The fragmentation of Yugoslavia in the ‘Balkan wars’ of the 1990s took place in the full glare of the international media spotlight. As successive republics declared their independence from the federation, conflict between different ethnic groups erupted over the right to self-determination. Journalists documented wide scale ethnic cleansing, city sieges, massacre, torture and rape across the contested territories and relayed these stories back to horrified and confused publics. It was one of the most dangerous conflicts that the media had ever been involved in – over seventy journalists were killed in the five year period of fighting, many were deliberately targeted (Times: 10th Aug. 1995a). Yet, journalists were also courted by all parties in an attempt to get international sympathy and support – Western states’ perceptions were crucial.

A four month-long visit to the former Yugoslavia woke my interest in the international portrayal of the region. Almost everything I encountered working with a peace building NGO and travelling around the region challenged my pre-formed perceptions. Many of these were drawn from the assumptions about the conflicts in the former Yugoslavia which are implicit in the term ‘the Balkans’. ‘The Balkans’ are synonymous with ethnic hatred, atrocities and fragmentation. We even have a verb ‘Balkanisation’ meaning to ‘divide (a region or body) into smaller, mutually hostile states or groups’ (Oxford English Dictionary, 2003, italics added). Glenny (2000: xxi) argues that these ‘generalisations about the peoples who inhabit the region, and their histories, were spread [during the 1990s] by media organisations that had long ago outlawed such clichés when reporting from Africa, the Middle East or China’. ‘The Balkans’ were not what I found in the former Yugoslavia.

Much of our perception of other countries is drawn from media representation; we subconsciously absorb attitudes and stereotypes which, cognitively speaking, organise our understanding of the world (Fowler, 1991: 17). The text that we draw our ideas from is itself shaped by the (culturally constructed) understanding of journalists and the constraints acting upon them. My experiences in the former Yugoslavia have led me to explore this area of representation and attitude formation. There are a number of theories which apply: those analysing cognitive processing of texts, the institutional impact on news collection, what makes ‘news’, and the analysis of the news text itself.

This study will examine these issues with regards to the conflict within Croatia’s current borders. I have chosen this conflict because, while there are many studies of press representation of the wars in Bosnia-Herzegovina and Kosovo, the only analyses which address the war in Croatia have examined primarily US press coverage. I have a personal interest in this particular conflict because I was based in Croatia on my visit, in a region devastated by the war. In addition, the war in Croatia was more limited than that in neighbouring Bosnia-Herzegovina and not quite as complex; these factors narrow the scope of the study. In order to conduct an analysis of the general themes of representation in the press, I will focus my study on ‘frames’.

Gamson and Mogdigliani (in Yang, 2003: 232) describe a frame as ‘a central organising idea or storyline that provides meaning to an unfolding strip of events, weaving a connection among them’. Framing is to select “some aspects of a perceived reality, to highlight connections among them, and thereby to make a particular interpretation and evaluation more salient than others” (Entman in Livingston and Bennett, 2003: 366). It is an essential element of any discourse; we cannot avoid selecting information for inclusion and exclusion, emphasising some elements over others, or explaining and organising things. Thus, ‘anything that is said or written about the world is articulated from a particular ideological position’ (Fowler, 1991: 10).

Theories of discourse analysis provide us with a number of tools which can be used for a detailed study of framing in the press. Frames can be found most obviously in the structures of news reports, the topics they cover and omit, the sources they use and quote, and the individual words and phrases that make up the report. These tools make it possible to compare across reports, which is important if general framing themes are to be discovered.

Through an analysis of contemporary news reporting, I wish to reveal these themes in the war on Croatia to see if the coverage favoured any particular position – support of the Croatian or Serbian governments or a particular strategy of intervention/non-intervention. Coverage of the conflict in the USA has been shown to have taken an anti-Serb stance throughout the wars in Slovenia, Croatia, Bosnia-Herzegovina and Kosovo (Kavran, 1991; Djordjevich, 1992; Vincent, 2000; Brock, 1993). This was also the case for a number of Western European papers (including British titles) in coverage of the Kosovo conflict (Nohrstedt, Kaitatzi-Whitlock, Ottosen and Riegart, 2000). My experience, as well as the articles referenced above, suggests that the particular demonisation of the Serbs was not fully deserved and I wish to see if this pattern emerges in British coverage of the war in Croatia. It will also be interesting to discover whether press coverage supports the position eventually taken internationally on the conflict. In order to limit the scope of the study, I will first come to a conclusion on what the most politically salient framing contests were and then examine the extent to which these are used in the press coverage.

This approach could be seen as a search for ‘bias’. However, in light of the fact that taking an objective position is impossible, this study will follow Fowler’s (1991: 12) lead in separating deliberate or cynical ‘bias’ from inevitable ‘skew’ – a much less judgemental and emotive term. Skew in reporting will be used to refer to broad patterns in which a frame or frames are consistently preferred.

The representation of Croatia and Serbia has become politically salient in recent years with their anticipated accession to the European Union. It is important that old stereotypes and judgements are re-evaluated in preparation for closer ties. This study will attempt to add to this process by examining the attitudes embedded in British press coverage. I will do this by using a qualitative discourse analysis of a selection of articles to identify what may be the main framing themes and to look at specific areas of skew. This cannot pretend to be a full analysis of these issues; indeed, that would be a virtually limitless task given the potential depth of analysis and the huge scope of reporting from the conflict. This study will, however, seek to identify and introduce the main issues regarding framing of the conflict and suggest some trends and areas of skew in the coverage. I hope to be able to suggest important areas for future study as a result of this work.

To clarify, my research question is: what were the major framing contests in the war in Croatia and how were these represented in the British press? This needs to be broken down into a number of areas. Firstly, I will present an outline of the war in Croatia, which is necessary in order to understand later arguments. I will then go on to address the influences on journalists’ framing of the conflict in Chapter 2. In particular, the role of propaganda and of different interpretations of the conflict will be discussed. This, along with historical and theoretical knowledge of the conflict will allow me to distinguish the most politically salient framing contests in the conflict. Chapter 3 will discuss the methods of analysis to be used in examining articles; these will be drawn from theories of discourse analysis. It will also include an explanation and outline of my methodology. Chapter 4 will present the framing patterns and trends within the selected articles and I will conclude with an examination of the wider implications of these findings and the identification of areas for further study.

 

Figure 1. The contested territory, 1991-1995 (Wikipedia, online)

Pink = Croat-held areas, Red = Serb-held areas (The Serbian Republic of Krajina)

The irony in attempting to describe the war in Croatia at the start of a report about representative skew is clear. However, a basic level of understanding is necessary so I will attempt to sketch the outline of the conflict.

The war in Croatia was the second secessionist conflict in Yugoslavia after Slovenia’s. Yugoslavia was a multi-ethnic federation of republics under a communist government led by the president-for-life, Marshall Tito. Many reasons have been given for the break up of the federation but these need not be addressed in this study. Briefly, the conflict can be directly linked to the rise of exclusivist nationalisms and the breakdown of the structures formerly used to resolve ethnic conflict (Duffy and Lindstrom, 2002; Pavkovic, 2000). After Tito’s death few leaders had strong reasons for loyalty to the Yugoslav state and, in the midst of severe economic crisis, promoted and encouraged divisive nationalisms to maintain their legitimacy within the different republics (Pavkovic, 2000). The growing influence of the Serbian president, Slobodan Milosevic, in federal affairs was seen as a threat by both Slovenian and Croatian elites (Glenny, 2000).

The war in Croatia could be said to have began in 1990 with the election of the Croatian Democratic Party, the HDZ, to government in the Croatian Republic. Elected on a nationalist, secessionist mandate, the HDZ, led by President Franjo Tudjman, began to create an armed force independent of the Yugoslav National Army (JNA) and to push for increased sovereignty, if not outright independence from Yugoslavia (Pavkovic, 2000: 122, 94). Ominously, it also proceeded to remove Serbs from influential positions, delete the special status of the Serb minority from the Croatian constitution, and reinstate historic Croatian symbols linked to the World War II genocide of Serbs (ibid.). In January 1991, finding themselves likely to be a national minority in a seemingly hostile state, the group of Croatian municipalities with Serb majorities or near-majorities declared their independence from the government of Croatia and their intention to remain part of Yugoslavia in the event of Croatian secession. They formed two political blocks, the Serbian Autonomous Region of Krajina and the Autonomous Region of Slavonia, Baranja and Western Srem. These were later united as the Serbian Republic of Krajina, which will be referred to as ‘the Krajina’ throughout.

The attempted secession of the Krajina Serbs from the Croatian republic sparked the transition to violent conflict as the Croatian police attempted to regain control of Serb-held areas and Serbian militias resisted. The violence was sporadic and localised but was accompanied by intensive propaganda campaigns in both the Croatian and Serbian media. These spread ‘fear and hatred’ by highlighting each other’s potential for genocide, based on Second World War experience, and led to the beginnings of ethnic Serb and Croat refugee movements into and out of the Krajina, respectively (Kaldor, 2001; 8).

Ostensibly to keep the peace, the Yugoslav presidency (by this time largely controlled by Serbian interests (ibid.: 39)) sent the JNA to separate the factions. This soon came to be seen by the Croatian government as a move to support the Krajina Serbs and prevent the Croats from establishing control over the disputed territory. When Croatia declared its independence from Yugoslavia on the 25th June 1991, the status of the JNA within Croatia was further compromised. On the 22nd August, Tudjman declared that, if it did not stop aiding the Croatian Serbs and disarm them, the JNA would be considered an army of occupation (Pavkovic, 2000: 143). The mobilisation of the newly-formed Croatian National Guard and the blockade of JNA barracks within Croatia duly followed.

Despite a desire to protect the integrity of the Yugoslav federation by preventing Croatia’s secession, the JNA high command were forced, by strong resistance in Croatia, Serbia and Montenegro, and in the Yugoslav Presidency itself, to restrict their efforts to only parts of Croatia (Pavkovic, 2000: 144). The ensuing war, fought by the Croatian government, the JNA, Croat and Serb militias, and paramilitary groups on both sides, was therefore territorially restricted to existing Serb-held areas and their surroundings, and to strategically important areas for the remaining Yugoslav state. The frontline stabilised in October 1991 as each side consolidated their gains (Pavkovic, 2000: 145) and, in December, the UN was able to negotiate a ceasefire, the withdrawal of the JNA, and the deployment of Peacekeepers to the contested territory (see Figure 1.). By this time, refugee movements had come close to homogenising the populations of Croatia and the Krajina along ethnic lines (ibid.: 146).

The outbreak of war in Bosnia-Herzegovina in April 1992 complicated the political situation in Yugoslavia; Croatia and the Krajina Serbs both became heavily involved in the conflict but, perhaps partially as a result, this left the situation in Croatia and the Krajina relatively stable until May 1995. At this point, safe in the knowledge that the EU had decided that it could not recognise the secession of the Krajina Serbs and, supported by the US, the Croatian government decided to ignore the UN and regain control over the Krajina. Western Slavonia was taken in a 48-hour period at the start of May 1995 in which time almost the entire Serb population fled to Bosnia and Serbia (Pavkovic, 2000: 153). In a much larger offensive three months later named Operation Storm, Croatian forces overran most of the remaining Krajina in just four days, creating between 150,000 and 200,000 Serbian refugees – at the time, by far the largest refugee movement of the Yugoslav conflicts (Glenny, 2000: 650). The Krajina Serbs’ resistance to these moves had been severely constrained by the Serbian President, Milosevic, through loyal commanders in the Krajina forces who were ordered to withdraw and evacuate. This abandonment of the Krajina Serbs was a condition of the partial lifting of UN sanctions on Serbia (Pavkovic, 2000: 154).

The conflict was finally ended with the UN transfer of Eastern Slavonia, the only remaining area of the Krajina, to Croatian control in 1998. By this time, Croatia’s population was 94% Croat and it was one of the most ethnically homogeneous states in the region (Pavkovic, 2000: 154).

Chapter 2: Framing and Propaganda

To identify the main framing contests it is necessary to examine the influences on journalistic framing of the war in Croatia. In this chapter, I will argue that framing is unavoidable in any form of communication and that news reports, through selecting and interpreting ‘reality’, must necessarily present an ideologically skewed representation of events. This selection and interpretation is specifically driven in news by ideas of what is newsworthy and by the influence of certain powerful interests in society. In addition, the war in Croatia may have presented significant problems for journalists as it was complex, ambiguous, and because the parties to the conflict conducted extensive propaganda campaigns.

I will examine the political significance of framing the war in Croatia and identify four frames that will form the focus of my analysis of news reports. The warring parties perceived the media as important in influencing international governmental responses which, in turn, proved to be very significant in determining the outcome of the war. There were two main areas of contention in which the international adoption of different frames had the potential to promote radically different approaches to the conflict. The first was the struggle for international support between Croat and Serb authorities, and the second, raised by Mary Kaldor (2001), was the contest between two opposing views as to the nature of the conflict. It is in these contested areas that the skew of news reports was most relevant.

Reasons for Framing

Faming is unavoidable. If a frame provides the ‘meaning’ to a set of events (Gamson and Modigliani in Yang, 2003: 232) or makes them ‘comprehensible’ (Fowler, 1991: 17), it is absolutely fundamental to effective communication. However, a frame represents one interpretation of ‘reality’ and must necessarily marginalise others, thus the choice of frame is a political act. ‘Anything that is said or written about the world is articulated from a particular ideological position: language is not a clear window but a refracting, structuring medium’ (ibid.: 10).

It is possible to identify some of the manners in which frames are used specifically in relation to the reporting of news. If we understand that any set of events can be framed in numerous ways, it follows that journalists will frame events to make them ‘newsworthy’ and likely to be published (ibid., 19). Galtung and Ruge’s seminal article, ‘Structuring and Selecting News’ (1973), is perhaps the best analysis of the influences on this process of selection and framing in the Western press. They identified twelve factors:

(F1) frequency

(F2) threshold

(F2.1) absolute intensity

(F2.2) intensity increase

(F3) unambiguity

(F4) meaningfulness

(F4.1) cultural proximity

(F4.2) relevance

(F5) consonance

(F5.1) predictability

(F5.2) demand

(F6) unexpectedness

(F6.1) unpredictability

(F6.2) scarcity

(F7) continuity

(F8) composition

(F9) reference to elite nations

(F10) reference to elite people

(F11) reference to persons

(F12) reference to something negative (Galtung and Ruge, 1973: 62-72)

These factors describe what is perhaps common sense; F2, for example, tells us that events must be of a certain size to be included, and F4, that things occurring in places similar to ‘us’ or that will affect this country directly are given higher priority than others. The greater the number of factors that can be satisfied by the journalist’s frame of an event, the more newsworthy it becomes. Thus there is significant (although perhaps unnoticed) pressure on journalists to frame events in certain ways which will be acceptable to their editors and of interest to their readers.

Rock (1973: 73) argued that the types of news we receive are also ‘the result of an organised response to routine bureaucratic problems.’ The necessity of having enough news for daily publication, for example, leads journalists to constantly monitor certain sources such as parliament, the police, courts, and royalty. These institutionalised sources can be very useful and provide news at suitable intervals (in accordance with criterion F1) but will frame events according to their interests – those of the dominant forces in society (Fowler, 1991: 22). Organised press conferences also ease the pressure on journalists yet only people or organisations with money and influence are able to present their perspectives and influence news frames in this way (Manning, 2001). The use of official sources may, in addition, be sought by journalists in order to emphasise their ‘objectivity’ and to lend legitimacy to the report (ibid.: 68-9).

Dominant interests in society also affect journalists because the mainstream newspapers are ‘first and foremost businesses’ and are intimately tied to the capitalist economy (Sparks, 1999: 45). It is not in their interest to print reports that would undermine their capitalist interests or that would not appeal to the public (thereby damaging sales). Thus, more radical views tend to be marginalised or excluded (Fowler, 1991: 20). Market competition among similar papers, such as that which occurs among the broadsheets in the UK, leads to the need for ‘product differentiation’. A small but significant element of this differentiation is the intentional political bias of the paper (Sparks, 1999: 49). While this may not be enforced or made explicit to journalists, Breed (1999) has demonstrated that the social pressure to conform is high.

By examining the factors influencing the framing of events in the press, it becomes clear that the news will always be skewed, usually towards ‘common sense’, powerful interests, and the status quo (Manning, 2001). However, some situations lead to a greater skew in reporting than others. The war in Croatia may be an example of just such a situation.

Factors Influencing Framing of the War in Croatia

Two factors mark out the war in Croatia as particularly difficult to report – the barrage of propaganda from different parties seeking to impose their interpretation of the conflict, and the complexity and ambiguity of the situation, which made the avoidance of propaganda more difficult. For the purpose of this study propaganda will be understood as ‘the deliberate and systematic attempt to shape perceptions, manipulate cognitions, and direct behaviour to achieve a response that furthers the desired intent of the propagandist’ (Jowett and O’Donnell, 1992: 4).

Official propaganda campaigns were certainly ‘deliberate and systematic’ (Poggioli, 1994; MacDonald, 2002) and their ‘desired intent’ was international acceptance of their victim status and thus the legitimacy of their struggle. Journalists were faced with huge quantities of propaganda from both governments and populations as everyone responded to the press with their vested interests in mind (Nohrstedt et al., 2000: 384). Lord Owen (Co-Chairman of the Steering Committee of the International Conference on the former Yugoslavia) claimed that “[n]ever before in over thirty years of public life have I had to operate in such a climate of dishonour, propaganda and dissembling”. (quoted in Malesic, 2000: 25). While journalists are frequently capable of recognising propaganda, the complexity and ambiguity of the conflict may have increased journalistic susceptibility to manipulation.

Galtung and Ruge’s (1973) third criterion for news selection is ‘unambiguity’; overly complex or ambiguous events are not generally considered newsworthy. How then are journalists to explain in simple terms the reasons for the conflicts in Yugoslavia or the complex legal issues surrounding who has the right to secede from who? Framing these issues in a way to fit selection criteria and appeal to newspaper editors might require an oversimplification which skews the reporting towards a political stance. Where territory is contested, describing actions simply, as ‘aggression’ or ‘defence’, may involve legitimising one party at the expense of another.

The pervasiveness of propaganda aimed internally to mobilise populations around nationalist goals added to the complexity of the situation. The concept of ‘objective’ reporting was confounded by the apparent merging of propaganda with reality; events were frequently driven by what people believed to have happened (often propaganda) rather than any kind of ‘reality’ (MacDonald, 2002). The propaganda campaigns were such that Malesic (2000) was able to identify significant psychological effects on those exposed to it. His study found clear evidence of ‘mass psychosis’, where people chose to hear only that which confirmed their already polarised view, and of addiction – people ‘demanded a daily dose of threats, hatred, hostility and even fear’ (ibid., 27). One Western observer claimed that “It was as if all television in the USA had been taken over by the Ku Klux Klan” (Malcolm in Malesic, 2000: 25-26).

The salience of propaganda and perception in this very ambiguous conflict suggests that there may have been problems for those attempting to report the ‘facts’, and raises the possibility that propaganda was unintentionally transmitted by journalists.

The Role of the Media in the Conflict

The extent of the propaganda campaigns in the war in Croatia hint at the importance accorded to international opinion. In addition to the standard methods of reaching the press through briefings and statements, both the Croatian and Serbian governments attempted to hire public relations firms (only Croatia was successful) (Pavkovic, 2000: 147; Times: 18th Oct. 1991); local populations and diaspora communities mobilised for huge, ‘spontaneous’ propaganda campaigns (MacDonald, 2002: 210-1); and the Croatian government went as far as to deliberately draw fire to the historic ramparts of Dubrovnik (Pavkovic, 2000: 147). International opinion was seen to be important in two broad areas: whether or not the major powers would intervene to support Croatia; and whether nationalist leaders and their ideas would be granted legitimacy and influence in negotiations.

That so much of this propaganda was targeted at journalists demonstrates a belief that the framing of news would affect the actions of governments. It is commonly argued that the media is very influential in shaping public opinion which, in turn, may force governments to act in a certain way – the so-called ‘CNN effect’. It is not the intention of this study to examine the debate over the accuracy of this belief as the effects of reporting will not be addressed. However, understanding the political salience of framing makes it possible to identify the most important frames to search for when analysing articles.

It is possible to identify two major areas in which there were more than one means of interpreting events, and where the privileging of one interpretation by the major powers had fundamental impact on the course of the war. These can be described as ‘framing contests’. These two areas of debate will be fundamental to the analysis of news frames in Chapter 4.

Framing Contest 1: ‘Serb’ vs. ‘Croat’

The (externally oriented) propaganda war between Croat and Serb positions centred on the issue of international involvement. The model of secession aimed for by Slovenia and Croatia (successfully) and Bosnia-Herzegovina (unsuccessfully) relied very heavily on international influence to neutralise the opponent, in Croatia’s case the JNA (Pavkovic, 2000: 140). As it could not win a direct conflict with the JNA, victory was to be achieved ‘by restricting [the JNA’s] movements through international monitoring and then, by international mediation or pressure, to effect its peaceful withdrawal’ (ibid. 155). International recognition of Croatia’s borders to include the Krajina was also essential to legitimise the Croat claim to the region.

The strategy for obtaining international involvement was to appeal to Western states, Germany and the US in particular, by presenting the Croatian struggle as a war of national liberation from Serbian dominance and oppression (a position vigorously refuted by Serbs). In addition, Croatia’s European identity was stressed and distanced from the brutal, Eastern Serbs (Glenny, 2000: 637; MacDonald, 2002: 211). To do this, Tudjman travelled extensively seeking sympathy from Western governments and mobilising diaspora groups in support of his cause. The public relations firm Rudder Finn was engaged to present Croatia as a victim of aggression, and tales of destruction and genocide were used to create the ‘correct’ impression in the press (Pavkovic, 2000: 147).

The Serbs used propaganda in much the same way as the Croats but with the intention preventing Croatia’s secession entirely, or at least preventing it taking majority-Serb areas if it did secede. Like the Croats, a history of suffering was drawn on and particular emphasis placed on the Croatian regime’s genocide of Serbs during the Second World War (MacDonald, 2002). The secession of the Krajina and JNA intervention on its behalf was justified in humanitarian terms, by the claim that this genocide was about to be repeated by Tudjman’s government (ibid.: 201-2). Croatia’s claim to the Krajina territory on the basis of its Yugoslav borders was refuted. These borders were seen to be merely administrative and the right to self-determination by the majority Serbs living there was championed (ibid.: 188-9).

The more sophisticated Croatian strategies ultimately won the day and Germany successfully took up the cause of Croatian independence in the EU. In November 1991 the EU decided that, legally, the existing borders of the six Yugoslav republics were international borders as the federal state was ‘in the process of dissolution’ (Pavkovic, 2000: 149-150). Independence was thus recognised for Croatia but denied to the Krajina Serbs. With this simplification of the territorial ‘rights’ in the region, the UN ‘neutralised’ Croatia’s enemy by negotiating JNA withdrawal in return for a peacekeeping force (ibid. 156). In a similar strategy in 1995, Croatia secured the support of the US government for its offensive against the Krajina. Much of the US media accepted Croatian arguments and presented the conflict as between Croatian capitalism/democracy and Serbian communism/totalitarianism (Kavran, 1991). In addition, the Croatian diaspora in the US was large, active and supportive of the nationalist cause. US training, logistical support and equipment was vital in allowing the Croat forces to take the Krajina before any UN action could prevent it (Pavkovic, 2000: 156).

Framing Contest 2: the Nature of the Conflict

Whilst international perceptions favoured Croatia in the conflict, Kaldor (2001) stresses that all nationalist parties benefited in some manner from the portrayal of the nature of the conflicts in the former Yugoslavia. In order to make clear the benefits gained by nationalists, she presented two different ways in which the conflict could be framed. The first is the commonly held view that the conflict was between different ethnic groups with a history of hatred and violence and who were fighting over the control of territory. The high levels of civilian casualties and refugees were by-products of this dispute, and a territorial settlement was necessary to divide the groups and end the fighting.

The alternative view frames the conflict in an entirely different way. Recognising the very similar aims and tactics of the competing nationalist movements, Kaldor described a conflict of exclusive and brutal ethnic nationalisms against civil society and cosmopolitan humanitarianism. The nationalist groups had in common a desire to control a given territory and population and to be recognised as the legitimate leaders of their nations. The general populations were the main targets of attack in this reading of the conflicts; they were forced, by the ‘manufacture’ of fear and hatred, to abandon civil, inclusive values and to reorganise themselves into ethnically homogeneous territorial units. The problem, then, was of ‘political and social organisation’ (ibid.: 59) not territory. The solution? To reject the legitimacy of nationalist leaders, protect the civilian population, and actively support cosmopolitan forces within civil society.

The apparent marginalisation of this second framework for understanding the war can be seen to favour the nationalist parties. ‘The international community fell into the nationalist trap by taking on board and legitimising the perception of the conflict that the nationalists wished to propagate’ (Kaldor, 2001: 58). Tragically, ‘the only possible solution [within this framework] was one which accepted the results of ethnic cleansing’ (ibid.: 59).

It is not the intention of this report to make any judgements on the relative merit of these different interpretations. Rather the aim is to distinguish the most politically salient aspects of the struggle for representation in the media. The success of each frame will be examined in Chapter 4.

Chapter 3: Discourse Analysis and the Selection of Articles

This study is only a preliminary examination of an issue that could be virtually limitless due to the extent of the reporting and the depth in which each text can be analysed. I will use a sample of articles and a qualitative discourse analysis to suggest some of the framing themes and areas of skew that may be significant in the reporting as a whole.

Analytical Approach

I will look for framing in the British press using discourse analysis; a cross-disciplinary area of study which seeks to identify, understand and explain the manners in which people interact with texts (a broad term referring to spoken, written and visual communication). Discourse itself can be defined as ‘a complex communicative event that also embodies a social context, featuring participants (and their properties) as well as production and reception processes’ (van Dijk, 1988b: 2). News is, therefore, a discourse embodying a certain national and socioeconomic culture with active participants producing it (journalists working under the framing conditions discussed above) and interpreting it (the readership).

News analysis can be considered a sub-category within discourse analysis. Theorists have applied the principles of analysis to newspaper articles and explored the specific structures, meanings and techniques which are used in the communication and comprehension of news. Central to this field are van Dijk (1988) and Fowler (1991), both of whom will be drawn on heavily. A number of tools are available in discourse analysis for a measured and detailed study of framing in the news. The most prominent of these will now be discussed.

Semantic Macrostructures

The most obvious place in which frames can be identified is the semantic macrostructure. This ‘makes explicit the overall topics or themes of a text and at the same time defines what we could call the overall coherence of a text as well as its upshot or gist’ (van Dijk, 1988b: 13). It is at this level that we can instinctively grasp elements of the frame through describing what the text ‘is about’. We derive the semantic macrostructure, not from any explicit source, but by adding up and generalising the principal elements in order to reduce the entire text to one or two sentences (van Dijk, 1988a: 32). This takes place at a number of levels, allowing analysts to describe the macrostructure as a tree-like formation made up of micro- and macropropositions which are drawn together into increasingly large generalisations (ibid.: 33).

In cognitive terms, the macrostructure has a particular importance as it is the way in which we store the information in a text. After reading a news report, most people remember only the most generalised levels of the macrostructure and this will form the basis of their impression of the event (van Dijk, 1988b: 14). The process of deriving the macrostructure is subjective (within the limits of the text) and depends on the importance accorded to information by each reader. However, the concept is still useful, particularly when linked to the schematic superstructure.

Schematic Superstructures and Relevance Structuring

The schematic superstructure is the ordering or structuring of the text which makes it easily comprehensible. There is a specific form of superstructure for a news report which allows us to recognise what it is before beginning to read. Articles are organised in categories beginning with the headline, a lead sentence, and then the body of the article. The body is made up of a number of sections, for example main events, context, opinions, and predictions. The most generalised levels of the macrostructure are usually present in the headline and lead meaning that ‘in principle the beginning of the text always contains the most important information’ (van Dijk, 1988b: 15).

The rule that important information will be schematically privileged in newspaper reports is known as relevance structuring. It is applicable to all levels of a text, from macro- to micro-. Information at the start of paragraphs or sentences tends to be accorded greater importance by the reader because of their (culturally derived) schematic expectation (ibid.: 16). Relevance structuring holds an important framing element because information the reporter believes is salient can be privileged.

Topics

The choice of topics for inclusion in a newspaper or a single report is an important element of framing (Yang, 2003). Journalists cannot report on everything so only certain topics will be selected; this then determines what readers can know about the situation. The omission of topics can lead to heavily skewed reporting. For example, Chomsky highlighted a significant flaw in US coverage of the Kosovo conflict in which very little mention was made of the Serbian parliament’s peace proposal prior to NATO bombing (in Nohrstedt et al., 2000: 399). Nohrstedt et al. confirm that the same was true for British and Norwegian papers with only a very brief mention being made in the Swedish press (ibid.). The omission of this information can be said to have strengthened NATO’s justification for war.

Sources

A high proportion of news reports contain quotes from or references to statements made by officials. This results from the routine practice and the need to provide evidence, discussed in Chapter 2. Livingston and Bennett (2003: 366) state that:

‘we expect that the presence of officials in news coverage matters, particularly when credible officials take strong policy positions in the media, and the stakes seem important for domestic peace, economic prosperity, or for easing moral outrage in extreme cases of ethnic cleansing or other human suffering.’

In other words, official voices in the news can be very influential. Their inclusion often implies authority and truth as well as giving institutions an opportunity to present their frame on events (Manning, 2001). The choice of officials can heavily influence the portrayal of events. For example, Yang (2003) demonstrates that American and Chinese journalists looked to almost entirely different groups of officials in their coverage of the Kosovo conflict. This played a significant role in their differing frames.

Non-official sources can also be important in framing. Galtung and Ruge’s (1973) 11th ‘news value’ is ‘reference to persons’, which Fowler (1991: 16) describes as an ‘obsession’ of the press. Individuals may be used as symbols of a wider group or set of beliefs allowing the narrative to be heavily simplified (ibid.: 15-6). Their inclusion can also ‘promote straightforward feelings of identification, empathy or disapproval’ (ibid.: 15). The disproportionate use of public voices promoting any particular frame can lead to skew in reporting.

Keywords, Scripts and Metaphors

At the micro-level of words and phrases there are again ways in which frames are defined. A keyword can be any word that is important in labelling or judging information/events/actors (Yang, 2003). Keywords can influence the frame when they are particularly strong or when a number of words reinforce each other. They often act as ‘consensual values’ meaning that they have a social judgement implicit in them (Fowler, 1991: 51-2). Keywords can be part of ‘scripts’. These are culturally shared ‘clusters’ of knowledge ‘which contain all we know in our culture about a specific stereotypical type of episode’, including judgements and ‘social attitudes’ (van Dijk, 1988b: 13). We have scripts for having a picnic, gang violence, clubbing, and every other stereotypical occurrence. Keywords and scripts may also be used as metaphors. Van Dijk (1988b: 16) uses the example of a demonstration being described in military terms, thus activating a script of violence. Because of the judgements inherent in our language, whether used metaphorically or not, keywords and scripts have an important ideological role.

Rhetoric

Finally, rhetorical structures are those ‘special operations’ which ‘are used to make the text more persuasive’ (van Dijk, 1988b: 16). These include alliteration, repetition, rhyme, parallelisms, exaggeration and understatement. They draw attention to, and help the reader to remember, certain elements of the text (ibid.). Where this emphasis is used it can have significant effect on framing.

Taken together these elements of a text can indicate the frame and the ideological skew. Analysis of these will therefore be applied to a selection of articles written on the conflict.

Selection of Articles

To limit the scope of this project it was necessary to examine only a sample of reports. Initial research using LexisNexis Professional showed that stories on the war in Croatia were regularly carried in only three British daily papers, The Times, The Guardian and The Independent. Of these I selected two for comparison – The Times and The Guardian, as these are generally considered to have quite different ideological stances. Sparks (1999) argues that the British press are ‘obviously and stridently partisan’ and that their different political stances act as a means of product differentiation. The Times and The Guardian are in competition with each other in the ‘quality’ market (ibid.: 47); The Times tends to take a more conservative or right-wing view while The Guardian targets liberals and the left. However, McNair stresses that, at the time of the war in Croatia, the British press as a whole had a ‘pro-Tory bias’ (1999:59).

This ideological element is an interesting aside to my main focus: If differences in coverage of the war appear between the two papers it may suggest that ideological and institutional factors have a strong impact on journalistic framing; If they do not, wider cultural factors and possibly propaganda may be the most salient factors.

The articles have been selected in LexisNexis Professional using the search terms ‘Croat’ and ‘Serb’. This search highlights 129 articles, approximately half in each paper. The search terms do not bring up all relevant articles and, indeed, using the LexisNexis system it would not be possible to discover all articles on the conflict. As a result of this I will make no pretensions to quantative analysis and will treat the articles merely as a selection from the reporting. Although it may be necessary to stress that ‘most’, ‘many’, ‘a few’, ‘all’ or ‘none’ of the articles suggest certain framing patterns, this will be done with caution.

The articles have been selected in two months during the war – October 1991 and August 1995. These months have been chosen because they contained the height of the fighting and because they are roughly symmetrical in events. October 1991 saw a substantial part of the JNA’s campaign to gain control over strategic points bordering the Krajina as well as the full mobilisation of the Croatian forces on the 7th. The sieges of Dubrovnik and Vukovar continued throughout the month. August 1995 saw something of a reversal with the Croat forces launching Operation Storm on the 4th and the Serbian refugee exodus dominating the coverage for the rest of the month.

In many ways these periods are comparable. Both begin with relative peace but see a rapid escalation of events within the first week – the beginning of the JNA bombardment of Dubrovnik in 1991, and the launch of Operation Storm in 1995. Both also cover attacks on civilians, ethnic cleansing, and significant refugee movement, firstly of Croats and latterly of Serbs. In addition, both include a number of peace talks and broken agreements.

There are a number of evident problems in selecting articles and I make no pretence of overcoming these. However, by using broad search terms and focusing on two months in which different parties were on the offensive and defensive, I believe that the selection of articles will be sufficiently representative for a qualitative analysis to suggest general trends.

Organisation of Analysis

I will approach the articles by theme following the example of Philo and Berry (2004). This involves the selection of a number of themes believed to be important and an analysis of how these are presented in texts. For example, in Philo and Berry’s analysis of coverage of the Israeli-Palestinian conflict they selected the reporting of casualties as one theme and used techniques of discourse analysis to demonstrate a significant skew towards Israeli perspectives (ibid.: 144-156). I will select themes based on their relevance to the framing contests identified above between ‘Croat’ and ‘Serb’, and ‘ethnic-territorial’ and ‘nationalisms vs. civil society’ perspectives.

The first theme will be the reasons for the conflict. As Philo and Berry (2004: 258) have demonstrated, the way in which reasons for conflict are presented in the news can have a significant impact on audience attitudes and opinions. In this case I will first look for the inclusion of explanations, something often lacking in news because long and complex processes do not fit easily into the pattern of daily news production – Galtung and Ruge’s (1973) factor F1 (Fowler, 1991: 14). The omission of explanations may favour one or other party (the one with a weaker justification), or it may just give a sense of meaningless to the conflict.

Where explanations are included, the types of reasons are important. References to the increased influence of Serbian nationalism in Yugoslavia or the Croatian nation’s desire for self-determination would favour the ‘Croat’ frame, whilst the threat of Croatian nationalism and the right of the Croatian Serbs to self-determination would favour the ‘Serb’ frame. Similarly, explanations emphasising ethnic hatred or historical conflicts would support ‘ethnic-territorial’ perspectives and those describing elite manipulation and desire for power would favour ‘nationalisms vs. civil society’.

The second theme will be the portrayal of leaders. Differing descriptions of Croat and Serb leaders could support either the ‘Serb’ or ‘Croat’ frame. For example, a significant element of skew in reporting the Kosovo conflict was the unparalleled demonisation of Milosevic (Nohrstedt et al., 2000). The mention of leaders is also a crucial element of the ‘nationalisms vs. civil society’ frame. Attributing responsibility to the leaders and separating them from society in general is central to the presentation of these two groups as conflicting.

Thirdly, I will look for the inclusion or omission of references to propaganda in the text which will be important in determining the nature of the conflict. If propaganda is linked to people’s actions, their responsibility can be seen as reduced and blame is more focused on governments and leaders. The failure to mention propaganda in the text reinforces the dominant ‘ethnic-territorial’ view.

The fourth theme will be the extent of empathetic access to the actors. Galtung and Ruge’s (1973) 11th news value, ‘reference to persons’ tells us that events can be easier to understand when described through the experiences of individuals (Fowler, 1991: 15). We are much more able to empathise with one person’s story than with more abstract descriptions of events. The power of ‘personalisation’, as it is known (ibid.), influences our attitudes towards different actors and can be used in a skewed manner, allowing empathetic access to one side only. This theme may be found to support either the ‘Croat’ or ‘Serb’ frame.

The fifth theme will be the mention of violence. The violence of their opponent, in particular their genocidal tendencies, was key to the propaganda campaigns of each side. Reporting on this theme may provide the clearest demonstration of acceptance or rejection of this idea and may support the ‘Croat’ or ‘Serb’ frames.

Finally, I will look for the treatment of voices which dissented from the nationalist agendas. When these are included they might provide evidence for the civil society of the ‘nationalisms vs. civil society’ frame and undermine the idea of the homogeneous ethnic groups in the ‘ethnic-territorial’ frame.

The sample of reports will be examined for the inclusion (or absence) of each of these themes and the general framing trends within them. These trends will be found through the analysis of discourse outlined above and then be demonstrated through examples.

Chapter 4: Analysis

This chapter will draw together the tools of discourse analysis and the identification of politically salient framing contests to analyse the trends in the selection of articles. This will be done under six themes: reasons given for the conflict, description of leaders, references to propaganda, empathetic access to the actors, mention of violence, and treatment of dissenting voices.

Reasons Given for the Conflict

The reported reasons for the conflict in this sample can be broadly split into two groups; those which explain events by history and ethnic hatreds, and those that attempt to address the ambitions of each side and the fear and discrimination amongst the public. The former are almost exclusively found in The Times and the latter in The Guardian. However, these issues are only rarely mentioned in either paper and, where it is included, the background is often mentioned very briefly or partially, such as the use of the phrase, ‘the broader goal of achieving a “Greater Serbia”’, in a Times article (18th Oct. 1991b).

Those articles emphasising ethnic hatreds or historical causes fit closely with the dominant ‘ethnic-territorial’ frame described in Chapter 2. It can be summarised as the idea that

‘The Balkans … has always been characterised by ethnic divisions and rivalries, by ancient hatreds that persist just beneath the surface. These divisions were temporarily suppressed during the communist period, only to burst forth again in the first democratic elections.’ (Kaldor, 2001: 34)

This idea returns again and again in The Times and is the primary reason given for the conflict. For example, a report from the 9thTimes: 11th Aug. 1995a; 8th Aug. 1995c; 4th Aug. 1995c). One article even claims that ‘One can see the conflict as a prolonged civil war which started in 1914 at Sarajevo’ (Times: 3rd Aug. 1995b). The mention of history or ethnic hatreds in a number of headlines (Times: 9th Oct. 1991, 21st Oct. 1991, 8th Aug. 1995c, 9th Aug. 1995c, 11th Aug. 1995a) schematically emphasises this theme by placing it at the centre of the macrostructure. August 1995(c) describes the Krajina as: ‘an historic faultline that has long been both a buffer between worlds and a patch of territory fought over by empires now long gone.’ Other articles mention the depth of ‘Balkan ethnic hatreds’, ‘the 1,000-year-old dream of uniting Croatia into a nation state’ and the emergence of Serbian ‘Ancestral fears of being dominated by the Croats’ (

The above approach is barely mentioned in The Guardian, which chooses to emphasise very different factors. In particular, the conflict over the Krajina is explained by the Serbian population’s fear of living in a Croatian nationalist state. One article (Guardian: 26th Aug. 1995) legitimises this fear by taking the unusual (although not necessarily incorrect) stance of referring to ‘Croatian ethnic cleansing in 1991’ as the factor driving Serbian actions. A number of articles also mention (without explicitly supporting it) the Serbian government’s position that Serbs in Croatia should have the right to self-determination if the Croats do (Guardian: 12th Oct. 1991a; 15th Oct. 1991; 23rd Oct. 1991b). The Times also covers Serbian motivations. As well as noting the fear of the Krajina Serbs (Times: 4th Aug. 1995c, 8th Aug. 1995a), it mentions the desire for a ‘Greater Serbia’ a number of times (Times: 11th Oct. 1991, 18th Oct. 1991, 25th Oct. 1991). This latter explanation, expansionism, tends to be considered illegitimate in our culture so appears to support the ‘Croat’ position that the Serbs were in the wrong.

Explicit reasons for Croatia’s secession are largely absent from this sample. The Guardian (7th Aug. 1995f) makes only one mention of the centralisation of power around Milosevic in the final years of Yugoslavia which formed part of the reason for Croatian independence moves and The Times does not address this issue at all. While this appears to damage the Croatian cause, that may not be the case – independence for Croatia appears to be an assumed right. For example, ‘Glina [a Serb-held town] had already been liberated’ from the Serb Krajina (Times: 5th Aug. 1995d, italics added). Similarly, it is the Krajina Serbs, who wished to remain part of Yugoslavia (rather than the Croats, who wished to secede) who are consistently termed the ‘rebels’. ‘Rebel’ tends to be a negative keyword in our culture (a consensual value), suggesting a defiance of established or correct authority. The implicit normative assumption in this use of ‘rebel Serbs’ is that Croatia has a right to authority over the Serbs but that Yugoslavia does not have that right over the Croats.

Despite all this, the vast majority of articles give no reason for the conflict which perhaps gives the impression that the conflict is meaningless or irrational. As Philo and Berry (2004) demonstrated in their analysis of coverage of the Israeli-Palestinian conflict, if people do not understand the background to events, they do not have their own frame in which to make sense of events and are therefore more susceptible to the subtleties of news frames.

Description of Leaders

The leaders of the warring parties could be portrayed in a number of ways to suit different frames. As a general rule, the mention of leaders is rare across the sample and the actors are instead ‘Croats’ and ‘Serbs’. This homogeneous portrayal of the warring parties supports the common ‘ethnic-territorial’ view by reifying ethnic groups and treating them as the base units of the conflict. However, in the wake of Operation Storm, perhaps reflecting an increased disillusionment with the leaders’ roles in negotiations, there are many more articles which attribute responsibility to, and heavily criticise the leaders. These refer almost exclusively Presidents Milosevic and Tudjman who were seen to have most control and who were very important in all negotiations.

A number of Guardian articles clearly support the alternative frame of ‘nationalisms vs. civil society’ by contrasting leadership responsibility with public victims. For example, these keywords have been selected out of an article entitled ‘Go home to roost: Refugees in the Balkan conflict may one day take terrible revenge, says Ian Traynor’:

… politically-inspired hatred… black propaganda… Like millions of others in former Yugoslavia she is a victim… a victim like so many before her… she is also a toy and a weapon… the crazed cynics who pass for politicians… calculated and systematic policy of forced displacement… demographic games… deranged dreams of extremist intellectuals… cycle of politically-controlled hysteria… propaganda… any Croat with common sense… zealous settlers… distort and disfigure… frontier fanatics… reprobates… ruthless programmes… deranged dream… Milosevic’s insane scheme… demographic game of ethnic separation… Mira and millions like her are the innocent victims…. (Guardian: 14th Aug. 1995d)

A number of themes are evident. Firstly, responsibility is clearly attributed to leaders but not the public in general. Secondly, a contrast is made between the ‘common sense’ of ordinary people (further supported by examples) and the madness of politicians who are ‘crazed’, ‘deranged’ or ‘insane’. Finally, the description of the conflict as a ‘game’, the refugee as a ‘toy’ and the objective as a ‘dream’ suggest that those controlling the war are not living in the real world of adult rationality. The general themes of this report are similar to a number of other Guardian articles. The use of keywords activating a script of madness in particular is common in the August 1995 reports to suggest both the irrationality and obsession of both leaders.

In most cases the two presidents are described as equally bad in The Guardian; for example, the Bosnian Muslim president is quoted comparing the choice between them to ‘choosing between leukaemia and a brain tumour’ (7th Aug. 1995f, 18th Oct. 1991a). However, one reporter, Martin Woollacott, emphasises that the primary responsibility lies with (all of) the Serb leaders whilst condemning both governments and calling for the leaders to be marginalised (Guardian: 7th Aug. 1995e, 12th Aug. 1995). This skews The Guardian’s reporting slightly towards the ‘Croat’ frame.

The portrayal of leaders in The Times is much more ambiguous. There is a large amount of reporting on Tudjman but almost none on Milosevic, making them difficult to compare. Three articles in the sample focus specifically on the Croatian president (4th Aug. 1995c, 7th Aug. 1995e, 8th Aug. 1995e). All arise from exclusive access to a map drawn by Tudjman in which he showed the division of Bosnia between Serbia and Croatia. Despite mentioning some of the less savoury parts of Tudjman’s character, these articles display considerable sympathy towards him. The strongest support comes in the article printed on the 8th which describes Tudjman as an ‘academic of an old-fashioned kind’ and stresses that ‘his record is really very good in many ways’. Drawing the map indicating his territorial ambitions was ‘an absurd thing to do’ but an ‘indiscretion’ and no criticism is made of the ambitions themselves, merely the decision to display them. Importantly, no keywords are used to indicate that Tudjman was dangerous – ‘indiscretion’, ‘absurd’ and academic’ all point in the opposite direction. This is despite Operation Storm and the exodus of Serb refugees that were occurring at the same time. The other two articles present a less positive picture of the Croatian president, yet also avoid direct criticism or mention of his role in contemporary events.

Perhaps surprisingly given his later role as an international hate-figure, Milosevic makes almost no appearance in The Times’ coverage. He is mentioned a number of times as being influential yet only two descriptive keywords, ‘xenophobia’ and ‘hardline’ (8th Oct. 1991b, 11th Oct. 1991b), indicate how journalists felt about him. This seems to have the effect of making Milosevic largely invisible and minimising his perceived responsibility. The same can also be said of the coverage (of rather lack of coverage) of Tudjman in October 1991. The Times’ lack of coverage of the leaders suggests a significant skew towards the ‘ethnic-territorial’ position as differentiation of the leaders and people is a necessary condition for ‘nationalisms vs. civil society’. The balance between ‘Croat’ and ‘Serb’ frames is indistinguishable from this sample as there is both criticism and sympathy toward Tudjman but little mention of Milosevic.

References to Propaganda

The inclusion of information about the internal propaganda struggle is important in differentiating the ‘ethnic-territorial’ frame from the ‘nationalisms vs. civil society’ frame. This is because knowledge of the propaganda campaigns mitigates the responsibility of ordinary people for the conflict and focuses attention instead on a manipulative elite. In this way, mentioning propaganda can be seen as the opposing position to focusing on ethnic hatred, which does implicate the public.

Discussion of propaganda is very rare in both papers and occurs principally in articles from 1995. There are only two examples, both in The Guardian, where reference to propaganda appears schematically prominent enough to get into the lower levels of the macrostructure. Almost all mentions of propaganda in The Times are merely hints, such as ‘the rhetoric of the Croatian media had become more bellicose by the day’ (5th Aug. 1995d). While this may be true, what this example doesn’t say is that the media was state controlled and that it affected people’s perception of the conflict. Therefore it doesn’t change the focus of this particular article from the irrational hatred of ‘the old man’ who ‘was grinning wickedly as he watched Petrinja [a Serbian town] burn’ (ibid.). In addition the hatred is schematically privileged, while the propaganda hint is much lower in the relevance structure.

The Guardian mentions propaganda much more often than The Times and links it directly to the actions of the population on a number of occasions. For example, an article on the 14th August 1995(d) refers to ‘black propaganda’, ‘politically-inspired hatred’ and ‘politically-controlled collective hysteria’ as responsible for the conflict and for the fear driving refugee movements. Guardian articles also occasionally directly implicate the leaders in propaganda:

his [Tudjman’s] own propaganda (Guardian: 14th Aug. 1995d)

He blames…Slobodan Milosevic for starting the war, and for persistently misrepresenting it, to retain and expand his own power. “The whole situation was produced by lies. We in Serbia live in a civilisation of lies…”

(Guardian: 23rd Oct. 1991a)

A number mention Milosevic’s propaganda while only a couple implicate Tudjman. There are not, however, enough articles on this topic to draw any conclusions about differing portrayals of the two presidents. There are still very few references to propaganda overall.

The fact that journalists rarely mention propaganda suggests that Kaldor’s ‘nationalisms vs. civil society’ frame is fairly weak in this sample. Most reports, by omitting propaganda as a topic, imply that the publics were more or less willing participants rather than being manipulated by fear. This supports the dominant ‘ethnic-territorial’ idea of group hatred and is much stronger in The Times than The Guardian.

Empathetic Access

The way in which actors are presented affects our ability to empathise with them and to feel concern for their fate. Where actors are marginalised or distant in reports or where the individual humanity of participants is not obvious, empathising is more difficult. The different effects of ‘Tens of thousands of refugee Serbs fled Krajina’ (Times: 8th Aug. 1995b) and ‘Sitting solemn-faced behind the wheel of a tractor, his feet barely able to reach the pedals, a 10-year-old boy steered his family… into the safety of Serbia’ (Guardian: 26th Aug. 1995) illustrates this point. The majority of articles covering the Serbian refugee exodus in August 1995 take the former, more distant approach. In fact, empathetic access to the Serbs, marked by interviews, individual stories, and descriptions, is strikingly rare considering the extent of coverage they received. This does not mean that reports are necessarily unsympathetic, but the emphasis on the vast numbers of people and the clinical descriptions of atrocities make the refugees’ situation less comprehensible to readers than individual stories.

In contrast, the few stories covering the slow process of Croatian refugees returning to their former homes in the Krajina, focus entirely on individual stories. There is a marked difference in these articles between the distant ‘departed Serbs’ or ‘retreating Serbs’ (Guardian: 28th Aug. 1995; Times: 19th Aug. 1995) and the immediacy of the Croat stories. As a result, in terms of empathetic access, these two processes, refugees leaving and returning, are almost equal despite the very different scales of the processes and of the coverage accorded to them.

To take another example, empathetic access is skewed in coverage of the JNA and the Croat forces during the siege of Dubrovnik in October 1991. The JNA is almost completely inaccessible and is denied a human face. In one article it is even referred to as a ‘war machine’ and ‘battering ram’ (Times: 3rd Oct. 1991), activating an industrial script which completely removes the human element, but most speak of ‘the federal army’, ‘the army’, ‘the air force’ etc. While these latter descriptions imply little skew themselves, there is a significant contrast to the portrayal of Croat forces in the same articles. Croats are presented in a much more human manner, many are quoted and some individual stories are used to explain the situation as a whole. Two articles focus specifically on groups of Croatian soldiers, exploring their backgrounds, motivations and opinions (Guardian: 30th Oct. 1991; Times: 26th Oct. 1991). There is also considerably more space dedicated to them than to the JNA. With little exception (Times: 7th Oct. 1991b), these articles are skewed towards the ‘Croat’ position as they allow the reader to empathise with the Croatian position but deny empathetic access to the JNA/Serbs.

Reports describing the fighting in Croatia’s Operation Storm (not including later reports on the refugee movements) do not entirely reverse this picture. Like the JNA earlier, Croat forces are once referred to as a ‘war machine’ (Times: 5th Aug., 1995b) but tend to be called ‘Croat forces’, ‘the Croatian military force’, ‘the Croatian army’, or just ‘the Croatians’. As such they are depersonalised. However, in contrast to earlier coverage of the siege of Dubrovnik, a few Croatian civilian voices mitigate this alienation. For example, The Guardian reports a conversation with the families of fighting men. Their fears for their loved ones draw the reader’s attention to the soldiers’ individuality (5th Aug. 1995c). Similar effects can be seen in The Times (7th Aug. 1995b; 5th Aug. 1995d) although they are rare overall.

The combination of these examples can indicate a further area of skew; the empathetic access to attacking forces and defending militias and civilians is not symmetrical. In Dubrovnik, the distant JNA attacks often named Croatian individuals, whereas, in Operation Storm, the largely distant Croatian army attacks equally distant Serbs. The empathetic emphasis is therefore on the suffering of Croat victims more than Serb victims, placing the JNA in a worse light than the Croat forces. Taken together, these examples suggest a general empathetic skew towards the Croat position.

Violence

There are three areas in the reporting where the focus is on violence. These are the JNA sieges of Dubrovnik and Vukovar and attacks on their surrounding areas in October 1991, and the Croat forces attack on the Krajina in August 1995. All were treated with revulsion by the press but the portrayals differed in some significant ways. As there is a very large body of material on these themes, I will address only a selection of important points.

The treatment of JNA actions in October 1991 is particularly emotive in The Times. The siege of Dubrovnik, for example, is called ‘A Croatian Crucifixion’ in the headline of one article (Times: 26th Oct. 1991a) and others fully replicate the Croatian government’s position that the siege of the city was cultural genocide, despite evidence and JNA claims that Croat forces were drawing fire to historic buildings deliberately (Times: 3rd Oct. 1991c, 25th Oct. 1991). The Guardian reports tend to take a less emotive line, whilst still showing sympathy to the victims, and avoids The Times’ focus on cultural destruction. There is a particularly noticeable difference between the headlines of the two papers in articles reporting the same or similar events. For example, on the 26th October 1991 both published reports covering the JNA’s attacks on Dubrovnik and the negotiation of a temporary ceasefire to evacuate some of the inhabitants. The headline in The Times was ‘Croat defenders fear the sack of Dubrovnik’, with the ceasefire appearing low in the schema, in The Guardian it was ‘Evacuation plan for Dubrovnik’ which was privileged over news of the continued attacks. The Times’ use of the terms ‘fear’ and ‘sack’ stand in contrast to the more distant approach of The Guardian.

In reference to Vukovar, similar patterns are evident. Emotive keywords are used in The Times’ coverage of the siege which strongly imply a revulsion for JNA actions. These have been drawn from the article ‘Convoy forced to abandon sick in ‘medieval siege’:

… agony… horror… gruelling… perilous… grey, fixed expressions… horror… suffering… pain… medieval siege… bloodier… destructive power… begging… torture… exhaustion… “terrible”… “fear”… cruel choice… gruelling… mass of bodies… ghastly… bloodied… moaned… “savage devastation”…

(Times: 21st Oct. 1991)

In The Guardian’s (marginally shorter) coverage of a very similar event a week prior to this (also written by a journalist on the scene) only a few comparable words can be found:

… “Terrible odyssey”… battered… ferocious fighting… bodies… confusion… grotesque… (Guardian: 14th Oct. 1991)

This again shows a distance in The Guardian that is not present in The Times. The Times coverage tends to vilify the JNA and emphasise the suffering of Croat civilians (an approach supportive of the ‘Croat’ frame) more than The Guardian.

In the reversed situation, this pattern is less clear. The atrocities that occurred during Operation Storm are condemned but tend to be treated in a more distant manner. Perhaps in some ways this emphasises how bad things were – both papers carry reports listing evidence of atrocities which are quite striking (Guardian: 19th Aug. 1995, Times: 12th Aug. 1995a). Other reports also seem to let the events ‘speak for themselves’. For example, the journalist makes no comment on the assertion that “Croats had put a wooden cross by the roadside with a headless corpse on it – just a warning”, and changes the subject immediately afterwards (Guardian: 14th Aug. 1995c). However, the journalists’ disgust is much less explicitly evident which may give these events less prominence that those described using strong or emotive language.

There is little that is comparable to earlier (Times) descriptions of Vukovar, this description in a Guardian article is among the only emotive passages:

Up here it is clear what has been happening by night during the “liberation” of Krajina. Almost every house has been torched, charred rafters rising from the blackened masonry like protruding bones from a corpse… the villages were as silent as tombs. (Guardian: 28th Aug. 1995)

The general lack of judgemental keywords, such as the metaphors ‘corpse’ and ‘tombs’ above, suggests that journalists were fairly emotionally detached from these events. The contrast between The Times’ reporting of Serb (1991) and Croat (1995) atrocities is marked, suggesting a skew towards the ‘Croat’ frame, whilst The Guardian appears to be largely consistent. The replication of Croatian propaganda from Dubrovnik in The Times’ coverage is a particularly strong example of skew.

Treatment of Dissenting Voices

The use of sources which oppose nationalist agendas occurs in only a tiny fraction of reports in both papers. Quotes such as “We need to break the cycle of revenge” (Guardian: 9th Aug. 1995c) are far outnumbered in the reporting by those supporting the nationalist positions, “I don’t think I could ever live alongside Serbs again” (Times: 7th Aug. 1995b). Where such sources are used they tend to be placed very low in the relevance structure and often appear as an aside to the main focus of the report. In contrast, nationalist sources often appear much higher in the relevance structure. This schematic pattern occurs even where both perspectives are addressed in the same report (for example, Guardian: 23rd Oct. 1991a) and has the effect of marginalising non-nationalist sources. These sources are important in the ‘nationalisms vs. civil society’ frame and their rarity reduces the likelihood that readers will draw this perspective from the reporting.

Patterns of Framing

There is evidence for all four politically salient frames in the coverage of the war in Croatia and considerable difference between the papers in their selection of perspectives. The ‘ethnic-territorial’ frame is the clearest throughout the reporting. A few Times articles privilege this view in the highest levels of the macrostructure but most other articles support this frame implicitly. The general absence of background to the conflict allows the reader to assume that there are no real reasons for it, leading to the possible belief that the conflict was ‘natural’ and feeding into the idea of ancient ethnic hatreds which is so commonly found in The Times. The consistent homogenising of the populations as ‘Croats’ and ‘Serbs’ also supports this position as it makes ethnic identity the basis of the analysis. In addition, the almost complete omission of propaganda as a topic hides the role of leaders in manipulating events, allowing the conflict to be seen as the result of ‘real’ hatreds. This frame is particularly strong in The TimesThe Guardian. coverage which shows the above trends much more than

The alternative ‘nationalisms vs. civil society’ frame is generally weak in the coverage, appearing almost exclusively in The Guardian’s August 1995 reports. It can be seen most clearly on the few occasions where propaganda is mentioned as this reduces the responsibility of the population for their actions and focuses attention instead on the leadership. The mention of leaders as the actors, as opposed to the more general ‘Croats’ and ‘Serbs’, has the same effect. The use of non-nationalist sources provides evidence of peaceful desires within the population and undermines the idea that the region is consumed by hatreds. However, this frame appears in a small minority of articles in The Guardian 1995 sample and only a tiny percentage overall.

The treatment of ‘Croat’ and ‘Serb’ frames is less obviously skewed but the ‘Croat’ frame appears to be favoured, particularly in The Times. The ‘Croat’ position has the advantage of empathy as both forces and civilians are generally more accessible than their Serbian counterparts. Similarly, the atrocities committed by the JNA are treated more harshly in The TimesThe Guardian appears to make little distinction between the two). than those committed by Croat forces. (

The portrayal of the presidents of Croatia and Serbia, Tudjman and Milosevic, appears to follow a similar pattern but is much more subtle. The Guardian, where it mentions the leaders, commonly takes the approach of treating both as equally bad, whilst The Times makes little mention of either. However, on the basis of Woollacott’s reports in The Guardian and The Times’ coverage of Tudjman’s map, I might tentatively suggest that the reporting of the leaders shows a very slight skew towards the ‘Croat’ position.

The most ambiguous aspect of the ‘Croat’-‘Serb’ framing contest is found in the reasons for the conflict. Explanations supporting the ‘Serb’ perspective appear explicitly but infrequently (particularly in The Times), whereas those supportive of the ‘Croat’ frame are implicit but occur throughout. It is impossible to tell, on the basis of a textual analysis only, which of these will have a higher impact on readers.

The skew of the reporting is fairly subtle overall and would not be evident without the use of a discursive analysis. The reporting generally favours the ‘ethnic-territorial’ and ‘Croat’ positions but this is much stronger in The Times than The Guardian, and in October 1991 than August 1995.

Conclusion

This study has addressed the question: what were the major framing contests in the war in Croatia and how were these represented in the British press? This was answered by initially analysing the factors influencing journalists’ frames on the war in Croatia and then by a qualitative discourse analysis of a sample of newspaper reports. I have highlighted four main frames and described some of the ways in which they were evident in the reporting.

The first half of this study considered how to identify the most politically salient frames on the conflict. It was argued that journalists’ choices of frame are determined by ideological and institutional factors and that objectivity is impossible. In reporting from the war in Croatia, journalists became the targets of propaganda campaigns which aimed at influencing the framing of events. The pervasiveness of propaganda and the complexity of the war may have made journalists more susceptible to replicating these frames.

By drawing on the work of historians and through an analysis of the propaganda used, my study has identified two politically salient framing contests that occurred during the conflict. These were cases where the adoption of a different frame may have changed the nature of international involvement significantly. The first of these framing contests was between ‘Croat’ and ‘Serb’ propaganda positions and the second concerned the nature of the conflict – the choice between the ‘ethnic-territorial’ reading and the ‘nationalisms vs. civil society’ frame.

These four frames were used to guide the analysis of a selection of reports from two British daily papers. Using a qualitative approach and the techniques of discourse analysis, all four frames were identified in the reporting. The ‘ethnic-territorial’ position was the strongest overall and the ‘Croat’ frame seemed to get marginally more support than the ‘Serb’ frame. However, the framing trends were slightly different in the two papers and between the two months of coverage.

This analysis highlighted a number of interesting issues with regards to framing in the press. Firstly, there were noticeable differences in approach between The Times and The Guardian. The Times was generally focused on the ‘ethnic-territorial’ frame but showed sympathy to the ‘Croat’ position. It included barely any support for either the ‘Serb’ or the ‘nationalisms vs. civil society’ frames. In contrast, The Guardian supplied evidence for all four frames. These differences will, of course, be heavily linked to the opinions and experiences of the journalists working for each paper. However, they also suggest that the institutional factors driving frame selection (such as the ideology of the paper as a whole) may have played an important role in distinguishing their approaches.

Secondly, in The Guardian in particular, there is a marked difference in the coverage of the two months – October 1991 and August 1995. The ‘Serb’ and ‘nationalisms vs. civil society’ frames are all but invisible in the first month but become much more prominent in the second. This is perhaps a sign of increasing disillusionment with the conflict and the leaders in particular, following four years of war in Bosnia and numerous broken agreements.

Thirdly, the ‘ethnic-territorial’ frame, the most prominent in the reporting, corresponds closely to the idea of ‘The Balkans’ or ‘balkanisation’. This reputation of the region is not a product of the recent wars, it can be traced back at least to the First World War if not before (Glenny, 2000: xxiii; Mazower, 2001: 4) and can be considered as a culturally shared ‘script’ in our language. Fowler (1991: 13) emphasises that journalists ‘rely heavily’ on the use of scripts such as this in order to facilitate the reader’s understanding of complex events. The heavy use of the ‘Balkan’ stereotype found in this study appears to support this idea that culturally shared assumptions exert a powerful influence on media frames.

Finally, the extent of skew discovered in this initial analysis of British coverage differs significantly from that found in studies of US coverage. During the war in Croatia and other Yugoslav conflicts, US coverage was shown to be clearly and consistently skewed against the Serbs (Kavran, 1991; Djordjevich, 1992; Vincent, 2000; Brock, 1993). Whilst much of the British coverage does not favour the Serbs, this skew is subtle and much less significant than that found in the above studies. This difference broadly mirrors the divergent stances taken by the US and British governments towards the conflict. The idea that the media may adopt the frames of their governments is examined by Yang (2003) with reference to the Kosovo conflict.

My study was very limited in is scope and aims. I wished only to present some of the major issues and highlight the principle trends in two, politically salient framing contests. A much wider and more detailed analysis would be necessary to confirm my initial findings and to discover the trends outside of the four frames I have identified.

A logical extension to this project would be an audience study to determine the effects of this coverage on different people. Of particular importance is the issue of implicit support for different frames. For example, I described in the previous chapter how the reasons for Croatia’s involvement in the conflict were rarely addressed directly but were frequently hinted at. From a study of the texts only, it is not possible to judge the relative weight of this implicit frame compared to the more explicit, but less frequent, ‘Serb’ justifications. Similar issues arise in relation to the strength of the ‘ethnic-territorial’ perspective.

This study has demonstrated the links between propaganda, ideology, and the representation of events in the news. Skew in the reporting from the conflict in Croatia was a result of these links and demonstrates the limits of understanding gained from the press. The extent to which this skew was absorbed by the public remains to be seen.

Bibliography

Brock, P. (1993) ‘Dateline Yugoslavia: The Partisan Press’ in Foreign Policy No. 93 (Winter 1993-1994) pp.152-172.

Breed, W. (1999) ‘Social Control in the Newsroom: A Functional Analysis’ reprinted in Tumber H. (ed.) News: A Reader, pp. 79-84, Oxford: Oxford University Press

Cohen, S. and Young, J. (1973) ‘Introduction’ in Cohen, S. and Young, J. (eds.) The Manufacture of News: Social Problems, Deviance and the Mass Media, London: Constable and Co.

Djordjevich, D. (1992) ‘Death Camps and Desert Storm Frame Bosnia Coverage’ www.fair.org/index.php?page=1205 (accessed 29/10/06)

Duffy, G. and Lindstrom, N. (2002) ‘Conflicting Identities: Solidary Incentives in the Serbo-Croat War’ in Journal of Peace Research, pp 69-90, London: Sage Publications Ltd.

Fowler, R. (1991) Language in the News: Discourse and Ideology in the Press, London: Routledge

Galtung, J. and Ruge, M. (1973) ‘Structuring and Selecting News’ in Cohen, S. and Young, J. (eds.) The Manufacture of News: Social Problems, Deviance and the Mass Media, London: Constable and Co.

Glenny, M. (2000) The Balkans: Nationalism, War and the Great Powers, London: Granta Books

Höijer, B., Nohrstedt, S.A., Ottosen, R. (2002) ‘The Kosovo War in the Media’ in Conflict & Communication Online, Vol. 1(2), www.ccoregener-online.de.pdf (accessed 01/11/06)

Jowett, G.S. and O’Donnell, V. (1992) Propaganda and Persuasion, London: Sage Publications Ltd.

Kaldor, M. (2001) New and Old Wars: Organised Violence in a Global Era, Stanford: Stanford University Press.

Kavran, O. (1991) ‘Cold War Lives on in Yugoslavia Reporting’ www.fair.org/index.php?page=1200 (accessed 29/10/06)

Livingston, S. and Bennett, W.L. (2003) ‘Gatekeeping, Indexing, and Live-Event News: Is Technology Altering the Construction of News?’ in Political Communication Vol.20(4) pp. 363-380, Taylor and Francis Publishing Ltd.

MacDonald, D.B. (2002) Balkan Holocausts?: Serbian and Croatian victim-centred propaganda and the war in Yugoslavia, Manchester: Manchester University Press

Malesic, M. (2000) Peace Support Operations, Mass Media and the Public in Former Yugoslavia, Report 157 for the National Board of Psychological Defence, Stockholm, www.krisberedskapsmyndigheten.se/444.epibrw (Accessed 04/10/06)

Manning, P. (2001) News and News Sources: A Critical Introduction, London: Sage Publications Ltd.

Mazower, M. (2001) The Balkans: From the End of Byzantium to the Present Day, London: Phoenix Press

McNair, B. (1999) News and Journalism in the UK: A Textbook (3rd edition), London: Routledge

Nohrstedt, S.A., Kaitatzi-Whitlock, S., Ottosen, R., and Riegert, K. (2000) ‘From the Persian Gulf to Kosovo – War Journalism and Propaganda’ in European Journal of Communication Vol. 15(3) pp 383-404, London: Sage Publications

Oxford English Dictionary (Compact) (2003) Oxford: Oxford University Press

Pavkovic, A (2000) The Fragmentation of Yugoslavia: Nationalism and War in the Balkans (2nd edition) London: Macmillan Press Ltd.

Philo, G and Berry, M. (2004) Bad News From Israel, Glasgow University Media Group, London: Pluto Press

Poggioli, S. (1994) ‘Scouts Without Compasses: An NPR Reporter on the “Disinformation Trap” in Former Yugoslavia’

www.fair.org/index.php?page=1229 (accessed 29/10/06)

Rock, P. (1973) ‘News as Eternal Recurrence’ in Cohen, S. and Young, J. (eds.) The Manufacture of News: Social Problems, Deviance and the Mass Media, London: Constable and Co.

Sparks, C. (1999) ‘The Press’ in Stokes, J. and Reading, A. The Media in Britain: Current Debates and Perspectives, London: Macmillan Press Ltd.

Van Dijk, T.A. (1988a) News as Discourse, Hillsdale: Lawrence Erlbaum Associates Ltd.

Van Dijk, T.A. (1988b) News Analysis: Case Studies of International and National News in the Press, New Jersey: Lawrence Erlbaum Associates Ltd.

Vincent, R.C. (2000) ‘A Narrative Analysis of US Press Coverage of Slobodan Milosevic and the Serbs in Kosovo’ in European Journal of Communication Vol. 15(3), pp. 321-344, London: Sage Publications

Wikipedia, Map – Republic of Serbian Krajina,

http://www.answers.com/topic/republic-of-serbian-krajina

(accessed 12/02/07)

Yang, J. (2003) ‘Framing the NATO Air Strikes on Kosovo Across Countries: Comparison of Chinese and US Newspaper Coverage’ in Gazette: The International Journal for Communication Studies Vol. 65(3) pp. 231-249, gaz.sagepub.com/cgi/content/abstract/65/3/321 (accessed 01/11/06)

Newspaper Reports

The Times

1st Walker, C. ‘Yugoslav ceasefire near collapse as tanks cross border’

3rd(a) Walker, C. ‘Plea for help as Serbian guns shell Dubrovnik’

3rd(b) Boyes, R. ‘Conflict tests loyalties of Yugoslav Army’

3rd(c) no named author, ‘War of Art’

5th(a) Walker, C. ‘Air raids drown talk of peace’

5th(b) Walker, T. and Trevisan, D. ‘Croatia accord as Serbian leader relents’

7th(a) Walker, C. and Walker, T. ‘Zagreb sounds general alert as war nears

7th(b) Judah, T. ‘Montenegrins join city siege’

7th(c) Walker, C. ‘President’s call to arms electrifies a nation’

8th(a) McElvoy, A. ‘European Myth fails Yugoslavia’

8th(b) Boyes, R. ‘Oil embargo could worsen war’

9th Boyes, R. ‘Ethnic roots exposed in a nation’s ruins’

10th no named author, ‘Back-up for peace’

11th(a) Evans, M. ‘Serb leadership split jeopardises ceasefire’

11th (b) McElvoy, A. ‘Karlovac braces itself for ‘more madness’’

15th McElvoy, A. and Brock, G. ‘Relief of Vukovar thwarted as aid convoy turns back’

17th McElvoy, A. ‘Meat van bears grim testimony to conflict’

18th(a) no named author, ‘Selling of Serbia’

18th(b) Boyes, R. ‘Yugoslav generals stage last push to avoid humiliation’

18th(c) McElvoy, A. and foreign staff, ‘Serbs and Croats fight hand to hand in besieged town’

19th McElvoy, A. ‘Dodging Serbian mortars on the road to Vukovar’

21st McElvoy, A. ‘Convoy forced to abandon sick in ‘medieval siege’

22nd Trevisan, D. ‘Accusations fly in broken truce’

25th Boyes, R. ‘Art falls prey to federal army tactics’

26th(a) Wilkes, K. ‘A Croatian Crucifixion’

26th(b) Binyon, M. ‘Croat defenders fear the sack of Dubrovnik’

The Guardian

1st(a) no named author, ‘No decision on peacekeepers’

1st(b) Hearst, D. ‘Yugoslavia’s founding father gone but not forgotten’

1st(c) Hearst, D. and Sullivan, K. ‘War looms in Bosnia after Serbs ordered to mobilise’

5th Hearst, D. ‘Army sets tough terms for truce’

7th(a) Brown, D. ‘Serbs welcome royal ‘protector’ after 43 year exile’

7th(b) Hearst, D. ‘Young lovers heed Tudjman’s call to fight for Croatia’

7th(c) Palmer, J. ‘Community gives embargo ultimatum to Yugoslavia’

8th Hearst, D. and Brown, D. ‘Rocket attack on Croatian palace’

9th Hearst, D. ‘Serbs and Croats agree ‘permanent’ ceasefire’

10th(a) Hearst, D. ‘Truce in Croatia is a contract made to be broken’

10th(b) Hearst, D. ‘Croatian crusaders wait for the replay’

12th(a) Traynor, I. ‘Federal army rejects call for pull-out’

12th(b) Vulliamy, E. ‘Bombardments stop EC convoy taking food to besieged Croats’

14th Vulliamy, E. ‘EC convoy rolls onto siege city’

15th Brown, D. and Palmer, J. ‘Gorbachev calls warring Yugoslav leaders to talks’

16th Vulliamy, E. ‘Croatian extremist militia seizes military initiative’

17th Brown, D. ‘Grobachev truce consigned to Croatian dustbin’

18th West, R. ‘An apologist for Hitler’

23rd(a) Brown, D. ‘Serbian schizophrenia’

23rd(b) Traynor, I. ‘Army chief denounces EC plan’

24th Vulliamy, E. ‘Croats take aim as exhaustion and winter wear down federal advance’

26th Vulliamy, E., Traynor, I. And Palmer, J. ‘Evacuation plan for Dubrovnik’

28th Vulliamy, E. ‘Victims of war in Croatia lose homes but not spirit’

30th Vulliamy, E. ‘Denim pawns played in medieval chess game’

The Times

1st Loyd, A. ‘Croats resume drive for land with heavy barrage’

2nd(a) Prentice, E. and Kiley, S. ‘Rebel Serbs launch air attacks to halt Croats’

2nd(b) Trevisan, D. ‘Tudjman gambles on Milosevic shying away from rescue mission’

2nd(c) Evans, M. ‘Krajina forces line up for full-scale assault’

3rd(a) Rhodes, T. and Kiley, S. ‘Congress defies Clinton over Bosnia arms ban’

3rd(b) Rees-Mogg, W. ‘Will the Serbs succumb?’

4th(a) Evans, M. ‘Serbs launch shell attack on Dubrovnik’

4th(b) Fletcher, M. ‘Washington gives tacit support to Croat attack’

4th(c) Trevisan, D. ‘Former partisan shows tactical skill’

5th(a) Evans, M. ‘Croatia launches all-out offensive’

5th(b) Eyal, J. ‘Tudjman’s march to victory seals doom of Muslims’

5th(c) Kiley, S. ‘Leering fixer longs for the kill’

5th(d) Loyd, A. ‘Croat war machine rolls into Petrijna at dawn’

7th(a) Evans, M. ‘Thousands flee as Croats celebrate crushing victory’

7th(b) Kiley, S. and Sullivan, S. ‘UN fears new round of ‘ethnic cleansing’ in the wake of offensive’

7th(c) Evans, M. ‘Threat to Muslims grows as Croat victory pushes Serb forces into Bosnia’

7th(d) Fletcher, M. ‘Zagreb minister claims America gave advice on Knin attack’

7th(e) Evans, M. ‘Tudjman mapped out future on city menu’

7th(f) no named author, ‘Full circle’

8th(a) Glenny, M. ‘City that fell victim to wanton warmongers’

8th(b) no named author, ‘150,000 Serbs flee Croat conquerors’

8th(c) Trevisan, D. and Sullivan, S. ‘Triumphant Tudjman fulfils his nation’s 1,000-year-old dream’

8th(d) Prentice, E., Sullivan, S. and Beeston, R. ‘Krajina refugees pack roads as Croats end assault’

8th(e) Stone, N. ‘Strategy on the back of a menu’

9th(a) Evans, M. and Trevisan, D. ‘Troops build up on border of Eastern Slavonia’

9th(b) McElvoy, A. ‘Losers of power game condemned to a shotgun marriage’

9th(c) Binyon, M. ‘Land shaped by centuries of conflict endures another upheaval’

9th(d) Eyal, J. ‘The Bosnian carve-up has begun’

9th(e) no named author, ‘Columns of Misery’

10th(a) Prentice, E. ‘BBC man shot dead by Croat soldiers’

10th(b) Prentice, E., Beeston, R. and Rhodes, T. ‘Milosevic offered olive branch by US and Germany’

11th(a) Kiley, S. ‘Fear drives Serbs on the highway of hatred’

11th(b) Bone, J. ‘UN to pull out 10,000-strong peacekeeping force’

11th(c) McElvoy, A. ‘Eastern Slavonia lives on its nerves as frontline towns await fresh war’

12th(a) McElvoy, A. ‘Serb bodies trigger alarm over Croat atrocities’

12th(b) Eyal, J. ‘America’s search for peace may merely pour old wine into new bottles’

12th(c) no named author, ‘Buried bodies’

15th Prentice, E. ‘Refugees flee on all sides as fresh Balkan attacks are launched’

17th Prentice, E. ‘Croat troops mass around Dubrovnik for attack on Serbs’

18th Kiley, S. and Prentice, E. ‘Croatian advance puts pressure on divided Serbs’

19th McElvoy, A. ‘Refugees return to mixed fortunes’

25th Bone, J. and Sullivan, S. ‘UN seeks huge cut in Croatia peace force’

The Guardian

1st Borger, J. ‘Croatia and Serbs dismiss UN talks’

2nd Wheen, F. ‘Winners in a war of words’

3rd Steele, J. ‘Beleaguered Krajina Serbs find few friends in Belgrade’

4th(a) Borger, J. and Traynor, I. ‘Croats place peace deal in jeopardy’

4th(b) Borger, J. ‘Croatian general confident of victory’

5th(a) Borger, J. ‘Croats launch all-out war’

5th(b) Traynor, I. ‘Croatia’s war waged for blood and soil’

5th(c) Borger, J. ‘Wistful ballads to tune of death’

5th(d) Woollacott, M. ‘Another move into check’

7th(a) Borger, J. and Traynor, I. ‘Refugees panic in Croat blitz’

7th(b) Traynor, I. ‘Wily Milosevic holds the military ace’

7th(c) Huic, D. ‘’Work not war’ declared in Knin’

7th(d) Fairhall, D. ‘Even Croat planners surprised by Operation Storm’s quick success’

7th(e) Woollacott, M. ‘Way out from a dead end’

7th(f) Traynor, I. ‘Il Duce of the Balkans’

8th(a) Traynor, I. ‘Civilians in jaws of facing armies’

8th(b) Borger, J. ‘Croats wipe out all traces of Serb past’

8th(c) Freedland, J. ‘US denies giving nod to Zagreb’s lightning war’

8th(d) no named author, ‘Martial plans’

8th(e) no named author, ‘An unremitting tragedy’

9th(a) Solar, C. ‘In queues stretching back 40 miles the refugees were stuck in the heat, in cars behind tractors behind horse carts, with hours to contemplate their defeat’

9th(b) Traynor, I. ‘Last diehard Serbs start to surrender’

9th(c) Traynor, I. ‘Six women tremble as they see the ‘door of hell’ open’

12th(a) Traynor, I. And Borger, J. ‘Battle front returns to Bosnia’

12th(b) Steele, J. ‘Short cycles in the ethnic cleanser’

12th(c) Woollacott, M. ‘Slobodan’s strangulation’

14th(a) Borger, J. ‘Serb gunners shatter Dubrovnik’s tourism hopes’

14th(b) Traynor, I. ‘US proposes summit on new carve-up’

14th(c) Steele, J. ‘Old neighbours forget the hate on slow road to Serbia’

14th(d) Traynor, I. ‘Go home to roost’

16th(a) Borger, J. and Vulliamy, E. ‘General signals imminent offensive neat Dubrovnik’

16th(b) Steele, J. ‘The seaside deal with a whiff of Yalta’

19th Borger, J. ‘Victorious Croats ‘burned villages’’

21st Borger, J. ‘Victorious Croatia closes doors to ethnic minorities’

26th Steele, J. ‘Secret scars of families at war’

28th Vulliamy, E. ‘Sad homecoming in a land that is as silent as a tomb’

31st Glover, J. ‘Croat troops ‘given drugs’’

 —

Written by: Katie Smith
Written at: Aberystwyth University
Date Written : 2007

Further Reading on E-International Relations

• Re-Framing Gender Relations in Conflict Settings: UNSCR 1325 in Sierra Leone
• Were ‘Ancient Hatreds’ the Primary Cause of the Yugoslavian Civil War ?
• Narratives of Violence: The Hong Kong Protests Through Opposing Media Outlets
• How the Islamic State Weaponizes Imitation in Its Propaganda
• Computational Propaganda: Challenges and Responses
• Ideology and Economic Policy in European Social Democracy c.1890-2010