THE CONTROL OVER PERCEPTION OF MASS VIOLENCE THROUGH STRATEGIC LABELLING

: This paper examines the creation and use of names that refer to a mass violence in Indonesia that occurred in May 1998 in several cities. The media has dubbed the event the May 1998 riots . Alternative names have been widely used and each represents either a different portrayal of the event or social political stance towards the event. Using discourse analysis, this paper will demonstrate how the choice of names affects presentation of the event, recognition of what actually happened, and the identiﬁ cation of parties involved, both the perpetra-tor and the victim. The labels assigned by the media to the event serve to promote certain points of view, shape social narrative and inﬂ uence readers’ understanding of mass violence. This paper found that the use of May 1998 as the token has blurred the nature of the event that is arguably politically motivated and downplayed the violence suffered by a particular group targeted during the riots.


INTRODUCTION
Novelist George Orwell wrote of how language is used as a tool to manipulate our perception. 1 Such use of language has been known as This paper will look into how different name assignments and categorization of the event that occurred in May 1998 affects the whole portrayal, representation and perception of what transpired during the fateful month. The May 1998 event has cast a dark shadow over the enforcement of human rights and justice in Indonesia, with the spectre of the event still looming large. As the narrative of the event has been controlled by the media, the linguistic choices they made in their news reports need to be looked at with a critical eye. This paper will investigate the use of the label May 1998 riots by the local media to refer to the event.
To examine how different names encapsulate the description of an event and represent it, the author will consider and compare different choices of names and how they project a discrete angle or version of a single event. Names can be chosen by the media to captivate readers' attention and direct them to focus on a certain aspect of the event. In the same vein, alternatives to those names may also be chosen in order to defl ect readers from the main substance of the event.
Different expressions pointing to a single referent serve to highlight different aspects of the event. This results in a different presentation and evokes different sense and evaluation. For example, the name May riots has been alternatively referred to as the May tragedy. The word riot would entail a large mass of people and presumably involve an act of violence, while tragedy may entail either an individual or large group but does not necessarily involve violence. The word riot would give more information than tragedy as tragedy focuses on how one sees the end of something. Riots meanwhile focuses on what happens and not about how one perceives the whole event.
Several scholars have looked at the use of different references or names for a single referent. Lee 5 and Chilton 6 look at how different references generate different connotations. Fowler considers how reality is presented in news in a way that may favour a certain group and discredit the other. 7 Nossek et al. focuses on linguistic choices made for references in political violence discourse and also how news outlets project a certain standpoint when reporting political violence. 8 Lee wrote that "language is an instrument for the assignment of the phenomena of human experience to conceptual categories; it is clearly not simply a mirror that refl ects reality. Rather, its function is to impose structures on our perception of the world." 9 Different words denoting actions and situation, such as massacre versus tragedy and riot versus chaos, convey different ideologies and imply different levels of wrongdoing and different extents of social impact. In his study of word classifi cation, he examined the differences in the application of the term European settlement versus European invasion in the context of the European arrival in Australia at the end of the eighteenth century. 10 It is evident that the word European settlement was a term used by the Europeans to describe their peaceful migration to Australia in search of a better life. The latter term was used by the Aboriginal Australians who saw the arrivals as a threat to their existence. Lee stated that "whether speakers use settlement or invasion now provides some indication of their political viewpoint". 11 Fowler examines the concept of transitivity in the language of news reports to show how a certain reality is more prominently portrayed than others. 12 He states that "transitivity is the foundation of representation… and transitivity has the facility to analyze the same even in different ways". 13 Transitivity strategy covers the inclusion of certain modifi ers in the noun phrase as part of the agent or patient (recipient), omissions of certain participants in the form of nominalization and types of predicates (passive or active constructions). Other elements that Fowler looked into are lexical choices and structure in terms of how they affect meaning, both literal and connotative meaning. He also considers modality in the study of news language. The same author also conducted a study on names or titles used to refer to or address a person. He focuses on how a certain calling or addressing term may be either abusive or glorifying. He states that "newspaper…sorting people into categories, and placing discriminatory values on them". 14 He uses news reports from several newspapers on Libya and US confl icts in 1986 for his study of addressing terms. His analysis shows how opposing government members are addressed with either diminutive or honorifi c titles depending on whether the newsroom supports or disapproves of the discussed policy.
While many studies using the approaches outlined above have focused on references to persons or activities, there appears to be a lack of studies which analyze the use of event names. To fi ll the research gap, this paper will examine the choice of names in the media to refer to the event of May 1998 using several analytical tools. To analyze naming strategies in the media, this paper utilizes lexical semantics analysis and Critical Discourse Analysis (CDA). Lexical semantics analysis will help explore the multi-dimensions of meaning and the extent to which readers are led to different understandings by different names. Discourse analysis is utilized to uncover the ideologies behind name choices in the media. The visual aspects of all of those names will be compared and examined to see how visual aspects intervene with the semantic meaning and fi nally how they affect recognition and understanding of what happened.

THEORY
This paper will utilize several tools of analysis to look at strategic labelling to refer to an event and analyze the choice of names, the perceived impacts of such assignment and the motive behind it through the lens of discourse analysis.
In semantic analysis, reference means the acts of referring to or denoting an entity (which will be called referent) through means of linguis-tics features. 15 Theory of reference tries to attend to the questions of what relations a reference has to its referent. A reference does not only function to refer or point to an item but it also covers semantically relevant relations. 16 Saeed states that there are two approaches we use in the way people talk about the world: the fi rst is the referential approach and second is the representational approach. 17 The referential approach tries to answer questions regarding the relations between expression and the world they refer to. In other words, the reference approach considers how a certain word points to the world or the reality. Meanwhile, the representational approach tries to attend to the issue of how our mind projects the reality as expressed through the reference; what is the mental model of a situation or a thing. 18 The reality or the thing out there may be the same but there are different ways of expressing that reality or thing. Reference and representation are two different things. A reference assigns a name as the identity of a referent while a representation assigns a certain perception towards a referent.
Names are labels for people and places which we use to talk about individuals or locations. 19 The use of names is normally bound by context in the assumption that the audience can identify the referent. Names can work to identify something or someone because of the shared knowledge between the speaker, who uses the name, and the addressee. There are two approaches in looking at how names work. The fi rst is description theory and second is causal theory. 20 Description theory says that a name is a label for one or more defi nite descriptions about something. A name provides a description for an entity. Meanwhile, causal theory emphasiz-15 John I. Saeed, Semantics (Oxford: Blackwell Publishing, 2016 Discourse Analysis (CDA). CDA is a methodological tool to help examine the use of language that helps promote an ideology through strategic text production by social institutions and the controlled process of reading and understanding of texts by readers. Fairclough discusses the social conditions of production and social conditions of interpretation that constitute the meaning making process. 25 He said when people are analysing texts in their head as part of individual cognitive processes, they are bound by social practices ingrained in them. Similarly, the production mode of a text is also within the confi nes of social practices. He argues that we are never independent of the dictates of social context in our process of producing, reading. This paper will consider the social practices underlying the carefully chosen linguistic expressions.
Young and Fitzgerald state that "CDA focuses on linguistics analysis to expose misinterpretation, discrimination or particular position of power in all kinds of public discourse such as political speeches, newspaper and advertisements." 26 One of the CDA features that will be used for the analysis here is appraisal theory. Appraisal theory looks at how the choice of words reveals the attitude of the speakers. Appraisal theory allows a focus on the way a speaker evaluates a situation, whether it is in a positive or negative light. Evaluation can be expressed directly or implicitly.

RESEARCH METHODOLOGY
The study in this paper will be conducted using a content analysis method. The author will read articles across major newspapers containing different labels referring to the event and come up with a list of common labels. The author will then determine the frequency of each name and to confi rm that the popular label of "May riots" is indeed the most common name. To determine the most frequent words as part of quantitative studies in linguistics and language 27 , a researcher can use established cor-pus as the primary data or select a number of relevant articles from the media which the study will be based on. Corpus is a collection of texts on particular subjects developed either by an institution, for example, the British Library, or a team of researchers. This paper, however, will not use a corpus to determine the frequency between different names referring to the event because of the absence of a corpus that covers the terms surrounding it. The search for phrases, both in Indonesian, "Kerusuhan Mei 1998" and "Kerusuhan Mei" in a corpus developed by the Indonesian Ministry of Education came up with no results. As such, the frequency of the Indonesian phrases that become the labels for the event is determined using Google search. This method is used to ensure that the author is not relying on her impression about the most common label but has an empirical validation by means of Google search results.
As this research is focused on examining how different media or information portals have assigned different names to one single political event, the data will be taken from several different news outlets across the globe. The data is taken from transnational media to show how the media outside the country where the event occurred have translated or adopted new names to refer to the same event. The author will have one article as a reference to the usage of a particular label that becomes an alternative to the common name. She will refer to the content of the article and the context of its publication as a sample to compare and discuss the different representations of the event, what could have motivated the choice and how common name and its alternatives infl uence the perception of what transpired in the event.
The author of this paper may have to resort to the usage of a more famous name in referring to the event in this study. The author realizes this may lead to a slight bias as she chooses a certain name over other alternatives when referring to the events in this paper. However, she will try to maintain neutrality in analysis, although the names she chooses as the general reference in this paper may refl ect a certain stand point or ideology.

ANALYSIS BACKGROUND
The event occurred in Jakarta and several other major cities in Indonesia, such as Solo and Medan. The riots, which broke out on May 13, 1998 and lasted three days, started after students in Jakarta were shot in a protest against the then president a day earlier. The media published reports that the mob targeted Chinese residences and shops in Chinatown.
Shops and residences alike were broken into, ransacked, and raided. They looted everything from groceries to electronics, destroying the property in the wake of it. The racial violence in the Chinese enclaves has not only left the community with fi nancial loss but also subjected them to trauma, Although the name may suggest that the May riots was an isolated event, the May riots has been closely linked with the fall of the then president Suharto. In local media reports, the May riots were reported to be part of a string of political events that started with the shootings of students during the Trisakti tragedy or Trisakti shootings. As the May riots covered a wide political, economic and social spectrum, books or media reports adopted slightly different emphasis on defi ning what constituted the riots.
In a report prepared by a government-sponsored team formed shortly after the riot, the event was defi ned more as a political crisis derived from the 1997 presidential elections although it also made a reference to the economic crisis that gripped the nation at that time. 30 The report pro- In his book on Indonesian politics, Michael Vatikiotis, a former journalist who was in the country during the event to cover for the now-de- There is some evidence that the capital city is rarely used as a name even if that is where the event is taking place or it is affecting the whole city or is of a scale that affects the government located in the capital city.
The May riots occurred mainly in Jakarta and also affected the government stability. Thus, using the name Jakarta riots in the way that Vatikiotis did, just appears to be a common-sense choice. However, the name Jakarta riots never took off. For readers living in a capital city, the name of the capital city is an internal premodifi er that does not provide a clear frame of reference when being used as names for events. They need a premodifi er or an identity that can be looked at externally and pointed at when using the name. As a premodifi er, it has to be an entity beyond the reader (not an entity where the readers are positioned) such as a landmark, a month or another city. refer to the events. The fi rst event was dubbed the Sampit riots while the latter is referred to as the Ambon riots. Racial profi ling is not encouraged in the media to avoid retaliation from both confl icting sides. By contrast, foreign media has identifi ed races or origins to refer to those confl icts in their news reports because it provides better background information for foreign readers. Foreign readers are less likely to have a personal interest and may have little possibility of involvement in the confl ict and so retaliation issues are not a major concern. While local media tries not to provoke anger in their news reports by choosing toned down names, the foreign media has used stronger names or categorization to describe the events. An example is the BBC, which referred to the Sampit riots as the Borneo massacre. 35 Owing to its proximity with the subject of the coverage, the local media may subject themselves to self-censorship. The use of the noun massacre is not popular for the similar reasons of avoiding retaliation. In the example of the Borneo massacre, the name Borneo is chosen because it is more familiar to foreign readers rather than Sampit, which is just a small city in the island of Borneo.

INVOLVED
Semantically, the noun riot in the May riots does not specify whether it involves two opposite parties, in a way that a noun such as fi ghting or violence would. Riot can simply point to a chaotic situation where a large number of people gather and do something violent towards something.
The Oxford dictionary defi nes riot as: a violent disturbance of the peace by a crowd (Pearsall 2013 Anti-Chinese riots are derived through a nominalization process. In the transitivity model, the nominalization process shapes the way information is presented by taking away participants, either agent (the doer) or recipient (the target). The process allows certain aspects such as perpetrator to be omitted from the sentence. The noun phrase May riots is the nominalization from sentences "The riots occurred in May 1998". The The linguistic choice becomes the structure of our reality. The use of the name May riots may not conjure up a graphic description unless one has been widely exposed to the event. Unlike anti-Chinese, the pre-modifi er May can be replaced with any other month without changing or infl uencing understanding of the event. The use of date as a pre-modifi er removes the visual hook with which readers establish cognitive perception.
In comparison, the name Jakarta riot may actually be more encapsulating as it provides an immediate vision of a riot in a capital city.
Fairclough states that analysing a text involves considering texts, processes of interpretation and their social conditions. 38 The event, conveniently framed with a month and year as the token, discourages readers from an interpretation that will pit the minority against the majority. This brings us to the question of whether the assignment of labels to an event is a well thought out practice or is it a matter of convenience? Does the media, subjected to a particular social condition, inadvertently reproduce social narrative that shields parties implicated from public scrutiny? 39 Does the production of the text lean towards de-escalation of a burgeoning confl ict as an effort to control the processes of interpretation on the readers' part? As far as deciding on a name is concerned, different media may refer to it with different names before they fi nally concur on a single name.
Although in some cases, the media may not come to any consensus on a name, each media outlet uses different names to point to a single event.
Speaking of choosing names to identify an event immediately after it breaks out, it is inconclusive whether the degree to which a pre-modifi er encapsulates the event is a priority in the process of deciding on names in the newsroom. Certainly, in the nominalization process, a pre-modifi er is chosen not because of how it can capture the event but more on how it might represent the aspect that is to be highlighted. Clearly May riots and anti-Chinese riots represent two different perspectives. Readers may not be able to locate discrimination aspects in May riots the way they do in anti-Chinese riots. Although anti-Chinese riots can capture the event better, is more visual and provides a better understanding to readers, it is not seen as a preferred choice because most likely, among other things, the identifi cation of the target may stoke further tension in the already volatile society and the media is well aware of the consequences of pushing forward with an already sensitive racial issue in society. This pattern is also demonstrated when the local media refrained from identifying the ethnicity in other racial confl icts in order to avoid backlash or retaliation.
However, the author contends that there could be a different treatment in identifying local Chinese as the target in the riots due to the long-standing history of being an easy target during political and social crises. Retaliation from local Chinese has never been reported and so dismissing the concern of stoking tension. Still, this leaves us with the question of the scope of dynamic that infl uences the decision to label the event May riots.
Is there any other sentiment at play that helps settle the name as such? Is there any pattern that dictates the process of selecting a name to identify an event?
A quick examination of the names that are used in the media for other crisis events reveals that choices generally range from those imbued with hype to those with an understatement; and from those with descriptive to those with symbolic qualities. There is no evidence that names are These examples illustrate the diverse landscape of naming practice.
When a particular situation occurs, the media will decide on their standpoint and come up with a name. Thus, each name refl ects the newsroom's judgment (or lack of judgment) towards an event. Although each name indicates the sorts of judgment news media have, there are no conclusive patterns of how a certain name is chosen over another alternative that could similarly represent the event in the same judgment. It is important to note that the media is also under the spell of social conditions, and it is not far-fetched to argue that they had internalized a self-censorship practice that will see the newsroom being discreet with what kind of identification they make in their news reports.
As far as the May riots is concerned, the linguistic choices have been strategically aimed at defl ecting readers from the discriminatory violence that occurred. The choice of word 'riot' is elusive as it does not imply any active confl icting parties as the word 'attack' does. The word 'riot' semantically suggests an unplanned chaos amidst confusion and misinformation among the mob.
However there remains an unsolved riddle as far as these mobs go.
Were they really just mobs taking advantage of the situation or were they part of orchestrated groups to create chaos? There is some anecdotal evi- Nossek et al. indicates that political violence will often involve a message that is wanted to be covered by the media, in a way that the media itself can be exploited to serve the interest of a certain group as a part of the political agenda itself. 43 The violence occurring in May 1998 warrants a massive coverage by the media but at the same time they also refuse to discredit any parties. Instead, the labelling of the event by the media directs readers' attention to a horizon in which the plight of the Chinese minority is toned down.
There can be little doubt that the creation and usage of names in mainstream texts are controlled by the media. The readers' role is limited at the receiving end. Readers construct their understanding based on news reports that have been interpreted fi rsthand by reporters, either favorably or not. Echoing what Lee said, the language is not a mirror of reality, but a tool to help us organize information and perception accordingly. 44 Language is a tool that assigns a phenomenon a certain conceptual structure.
To tease out the conceptual structure of the event of May 1998 is to factor in the anti-Chinese sentiment prevalent in the country, mainly prompted by wealth imbalance. Taking this into consideration, the plight of the minority was not a favourable cause to be advocated, especially when the civil unrest was perceived to be triggered by a prolonged economic crisis.
In this regard, the atrocities occurring in May 1998 were treated as just a noise amid a bigger political agenda that should receive the focused attention. To this day, victims have never seen a day of justice. Over the course of time, the governments have been subjected to pressure to settle the human rights violations occurring in the event, but it has largely concentrated on the justice for students killed during the Trisakti shooting.
The loss of the Chinese community and the acts of violence committed against them have rarely been the centre of attention. On their part, there is no evidence of the Chinese community rallying for justice either shortly after the event or years afterwards. For many, the pursuit of justice has never come to the fore considering the history of the Chinese minority becoming an easy target that goes back a long way. As they are no stranger to racial discrimination, the local Chinese have tried to steer clear of further trouble that may come if they push the government and the law enforcement with demands of justice.
The label of May riots refl ects both the ideology pushed by the government, perpetuated by the media to control the narrative of the event and disempower the local Chinese. At the end of the day, May riots were just some chaos occurring in May. As victims are hardly indicated and perpetrators are not clearly marked, the violence and the manipulative nature of the event were well hidden behind the generic name. Labelling the event May 1998 riots deceptively turns readers' attention from the main violence that was perpetrated by certain groups against 45 Fairclough,Language and Power,p. 24. others. The name riots suggest the event was an uncontrolled activity by a mob and there was no designated target. The label was heuristically adopted as it was the popular choice of name, and the interpretation of the event was internalized through the controlled social productions.

CONCLUSION
The use of date as a token in the name is a strategic labelling for controversial events in order to tone down the meaning they carry. May riots is a softer, less aggressive and more inclusive version of anti-Chinese riots, which is chosen by the media to project the event as being sporadic and not engineered towards a certain group. The use of the latter would prompt a pushback since the government has never confi rmed that discrimination took place during the riots. The approach adopted by the media resulted in a narrative that has played down the violence at the expense of the victims. The tone that is carried by May riots and the message it relays to the audience is simply restricted to acknowledge that riots occurred during May 1998. The label chosen by the media has inadvertently perpetuated the ideology pushed forward by the government. 46 While the label may be accurate in terms that the event did happen in May 1998, it strategically promotes the undisputable, neutral aspect surrounding it, leaving obscure the more sensitive nature of the event that calls for accountability.