
Philippines
China
nationalism as both newspapers claim incompatible political rights and duties for their
respective national governments in their respective underlying storylines. News
reports in the Philippine Daily Inquirer cohere around a storyline that (a) international
laws give the Philippines the right to own Scarborough Shoal; (b) China is the aggressor against the Philippines fishing interests; and (c) the Philippines and the US have
the duty to defend the Philippines against China. On the other hand, Chinas Peoples
Daily consistently asserts a storyline that (a) ancient history gives China the right to
own Scarborough Shoal; (b) the Philippines is the aggressor against Chinas sovereignty; and (c) China has the duty to defend itself against the Philippines and the
United States, if the latter intervenes.
Discussion
Our findings confirm a quantitatively significant difference in the sets of words used
to report about the Scarborough Shoal conflict in China and Philippine newspapers.
Both media outlets convey different meanings to their local populations and the global
audience. China and Philippine newspapers discursively position their own country as
having the morally correct rights and duties to defend themselves as victims who
rightfully own Scarborough Shoal. Such local news accounts boost national identities
during international conflict. As governments of two countries posture against each
other on the global political arena, so also do their respective domestic media position
their own countries in a positive light.
The territorial-rights and victimology discourses on both sides of the conflict are
qualitatively different. Both countries claim territorial rights over Scarborough Shoal.
But China news reports delve deep into ancient historical narratives to lay its claim on
the contested shoal, while the Philippines emphasizes contemporary international law.
The victimology on both sides are likewise strikingly dissimilar. On one hand, Chinas
Peoples Daily frames its maritime dispute story as one of lost sovereignty. Raising the
15
Montiel et al.
mantle of sovereignty evokes images of superpower victimhood and a tortuous history

of political subordination to colonial powers. Such a narrative catapults the China
Philippine maritime dispute beyond regional boundaries and straight into a discourse
about a United StatesChina clash intended not only for Chinese audiences but also
for a global audience; and subtly yet powerfully includes a warning to the United
States not to interfere in the Scarborough Shoal dispute. On the other hand, the
Philippine Daily Inquirer talks about Philippine victimhood through an economic
rather than a political narrative, claiming the loss of local fishing livelihoods due to
intrusion into Philippine fishing grounds by Chinese vessels.
Our findings point to pragmatic implications in the Scarborough Shoal conflict. A
more careful study of the discursive positionings in Table 5 signals a potential trigger
to international conflict escalation in the Pacific. The conversation changer may lie in
a combination of political conditions, with the United States as global superpower
joining the conflict on the side of the Philippines and confronting a China that holds
national sovereignty as its pivotal discourse. Indeed, this kind of international scenario
is beginning to unravel in the Pacific. However, another more peaceful discourse is
likewise emerging, in the form of the Association of Southeast Asian Nations (ASEAN)
Code of Conduct in the South China Sea. ASEAN member-states have recently arrived
at a consensus to speak as a collective about issues concerning maritime claims in the
Pacific Ocean (Agence France-Presse, 2013). Such a collective move among ASEAN
governments may not be as threatening to Chinas sovereignty issues because ASEAN
is not perceived as a superpower.
Our research demonstrates that new theories and methods can be employed to study
international relations in general, and constructed media nationalism in particular. We
show the potency of positioning theory to illuminate the landscape of public politicized meanings during an international crisis. Because positioning theory analyzes
text, it is able to uncover meaning through media reports. Located at the epistemological juncture of language and social psychology, positioning theory detects discursive
storylines and identifies how rights and duties of conflict actorsor governments, in
the case of international disputesare allocated by news accounts.
Our research likewise shows how discourse can be analyzed quantitatively through
text mining and that an appropriate mixing of quantitative and qualitative strategies
can enrich discourse analysis. Our innovative methodological approach also demonstrates the power of positioning analysis in studying the subtleties of banal nationalism. Although thematic analysis of media news can be employed to describe
constructed nationalism, simply comparing themes may distort comparative discursive meanings. For example, a thematic principal components analysis in our research
suggested that both the Philippines and China held common interests in fishing livelihoods, but a subsequent positioning analysis showed that news reports favored either
Chinese or Philippine fishers, depending on whether the newspaper was from China or
the Philippines.
Finally, the contribution of our research to language and social psychology lies in our
findings affirmation that two contradictory and nationalistic versions of an international
conflict may be discursively constructed through domestic media reports. Positioning
16
theory assumes that meaning is relative rather than absolute, discursively constructed
within a relationship rather than formed inside the mind of an individual. In our research,
the meaning of the Scarborough Shoal dispute was constructed through media reports
that were audience-designed (Bell, 1984) for both domestic and global readers.
News accounts are expected to be impartial. But media can produce a biased public
discourse about a conflict without appearing overtly one-sided by its purposive use of
linguistic positioning. Discourses are not only constructed by but also constructive of
social reality (Potter & Edwards, 2001) and can help configure reality by positioning
nations within a nationalistic story of morally correct rights and duties.
During international conflicts, domestic media can churn out news accounts that
are not only morally ascendant but also entitle their homeland to clash with the other
country. In his discussion of medias power to produce nationalism, Billig (1995)
warned that
banal nationalism can hardly be innocent: it is reproducing institutions which possess
vast armaments . . . forces can be mobilized without lengthy campaigns of political
preparation. The armaments are primed, ready for use in battle. And the national
populations appear to be primed, ready to support the use of those armaments. (p. 7)
Indeed, as local newspapers shape the public mind (Hass, 2009; Hoffman, 2012;
Scheufele, 2000), discursive positioning in Chinese and Philippine news reports fortify the morally correct storyline among local populations, creating psychologically
nationalistic and war-ready publics in the two antagonistic countries.
Appendix
Technical Note on the Machine Learning Algorithms
Using RapidMiner, we applied each of the learning algorithms to the data set to create
classification models. Nave Bayes (NB), Support Vector Machine (SVM), and
k-Nearest Neighbor (K-NN) learning algorithms approach the creation of classification models differently. As such, each learning algorithm has different characteristics
and is expected to perform differently depending on the nature of the dataset.
NB is considered to be the simplest probabilistic classifier used for classification of
text documents. It severely assumes that each feature word or term is independent of
other feature words or terms in a document. NB basically uses the joint probabilities
of words and labels in estimating the class of a given document. It is considered popular as a machine learning classifier of text document due to its ease of implementation
and its good performance. However, it is only manageable for low dimensions (Chen,
Huang, Tian, & Qu, 2009; Goller, Lning, Will, & Wolff, 2000; Gupta, 2011; Harish,
Guru, & Manjunath, 2010).
SVM is considered to be a highly accurate machine learning method for classification problems. To classify a binary or multiclass classification problem, SVM tries to
find the optimal hyperplane within the input space. Training examples are separated
17
Montiel et al.
using the maximum margin between its support vectors. The high popularity of SVM
may be attributed to its less susceptibility to overfitting compared to other learning
methods because the model complexity is independent of the feature space dimension.
SVM is good at handling data with large feature spaces. It also does not take too much
time in training and classification. However, SVM is known to be difficult to implement and is not good at scaling with the number of documents in the text collections
(Gupta, 2011; Harish et al., 2010; Zhang, Yoshida, & Tang, 2008).
K-NN is also known as instance-based learning or lazy learning. Its algorithm is
simple, valid, and nonparametric. The algorithm is based on k, which is the number of
closest training examples. Nodes represent instances of labeled training data as well as
unlabeled text or documents for classification. When classifying, the distance between
each labeled node and unlabeled node is calculated. K-NN determines distance using the
Euclidian distance formula. The unlabeled text or document is assigned to the class of its
nearest neighbor, if k is 1. The performance of k-NN depends on the value of k and the
similarity or dissimilarity value. K-NN is a simple algorithm to implement because it
uses two parameters only. It is also considered as robust when dealing with noisy training data. K-NN is well-suited for multimodal classes because its classification decision
is based on a small neighborhood of similar documents. The complexity of k-NN in
determining the label of an unlabeled node is N * log k where N represents the number
of records in the training data. Because computing for similarity/dissimilarity is timeconsuming, it is highly impractical to implement k-NN for voluminous amounts of training data (Goller et al., 2000; Gupta, 2011; Harish et al., 2010; Wang & Wang, 2007).
Acknowledgments
The authors are sincerely grateful to Howie Giles and the two anonymous reviewers whose
valuable suggestions helped improve this article.
Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the authorship and/or
publication of this article.
Funding
The author(s) disclosed receipt of the following financial support for the research, authorship,
and/or publication of this article: This research was supported by an Ateneo de Manila University
Loyola Schools Scholarly Work Grant in 2013.
References
Agence France-Presse. (2013, August 14). ASEAN vows unity to press China on code of conduct
in West PH Sea. Retrieved from http://www.interaksyon.com/article/68623/asean-vowsunity-to-press-china-on-code-of-conduct-in-west-ph-sea
Banea, C., Mihalcea, R., & Wiebe, J. (2010). Multilingual subjectivity: Are more languages
better? In Proceedings of the 23rd International Conference on Computational Linguistics
(pp. 28-36). Stroudsburg, PA: Association for Computational Linguistics.
18
Banea, C., Mihalcea, R., Wiebe, J., & Hassan, S. (2008). Multilingual subjectivity analysis using
machine translation. In Proceedings of the Conference on Empirical Methods in Natural
Language Processing (pp. 127-135). Stroudsburg, PA: Association for Computational
Linguistics.
Beijing Journalism Association. (1986). An analysis of Renmin Ribao readers. Chinese
Sociology and Anthropology, 18, 114-122.
Bell, A. (1984). Language style as audience design. Language in Society, 13, 145-204.
Billig, M. (1995). Banal nationalism. Thousand Oaks, CA: Sage.
Chan, M. (2014). (Re)categorizing intergroup relations and social identities through news discourse: The case of the China Dailys reporting on regional conflict. Journal of Language
and Social Psychology, 33, 144-164.
Chen, J., Huang, H., Tian, S., & Qu, Y. (2009). Feature selection for text classification with
nave Bayes. Expert Systems with Applications, 36, 5432-5435.
Goller, C., Lning, J., Will, T., & Wolff, W. (2000). Automatic document classification: A
thorough evaluation of various methods. ISI, 2000, 145-162.
Gupta, V. (2011). Recent trends in text classification techniques. International Journal of
Computer Applications, 35, 45-51.
Harish, B. S., Guru, D. S., & Manjunath, S. (2010). Representation and classification of text
documents: A brief review. International Journal of Computer Applications, 2, 110-119.
Harre, R., & Moghaddam, F. (2003). Introduction: The self and others in traditional psychology and in positioning theory. In R. Harre & F. Moghaddam (Eds.), The self and others
(pp. 1-11). Westport, CT: Praeger.
Hass, R. (2009). The role of media in conflict and their influence on securitisation. International
Spectator, 44, 77-91.
Hermann, M. G., & Kegley, C. W. (1995). Rethinking democracy and international peace:
Perspectives from political psychology. International Studies Quarterly, 39, 511-533.
Hoffman, L. H. (2012). When the world outside gets inside your head: The effects of media
context on perceptions of public opinion. Communications Research, 40, 463-485.
Hornsey, M. J., Gallois, C., & Duck, J. M. (2008). The intersection of communication and
social psychology: Points of contact and points of difference. Journal of Communication,
58, 749-766.
Kanayama, H., & Nasukawa, T. (2006). Fully automatic lexicon expansion for domain-oriented
sentiment analysis. In Proceedings of the 2006 Conference on Empirical Methods in Natural
Language Processing (pp. 355-363). Stroudsburg, PA: Association for Computational
Linguistics.
Kim, J., Li, J., & Lee, J. (2010). Evaluating multilanguage-comparability of subjectivity analysis
systems. In Proceedings of the 48th Annual Meeting of the Association for Computational
Linguistics (pp. 595-603). Stroudsburg, PA: Association for Computational Linguistics.
Lee, N. (2010). Afterword. In F. Moghaddam, R. Harre, & N. Lee (Eds.), Global conflict resolution through positioning analysis (pp. 201-207). New York, NY: Springer.
Li, J. (2009). Intertextuality and national identity: Discourse of national conflicts in daily newspapers in the United States and China. Discourse & Society, 20, 85-121.
Louis, W. (2008). Intergroup positioning and power. In F. Moghaddam, R. Harre, & N. Lee
(Eds.), Global conflict resolution through positioning analysis (pp. 21-39). New York, NY:
Springer.
Lunt, P., & Livingstone, S. (2001). Language and the media: An emerging field for social psychology. In W. P. Robinson & H. Giles (Eds.), The new handbook of language and social
psychology (pp. 585-600). London, England: Wiley.
19
Montiel et al.
Moghaddam, F., & Harre, R. (Eds.). (2010). Words of conflict, words of war: How the language
we use in political processes sparks fighting. Santa Barbara, CO: Praeger.
Moghaddam, F., Harre, R., & Lee, N. (Eds.). (2008). Global conflict resolution through positioning analysis. New York, NY: Springer.
Ng, S. H., Ye, J., & Lee, C.-C. (2011). Media discourse on globalization in China: A social
psychological analysis. Journal of Language and Social Psychology, 30, 139-157.
Philippine Daily Inquirer. (2012, February 20). Philippine Daily Inquirer widens lead in
readership. Retrieved from http://newsinfo.inquirer.net/148811/philippine-daily-inquirerwidens-lead-in-readership
Pitts, M. J., & Nussbaum, J. F. (2006). Integrating the past and paving the future: Examining
current trends and extending boundaries of language and social psychology research.
Journal of Language and Social Psychology, 25, 197-202.
Potter, J., & Edwards, D. (2001). Discursive social psychology. In W. P. Robinson & H. Giles
(Eds.), The new handbook of language and social psychology (pp. 103-118). London,
England: Wiley.
Pruitt, D. G., & Kim, S. H. (2004). Social conflict: Escalation, stalemate, and settlement.
Boston, MA: McGraw-Hill.
Scheufele, D. A. (2000). Agenda-setting, priming, and framing revisited: Another look at cognitive effects of political communication. Mass Communication and Society, 3, 297-316.
Simon, B., & Klandermans, B. (2001). Politicized collective identity: A social psychological
analysis. American Psychologist, 56, 319-331.
Singelis, T. M. (1996). The context of intergroup communication. Journal of Language and
Social Psychology, 15, 360-371.
Slocum-Bradley, N. (2010). The positioning diamond: A trans-disciplinary framework for discourse analysis. Journal for the Theory of Social Behavior, 40, 79-107.
Snyder, G. H. (2002). Mearsheimers world-offensive realism and the struggle for security.
International Security, 27, 149-173.
Starr, J. B. (2010). Understanding China: A guide to Chinas economy, history, and political
culture (3rd ed.). New York, NY: Hill & Wang.
Stewart, C. O., Pitts, M. J., & Osborne, H. (2010). Mediated intergroup conflict: The discursive
construction of illegal immigrants in a regional U.S. newspaper. Journal of Language
and Social Psychology, 30, 8-27.
Tsaliki, L. (1995). The media and the construction of an imagined community: The role of
media events on Greek television. European Journal of Communication, 10, 345-370.
Van Dijk, T. A. (1988). News analysis: Case studies of international and national news in the
press. London, England: Lawrence Erlbaum.
Waltz, K. N. (1988). The origins of war in neorealist theory. Journal of Interdisciplinary
History, 18, 615-628. Retrieved from http://www.jstor.org/stable/204817
Wang, Y., & Wang, Z. (2007). A fast KNN algorithm for text categorization. In Machine
Learning and Cybernetics, 2007 International Conference (Vol. 6, pp. 3436-3441). New
York, NY: Institute of Electrical and Electronics Engineers.
Yao, J., Wu, G., Liu, J., & Zheng, Y. (2006). Using bilingual lexicon to judge sentiment orientation of Chinese words. In The Sixth IEEE International Conference on Computer and
Information Technology (p. 38). Washington, DC: Institute of Electrical and Electronics
Engineers Computer Society.
Zhang, W., Yoshida, T., & Tang, X. (2008). Text classification based on multi-word with support vector machine. Knowledge-Based Systems, 21, 879-886.
20
Author Biographies
Cristina Jayme Montiel is a professor of peace/political psychology and has been teaching at
the Ateneo de Manila University for more than 30 years. She is the recipient of the 2010 Ralph
White Lifetime Achievement Award from the American Psychological Associations Division
of Peace Psychology. She was managing editor of the Encyclopedia of Peace Psychology
(Wiley-Blackwell, 2012), editor of the volume on Peace Psychology in Asia (Springer, 2009),
and has recently published in Journal of Pacific Rim Psychology, Journal for the Theory of
Social Behavior, Asian Journal of Social Psychology, and Peace and Conflict: Journal of Peace
Psychology.
Alma Maria O. Salvador obtained her PhD in development studies at the Department of
Political Science of De la Salle University, Philippines. She was formerly the chair of the
Department of Political Science of Ateneo de Manila University, and she currently coordinates
the academic program of the Dual Masters Programme in Peace Studies of the University for
Peace and Ateneo de Manila. Her research is in the fields of international relations, security
sector reform, and civilmilitary relations.
Daisy C. See holds PhD units in Chinese philosophy at the Peking University, China. She currently chairs the Chinese Studies Program of Ateneo de Manila University where she also
teaches Chinese languages and culture courses.
Marlene M. De Leon studied at the Ateneo de Manila University where she received her bachelors degree in management information systems and masters and doctorate degrees in computer science. She is currently chair and assistant professor in the Department of Information
Systems and Computer Science, Ateneo de Manila University, where she teaches software engineering, systems analysis and design, and database management.

Journal of Language and Social Psychology-2014-Montiel-0261927X14542435

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Journal of Language and Social Psychology-2014-Montiel-0261927X14542435

Uploaded by

Copyright:

Available Formats

542435

Nationalism in Local Media

Cristina Jayme Montiel1, Alma Maria O. Salvador1,

de Manila University, Manila, Philippines

Downloaded from jls.sagepub.com by guest on May 30, 2015