Political agency of news outlets in a polarized media system: Framing the corruption probe in Turkey

Abstract

This article aims to determine the stances of media outlets during crises in a polarized media system such as Turkey. Adopting a content analysis methodology, this article analyses the framing strategies of three national newspapers affiliated with certain sociopolitical camps (namely, the pro-government Sabah, the anti-government Kemalist Sözcü and the pro-Gülen Zaman) to observe possible similarities/differences during the critical 17 December corruption probe. The findings not only confirm earlier studies on ‘press-party’ parallelism but also reveal ‘press-sociopolitical camp parallelism’ in Turkey’s polarized media system.

Keywords

Corruption (political)framing media media systems political parallelism power (political)

Introduction

Debates on the roles played by media outlets in shaping public opinion have been ongoing since Lippmann and Dewey in the 1920s (Schudson, 2008; Whipple, 2005). While both sides offer compelling arguments, we can safely assume that contemporary news media outlets are controversial political actors. According to framing theory (Entman, 1993), public perceptions of politics are strongly connected to messages communicated by media outlets whose issue characterization can influence public interpretations. Commentary-oriented journalism plays a significant role in characterization, such as the polarized perspectives of an issue when the media are biased and press–party parallelism is high. Seymour-Ure (1974) defines ‘press-party parallelism’ as a degree of partisanship and/or pluralism in mass media (p. 157). Hallin and Mancini (2004) broaden its scope with a four-dimensional characterization: (the newspaper industry or the development of mass press, ‘political parallelism’, journalism professionalism, state role in the media system). They identify three types of media systems: ‘democratic corporations’, ‘liberal’ and ‘polarized pluralist’.¹

Turkey’s media system matches the characteristics of the polarized pluralist model. This includes high media integration into party politics (or political parallelism) and state intervention, along with low media commercialization and journalistic professionalism (Akser and Baybars-Hawks, 2012; Çarkoğlu et al., 2014; Kaya and Çakmur, 2010; Oğurlu and Öncü, 2013). This study investigates Turkey as an under-studied mediascape with polarized characteristics, with particular reference to the political scandal caused by the 17 December 2013 corruption probe,² which is predicted to have increased political parallelism dramatically. The main research question motivating this study is whether there were systematic and predictable variations in how media outlets framed the corruption probe.

We will show that high party–press parallelism in Turkey extends beyond party political lines to follow the contours of the political landscape as shaped by major sociopolitical camps, which certainly overlap with but also transcend political party lines. We will provide partial evidence to support this claim by showing that two media outlets opposing the governing Adalet ve Kalkınma Partisi (Justice and Development Party (AKP)) but not organically tied to other political parties followed framing practices commensurate with their political allegiances.

The article will proceed as follows. The first part examines the existing literature in framing and the news frames that have been identified (such as corruption, smear campaigns, tampering, sinister powers, fractures within sociopolitical camps and fabricated news). The second part describes the 17 December corruption probe, which further differentiated the already highly polarized Turkish media system. In particular, the section identifies how three national newspapers (the pro-government Sabah, the anti-government Sözcü and the pro-Gülen movement Zaman) each strove to protect their affiliated sociopolitical camp’s ‘ontological security’,³ as reflected in their conflicting discourses concerning the probe. The third part discusses the qualitative content analysis we used to measure the frames. Following a discussion of the results of the frames adopted by selected media outlets, it concludes that the findings not only confirm earlier studies on ‘press-party’ parallelism but also reveal ‘press-sociopolitical camp parallelism’ in Turkey’s polarized media system.

The media’s dual role in a polarized media system

As Peters (2003) suggests, the media shape the climate of democratic debate and help establish and maintain good governance. On the other hand, interpretive journalistic style creates different perspectives, depending on the relationship between context and information (Minsky, 1975; Simon and Xenos, 2000). This power to create perspectives and differentially emphasize various aspects of an issue does not always act as a corrective tool in service of good governance. It can also be used to sway the public opinion in favour of a particular position. Journalists tend to portray an issue through dichotomies. For example, they provide interpretive cues regarding emerging ‘political scandals’, sometimes by holding their own parallel trials and by defining culprits and victims in line with their political allegiances. Thus, media outlets are not merely transmitters of news but are also generators of biased content and partisan discourses.

As a typical example of the polarized pluralistic media system typology, this article examines Turkey’s case during a highly contentious corruption probe, with an emphasis on media outlets’ use of frames. Before investigating this case, however, it is necessary to outline framing theory and clarify our proposed frame categories.

Frames in the representation of political corruption

As Patterson (1997) emphasizes, the media power works to a large degree through the way in which it frames situations. The particular frames of different news sources create a symbolic filter that adopts a particular discourse or focuses on a certain representation of events (Kalogeropoulos et al., 2015; Kostadinova and Dimitrova, 2012; Zamora and Albaladejo, 2010). Framing, as defined by Entman (1993), is the process by which journalists ‘select some aspects of a perceived reality and make them more salient in a communicating text, in such a way as to promote a particular problem definition, causal interpretation, moral evaluation, and/or treatment recommendation’ (p. 51). Framing is thus a critical tool in crises by enabling political powers to influence public opinion in order to secure their legitimacy.

The existing literature on corruption employs both generic and issue-specific frames. Generic frames can be applied to different topics in different times and cultural contexts (De Vreese et al., 2001: 109). To date, the framing researchers have proposed several generic frames commonly found in the news (Cappella and Jamieson, 1997; Iyengar, 1991). Zamora and Albaladejo (2010), for example, used these to analyse the mediatization of corruption cases. Issue-specific frames also play an important role in the framing literature. These are drawn from a detailed analysis of the subject matter to define a particular approach to the subject (Arroyo, 1997; Canel and Sanders, 2006; Jiménez, 1995; Protess et al., 1991; Sherman, 1978).

An issue-specific approach to the study of news frames allows the framing of particular events to be investigated with great specificity and detail. It may capture specific aspects of selection, organization and elaboration that are present in news coverage and pertain specifically to a well-defined issue (De Vreese et al., 2001: 109). The inductive approach involves analysing a news story with an open mind to reveal the array of possible frames, beginning with very loosely defined preconceptions. This approach can be used to detect the many possible ways that an issue can be framed (Semetko and Valkenburg, 2000). In this way, Zamora and Albaladejo (2010) identified the following five frames specific to corruption issues:

Personification. This focuses on a well-known individual as embodying and symbolizing a particular ideology. It can be regarded as a shortcut that journalists use to simplify the issue, making it easier to insert discussion into the limited space of a news story.

Popular reaction. This shows the public’s reaction (e.g. surprise, outrage or support) regarding a corruption case and its actors. Public opinion survey results testing public reaction to the crisis may form part of such news.

Conquered booty. This frame focuses on the assets illegally acquired by the individuals involved, generally providing statistics to present the issue more dramatically.

Comparisons. This compares the current corruption case with others involving different parties, politicians or governments.

Purification of responsibilities. This focuses on the reactions or punishment of individuals within political organizations as a political corruption crisis develops.

According to Thompson (2013: 166), issues and the ways they unfold are shaped by the particular set of conventions and expectations specific to the cultural form that is triggered. Since issues are shaped within the cultural context in which they occur, existing frames employed by earlier studies are insufficient for our particular case. As discussed in the next section, the 17 December corruption probe provides a useful case for studying the dynamics that exist between media outlets and political actors. By adopting an inductive approach while drawing on the existing literature’s methodologies, we propose the following specific frames relevant to the current context:

Corruption frame. This focuses on government members’ attempts to directly or indirectly abuse their power for their own material benefit. Coverage of such events either explicitly mentions or implies that legal norms and/or ethical principles are violated.

Smear campaign frame. This is the anti-thesis of the corruption frame. Its defensive intention leads related news articles to discredit the corruption scandal as an attempt to damage the reputation of the government, or even to be a coup attempt.

Tampering. This frame emphasizes government malpractice regarding the proper functioning of the political system. It implies that the government interferes with officers (e.g. police or district attorneys) and tampers with evidence.

Sinister powers. This frame focuses on the presence of secretly operating national or international powers that seek to bring the country into turmoil. It is offensive in intent and content. News reports adopting this frame may target specific individuals or groups by accusing them of preventing the country’s development.

Fractures within sociopolitical camps. This frame focuses on group members’ speeches that are inconsistent with their group’s official discourses or their leaving the group as an indicator of weakening. It also implies that the sociopolitical camp is somehow disunited.

Fabricated news. This frame focuses on other media outlets’ news to discredit them as offering baseless claims or fabricated interviews. News reports employing this frame accuse the given media outlet of manipulation or provocation through the distribution of fabricated messages (Table 1).

Table 1.

Proposed issue-specific frames for a political corruption scandal in Turkey.

Frame	Explanation	Example
Corruption	Violation of legal norms and/or ethical principles for their benefits Public officials’ misuse of public power	‘Tayyip, clean this (corruption) mess or resign and leave’ (Sözcü, 18 December 2015) ‘The bribery and corruption operation that shakes Turkey’ (Zaman, 18 December 2015)
Smear campaign	Discrediting the developments as an attempt to damage the reputation/image of the government	‘The objective of the operation: Discrediting politics’ (Sabah, 18 December 2015)
Tampering	Malpractices against proper functioning of the system	‘Deputy prosecuting attorney: Two of the prime minister’s men have threatened me’ (Zaman, 9 January 2014)
Sinister powers	Powers that operate secretly and seek to weaken the government and lead the country into turmoil.	‘Principal pro-coup mindset: Fethullah Gülen’ (Sabah 17 January 2014)
Fractures within sociopolitical camps	Contradictory speeches in a given camp	‘Another resignation from AKP: That much insult is inacceptable’ (Zaman, 24 February 2014)
Fabricated news	Baseless claims by other outlets	‘6 lies in a single news report’ (Zaman, 11 February 2014) ‘Response to Zaman: 6 truths’ (Sabah, 12 February 2014)

Turkey’s polarized media system

Since the first generation of journalists in Turkey, media and politics have had a fluctuating relationship. Media’s struggle to shape public opinion and the politicians’ treatment of media as a mere tool to disseminate propaganda have created tension between these two estates (Akser and Baybars-Hawks, 2012; Heper and Demirel, 1996). The roots of press repression and state censorship that is the distinctive difference between the Turkish and Western (especially Northern European and North Atlantic) media landscapes can be found in this ongoing tension. As outlined above, Turkey’s media system matches the polarized pluralist model. Çarkoğlu and Yavuz (2010) describe the intricate nature of government and media relations in Turkey with an emphasis on press–party parallelism:

Before the commercialization of the media, the Turkish state enjoyed substantial power over the media. However, after the deregulation of media markets, newspaper owners started to utilize the material benefits of a patrimonial/clientelistic relationship between the media [e.g. Sabah daily] and the state [e.g. the AKP government]. The result of this relationship is the instrumentalization of the media [examples added]. (p. 618)

Christensen (2010) also draws attention to media ownership and related constraints on freedom of expression in Turkey due to the concentration and conglomeration of Turkish media. In a more recent study on the 2011 Turkish elections, Çarkoğlu et al. (2014) confirmed earlier findings of high press–party parallelism in terms of major Turkish newspapers’ political favouring of either AKP or Cumhuriyet Halk Partisi (Republican People’s party (CHP)). Toker (2015) had similar findings from comparing media coverage of the 2007 and the 2011 elections. Against this backdrop of increased press–party parallelism, this study explores the differential coverage of the political crisis due to the corruption probe in terms of the different frames that three competing media outlets preferred to adopt.

The 17 December 2013 political corruption scandal

Corruption allegations made on 17 December 2013 against several AKP government ministers and their sons turned the local elections on 30 March 2014 to one of the tensest in Turkey’s political history. Turkey subsequently witnessed an election in which allegations of corruption and counter allegations of a coup attempt set the agenda much more than issues of local governance or the candidates’ visions.

Differentiation in the Turkish media over the scandal mainly stemmed from news outlets’ particular affiliations with conflicting groups (e.g. the pro-AKP Sabah versus the pro-Gülen movement Zaman) in their ontological struggle concerning the corruption probe. As former allies,⁴ the AKP⁵ and the Gülen/Hizmet movement (Fethullahçı)⁶ had ‘domesticated’ the Kemalist⁷ military establishment through the institutionalization of the economy and the military’s marginalization. However, they then began to perceive each other as threats to the constitutional order.

Expectations and hypotheses

The crises leading up to the 2014 local elections provided a fertile ground for exploring differential use of media frames by three newspapers that have been reliably demonstrated to favour specific political camps, either through organic ties or past coverage. However, this study does not aim just to describe the issue frames employed by the newspapers. Rather, we wish to show that, given the political alliances, frame differentiation demonstrates a dynamic that can be inferred from the nature of the crises and the political alliances (implicit or explicit) between media groups and political movements or parties. We therefore propose the following interconnected hypotheses based on previous framing literature and the dynamics of party–press parallelism in Turkey:

H1. Outlets will adopt framing practices that make it easier to defend their political allegiances to help maintain the ontological security of their political camp.

This main hypothesis can be broken down into three specific sub-hypotheses:

H1A. Pro-government Sabah will focus on helping to defend AKP but, given the compelling nature of the evidence, will refuse to adopt the corruption frame.

That is, we expect Sabah to choose frames that deflect attention from the evidence presented by other news outlets:

H1B. Kemalist Sözcü will frame the issue in order to more easily attack and assign blame to AKP and its leaders, that is, through the corruption and personification frames.

H1C. Anti-AKP government, pro-Gülen Zaman will similarly focus on using the corruption probe to undermine the government’s authority. However, having until recently been allied to AKP, they will refrain from frames that can reflect negatively on the Gülen movement itself.

That is, we expect Zaman to adopt the corruption frame but not the personification frame, keeping its discussion at an institutional level lest it draws pro-Gülen politicians and bureaucrats into the corruption debate.

In addition to these three closely related hypotheses, we also hypothesize that, facing an abrupt, probably unexpected political assault, Sabah will initially adopt a defensive posture to deflect the assaults before later going on the offensive by attacking its political adversaries in AKP.

Methodology

This study analyses the frames used in three Turkish national newspapers from 18 December 2013, when the investigations began, to 30 March 2014, when the local elections took place.

Data collection

Three daily newspapers were selected for the analysis: Sabah, Sözcü and Zaman. They were chosen because they have high circulation rates⁸ and their editorial policies closely follow the conflicting sides’ ideologies. Because readers may not be familiar with these media outlets, we briefly describe their styles of reporting.⁹

Sabah

Sabah is a pro-AKP national newspaper. After its acquisition by the pro-government Calik Group, the paper became critical of CHP and a staunch defender of AKP. During the corruption probe, Sabah depicted the government ministers and their family members who were under investigation as victims of the actions of a parallel state (paralel devlet), which it claimed was carrying out a vicious campaign to smear the government. Sabah depicted this as an attack on the state itself.

Sözcü

This is a Kemalist nationalist paper with strong adherence to Republican ideals, particularly secularism, which makes it distinctly anti-AKP. During the corruption probe, Sözcü harshly criticized members of the government, particularly Erdoğan, who was prime minister at the time, by presenting them as being guilty of corruption.

Zaman

Since its inception, Zaman was an advocate of moderate Islam, as taught by Fethullah Gülen.¹⁰ Allegedly financed by Gülen, Zaman supported the AKP government between 2002 and 2009 and was critical of the MHP and CHP. However, as tensions rose between Erdoğan and Gülen, Zaman began to oppose the AKP and criticized the government during the probe.

Front page articles, as providing the most prominent coverage of the issue, were manually selected for the content analysis. All selected articles dealt with the corruption probe in terms of the investigation itself, detentions, recordings of phone conversations between government officials and the speeches of people involved with the parties. In total, 708 newspaper articles were coded.

Coding procedure

For operationalization of the news frames, we drew heavily on measures developed by Semetko and Valkenburg (2000). Question items were developed and applied to all stories and had to be answered by coders with yes or no. Coding was carried out by two native speakers of Turkish. Both coders read and coded the articles independently. However, time and resource constraints prevented us from employing both coders for the entire article pool. Instead, the article pool was divided into three groups. Each coder was assigned to one of two groups, while the third group was coded by both for reliability checks. The commonly coded articles were analysed for intercoder reliability, yielding a Krippendorff’s (2004) alpha coefficient of .82. Considering the sheer number of discrete categories to be coded (i.e. the total number of different frames), the reliability indicates that intercoder reliability was at more than acceptable levels.

Results and discussion

Table 2 summarizes the frame usage of the three newspapers for the entire period under study (18 December 2013 to 30 March 2014). The first pattern to be observed is the concentration in frame employment. For Sabah, the most frequently used three frames (sinister powers 65.56%, smear campaign 17.43% and personification 8.30%) account for 91.30% of all frames used. For Sözcü, the same figure (personification 15.98%, corruption 36.89% and tampering 30.33%) account for 83.20%. For Zaman, corruption 25.70, tampering 56.33% and fracture within sociopolitical camp 6.53% account for 88.57 %. For each newspaper, the next most frequently used frame is employed less than 5%of the time. This demonstrates that all three newspapers were indeed focussed on relatively few framing tactics. While we identified and coded 11 separate frames, our focus is on frames that are directly related to the political crisis of our study period.

Table 2.

Total number of frames adopted by news outlets.

	Sabah (pro-AKP)		Sözcü (pro-Kemalist)		Zaman (pro-Gülen)
	N	%	N	%	N	%
Personification	20	8.30	39	15.98	5	2.04
Popular reaction	9	3.73	12	4.92	4	1.63
Conquered booty	0	0.00	5	2.05	2	0.82
Comparisons	2	0.83	1	0.41	1	0.41
Corruption	0	0.00	90	36.89	63	25.71
Smear campaign	42	17.43	1	0.41	2	0.82
Tampering	2	0.83	74	30.33	138	56.33
Sinister powers	158	65.56	6	2.46	0	0.00
Fracture within sociopolitical camp	0	0.00	4	1.64	16	6.53
Fabricated news	5	2.07	6	2.46	12	4.90
Economic consequences	3	1.24	6	2.46	2	0.82

Figures 1 to 3 show the evolution of the most frequently used frames for the newspapers. In these figures, frame usage is presented as a percentage of total frames, which allows us to control for the newspapers’ stylistic choices, such as page size and number of stories run on the first page. In order to control for daily spikes in content (e.g. a day with no frames must be coded as 0%, creating a sharp spike in the graph), we used 5-day rolling averages. Considering the three figures together provides a temporal map of the period in terms of most frequently used frames.

Testing the hypotheses

H1A. Pro-government Sabah will focus on helping to defend AKP, but given the forceful and graphic nature of alleged evidence will refuse to adopt or function within the corruption frame. Sabah is expected to choose frames that deflect attention from the evidence presented.

For Sabah, the sinister powers frame was the most significant, being adopted from the very first day of the corruption probe and utilized continuously until the end of our research period. The smear campaign frame was the second most prevalent frame in Sabah.

When the investigation started, Sabah presented state officers and government officials and their families as victims of a smear campaign. However, in the first weeks of the investigation, the newspaper did not clearly identify the supposed powers behind this campaign. The term ‘parallel state’ started to emerge in mid-January. As Fethullah Gülen and the Hizmet movement began to be depicted as the once allied but now treacherous power that was supposedly behind all the attempts to undermine the government.

Our longitudinal analysis reveals that in December and in early January, Sabah mainly adopted the smear campaign frame. The defensive nature of the smear campaign frame helped Sabah gain time to articulate its own interpretation of the event. At the onset of the corruption crisis, Sabah arguably displays a reactionary and defensive strategy by adopting what we call the smear campaign frame as much as the sinister powers frame. As the initial shock of the corruption claims is overcome, the newspaper’s emphasis switches towards distributing blame, particularly by focusing on outside sources of influence. Thus, after mid-January, the newspaper began regularly using the sinister powers frame to ensure the group’s ontological security.

Personification was the third most commonly used frame in Sabah. Although this frame was employed by all the news outlets analysed in this paper, its content varied significantly. Sabah placed much emphasis on officials, such as district attorneys, who were investigating the probe,¹¹ depicting them as abusing their authority. However, Fethullah Gülen himself was the focal point of Sabah’s personification frame. Sabah described Gülen and the Hizmet movement as a sinister power being used as a subcontractor by larger international ‘evil powers’ that are ‘eager’ to prevent the rise of the ‘New Turkey’.

Equally importantly though, Sabah never used a corruption frame, even in defence against an attack. While a hesitant stance, especially initially during the campaign, is to be expected, the total lack of the corruption frame suggests a very high degree of constraint and discipline within the media group. Our data do not allow us to offer a deeper understanding of this observation nor does it provide any evidence as to the motivation behind it, but we might offer an educated guess. The corruption probe and its coverage brought a frenzy almost unprecedented in Turkish politics. The evidence (voice recordings, photographs, etc.) was quite detailed and sometimes quite disturbing, with some being very difficult to challenge. If the newspaper had been drawn into a discussion of specifics, it could have damaged the AKP. From an ontological security perspective, therefore, the newspaper would have both weakened the AKP’s position against its political adversaries and damaged its image in the eyes of its supporters if it had adopted the corruption frame. Accordingly, Sabah, reflecting the government’s strategy, first assumed a defensive posture that used the smear campaign frame as frequently as the sinister powers frame before gradually adopting a more offensive (hence deflecting attention from the evidence) stance that emphasized sinister powers (both domestic and international).

H1B. Kemalist Sözcü will frame the issue that allows attacking AKP and assigning blame on AKP and its leaders easier, that is, corruption and personification frames.

Lacking a clear connection to a political party, Sözcü seems to have felt relative freedom in attacking the AKP government and its leaders during the period under study. Initially, corruption and personification frames together accounted for almost 80% of the frame usage. Sözcü also started using the tampering frame early on. Unsurprisingly, it used the corruption frame most often, immediately accepting the allegations made during the investigations and presenting them as completely true to argue that the suspects were guilty. For most of the period under study, the corruption frame saw steady usage. However, as the election approached, Sözcü refocussed almost entirely on the corruption frame, probably having decided that this would be most effective at swinging votes away from AKP.

Whereas Sabah increasingly utilized the sinister powers frame to explain AKP actions regarding the government officials involved in the investigation, Sözcü increasingly adopted the tampering frame as the antithesis of the sinister powers frame. From Sözcü’s perspective, the AKP was trying to prevent the corruption investigation functioning properly. Indeed, these same state officials (Tayyip Erdoğan and the government ministers at the time), their families (including Bilal Erdoğan and the sons of other ministers) and business circles (Reza Zarrab, Süleyman Aslan, etc.) were frequently presented via a frame of personification as the embodiment of corruption itself. Indeed, Sözcü utilized personification most often of the three newspapers. To some extent, Sözcü mentioned the same individuals as Sabah. However, while Sözcü used this frame in relation to a large pool of actors, the other outlets were much more selective. As described earlier, as regards Sabah and as will be described for Zaman, those outlets preferred to restrict their targets to those representing opposing ideologies. Like Sabah, the popular reaction frame was used by Sözcü, but in this case as a way to undermine the AKP’s legitimacy. Riots, demonstrations, protests and interviews with anti-AKP and Kemalist members of the public were Sözcü’s main sources of content as it adopted a popular reaction frame. In terms of the ways that outlets dramatized the events through narratives, Sözcü used the frame of conquered booty to make the case even more dramatic. The money counters, the money found in shoeboxes at the home of a bank manager close to AKP business circles and a watch worth US$300,000 that had been given to a former economy minister are prominent examples of the content of Sözcü’s conquered booty frame.

H1C. Pro-Gülen Zaman, also opposing AKP government, will similarly focus on using the corruption probe to undermine the government’s authority but having been allied to AKP until recently they will refrain from frames that can reflect negatively on the Gülen movement itself.

The pro-Gülen Zaman primarily used the frames of tampering and corruption, and, to a lesser extent, fractures within the sociopolitical camp. In December 2013, Zaman used corruption and tampering almost equally (23 and 27 times, respectively). In January 2014, however, when the reappointment and detention of officials investigating the probe intensified and Fethullah Gülen and the Hizmet movements were presented as enemies of the state in Sabah, Zaman began to increase its usage of tampering. Arguably, this was the outcome of attempts to secure its ontology against Sabah’s use of the smear campaign frame against the Hizmet movement. Zaman also adopted this frame to present those affected by the government’s actions as victims. Although Zaman and Sözcü used the corruption frame fairly equally during first month of the probe, Zaman used it much less than Sözcü after mid-January.

In utilizing the fracture within the sociopolitical camp frame, Zaman tried to suggest that the AKP’s formal discourse is inherently flawed. This frame also served as an indicator to claim that there are people even within the AKP’s ranks who sympathize with the Hizmet movement. As discussed above, Zaman and Sabah accused each other of unethical journalism. Zaman’s adoption of the fabricated news frame was used primarily to attack Sabah to claim that the outlet was manufacturing interviews and news to draw attention away from the inconsistency of its position.

Unlike Sözcü, however, Zaman was not free to develop its corruption probe coverage without restraint. Having been organically tied to the Gülen movement, which was itself a close ally of the AKP until recently, the newspaper had to approach the corruption scandal with more care and constraint than Sözcü (and other opposition outlets). Therefore, despite increasing personal attacks towards Fethullah Gülen himself, Zaman refrained from using the personification frame to the extent that it was never used after the first week of January 2014. On the other hand, relying too heavily on the corruption frame was not feasible either since the Gülen movement had members positioned in senior bureaucratic and political positions ever since AKP’s first election victory in 2002. Moving the debate onto personal or institutional grounds could therefore reflect negatively on people associated with the Gülen movement who were also influential within AKP. Thus, this limited Zaman to frequent use of the only remaining frame suitable for attacking the AKP, namely tampering.

Conclusion

We selected the 17 December corruption probe in Turkey’s highly polarized media system as our case on the basis that it was an event in which a variety of actors (the AKP, the Gülen movement and the Kemalists) would struggle to protect their ontological security via their affiliated media outlets. Indeed, our findings demonstrate that, due to overt partisanship within the Turkish mediascape at this critical conjuncture, there was significant political parallelism in line with each news outlet’s affiliated political actors. Overall, our findings not only confirm earlier studies on press–party parallelism (Sabah-AKP) but also reveal press–sociopolitical camp parallelism (Sözcü-Kemalists and Zaman-Gülenists) in Turkey’s polarized media system.

Although party–press parallelism is not a novel phenomenon in Turkish politics, the way it transcends party lines has not been studied extensively before. While others have shown parallels between newspapers and parties, our primary goal in this study was to show that party–press parallelism in Turkey extends further than newspapers demonstrating a bias towards certain political parties. First, we have demonstrated that newspapers adopt framing strategies commensurate with their political alliances. While this is not a novel finding, we also showed that, during a severe political crisis, the newspapers focussed on a narrow band of framing tactics designed to respond to the crisis in support and defence of a specific political group or party. In addition, and probably more importantly, we have demonstrated how frame usage of newspapers that are considered to oppose the government varies.

This can be considered as proving our ontological security argument. In choosing frames, the main goal of an opposition newspaper is not only to oppose the government but also to do it in a way to maintain the ontological security of the political camp it belongs to. Thus, lacking organic ties to any political party, Sözcü chose more freely among the available frames while Zaman, having allied itself organically with the Gülen movement, had to refrain from using frames that could highlight the quite recent alliance between the AKP and the Gülen movement.

Further studies concerning issues of corruption could be used to test the explanatory power of the frames proposed in this study. They could also offer ways to compare and contrast the state institutional, political and social contexts in which news is produced. An intra-national comparative perspective could also be helpful in further testing the validity of our proposed frames in differing cultural settings (e.g. democracies and semi-democracies) and applying qualitative methods, such as discourse analysis, could result in more insightful research outputs.

Footnotes

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

Notes

References

Akser

Baybars-Hawks

(2012) Media and democracy in Turkey: Toward a model of neoliberal media autocracy. Middle East Journal of Culture and Communication 5(3): 302–321.

Allen

Savigny

(2012) Selling scandal or ideology? The politics of business crime coverage. European Journal of Communication 27(3): 278–290.

Amundsen

(1999) Political corruption: An introduction to the issues. CMI Working paper 1999:7, December. Bergen: Chr. Michelsen Institute.

Arroyo

(1997) Fábulas y Fabuladores; el Escándalo Politico Como Fenómeno de los Medios de Comunicación [Stories and Tellers; Political Scandal as a Phenomenon of the Media]. In: Laporta

Alvarez

(eds) La Corrupción Política [Political Corruption]. Madrid: Alianza Editorial, pp. 335–359.

Bacık

Kurt

(2011) New Islamic movements and a modern networks. Culture and Religion 12(1): 21–37.

Bilici

(2006) The Fethullah Gülen movement and its politics of representation in Turkey. The Muslim World 96(1): 1–20.

Canel

Sanders

(2006) Morality Tales: Political Scandals and Journalism in Britain and Spain in the 1990s. Cresskill, NJ: Hampton Press.

Cappella

Jamieson

(1997) Spiral of Cynicism: The Press and the Public Good. New York: Oxford University Press.

Çarkoğlu

Yavuz

(2010) Press–party parallelism in Turkey: An individual level interpretation. Turkish Studies 11(4): 613–624.

10.

Çarkoğlu

Baruh

Yıldırım

(2014) Press-party parallelism and polarization of news media during an election campaign: The case of the 2011 Turkish elections. The International Journal of Press/Politics 19(3): 295–317.

11.

Christensen

(2010) Notes on the public sphere on a national and post-national axis Journalism and freedom of expression in Turkey. Global Media and Communication 6(2): 177–197.

12.

De Vreese

Peter

Semetko

(2001) Framing politics at the launch of the Euro: A cross-national comparative study of frames in the news. Political Communication 18(2): 107–122.

13.

Entman

(1993) Framing: Towards clarification of a fractured paradigm. Journal of Communication 43(4): 51–58.

14.

Giddens

(1991) Modernity and Self-Identity. New York: Polity Press.

15.

Goldman

Mutz

(2011) The friendly media phenomenon: A cross-national analysis of cross-cutting exposure. Political Communication 28(1): 42–66.

16.

Hallin

Mancini

(2004) Comparing Media Systems: Three Models of Media and Politics. Cambridge: Cambridge University Press.

17.

Hallin

Mancini

(ed.) (2012) Comparing Media Systems beyond the Western World. Cambridge: Cambridge University Press.

18.

Hanioğlu

MŞ

(2013) Kemalizm’ in Tarihi Kökenleri [Historical roots of Kemalism]. In: Kuru

Stepan

(eds) Türkiye’de Demokrasi, İslam ve Laiklik [Democracy, Islam and Secularism in Turkey]. İstanbul: İstanbul Bilgi University, pp. 31–58.

19.

Heper

Demirel

(1996) The press and the consolidation of democracy in Turkey. Middle Eastern Studies 32(2): 109–123.

20.

Iyengar

(1991) Is Anyone Responsible? How Television Frames Political Issues. Chicago, IL: University of Chicago Press.

21.

Jiménez

(1995) Detrás del Escándalo Político. Opinión Pública, Dinero y Poder en la España del Siglo XX [Behind the Political Scandal. Public Opinion, Money and Power in the Twentieth Century Spain]. Barcelona: Tusquets.

22.

Kalogeropoulos

Albæk

De Vreese

. (2015) The predictors of economic sophistication: Media, interpersonal communication and negative economic experiences. European Journal of Communication 30(4): 385–403.

23.

Kaya

Çakmur

(2010) Politics and the mass media in Turkey. Turkish Studies 11(4): 521–537.

24.

Kostadinova

Dimitrova

(2012) Communicating policy change: Media framing of economic news in post-communist Bulgaria. European Journal of Communication 27(2): 171–186.

25.

Krippendorff

(2004) Reliability in content analysis: Some common misconceptions and recommendations. Human Communication Research 30(3): 411–433.

26.

Laing

(1990) The Divided Self: An Existential Study in Sanity and Madness. New York: Penguin Books.

27.

Lorasdağı

(2013) Türkiye’de Küreselleşme, Modernleşme ve Demokratikleşme: Fethullah Gülen Hareketi [Globalization, modernization and democratization in Turkey: Fethullah Gulen movement]. In: Keyman

(ed.) Türkiye’nin Yeniden İnşası: Modernleşme, Demokratikleşme, Kimlik [Turkey’s Reconstruction: Modernization, Democratization, Identity]. İstanbul: İstanbul Bilgi University, pp. 149–174.

28.

Mateescu

(2006) Kemalism in the era of totalitarianism: A conceptual analysis. Turkish Studies 7(2): 225–241.

29.

Minsky

(1975) A Framework for Representing Knowledge. In: Winston

(ed.) The Psychology of Computer Vision. New York: McGraw-Hill, pp. 211–277.

30.

Nye

(1967) Corruption and political development. American Political Science Review 61(2): 417–427.

31.

Oğurlu

Öncü

(2013) The Laic-Islamist Schism in the Turkish Dominant Class and Media. In Balkan

Balkan

Öncü

(eds) The Neoliberal Landscape and the Rise of Islamist Capital in Turkey. Oxford: Berghahn Books, pp. 272–287.

32.

Patterson

(1997) The news media: An effective political actor? Political Communication 14(4): 445–455.

33.

Peters

(2003) The Media’s Role: Covering or Covering Up Corruption? Transparency International Global Corruption Report. Berlin: TI, pp. 44–56.

34.

Protess

Cook

Doppelt

Ettema

Gordon

Leff

Miller

(1991) The Journalism of Outrage: Investigative reporting and agenda building in America. New York: Guilford Press.

35.

Schudson

(2008) The Lippmann-Dewey Debate and the Invention of Walter Lippmann as an Anti-Democrat 1986–1996. International Journal of Communication 2(12): 1031–1042.

36.

Semetko

Valkenburg

(2000) Framing European politics: A content analysis of press and television news. Journal of Communication 50(2): 93–109.

37.

Seymour-Ure

(1974) The Political Impact of Mass Media. London: SAGE.

38.

Sherman

(1978) Scandal and Reform: Controlling Police Corruption. Berkeley: University of California Press.

39.

Simon

Xenos

(2000) Media Framing and Effective Public Deliberation. Political Communication 17(4): 363–376.

40.

Şık

(2014) Paralel Yürüdük Biz Bu Yollarda [We Walked in Parallel through All Those Ways]. İstanbul: Postacı.

41.

Strömbäck

Dimitrova

(2006) Political and media systems matter a comparison of election news coverage in Sweden and the United States. The Harvard International Journal of Press/Politics 11(4): 131–147.

42.

Thompson

(2013) Political Scandal: Power and Visibility in the Media Age. Hoboken, NJ: John Wiley & Sons.

43.

Toker

(2015) More national less European? The comparison of the last two general elections in Turkey. European Journal of Communication 30(2): 188–208.

44.

Whipple

(2005) The Dewey-Lippmann Debate Today: Communication Distortions, Reflective Agency, and Participatory Democracy. Sociological Theory 23(2): 156–178.

45.

Yavuz

(ed.) (2006) The Emergence of a New Turkey: Democracy and the AK Parti. Salt Lake City, UT: The University of Utah Press.

46.

Yavuz

(2013) Toward an Islamic Enlightenment: The Gülen Movement. Oxford: Oxford University Press.

47.

Zamora

Albaladejo

JAM

(2010) La representación simbólica del escándalo político. Hacia una tipología de los marcos periodísticos (frames) utilizados en la narración del escándalo de corrupción política [The symbolic representation of political scandal. Towards a typology of journalistic frames used in the narration of political corruption scandal]. Razón y Palabra 15(73): 1–27.