Hacettepe University Graduate School of Social Sciences Department of English Linguistics EVALUATION AND CONSTRUCTION OF POLITICAL REALITY IN THE SPEECHES OF TURKISH POLITICAL PARTY LEADERS: ‘JUDGEMENT’ AND ‘ENGAGEMENT’ ANALYSIS Ayşe Dilek DEMİRTAŞ Phd Dissertation Ankara, 2017         EVALUATION AND CONSTRUCTION OF POLITICAL REALITY IN THE SPEECHES OF TURKISH POLITICAL PARTY LEADERS: ‘JUDGEMENT’ AND ‘ENGAGEMENT’ ANALYSIS Ayşe Dilek Demirtaş Hacettepe University Graduate School of Social Sciences Department of English Linguistics PhD Dissertation Ankara, 2017 v     to my parents, Gülten & İsmail Hakkı for their endless love, support and encouragement vi     ACKNOWLEDGEMENTS I am grateful to many people for supporting me with their academic expertise and moral support in the writing process of this dissertation. Firstly, I would like to thank my supervisor, Prof. Dr. Işıl Özyıldırım for her invaluable guidance, encouragement, patience and insightful comments. I am deeply grateful to her for being with me in this process. I want to point out that I have always valued her detailed and attentive feedback on the drafts of this work. I would like to express my gratitude to Prof. Dr. Nalan Büyükkantarcıoğlu and Prof. Dr. Gülsün Leyla Uzun, who shared their invaluable suggestions with me during the Thesis Supervising Committee meetings. Prof. Uzun`s feedback and encouraging comments enlightened this study and motivated me to do more critical work. More particularly, she helped me expand my knowledge of modality phenomenon in linguistics. Besides, Prof. Büyükkantarcıoğlu`s inspiration and encouragement contributed to my understanding and helped me develop my ideas. She helped me realize the significance of my research within the linguistic world. I want to thank Assoc. Prof. Dr. Emine Yarar for her generous time, attention and advice. With her sophisticated manner, she helped me categorize my data whenever I got lost. I also want to express my thanks to Prof. Dr. Yeşim Aksan for accepting to be one of my jury members, reading this work and giving constructive feedback on it. Additionally, I would like to thank to my friends and colleagues for their encouragement and support they provided. They tolerated me during this period and shared their worthwhile friendship with me. Finally, my greatest thanks go to my beloved parents, Gülten and İsmail Hakkı, and my brother Onur, who have always been at my side with their pure and endless love. I cannot find words to describe the immense support they have offered not only during my years of study only, but also my whole life. I owe you a lot and I love you. vii     ÖZET DEMİRTAŞ, Ayşe Dilek. Türk Liderlerin Parti Grup Konuşmalarında Değerlendirme ve Gerçekliğin Kurgulanışı: ‘Yargılama’ ve ‘Konumlanma’ İfadelerinin Çözümlenmesi, Doktora Tezi, Ankara, 2017. Bu çalışmada, Değerlendirme Kuramı`ndan (Appraisal Theory) yola çıkılarak, Türkçe politik parti grup konuşmalarında kullanılan Yargılama (Judgement) ve Konumlanma (Engagement) stratejilerinin incelenmesi amaçlanmıştır. Siyasi parti liderlerinin değerlendirme ifadelerinde kullandıkları dilsel stratejiler konuşmalarındaki gerçekliğin kurgulanış ve aktarılış yollarını göstermesi bakımından önem taşımaktadır. Türk siyasi hayatında çok büyük kırılma noktalarını oluşturan önemli politik olayların yaşandığı 8 Kasım 2013 - 25 Şubat 2014 tarihleri arasında yapılmış ve ilgili partilerin internet sitelerinde kamuoyu ile paylaşılmış olan ‘57’ parti grup konuşması çalışmanın derlemini oluşturmuştur. Bu bağlamda konuşmaların yapıldığı tarihlerde parti genel başkanları olarak görev yapan Recep Tayyip Erdoğan (AKP), Kemal Kılıçdaroğlu (CHP), Devlet Bahçeli (MHP) ve Selahattin Demirtaş`ın (BDP) haftalık parti grup konuşmaları çözümlenmiştir. Aynı konuları incelemek ve parti liderleri arasında eşitliği sağlamak adına seçilen konuşmalardan 10 söylem alanı belirlenmiştir. Bunlar ‘Gezi olayları’, ‘ekonomi’, ‘demokrasi’, ‘yargı’, ‘yolsuzluk’, ‘terör ve Kürt sorunu’, ‘Orta Doğu’, ‘30 Mart yerel seçimleri’, ‘eğitim’ ve ‘din’ alanlarından oluşmaktadır. Liderlerin yargılarında kullandıkları dilsel araçlar incelenerek, ilk olarak, açık şekilde ifade edilen Yargılamaların sözcüksel-dilbilgisel kategorileri belirlenmiştir. Ayrıca politik söylemde örtük ifadeler de çok yaygın kullanıldığı için, eğretileme (metaphor) ve ilişkilendirilmişlikler (invoked Judgement) aracılığıyla yargılama ifade eden yapılar da belirlenmiştir. Yargılama ifadeleri dışında, Türk Parlamentosu`ndaki parti liderlerinin değerlendirmelerini ‘söyleşimsel’ (dialogic) olarak nasıl ifade ettiklerini göstermek için Konumlanma bildiren dilsel yapılar da çözümlenmiştir. Çalışmanın sonuçları incelenen parti grup konuşmalarında değerlendirme dili için belli yapısal kategoriler bulunduğunu göstermiştir. Ayrıca parti liderlerinin ‘öz değerlendirme’ viii     (self-evaluation) ve ‘diğerini değerlendirme’ (other-evaluation) süreçlerinde kullandıkları ifadelerde de farklılıklar bulunduğu gözlemlenmiştir. Konumlanma gösteren dilsel yapıların çözümlenmesinde de liderlerin bir takım dilsel yapıları daha ön planda kullandıkları ortaya çıkmıştır. Muhalefet partilerinin liderlerinin iktidar partisi lideri ile aynı politik güce sahip olmadıkları için olumsuz eleştirilerinde daha çok ‘söyleşimsel açılım’ (dialogic expansion) gösteren dilsel yapıları tercih ettikleri, ve bu şekilde önermelerinin sorumluluğunu dinleyicileriyle paylaştıları görülmüştür. Bu liderlerin öz değerlendirme süreçlerinde ise daha çok ‘söyleşimsel daralma’ (dialogic contraction) ifadelerini kullandıkları ve değerlendirmelerinin sorumluluğunu alarak öznelliklerini daha çok ifade ettikleri gözlemlenmiştir. İktidar partisi liderinin ise hem ‘öz değerlendirme’ hem de ‘diğerini değerlendirme’ süreçlerinde söyleşimsel daralma ifadelerini daha çok tercih ettiği ortaya çıkmıştır. Tüm bu sonuçlar liderlerin politik güç durumlarının ve ideolojik duruşlarının, dilsel değerlendirme ve gerçekliğin kurgulanması süreçlerinde kullandıkları retorik stratejilerine etkisini göstermesi bakımından önem taşımaktadır. Anahtar Sözcükler Değerlendirme Kuramı, Türk politika söylemi, değerlendirme dili, kişilerarası anlam, örtülü anlam, açık anlam, söyleşimsel yaklaşım, Yargı, Konumlanma, gerçekliğin kurgulanması ix     ABSTRACT DEMİRTAŞ, Ayşe Dilek. Evaluation and Construction of Political Reality in the Speeches of Turkish Political Party Leaders: ‘Judgement’ and ‘Engagement’ Analysis, PhD Dissertation, Ankara, 2017. Based on the Appraisal Theory, this study attempts to explore the Judgement and Engagement strategies employed in Turkish political party group speeches. The strategies that political party leaders use in their evaluative language have a significant importance in demonstrating the ways in which they construct and express realities. ‘57’ group speeches given and shared with the public through the parties`websites on the dates between October 8th, 2013 and February 25th, 2014, during which many significant political issues considered as breaking points in Turkish politics occurred, formed the data of the study. In this sense, the weekly-held party group speeches of Recep Tayyip Erdoğan (AKP), Kemal Kılıçdaroğlu (CHP), Devlet Bahçeli (MHP), and Selahattin Demirtaş (BDP), who were the party leaders on the dates the speeches given were analysed. In order to analyse similar concepts and satisfy equal conditions among party leaders, 10 discourse areas were determined. These areas were ‘Gezi events’, ‘economy’, ‘democracy’, ‘justice’, ‘corruption’, ‘terror & Kurdish question’, ‘Middle East’, ‘30 March elections’, ‘education’, and ‘religion’. By analysing the linguistic devices employed by party leaders in their evaluations, lexico-grammatical categories of overt Judgemental expressions were identified first. Besides, as implicitly structured items are commonly used in political discourse, certain implicit Judgemental realisations were also detected, including metaphors and invoked judgements. Apart from these Judgemental expressions, Engagement resources were also analyzed in order to reveal how party leaders in Turkish Parliament express their evaluations dialogically. Findings of the study indicated certain structural categories for evaluative language within the examined party group speeches. There were also differences among x       leaders` evaluative expressions in their self-versus-other-based evaluations specifically. The analysis of linguistic resources showing Engagement has also indicated that leaders employ some linguistic devices more. It was realized that as the opposition parties` leaders do not have the same political power as the ruling party leader, they generally tend to use ‘dialogic expansion resources’ in their negative criticisms and share the responsibility of their propositions with their listeners. In their self evaluation processes, on the other hand, it was observed that they employ ‘dialogic contraction’ resources and express their subjectivity more by getting the responsibility of their evaluations. On the contrary, it came out that the ruling party leader prefers dialogic contraction resources more both in his ‘self-evaluations’ and ‘other-evaluations’. All these findings have great importance in demonstrating the influence of leaders` political power and ideological stance on their rhetorical strategies through which they make evaluations and construct reality. Keywords Appraisal theory, Turkish political discourse, evaluative language, interpersonal meaning, implicit language, explicit language, dialogic approach, Judgement, Engagement, construction of reality xi     TABLE OF CONTENTS KABUL VE ONAY…………………………………………………………………….. i BİLDİRİM………………………………………………………………………………. ii YAYIMLAMA VE FİKRİ MÜLKİYET HAKLARI BEYANI………………………… iii ETİK BEYAN……………………………………………………………………….….. iv ACKNOWLEDGEMENTS…………………………………………………………..… vi ÖZET…………………………………………………………………….…………….... vii ABSTRACT……………………………………………………….………………….… ix TABLE OF CONTENTS……………………………………….………………....…… xi LIST OF ABBREVIATIONS……………………………………….………….….…... xvii LIST OF TABLES…………………………………………….………………….…….. xviii LIST OF FIGURES……………………………………………………………….……. xix CHAPTER 1: INTRODUCTION ................................................................. 1 1.1. CLEARING THE GROUNDS: FROM EVALUATION TO REALITY CONSTRUCTION…………………………................................................ 1 1.2. INTRODUCING THE STUDY…………………………………….………… 4 1.2.1. Statement of the Problem.............................................................. 7 1.2.2. Purpose of the Study…………………………………………………. 8 1.2.3. Research Questions ………………………….................................. 10 1.2.4. Significance of the Study ……………………………………………. 10 1.2.5. Limitations of the Study…………………………..……………….….. 12 1.2.6. Organization ………………………………..…………………….…… 13 CHAPTER 2: BASIC CONCEPTS AND RELATED LITERATURE........... 15 2.1. POLITICAL DISCOURSE ....................................................................... 15 2.1.1. Political Speech as a Specific Genre ............................................. 18 2.1.2. Studies on Turkish Political Language ........................................... 21 2.2. LANGUAGE OF EVALUATION .............................................................. 23 2.2.1. Evaluation ....................................................................................... 25 2.2.2. Local Grammar Approach............................................................... 25 2.2.3. Stance............................................................................................. 27 2.2.4. Parameter-based approach............................................................. 28 xii     CHAPTER 3: THEORETICAL FRAMEWORK........................................... 31 3.1. APPRAISAL THEORY…………………………..………………….............. 31 3.1.1. Roots of Appraisal Theory................................................................ 32 3.1.1.1. Language as a System: Metafunctions in SFL………..…. 34 3.1.2. Components of Appraisal Model ………………..…….………….….. 35 3.1.2.1. Attitude ………………………………………………….….… 36 3.1.2.1.1. Affect……………………..……………………..…. 37 3.1.2.1.2. Appreciation……………..................................... 37 3.1.2.1.3. Judgement……………………………………….. 38 3.1.2.1.3.1. Inscribed Judgement…………………... 39 3.1.2.1.3.2. Invoked Judgement……………………. 40 3.1.2.2. Engagement …………………………….............................. 41 3.1.2.2.1. Dialogic Contraction………………..………….…. 43 3.1.2.2.1.1. Proclaimers…………............................ 44 3.1.2.2.1.2. Disclaimers……………………………... 44 3.1.2.2.2. Dialogic Expansion…………………………….… 45 3.1.2.2.2.1. Entertainment…………………………… 46 3.1.2.2.2.2. Attribution……………………………..… 46 3.1.2.2.3. Relationship between Engagement and Modality……………………………………………. 47 3.1.2.3. Graduation …………………………………………….…...... 48 3.2. APPRAISAL IN DISCOURSE STUDIES…………………………………. 49 3.2.1. Appraisal in Related Literature………….………………………….… 49 3.2.2. Appraisal and Political Discourse ………………….…………….….. 52 CHAPTER 4: METHODOLOGY................................................................. 56 4.1. RESEARCH DESIGN…………..………………………….…………….….. 56 4.1.1. Focus of the Analysis ..................................................................... 56 4.1.2. Data……………............................................................................... 58 4.1.2.1. Party Information.............................................................. 59 4.1.2.2. Data Selection…………………………………………...….. 61 4.2. DATA ANALYSIS………………………..………………….………………. 62 4.2.1. Identification of Discourse Areas ..................................................... 62 4.2.1.1. Gezi Events …………………….………………………...…. 63 4.2.1.2. Economy …………………………………………………….. 64 xiii     4.2.1.3. Democracy …………………………………………..……… 65 4.2.1.4. Justice …………………………………………………......... 65 4.2.1.5. Corruption ………………………………………………....... 66 4.2.1.6. Terror and Kurdish Question……………….…..……….…. 67 4.2.1.7. Middle East …………………………………..…….............. 68 4.2.1.8. 30 March Elections ……………………....………………… 68 4.2.1.9. Education ………………………………….….…………….. 69 4.2.1.10. Religion ………………………………………………......... 69 4.2.2. Identification of Evaluative Linguistic Items in Turkish ………….… 71 4.2.2.1. Judgement Analysis .......................................................... 71 4.2.2.1.1. Evaluative Phrases............................................. 71 4.2.2.1.2. Evaluative Clauses.............................................. 72 4.2.2.2. Engagement (source of evaluation) Analysis……………. 74 4.2.2.2.1. Dialogic Contraction Categories………………… 75 4.2.2.2.2. Dialogic Expansion Categories……………….…. 76 CHAPTER 5: FINDINGS AND DISCUSSION........................................... 82 5.1. FINDINGS OF JUDGEMENT ANALYSIS…………….……………….….. 82 5.1.1. Explicit Judgemental Resources: Lexico-grammatical Analysis…. 82 5.1.1.1. Overall Analysis of Explicit Lexico-grammatical Categories…………………………………………………... 82 5.1.1.1.1. Evaluative verb phrases…………………………. 83 5.1.1.1.2. Evaluative postpositional phrases……………… 85 5.1.1.1.3. Evaluative relative clauses………….…………... 87 5.1.1.1.4. Evaluative adverbial clauses.............................. 88 5.1.1.1.5. Evaluative noun phrases……...….……………... 90 5.1.1.1.6. Evaluative adjective phrases………………….… 91 5.1.1.1.7. Evaluative complement clauses………………… 92 5.1.1.1.8. Evaluative adverb phrases……………………… 93 5.1.1.2. Leader-Based Explicit Judgemental Analysis …………... 95 5.1.1.2.1. Evaluative verb phrases in leader-based use.… 96 5.1.1.2.2. Evaluative postpositional phrases in leader-based use………………………………… 98 5.1.1.2.3. Evaluative relative clauses in leader-based use…………………………………. 99 xiv     5.1.1.2.4. Evaluative adverbial clauses in leader-based use……………………………………………….. 100 5.1.1.2.5. Evaluative noun phrases in leader-based use…………………………………………………. 101 5.1.1.2.6. Evaluative adjective phrases in leader-based use…………………………………………………. 102 5.1.1.2.7. Evaluative complement clauses in leader-based use…………………………………………………. 103 5.1.1.2.8. Evaluative adverb phrases in leader-based use…………………………………………………. 103 5.1.1.3. Detailed Analysis of Lexico-grammatical Resources based on Leaders` Self and Other Evaluations………….. 104 5.1.1.3.1. Verb phrases in leaders` self-vs-other evaluations………………………………………… 105 5.1.1.3.2. Postpositional phrases in leaders` self-vs-other evaluations………………………………………… 106 5.1.1.3.3. Relative clauses in leaders` self-vs-other evaluations………………………………………... 108 5.1.1.3.4. Adverbial clauses in leaders` self-vs-other evaluations………………………………………… 110 5.1.1.3.5. Noun phrases in leaders` self-vs-other evaluations…….……………………………….…. 112 5.1.1.3.6. Adjective phrases in leaders` self-vs-other evaluations………………………………………… 113 5.1.1.3.7. Complement clauses in leaders` self-vs-other evaluations..………………………………………. 114 5.1.1.3.8. Adverb phrases in leaders` self-vs-other evaluations…….............................................….. 115 5.1.2. Implicit Judgemental Categories..................................................... 117 5.1.2.1. Overall Analysis of Implicit Judgemental Expressions….. 118 5.1.2.1.1. Evaluative metaphorical structures…………..... 119 5.1.2.1.2. Evaluative invoked Judgemental structures…… 120 5.1.2.2. Leader-Based Implicit Judgemental Analysis……………. 122 5.1.2.2.1. Evaluative metaphorical structures in leader-based use………………………………….. 122 xv     5.1.2.2.2. Evaluative invoked Judgement structures in leader-based use………………………………….. 124 5.1.2.3. Detailed Analysis of Implicit Judgemental Resources based on Leaders` Self and Other Evaluations…..……... 125 5.1.2.3.1. Metaphors in leaders` self-vs-other evaluations..125 5.1.2.3.2. Invoked Judgements in leaders` self-vs-other evaluations…………………………………………. 127 5.1.3. Comparison of Explicit and Implicit Judgemental Expressions....... 130 5.2. FINDINGS OF ENGAGEMENT ANALYSIS………………………………. 133 5.2.1. Analysis of Dialogic Contraction Resources………………………… 133 5.2.1.1. Overall Analysis of Dialogic Contraction Resources……. 134 5.2.1.1.1. Deny category …………………………………….. 135 5.2.1.1.2. Pronounce category ………………………. 136 5.2.1.1.3. Counter category …………………………….. 138 5.2.1.1.4. Concur category…………………………………... 139 5.2.1.1.5. Endorse category…………………………………. 141 5.2.1.2. Detailed Analysis of Dialogic Contraction Resources based on Leaders` Self and Other Evaluations……….… 143 5.2.1.2.1. Deny category in leaders` self-vs-other evaluations…………………………………………. 143 5.2.1.2.2. Pronounce category in leaders`self-vs-other evaluations…………………………………………. 145 5.2.1.2.3. Counter category in leaders`self-vs-other evaluations…………………………………………. 147 5.2.1.2.4. Concur category in leaders`self-vs-other evaluations…………………………………………. 149 5.2.1.2.5. Endorse category in leaders`self-vs-other evaluations…………………………………………. 150 5.2.2. Analysis of Dialogic Expansion Resources……………………….…. 153 5.2.2.1. Overall Analysis of Dialogic Expansion Resources…..…. 153 5.2.2.2. Closer Look at Dialogic Expansion Categories………..… 154 5.2.2.2.1. Epistemic modality……………………………..…. 155 5.2.2.2.2. Rhetorical questions………………………….…... 158 5.2.2.2.3. Attribution………………………………………….. 160 5.2.2.2.4. Deontic modality…………………………………... 161 xvi     5.2.2.2.5. Evidentials…………………………………………. 163 5.2.2.3. Detailed Analysis of Dialogic Expansion Resources based on Leaders` Self and Other Evaluations………………..... 164 5.2.2.3.1. Epistemic modality in leaders`self-vs other evaluations………………………………………… 165 5.2.2.3.2. Rhetorical questions in leaders`self-vs other evaluations………………………………………… 167 5.2.2.3.3. Attribution in leaders` self-vs-other evaluations………………………………………… 170 5.2.2.3.4. Deontic modality in leaders`self-vs-other evaluations………………………………………… 172 5.2.2.3.5. Evidentials in leaders`self-vs- other evaluations. 174 5.2.3. Comparison of Dialogic Contraction and Dialogic Expansion Resources…………………………………………………………….... 176 5.2.3.1. Leader-based Analysis of Dialogic Contraction and Dialogic Expansion Categories…………........................... 178 5.2.3.2. Detailed Analysis of Dialogic Contraction and Dialogic Expansion Categories based on Leaders` Self and Other Evaluations………………………………………………….. 179 5.2.3.2.1. Dialogic contraction resources in leaders` self-vs-other evaluations………………………..… 179 5.2.3.2.2. Dialogic expansion resources in leaders` self-vs-other evaluations……………….…………. 180 CHAPTER 6: CONCLUSION..................................................................... 183 6.1. SUMMARY OF THE MAJOR FINDINGS…………………….…………..…. 185 6.2. SUGGESTIONS FOR FURTHER RESEARCH……………………………. 194 REFERENCES……………………………………………………………….…. 197 APPENDIX I. ENGLISH TRANSLATIONS OF LEADERS` UTTERANCES……. 228 APPENDIX II. ORIGINALITY REPORT……………………………………………… 239 APPENDIX III. ETHICS BOARD WAIWER FORM………………………………….240 xvii     LIST OF ABBREVIATIONS AKP Adalet ve Kalkınma Partisi (Justice and Development Party) CHP Cumhuriyet Halk Partisi (Republican People`s Party) MHP Milliyetçi Hareket Partisi (Nationalist Movement Party) HDP Halkların Demokratik Partisi (People`s Democratic Party) SFL Systemic Functional Linguistics EAP English for Academic Purposes [+j] positive explicit Judgement [-j] negative explicit Judgement [i+j] positive invoked Judgement [i-j] negative invoked Judgement [metaphor+j] metaphor expressing positive Judgement [metaphor-j] metaphor expressing negative Judgement [e-deny] deny category of Engagement [e-pronounce] pronounce category of Engagement [e-counter] counter category of Engagement [e-concur] concur category of Engagement [e-endorse] endorse category of Engagement [e-epistemic] epistemic modality category of Engagement [e-RQ] rhetorical question category of Engagement [e-attribute] attribution category of Engagement [e-deontic] deontic modality category of Engagement [e-evidential] evidential modality category of Engagement xviii     LIST OF TABLES Table 1. Judgement – social esteem………………………………………….. 39 Table 2. Judgement-social sanction…………………………………………… 40 Table 3. Speeches in the data…………………………………………………. 61 Table 4. Total word count and analyzed word counts of leaders` other and self evaluations…………………………………………………… 70 Table 5. Categories showing epistemic modality…………………………….. 80 Table 6. Sample representation of Appraisal Framework for Turkish data... 81 xix     LIST OF FIGURES Figure 1. Metafunctions of language……………………………..…………….. 35 Figure 2. Main Categories of Appraisal Model…………………..…………..… 36 Figure 3. Strategies for Implicit Judgement…………………….……………… 40 Figure 4. An overview of Engagement………………………….……………… 43 Figure 5. Dialogic Contraction Categories…………………….………….……. 44 Figure 6. Dialogic Expansion categories……………………….……….……… 45 Figure 7. Palmer`s Modality Classification………………………………….…. 47 Figure 8. Graduation categories…………………………………….………….. 49 Figure 9. Percentage of Lexico-grammatical Categories of Explicit Judgemental Expressions in Turkish Politics…………….………… 83 Figure 10. Percentages of Lexico-grammatical categories of Explicit Judgemental expressions employed by party leaders……..……… 95 Figure 11. Percentages of Verbal and Nominal predicates in the Verb Phrase category……………………………………………..………… 97 Figure 12. Percentages of Verb Phrases employed in Other- versus Self- Evaluations……………………………………………………..……… 105 Figure 13. Percentages of Postpositional Phrases employed in Other- versus Self-Evaluations………………………………………..…….. 107 Figure 14. Percentages of Relative Clauses employed in Other- versus Self-Evaluations……………………………………………………….. 108 Figure 15. Percentages of Adverbial Clauses employed in Other- versus Self-Evaluations…………………………….…………………………. 110 Figure 16. Percentages of Noun Phrases employed in Other- versus Self-Evaluations………………………………….……………………. 112 Figure 17. Percentages of Adjective Phrases employed in Other- versus Self-Evaluations…………………………………………………..…... 113 Figure 18. Percentages of Complement Clauses employed in Other- versus Self-Evaluations……………………………………………..... 114 Figure 19. Percentages of Adverb Phrases employed in Other- versus Self- Evaluations………………………………..…………………..… 116 Figure 20. Percentages of Implicit Judgemental Expressions in Turkish Political discourse………………………………..……………..…….. 118 xx       Figure 21. Percentages of Judgemental Metaphor Categories in Leaders` Speeches………………………………………………...…. 123 Figure 22. Percentages of Invoked Judgement Categories in Leaders` Speeches………………………………………………...…. 124 Figure 23. Percentages of Judgemental Metaphor Categories in Leaders` self-versus-other evaluations………………………………….…..… 125 Figure 24. Percentages of Invoked Judgement Categories in Leaders` self-versus-other evaluations…………………………………..….… 127 Figure 25. Judgemental Expressions in Turkish Political Discourse……...….. 130 Figure 26. Explicit versus Implicit Judgemental Expressions employed by party leaders……………………………………………………….. 131 Figure 27. Explicit versus Implicit Judgemental Expressions in Self versus Other – Evaluations…………………………………………….…..…. 132 Figure 28. Dialogic Contraction resources employed in Turkish Political Discourse………………………………………………………………. 134 Figure 29. Percentages of ‘Deny’ category in self-versus-other evaluations of party leaders…………………………………………………...…… 143 Figure 30. Percentages of ‘Pronounce’ category in self-versus-other evaluations of party leaders…………………………………...…….. 145 Figure 31. Percentages of ‘Counter’ category in self-versus-other evaluations of party leaders………………………………………….. 147 Figure 32. Percentages of ‘Concur’ category in self-versus-other evaluations of party leaders…………………………………..….….. 149 Figure 33. Percentages of ‘Endorse’ category in self-versus-other evaluations of party leaders……………………………………….…. 150 Figure 34. Percentages of Dialogic Expansion resources employed in Party Group Speeches…………………………………………….…. 154 Figure 35. Detailed Analysis of Expansion categories in the whole data……. 155 Figure 36. Percentages of ‘Epistemic modality’ categories in self- versus-other evaluations of party leaders………………………..…. 165 Figure 37. Percentages of ‘Rhetorical question’ category in self- versus-other evaluations of party leaders……………………...…… 168 Figure 38. Percentages of ‘Attribution’ categories in self-versus-other evaluations of party leaders………………………………………..… 170 xxi     Figure 39. Percentages of ‘Deontic modality’ categories in self- versus-other evaluations of party leaders………………………...… 172 Figure 40. Percentages of ‘Evidentiality’ categories in self-versus-other evaluations of party leaders………………………………………..… 174 Figure 41. Percentages of Engagement Expressions in the sample ……… 177 Figure 42. Leaders` Use of Dialogic Contraction and Dialogic Expansion resources in their speeches…………………………………….…… 178 Figure 43. Leaders` use of dialogic contraction categories in their self- versus other-evaluations………………………………………….….. 179 Figure 44. Leaders` use of dialogic expansion categories in their self- versus other-evaluations………………………………………..……. 181 1     CHAPTER 1 INTRODUCTION This study derives from a longstanding interest on the evaluative dimension of language and its identification in discourse in that the analysis of evaluative property is a good way to explore not only the interpersonal function of a language, but also the ways in which social realities are constructed. Within the past two decades, a number of linguists from various fields have focused on the interpersonal evaluative phenomenon, from different perspectives or terminologies, including evaluation (Hunston, 1989, 2000, 2012; Hunston & Thompson, 2000), appraisal (Martin, 2000; Martin & Rose, 2003; Martin & White, 2005), modality (Halliday, 1994), local grammar patterns (Hunston & Sinclair, 2000), voice (White, 2003, 2004, 2006, 2008, 2009; White & Thomson, 2008), stance (Biber & Finegan, 1989; Biber et al., 1999; Conrad & Biber, 2000), parameter-based approach (Bednarek, 2009a, 2009b, 2010, 2014, 2015), and evidentiality (Chafe & Nichols, 1986). Generally speaking, this study aims to look at the ‘evaluation’ phenomenon in Turkish political language from a socio-cultural perspective, which is related to a wide range of concepts, including persuasion, Appraisal, construction of (political) reality, maintenance of solidarity and opposition, expression of individual and group ideologies, and intentional meaning. Although the evaluation and reality construction strategies of Turkish politicians are analyzed based on the politicians` linguistic choices, this study goes beyond the structural explorations as it tries to interpret these structural findings in relation to socio-cultural reflections. 1.1. CLEARING THE GROUNDS: FROM EVALUATION TO REALITY CONSTRUCTION People use language to express their ideas and thoughts; i.e., language reflects individuals` attitudes and beliefs. As social reality is constituted through social interactions among individuals in a particular context, there is a direct influence of society on people, and this influence can be observed within individuals` perceptions affected by the society around them. 2     Durkheim (1965) argued that man is composed of two main parts; one is the individual being within the body and limited by his own fact; and the other one is the social being within the reality of the society (pp. 15-16). In this sense, man cannot be thought in isolation from the community around. Therefore, reality construction should be considered as a process occurring in a particular society and through language. While studying the construction of reality, language has the key role as it is directly related to the society and culture. Berger & Luckmann (1966) argue that as the most important sign of human society, the analysis of language is essential for understanding any kind of reality (pp. 51-52). Moreover, language use not only represents the speaker`s perspective, but it also reflects the ideologies of other individuals in a collective manner; and linguistic styles can influence the persuasiveness of the conveyed messages in discourse (Arcimaviciene, 2014; Berger & Luckmann, 1966; Krauss & Chiu, 1998; Shi-xu, 2005; Tappan, 1997). The relationship between language and the construction of reality goes back to de Saussure`s (1916) ‘structuralism’, and to Whorf`s (1956) ‘language is thought and culture’ perspective. They are followed by Austin`s (1962) ‘performatives’ and Searle`s (1995) ‘speech acts’, claiming that people do certain things by saying. In other words, discourse is not only what we say, but also what we do by saying certain things. For Searle, by saying things, individuals perform social acts, and these performative utterances create new institutional / social facts at the end (pp. 54-55). He points out that reality is socially constructed since every utterance requires a “publicly accessible reality” (p. 190). As a constructivist, Vygotsky (1978, 1986) argued that there are certain steps genetically decided in the process of constructing reality. Firstly, a reality appears within a society. Then, it is internalized within the individual, which is called ‘mediated activity’ or ‘internalization’; i.e., “internal reconstruction of an external operation” (Vygotsky, 1978, p. 56). In this internalization phase, individual beliefs or thoughts are re-constructed as a result of cultural sharings. Finally, this social reality turns into an intra-personal reality. In other words, although it starts as a cultural, contextual and social concept, as people socially interact, reality turns into an individual, personal and cognitive notion at the end. 3     Vygotsky focused on the concept ‘dialectic’ (mono individual; intra-psychological) and tried to explain the development of mental functionings through an intra-psychological dialectical process. He ignored the notion of ‘addressivity’. Instead, he emphasized the importance of activity mediation, through which social realities are internalized and considered as individual realities. He saw language both as a personal and a social human process; and he believed that human thinking develops from the social to the individual (Beaty; 2015; Matusov, 2011; van Deer, 1996; White, 2011). As a constructionist, Bakhtin (1981, 1986), on the other hand, introduced the notions ‘dialogue’ (the existence and co-operation of various perspectives, multiple participating voices or realities; multivoicedness), and ‘voice’ (a way of speaking that reflects the speaker`s own perspective or reality). For him, “…the utterance is filled with dialogic overtones” (1986, p. 102), and meaning is a social production of a joint activity (Krauss & Fussell, 1996). Bakhtin viewed speech as a social concept, as all language is interanimated by others (White, 2011, p. 6), and he developed a dialogic, genre- based and pluralistic approach to study discourse and social reality. In other words, he emphasized the social side of an utterance. Regarding Bakhtin`s ‘voice’ concept, it has been argued that each voice is a representation of a particular ideology or attitude to reality (Fernhough, 1996, p. 49). Therefore, both the speaker and the addressee are important in the meaning-making process and the social construction of reality (Baxter, 2006, p. 106). Criticising Vygotsky and his socio-constructivist approach due to his reducing cultural differences among people to fundamental evolutionary properties, Wertsch (1991, 1994) argues that in order to formulate a more comprehensive sociocultural approach to meaning making and reality-construction processes, one should identify historically, culturally and institutionally situated forms of mediated action, which is directly related to Bakhtinian perspective. Looking at the construction of reality from a ‘sociological perspective’ like Bakhtin and Wertsch, Berger & Luckmann (1966) argue that reality is socially constructed (p. 13). They point out that an individual cannot be totally understood without taking the social context around him into consideration (p. 68). For them, people interacting in a certain social setting establish particular concepts or mental representations of each other`s 4     actions. These concepts turn into realities through time. In other words, people`s beliefs and thoughts are constructed within institutional settings in a society. In this process, language plays a significant role as it functions as a tool to express the shared meanings and makes them available to all participants within the linguistic community (p. 85). In the light of all these approaches, it can be argued that there is a direct relationship between evaluation and construction of political reality in political discourse. As language is symbolic and used as a mediating tool while constructing and expressing reality, the linguistic resources that politicians employ in their evaluations play a significant role in the exploration of strategies through which they construct reality, maintain solidarity, express their commitment to their claims and persuade their listeners. Politicians tend to persuade others to change their ideas. Besides, they try to control their behaviors, ideas or thoughts most of the time. As politics is directly related to power and persuasion, the analysis of reality construction in evaluative political discourse can provide certain insights regarding the relationship between construction of political reality and language, by asking questions like how political meanings are constructed, or how political situations are described (Burnier, 1994, p. 242). 1.2. INTRODUCING THE STUDY This study mainly focuses on the analysis of ‘evaluative language’ and reality construction strategies of politicians in Turkish political setting, more particularly in their party group speeches. In general, evaluative language is related to the analysis of attitudes, judgements or assessments of individuals within their speeches or texts. It has been argued in literature that it is impossible to make a wholly objective utterance as everything we say or write expresses a kind of attitude. Therefore, whenever we say something, we express our personal feelings, attitudes, values, judgements or assessments (Biber et al., 1999, p. 966; Quirk et al., 1985, p. 613). Hunston (1989, p. 2) states that evaluation is an important phenomenon for linguists, because it plays a significant role in the analysis of particular discourse properties, like persuasion or expression of a point of view. Regarding the role of evaluation in 5     discourse analysis, Hunston and Thompson (2000, pp. 6-8) provide three basic functions: (1) evaluation expresses the speaker`s or writer`s opinion, and while doing so, it reflects the belief system of that person together with his / her community; (2) evaluation establishes and maintains relations between the speaker / writer and the listener / reader; and (3) evaluation plays a significant role in the organization of discourse. They conclude that expression of the writer`s or speaker`s opinion is an important feature of language, and it should be examined in a detailed manner in order to realize the evaluative meanings within texts and speeches accurately. Similarly, Hunston (2011, p. 12) and Lemke (1998, p. 1) argue that evaluation is an intersubjective phenomenon as it serves to interact, align / disalign and establish relationships with others, and in this respect, one of the basic functions of language is to create interpersonal relationships between the addressees and the addressors. In other words, as language is a resource for taking a stance, it is necessary to have a better understanding of what evaluative language is and how attitudinal expressions enable individuals to make propositions, proposals, actions or things, and construct reality. The analysis of evaluative language should be accompanied with critical discourse analysis or socio-pragmatic approaches. At this point, ‘Appraisal model’ (Martin & White, 2005) tries to combine different approaches while providing a discourse- semantic framework. Based on Systemic Functional Linguistics (Halliday, 1994), Appraisal model provides a theoretical work on the language of evaluation by describing various ways of linguistic realization of interpersonal meaning (Martin, 2000; Martin & White, 2005; Martin & Rose, 2003). Although it was primarily employed in the field of psychology, through time, this model attracted attention in the field of linguistics. In general, Appraisal model is related to the interpersonal metafunctioning of language, as it tries to explore interpersonal functionality of language, by which speakers or writers construct their identities and position themselves (Martin, 2001; White, 2003). In this sense, it is an effective model not only to explore the interpersonality in language, but also to analyze the process of reality construction, as language users have particular linguistic preferences depending on their status, power, community background and ideological perspectives. As the main concern of this study is 6     evaluative language and reality construction in Turkish politics, Appraisal provides a comprehensive and systematic model for such an exploration. As Gales (2011a) argues, Appraisal provides a model to examine how speakers / writers express their attitudinal positioning based on their ideologies and main concerns, and use explicit and implicit resources to do that. Appraisal also helps to reveal the rhetorical devices used by speakers or writers in order to create their desired stances and to express commitment to their propositions (p. 266). Apart from the relationship between evaluation and reality construction, it is also necessary to stress the importance of evaluative functioning within political language. It is obvious that politics and evaluative language are highly interrelated phenomena, as political affairs are expressed via language. Chilton (2004, p. 4) states that language and politics are intimately linked at a fundamental level as political activity cannot exist without language, and the doing of politics is largely constituted through language. As politics covers a wide range of subject matters, including self-reflection, individuation, subjectivity and analytic methods, it is a rich discourse with regard to evaluative properties and social construction of reality. Concerned with the dynamics of power in society, and particularly the ways in which power in transferred, sociologist Bourdieu (1991) emphasized that ideologies are created and transferred among social actors who share certain semiotic meanings. It is clear that evaluation is not a bare linguistic phenomenon, but it is directly influenced by sociological and ideological patterns within particular political parties. Political speeches are generally considered as a part of ‘subjective’ and ‘ideology- driven’ language, which are accepted to contain a great number of attitudinal expressions. There are many linguists interested in political discourse analysis from a wide range of perspectives, including Critical Discourse Analysis (Chilton, 2004, 2008; Fairclough, 1995, 2001; van Dijk, 2002a, 2002b; Wodak & Reisigl, 2001); pragmatics (Adamkova, 2011; Bull, 2012; Fetzer, 2006; Fetzer & Bull, 2006; Fetzer & Lauerbach, 2007; Johansson, 2006; Matic, 2012), and functional linguistics (Feng & Liu, 2010; Savoy, 2010; Simon-Vanderbergen, White and Aijmer, 2007). 7     In addition to that, political speeches also provide a good basis for the study of evaluation in that politicians tend to assess many things in their speeches, within a wide range of issues from political matters to social and emotional conditions. By using evaluative resources, they tend to express their community-based realities and ideologies. In this respect, studying the evaluative dimension of political discourse is one of the core issues for discourse analysts (A`Beckett, 2009; Becker, 2009; Bednarek, 2015; Chingware, 2014; Coffin, 2003; Helander, 2014; Martin, 2004a; Yalçın & Uçar, 2014; Vukovic, 2014; Zhang, 2012). As politicians` language use is a reflection of intergroup reality within a party community, the examination of this evaluative language while constructing and expressing particular institutional realities would contribute a lot to the discourse studies. The analysis of appraisal categories within political discourse can provide certain insights regarding the functions and properties of evaluative language in politics. For the analysis in this dissertation, Appraisal framework (Martin & White, 2005) has been employed as it is highly systematic and provides a rather comprehensive analytical method to deal with interpersonal meaning through a wide range of evaluative resources. Besides, it focuses on how meanings are realized through interpersonal positioning by providing a discourse-semantic approach, and it is an effective framework to study reality construction in political discourse. 1.2.1. Statement of the Problem So far, it has been argued that analysis of evaluative language can provide certain insights regarding individuals` reality construction strategies, and the influence of socio- cultural backgrounds on their community-based ideas and thoughts. At this point, it can be stated that although there are many linguistic studies exploring evaluative language in various languages, less work on this phenomenon has been carried out in Turkish. To the knowledge of the researcher, only Yalçın & Uçar (2014) and Yalçın (2014) applied Appraisal model in their studies in Turkish political setting. Besides, although considerable investigation in political discourse has been carried out in Turkish, they have primarily focused on certain phenomena, such as discursive strategies and critical discourse analysis (Bayram, 2010; Büyükkantarcıoğlu & Yarar, 8     2006; Güngör, 2014; Küçükali, 2014; Tanıyıcı, 2003; Terkan, 2010; Tok, 2012), or politeness (Yetkin, 2006). More recent studies have focused on thematic concepts like the ones in this dissertation, including Gezi events (Dedeoğlu, 2015; Güven, 2014; Özel & Deniz, 2015; Sommer, 2014), democracy (Doğanay, 2007) and religion (Efe, 2013; Terkan, 2010). In order to broaden the linguistic analysis of Turkish political language in literature, through focusing on its functional side, and especially the evaluative property in this respect, such an evaluative language analysis is necessary. In a broader perspective, this study enables the exploration of interpersonal resources which contribute to the rhetorical potential in Turkish political discourse. 1.2.2. Purpose of the Study The aim of this dissertation is to examine evaluative properties in Turkish political language, particularly party group speeches, based on Appraisal model (Martin and White, 2005), so that Turkish politicians` reality construction strategies can be observed through their evaluative language. More specifically, this study focuses on the reflection of Judgemental positioning and Engagement strategies of Recep Tayyip Erdoğan (AKP), Kemal Kılıçdaroğlu (CHP), Devlet Bahçeli (MHP) and Selahattin Demirtaş (HDP), who were political party leaders on the dates speeches were given, so that their use of different evaluative linguistic items to report certain issues, convey their ideas, make criticisms concerning themselves and also other party leaders, and express their evaluative meanings together with in-group realities can be explored. Based on the Appraisal theory pointing out that speakers employ evaluative resources to express how they feel about things or people within a particular discourse, and negotiate their social relationships, the purpose of this study is threefold: 1) to provide a detailed account of the lexicogrammatical resources employed to make Judgemental evaluations, i.e., assessing people`s behaviours (from both self and other-evaluative perspectives) in Turkish party group speeches, 9     2) to draw conclusions with regard to overt and covert evaluative strategies of Turkish political party leaders in their group speeches, and to explore whether they have any distinctive tendencies or styles regarding the use of expicit and implicit devices, 3) to explore the use of ‘dialogic’ concept within party leaders` speeches by analysing their stance-taking strategies, through which they negotiate their ideas and share their political opinions with their listeners; and to examine how authoritatively and assertively the evaluations in Turkish party group speeches are presented to the listeners, focusing on the use of dialogic contraction versus dialogic expansion resources, expressing power and authority to some extent. In this respect, the analysis of Judgement strategies employed by Turkish party leaders in their group speeches can give certain insights regarding not only their evaluative preferences, and also their ideologies and belief systems. More than that, it is expected that through the analysis of Engagement resources employed by Turkish politicians, certain linguistic structures like epistemic modality, evidentiality, rhetorical questions, attributional items and deontic modality categories, and also their functions in self versus other evaluations of the party leaders can be examined. Analysis of attitudinal resources, particularly the Judgemental resources and dialogic negotiation strategies of Engagement within Turkish political setting can enable the exploration of the links among politicians` personal identities, their social actions and the culturally-oriented realities in their evaluations, as argued in related literature (Jaffe, 2009; Mueller, 1973). Each political party has its own inter-group reality, and party leaders tend to express these realities in their speeches, so that they can share them with their audience and convince them of their claims. In this sense, by studying evaluative language through Appraisal model, this study also aims to explore Turkish politicians` reality construction and maintenance strategies while they are addressing their own party members. Political realities can be about any issue, such as ‘education’, ‘economics’, or ‘religion’; and party leaders express their party-based ideologies while talking about these issues most of the time. The analysis of evaluative language can also indicate how party- based realities regarding current political issues are expressed. In general, how reality 10     is constructed in Turkish political discourse and where the evaluation is deployed in this construction are the other concerns of this study. 1.2.3. Research Questions As one of the basic purposes of political discourse is persuasion and as evaluation is directly related to this persuasive dimension, through which strategies it is achieved in the speeches of Turkish political party leaders carry an importance. In this respect, the research questions put forward in this thesis are as follows: 1. What kinds of language strategies (explicit-lexicogrammatical items & implicit resources) do Turkish political party leaders employ to construct and maintain their political realities while they are making Judgemental evaluations (self/other) in their weekly-held group speeches? 2. In this reality construction process, how do Turkish political party leaders make use of Engagement strategies in order to maintain solidarity and also opposition to persuade their listeners? 3. What insights do evaluation and reality construction strategies of Turkish politicians provide regarding the influence of their status and positionings in Turkish political life? 1.2.4. Significance of the Study This study constitutes a linguistic approach to the lexical and grammatical nature and properties of evaluative language in Turkish political discourse, specifically weekly-held group speeches of party leaders. In this sense, it is expected to make significant contributions to functional linguistic dimension of Turkish political discourse, as the aim is to look at the political discourse analysis from a different perspective by focusing on the evaluative dimension, including the analysis of interpersonal meaning, dialogic expressions and dynamic explanations in the construction of politicians` arguments; and how they maintain solidarity with their listeners. Mainly, it contributes to the analysis of Judgemental positioning and Engagement strategies (source of evaluation) of party leaders to communicate their opinions in a particular political agenda. As a study conducted within a functional Appraisal 11     approach, it is hypothesized to provide an important alternative to the interpretations of evaluative semantics and interpersonal meaning in Turkish political language. With the help of this study, it is expected that more comprehensive and theoretical understanding of evaluation in Turkish political party group speeches can be provided. In other words, the analysis of evaluative resources and the ways in which evaluative stance taken by the politicians can provide certain insights regarding how evaluative meanings are encoded in Turkish politics. More specifically, the findings can reveal the ways in which party leaders manage their self and other evaluations in their positive and negative Judgemental expressions while arguing for their own points of view. Besides, they may indicate the strategies in which politicians manage to maintain solidarity with their listeners and they orchestrate other voices and cooperate with them into their speeches, as the propositions are considered as dialogic phenomena (Bakhtin, 1981). In other words, this study may enable the exploration of speaker stance in Turkish political speeches. As evaluative language and reality construction are directly related to intentional meaning, this study is also significant in terms of exploring meaning-making processes in the genre of Turkish party group speeches. In this process, the analysis is carried out from the discourse-semantics and socio-cultural perspectives. As persuasion is in the minds of the speakers, politicians` Judgemental evaluations and solidarity- maintaining expressions can guide us to explore not only the constructed realities existent in party group ideologies, but also the ways in which these realities are expressed to the audience. In general, discourse is used to convey certain ideological thoughts and realities. Although language is composed of many lexical choices and utterances, it also represents socially conditioned communicative intentions at the same time. In this sense, by analysing how politicians employ language, one can explore their perceptions, ideologies, and community-based realities. As Garcia (2015) argues, “language is considered the main way to the construction of reality, on one side, and for the mediation of the reality socially constructed, on the other” (p. 33). Taking into consideration Bazerman`s (1990, p. 81) emphasis on the examination of a particular language for finding out how realities are socially constructed and how linguistic 12     categories are intertwined with social relations; this study can also indicate the means Turkish political party leaders employ for reconstructing knowledge socially. By understanding the linguistic and social resources, we can understand the social construction of realities expressed in their speeches. Moreover, studying discourses about ‘us’ and ‘them’ - expressed as ‘self-evaluation’ and ‘other-evaluation’ in this study - can provide certain insights regarding the discursive practices of power and status constructed in Turkish party group speeches. In other words, this study can give information regarding how practices of power are expressed in this genre. Finally, this study enables the analysis of in-group ideologies, conceptions, and shared attitudinal values within the framework of evaluation. Therefore, attitudinal analysis of Turkish political speeches not only provides information regarding the linguistic resources preferred by the politicians, but it also explains how the political speeches as a specific discourse genre are organized within Turkish context. Besides, it is anticipated to figure out certain value-based expressions employed explicitly and implicitly by the party leaders based on their ideological positioning, political concerns and values. 1.2.5. Limitations of the Study While exploring evaluative language in a particular discourse, apart from the verbal elements including lexico-grammatical categories and implicit expressions, there are also nonverbal resources, including gestures, body movements, facial expressions, pitch in the speaker`s voice, and eye contact that can be studied. In this dissertation, only verbal elements were analysed, and nonverbal evaluative categories were excluded. They may have influenced the overall evaluative properties of the party leaders if they had been included. However, these nonverbal categories were not in the scope of the current study. Besides, as the main concern was on the behavioural evaluations of the party leaders, only Judgement category of the Appraisal model was included. In this sense, Appreciation and Affect sub-categories of the Attitude element were excluded in the data analysis process. Similarly, the Graduation sub-component of the Appraisal model 13     was excluded, as well, since the aim was not to explore how Turkish party leaders scale their evaluations, but to explore their overall Judgemental evaluations. Finally, only party group speeches, which are known as planned talks, were analysed in this study. However, other speeches of the same leaders, such as the ones in interviews, TV programmes or spontaneous declarations might lead to other findings. Besides, findings in this study are restricted to the political situations at the time of data collection; i.e., as politics is a dynamic phenomenon and there may occur changes regarding politicians` evaluative attitudes. 1.2.6. Organization This dissertation is structured into six chapters: Following the Introduction Chapter, Chapter 2 gives information regarding major concepts. The aim is to provide some background before moving to Appraisal model. It is organised in two main sections. First of all, general explanations regarding ‘political discourse’ and more specifically ‘party group speeches’ are provided. This is followed by related political discourse studies in literature. Then, ‘language of evaluation’ phenomenon is introduced in detail, as evaluation is the starting point of the study. Chapter 3 is designed to introduce the theoretical framework, ‘Appraisal’. The chapter starts with a general overview to provide a general outlook towards this model. Then, some background information regarding the introduction of this model is provided together with Halliday`s ‘Systemic Functional Linguistics’, which is the main core of Appraisal model. After this general background, components of Appraisal framework are mentioned in detail, including Attitude, Engagement and Graduation categories. In a specific part, the relationship between Engagement and Modality is provided, and the chapter ends with Appraisal in related literature, and more specifically Appraisal studies in the field of political discourse. In Chapter 4, methodology of the thesis is provided. In this respect, data selection and the identification of certain discourse areas are provided. Then, each discourse area is mentioned in detail so that the political case in Turkish setting can be understood better, and it can give the general idea why these discourse areas are important within the data analysis process. This methodology chapter ends with the information with 14     regard to how data was analysed, including Judgemental and Engagement data analysis processes. In Chapter 5, the findings are provided with relevant discussion, in both quantitative and qualitative ways. The presentation of the findings starts with the Judgemental categories, including not only explicit and implicit resources, but also self and other- based ones. Each lexicogrammatical resource is explained in detail with its functions and features in Turkish political setting. The chapter continues with the detailed analysis of Engagement resources, together with their percentages in party leaders` evaluative expressions and examples from the data. This dissertation is completed with Chapter 6, in which conclusions are provided. In this final chapter, certain observations made in the previous chapters are handled together with a brief summary of the main findings, regarding the evaluative dimension in Turkish political language. This chapter also re-stresses the importance of the current study and gives some suggestions for further research. 15     CHAPTER 2 BASIC CONCEPTS AND RELATED LITERATURE This chapter provides information regarding basic concepts including political discourse, party group speeches as a specific genre and related studies, accompanied by studies in Turkish political discourse. Then, fundamental properties of evaluation are mentioned, together with key evaluative approaches and linguists. In this respect, this chapter aims to look at the relationship among politics, language, and evaluation phenomena; and this chapter functions as a bridge to the main model, Appraisal. 2.1. POLITICAL DISCOURSE Politics has been a subject field in linguistics for many years, and it is nearly as old as the notion of rhetoric. Chilton and Schaffner (2002) argue that one can understand how politicians influence their listeners and convey their opinions to the society by analysing their talks. They define politics as a struggle for power, between those who seek to assert and maintain their power and those who seek to resist it on the one hand; and as cooperation, as the practices and institutions a society has for resolving clashes of interest over money, power, liberty and the like on the other hand (p. 5). In this sense, the analysis of language in political discourse can provide certain insights regarding the institutional properties of social groups, and more specifically politicians with different ideological stances. According to Schaffner (1996), language plays an important role in the process of manifesting a political will. Therefore, it is quite natural that politicians tend to express or reflect their ideological stance in their speeches and acts. The speeches of the politicians play a crucial role in realizing political values, ideas and political acts. With his main interest in the relationship among politics, discourse analysis and rhetoric, van Dijk (2002a, p. 20) states that “discourse is political when it accomplishes a political act in a political institution, such as governing, legislation, electoral campaigning and so on”. He believes that political discourse is not a genre, but a class 16     of genres defined by a social domain, namely that of politics (van Dijk, 2002b). Certain genres such as parliamentary debates, propaganda leaflets, campaign speeches or slogans can be considered as some of the genres within political discourse. It is crucial to emphasize the constitutive properties of language while talking about political discourse and its ideological side. Kress (1989) argues that language is a social concept, and communities express their specific meanings through language in systematic ways. Thus, language is not only an individual phenomenon, but it also has institutional functions. Similarly, Fairclough & Wodak (1997) see language as social practice; i.e., it plays a crucial role in the construction of situations, identities and relationships among people. As discourse is socially consequential, use of language may provide certain insights regarding the distribution of power between social classes, women and men, or cultural majorities and minorities (p. 258). Wodak & Busch (2004) state that “Power does not derive from language, but language can be used to challenge power, to subvert it, to alter distributions of power in the short and the long term” (p. 109). At that point, Fairclough (1995) believes that “Language is a material form of ideology, and language is invested by ideology” (p. 73). He argues that language is not only representational, but it is also constitutive, in that it contributes to the construction of relations in the social world. For him, discoursal practices are ideologically constructed based on power relations, and these power relations are affected by discoursal practices in return (p. 82). Wodak (2001) similarly asserts that language plays a significant role in establishing and maintaining power relations, as it mediates ideology in various social situations. She also argues that it is not the language that is powerful on its own. Rather, it gains power by the use of powerful people (p. 10). For van Dijk (1995b), ideologies are abstract, mental systems that organize socially shared attitudes, and they affect the personal perceptions of members in a community (pp. 18-19). About the relationship between language and ideology, van Dijk (2006b) believes that talking about others includes both positive self-presentation and negative other-presentation. In line with all these arguments focusing on language, ideology and politics, it is obvious that political discourse has a manipulative side. According to Orwell (1969), 17     language may be used to manipulate certain thoughts, so politicians tend to employ linguistic means strategically to convey their thoughts to their audience, and these linguistic options form the central issues in the analysis of political discourse. Wilson (2001, p. 10) suggests that one of the basic objectives of political discourse analysis is to indicate the ways in which linguistic resources are manipulated in order that they can have a specific manipulative effect. Moreover, he argues that “the study of political discourse is a highly vast issue, which includes various subject matters, ranging from political ideologies, racism and values to face-saving strategies, politeness and speech acts” (p. 399). Politicians tend to convey their perspectives, thoughts and realities to their listeners. While doing this, they employ various persuasive linguistic devices from explicit lexical categories to the implied ones. Certain words or expressions, such as some technical words, loaded words or pronouns may be employed strategically. Similarly, some syntactic forms may be used depending on the ideological purposes; or political language may be manipulated at the phonological level as well. Besides, some pragmatic terms such as implicatures, metaphors or speech acts may be used for political purposes, as well. At this point, Chilton (2008, p. 239) argues that in order to study political language, especially cognitive and pragmatic theories of meaning should be taken into consideration. Linguistic research on political language can be divided into three main categories. The studies on the first category explore the relationship among political discourse, language and ideology representation, and they try to indicate the ties between discourse and power. The researchers in this group are usually interested in identifying the means by which politicians express crucial issues such as class, gender, racism, and so on; and they employ critical discourse analysis as their main framework most of the time. They tend to indicate that language is an instrument to gain and express power and it can be analyzed in ideology-driven studies (Al-Faki, 2014; Chilton, 2004, 2008; Fairclough, 1995, 2001; Lande, 2010; Matic, 2012; Post, 2009; Reyes, 2011b; Ricento, 2003; van Dijk, 1995a, 2001; Wenden, 2005; Wilson, 1990, 2001; Wodak and Reisigl, 2001). van Dijk (1995a) is interested in the phenomena of ideology and political discourse, and he examines how semantic structures of discourse, such as topic, focus, 18     propositional structure, local coherence, level of description, implications and macrostructures are monitored by underlying ideologies behind the utterances and the politicians. Similarly, Chilton (2004) explores the relationship betwen language and power through the analysis of implicatures as the components showing authority, legitimacy and consensus. He states that the use of implicatures enables political actors to convey more than they say, and in political discourse, this frequently happens (p. 37). The second group of studies on political language has a more ‘socio-pragmatic’ motivation (Adamkova, 2011; Ekström, 2001; Fetzer, 2006; Fetzer & Bull, 2006; Fetzer & Lauerbach, 2007; Jagtiani, 2012; Johansson, 2006), and they explore the communicative functions of expressions within this political discourse. Adamkova (2011) explores how the pragmatic perspective of language, defined mostly by maxims and principles, is interconnected with segmental planes of language. Similarly, Jagtiani (2012) and Ekström (2001) focus on political discourse from a Conversation Analytic Perspective, and they examine institutionalized interaction within a political community. The third group of studies interested in the field of political discourse takes a more ‘functional approach’ and concentrate on the linguistic means of persuasion. They particularly focus on linguistic choices employed by the politicians to demonstrate how speakers react towards face-threatening questions, deny accusations and strengthen their own arguments. Examining to what extent taken-for-grantedness is used as a strategy in political media language as a genre at the interpersonal level, Simon- Vanderbergen, White and Aijmer (2007) focus on the lexico-grammatical means that British, Flemish and Swedish politicans use to persuade the others in political debates. They show that various markers of presupposition are typically used in the three cultures, and therefore the genre of political media debate is to a large extent conventionalised at the interpersonal level and that the conventionalisation operates in similar ways in the three cultures. 2.1.1. Political Speech as a Specific Genre Political speech is considered as an important genre in political discourse. It is known as one of the most manipulative (Chilton, 2004; Orwell, 1969; Wilson, 2001) and figurative genres. According to Nur (2015, p. 52), political speech is a kind of text presented by concerned authorities. While speaking, politicians tend to perform more 19     speech acts than they talk, as they interact with language and employ it to express intended messages. Feng & Liu (2010, p. 825) assume that although public speeches are delivered orally, they are often well-prepared in writing and presented on formal occasions. Sarnackaite (2011) argues that making speeches is a significant part of a politician`s role in accouncing a policy and persuading people to accept it. In this process, rhetorical skills needed for persuasive public speaking have always been a vital factor of political speeches. Thus, politicians use particular persuasive devices that make their speeches compelling, clear and more effective (p. 21). While giving a talk, politicians not only express their personal political beliefs regarding personal and social affairs, but also speak as members of a political party and represent their party`s ideological positioning and orientations; i.e., intergroup realities (Chilton, 2004). Therefore, they act as a member of a particular community holding a particular point of view and trying to convey this view to others. The analysis of political speeches can provide certain insight regarding the underlying ideologies of parties, their values, targeted outcomes and thoughts. There are many studies focusing on the relationship between Discourse Analysis and Political Discourse with specific focus on political speeches. While some of these studies focus on the analysis of lexical categories encountered in political speeches (Feng & Liu, 2010; Orwenjo, 2009; Savoy, 2010; Wilson, 1990), some others explore this genre from a more pragmatic and critical perspective (Bull, 2012; Matic, 2012). Studies looking at the relationship between the functioning of lexical elements within political speeches and their ideological reflections have indicated that politicians employ certain lexical resources intentionally. For instance, the frequent use of ‘we’ pronoun enables the politicians to have intimate relationships with their listeners (Muqit, 2012). Additionally, analysing the pronoun uses in political speeches, Wilson (1990) concludes that in this genre, the most salient pronominal distinction can be seen between the pronouns ‘I’ versus ‘we’, or ‘us’ versus ‘them’. He similarly points out that pronouns are employed in political talks in order to show degrees of distance in general. 20     In a different study exploring the most frequent lexical items and word count in US Political speeches, Savoy (2010, p. 123) states that words and expressions used in politicians` discourses are not chosen randomly, but rather employed intentionally to reflect certain objectives of these politicians. Matic (2012) compares and contrasts discourse structures within ideological strategies used in the speeches delivered by two presidential candidates of two ideologically opposed political parties. She investigates semantic macrostructures (topics), local meanings and lexical styles obtained from speech acts, rhetorical devices, forms of indirectness and strategies especially aimed at positive self-presentation and negative other-presentation. She concludes that political speeches are social and representative in terms of ideology, but they are also personal to some extent. Some other studies have employed critical discourse analysis in order to explore the properties of political speeches as a specific genre in political discourse. Wang, for instance, (2010) analyzes the speech of Barack Obama from the pespectives of critical discourse analysis and systemic functional linguistics in order to explore the effects of power and ideology in the transitivity and modality of his speech. He concludes that his utterances are trying to arouse the American people`s confidence towards the president and his government, and while doing this, Obama makes his audience understand and accept his political perspective by means of modal verbs, tense and first person pronouns, as they all help him persuade the public to accept and support his policies. Similarly, Shayegh & Nabifar (2012) explore Obama`s speeches to demonstrate the ideology-based linguistic occurences and they examine how the phenomenon of power exists in Obama`s words on the base of Hallidayian Systemic Functional Grammar and the critical perspective of Fairclough. At the end, it is concluded that Obama, as the dominant character manipulating particular belief systems and values, mostly uses ‘I’ and ‘we’ pronouns, more religious statements, more persuasion and longer turns in his speeches. Studies exploring the pragma-semantic features of political speeches, on the other hand, emphasize the importance of communication skils, facework strategies and social context while conveying the intended message in political speeches. For example, a speech made by a politician after an election victory may differ considerably 21     from the one made in the case of a political scandal. Social context, therefore, should be taken into consideration in the process of political speech analysis (Bull, 2012). 2.1.2. Studies on Turkish Political Language In Turkish, there are many discourse-oriented studies focusing on a wide range of issues, from the lexical choices of Turkish politicians to their discursive strategies. For example, Küçükali (2014) focuses on the discursive strategies in Turkish politics, mainly in the AKP. He looks at Turkish politics from an interdisciplinary discourse- analytical perspective by combining the discourse-historical and critical approaches in order to reveal how AKP, as the ruling party, establishes and maintains its political hegemony on the discursive level. Similarly, Büyükkantarcıoğlu and Yarar (2006) analyse discursive structures in Turkish politics in order to demonstrate the propositional structures employed by the politicians for persuasion. They look at the propaganda speeches delivered before the 2002 election, and they reveal that all party leaders tend to employ similar types of discursive strategies; like positive self- representation and negative other-representation, referencing to future acts, blaming the opposing, and so on. In a different study, Tanıyıcı (2003) focuses on the discourses of political parties by analysing the parliamentary debates regarding the European Union, and he draws the conclusion that party leaders tend to emphasize democratization and human rights themes most. In a similar study, Bayram (2010) analyses the political speech of Recep Tayyip Erdoğan given at the World Economic Forum in Davos in 2009 from a critical perspective. He explores the strategy Erdoğan uses in a foreign policy to influence the voters in domestic politics. He concludes that Erdoğan`s attitude and linguistic behaviour are the reflections of a particular social group; i.e., party ideology. Güngör (2014) tries to define the features of political language in Turkish setting. He analyses the speeches of political party leaders delivered in-group meetings in terms of speech acts, usage of natural language, rhetorical aspects and themes. He draws the conclusion that patterns ‘I’ and ‘we’, certain persuasive methods and common agenda topics are used by the leaders. Besides, he states that political speeches are highly rich in vocabulary choice (p. 66). 22     There are also some other studies focusing on only one aspect of political discourse in Turkish language. For instance, looking at the impoliteness strategies in Turkish Parliament, Yetkin (2006) asks how the ruling party and the opposing party leaders express derogation from a pragmatic perspective. Through a pragmatic analysis, she draws a conclusion that politicians tend to employ various instances of positive impoliteness as well as off-record impoliteness strategies through threats to Public Face most of the time. In a different study, Terkan (2010) investigates the women issue in political language, by analysing the political discourse of the AKP and the CHP. He concludes that the AKP mostly depicts women as part of a family and important for society, whereas the CHP forms a connection between women and modernity, secularism and participation in public. As a different political discourse theme, Tok (2012) focuses on education policies of the parties by analysing the party programmes and election manifestos. In order to identify the educational policies of the parties, she investigates the education concept and deciphers some sub-components of it, including educational goals, language of education, formal versus informal education, education of religion, assessment in education, teaching staff, student support, education environment and education programs. Based on the analysis in terms of these sub-components, she reveals that parties` policies to address educational issues are quite general and abstract. She draws a conclusion that in the educational view of political parties, political ideologies they defend and the factors that determine their political identity come to forefront. The theme ‘Gezi Park protests’ has also been handled in different studies in Turkey. For instance, Özel and Deniz (2015) analyse how the national newspapers in Turkey presented these protests on their front pages. Doing a frequency analysis, they conclude that most national newspapers in Turkey present news related to these protests in their front pages, and related to the ideology of the newspapers, they tend to be the supporters or against the protests in their news. In a similar study, Dedeoğlu (2015) investigates how the Gezi protests have been represented in Turkish print press by using van Dijk`s method of critical discourse analysis. She draws the conclusion that although the protests have been represented in the print press, people`s reactions against the government have not been reflected sufficiently in the news. 23     Güven (2014) makes a semiotic analysis focusing on the language used by the protestors at Gezi Park events. He analyses the metaphors and metonymies in slogans and giraffities from a discourse-semiotic perspective, and concludes that there is a tendency from metaphoric to metonymic form, which shows that there has been a social transformation among people. Besides, he reveals that the slogans are quite ideological. The theme ‘democracy’ has also been studied in Turkish linguistics. Focusing on the democracy and conservatism discourses of the AKP, Doğanay (2007) analyses certain concepts emphasized by this party, such as deliberation, dialogue and participatory democracy. She states that in many speeches of the party, the term ‘conservative democracy’ is used in relation to the terms ‘tolerance’, ‘participation’, ‘civil society’, ‘deliberation’ and ‘multiculturalism’. She concludes that these terms are used as a pragmatic strategy. As a different discourse area, Tekin (2012) examines Turkish political discourse on ‘ethnic and religious minorities’ in the context of recent debates on a new Law on Foundations. She states that debates in the Turkish parliament provide a valuable opportunity to decipher the constructions of identity of the Self and the Other among these parties. She analyzes arguments, discursive strategies and rhetorical moves used in political debates about ethnicity and religion. Using the steps of critical discourse analysis, she demonstrates that minority rights reforms are presented via rhetorical language, metaphors, argumentative devices as topoi and fallacies. Similarly, Efe (2013) believes that the discourse on and around ‘turban’ (turban) and ‘başörtüsü’ (headscarf) is socially constructed. Exploring the use of these two terms in political discourse, he argues that preferring one word to the other can indicate the ideological position of the speaker / writer. In his study, Efe reveals that ‘turban’ is used remarkably more frequently in the newspapers Cumhuriyet and Hürriyet, whereas ‘başörtüsü’ takes precedence in the newspapers Zaman and Vakit. 2.2. LANGUAGE OF EVALUATION Language of evaluation has been studied from different perspectives with different terminologies, including ‘evaluation’ (Hunston, 1989, 2000, 2011; Hunston & Thompson, 2000), ‘stance’ (Conrad & Biber, 2000, 2009) ‘hedging’ (Myers, 1989; 24     Hyland, 1996, 2005), ‘modality’ (Halliday, 1994), ‘evidentiality’ (Chafe & Nichols, 1986), ‘politeness’ (Brown & Levinson, 1987), ‘parameters’ (Bednarek, 2006a, 2006c, 2006e, 2008b, 2010,) and ‘appraisal’ (Martin, 2000; Martin & Rose, 2003; Martin & White, 2005). However, all these concepts refer to the same phenomenon; i.e., ‘interpersonal meaning’. In general, evaluative language studies try to explore what kind of language resources are used to express evaluative meaning; how evaluative meaning is modelled in different genres, including narratives, academic writings, conversations, corpus studies and so on; and the functions of evaluation within particular discourses. Taboada & Carretero (2012) argue that “the study of evaluative language has intrinsic motivation as all individuals use language to evaluate, appraise and classify objects and other people on an everyday basis” (p. 276). Downes (2000) points out that “evaluations are semiotic, because the meaning, positive or negative, is mapped onto the substance of an experience in contexts of situation” (p. 104). Similarly, Bednarek (2006a) emhasizes that evaluative meanings should be analysed within context, as propositions perform certain discourse functions, such as expressing a criticism, involvement or mitigation based on the context in which they are uttered (p. 211). According to Hyland (2005), evaluations depend on certain standards. Thus, Judgemental evaluations can be persuasive or meaningful if they are made according to communal ideologies or belief systems within a society. In this sense, they may hold interesting, relevant, novel, useful, good or bad meanings dependent on the community (p. 175). Although there are different points of view towards evaluation, Hunston (2011) provides the commonly accepted features regarding evaluative language: (a) Evaluation is both subjective and intersubjective. Evaluative utterances express a personal opinion, and therefore evaluation is something personal, private, and subjective, but, at the same time, evaluation has the function of interacting with a social other; and (b) evaluation construes an ideology that is shared by the writer and the reader (or the speaker and the hearer), as it takes place within a social and ideological framework (p. 12). 25     In literature, there are various terminologies introduced for evaluation and evaluative language. In this part, some key terminologies are provided in order to demonstrate the similarities among them, and also to emphasize their relations with Appraisal model. 2.2.1. Evaluation Within different terminologies regarding evaluative language, Hunston (1989, 2000, 2011) and Hunston & Thompson (2000) use the term ‘evaluation’. According to Hunston & Thompson (2000), “Evaluation is the broad cover term for the expression of the speaker or writer`s attitude or stance towards, viewpoints on, or feelings about the entities or propositions that he or she is talking about” (p. 5). According to Hunston & Thompson (2000), there are certain functions that evaluation performs; including (a) expressing the speaker`s or writer`s opinion; (b) constructing and maintaining relations between the speaker or writer and hearer or reader; and (c) organizing the discourse. They believe that evaluation can be used to persuade the reader / listener to accept things the writer / speaker wants. This is usually achieved by manipulating linguistic resources to assess people or things negatively or positively, and that`s why, evaluation has a function of organising discourse based on this persuasion objective (pp. 6-8). Depending on these functions, it can be argued that an examination of political texts or speeches can reveal ideologies and underlying values of politicians and parties. They also state that some lexical items such as adjectives (like splendid, terrible, surprising, obvious, important), adverbs (like happily, unfortunately, interestingly, necessarily), nouns (like success, failure, tragedy, likelihood, triumph), and verbs (like succeed, fail, win, lose, doubt) can have evaluative functions within contexts they are deployed. Besides, they argue that evaluation in texts / speeches can be identified by exploring the comparators, markers of subjectivity (modals), and markers of value (evaluative lexis, adjectives, adverbs, nouns, verbs, indicators of the existence of goal achievement) (pp. 15-21). 2.2.2. Stance Similar to the concept of Evaluation, and even functioning as its synonym, ‘stance’ is considered as a speaker`s or writer`s internal thoughts, opinions or attitudes about a 26     person being assessed or a topic being conveyed through lexico-grammatical choices that s/he makes. It is generally considered that stance is related to valuing and taking a position towards entities or propositions, and some researchers take a grammar-based analysis perspective as their starting point in the identification of this value taking process (Biber & Finegan, 1989; Conrad & Biber, 2000; Halliday, 1994; Hunston, 2011; Hunston & Sinclair, 2000; Labov, 1981). According to du Bois (2007, p. 139), “one of the most important things we do with words is take a stance”. He argues that as an evaluative property, stance enables individuals to give value to people and entities. Similar to Bakhtin`s dialogism perspective (1981, 1986), du Bois argues that stance is achieved dialogically as each stance is in relation with previous, already existing stances within a community during interpersonal activities (p. 172). Biber and Finegan (1989) argue that stance is related to “the lexical and grammatical expression of attitudes, feelings, judgements or commitments concerning the propositional content of a message” (p. 93). In expressing stance, Biber et al (1999) argue that stance can be expressed lexically and grammatically (p. 968). They point out certain major categories expressing stance, such as stance adverbials (e.g., unfortunately), stance complement clauses (e.g., It is amazing that…), modals and semi-modals (e.g., I might); stance noun + prepositional phrase (e.g., They deny the possibility of); and pre-modifying stance adverbs (e.g., I am so happy for you) (pp. 369- 370). Similarly, Conrad & Biber (2000) argue that the use of adverbials as a grammatical category can give certain insight with regard to how feelings and evaluations are conveyed. They believe that stance can be encountered in three major domains; ‘epistemic stance’, ‘attitudinal stance’, and ‘style stance’. Exploring the ‘lexical bundles’ which are multi-word sequences, Biber, Conrad & Cartos (2004) provide two major kinds of meaning: ‘epistemic’ and ‘attitude / modality’. Epistemic stance bundles comment on the knowledge status of the information in the following propositions: certain, uncertain, or probable, possible (e.g. I don`t know if, I don`t think so…); whereas attitudinal modality stance bundles express speaker attitudes towards the actions or events described in the following proposition (e.g., I 27     want you to, I am not going to…). Attitudinal / modality stance bundles include ‘desire bundles’, ‘obligation/directive bundles’ and ‘intention/prediction bundles’ (pp. 389-391). 2.2.3. Local Grammar Approach Apart from the analysis of modality and adverbial markers to express evaluation, Hunston (2011) and Hunston & Sinclair (2000) handle grammar and evaluation relationship from a different perspective and introduce the ‘local grammar pattern’ approach for the analysis of evaluative language. In this process, Hunston (2011) explores the relationship between recurring patterns and evaluative meaning. She discusses grammar patterns and local grammars of evaluation, and asks whether grammar patterns can be used as a diagnostic to distinguish between types of evaluative meaning or to parse evaluative statements in running texts. Exploring patterns such as “V+n+into+–ing”; “v+way+prep / adv”, and so forth, she concludes that patterns make some contribution towards distinguishing evaluative meaning. She argues that grammatical patterns have particular functional roles, and they play a significant role in the study of evaluative language (p. 121). In their analysis, Hunston and Sinclair (2001) try to provide evaluative patterns, which cluster around evaluative adjectives and nouns, and they try to set them up in large comprehensive grammars. They try to identify and parse evaluation by considering some patterns which are typically used to evaluate, such as: 1. It + linking verb + Adjective group + clause (e.g., It seemed important to trust her judgement. 2. there + linking verb + something / anything / nothing + adjective gorup + about / in + noun roup / -ing clause (e.g., There is something rather appealing about being bale to spend) 3. linking verb + adjective group + to – infinite clause (e.g., This book is interesting to read) 4. linking verb + adjective group + that - clause (e.g., Doctors are optimistic that he would make a full recovery) 5. Pseudo – Clefts (e.g., What is interesting is the tone of the statement) 6. Patterns with general nouns (e.g., The surprising thing about chess is that computers can play it so well) They conclude that based on Local Grammar Approach, it is possible to identify some patterns whose primary purpose is to evaluate, or to attribute evaluation to another speaker, and which therefore tend to select evaluative adjectives. These patterns may be used as a diagnostic for evaluative positioning. 28     Bednarek (2009b) similarly investigates how far linguistic patterns express evaluative meaning and whether there are any specific pattern used to evaluate things, persons and emotions, through the analysis of utterances in the 100 million Word British National Corpus. She aims at finding instances where the patterns occur with sub- categories, as in ‘I feel adj about / that’, ‘It was adjective of X to’, ‘I consider it adj (that)’ or ‘I find it adj (that)’. She suggests that a researcher should classify attitudinal lexis first, and then s/he should look at the pattern in which it is used and what the effect of this usage is (p. 26). Similar to Hunston, and Hunston & Sinclair, Bednarek also agrees that patterns have an important meaning-making role within evaluative language. 2.2.4. Parameter-based Approach Bednarek (2009a, p. 148) defines evaluation as “the linguistic expression of speaker / writer opinion”. She argues that when speakers / writers evaluate something or someone, they make negative or positive assessments (Bednarek, 2006a). In general, she tries to establish her own framework of evaluative parameters to study evaluative meaning, and she provides her parameter-based approach. She argues that “Evaluative parameters refer to the standards, norms and values according to which we evaluate something through language” (Bednarek, 2010, p. 18). Parameter-based theory of evaluation is based on the assumption that there are different parameters along which speakers can evaluate aspects of the world. What speakers are talking about can be evaluated in relation to a wide range of norms: do we feel that we are talking about is ‘good news’ or ‘bad news’; do we evaluate the information we have as reliable or unreliable; is what we are talking about presented as expected or unexpected; obvious or surprising, important or unimportant, and appropriate or inappropriate. This framework is composed of ten parameters; ‘comprehensibility’, ‘emotivity’, ‘importance’, ‘seriousness’, ‘expectedness’, ‘mental state’, ‘evidentialty’, ‘possibility / necessity’, ‘reliability’, and ‘style’ (Bednarek, 2006e, pp. 1888-189). Speakers can evaluate aspects of the world using these parameters. Besides, Bednarek argues that there are ‘core evaluative parameters’, which are concerned with the evaluative qualities ascribed to the entities, situations or propositions that are evaluated within two scales; and ‘peripheral evaluative parameters’, which do not involve any evaluative scales, and do not indicate the same 29     kind of qualitative evaluation of entities, situations or propositions as do core evaluative parameters (2008b, pp. 10-12). Other parameter-based approaches include Francis (1995, p. 9), providing eight main parameters of evaluations, as modality, ability, importance, predictability, obviousness, value and appropriacy, rationality and truth. Moreover, Thompson & Hunston (2000, p. 22) use the term ‘parameter’ and describe four of them as: good-bad / positive-negative parameter; certainty parameter; expectedness / obviousness parameter; and relevance / importance parameter. Similarly, Lemke (1998) describes seven dimensions of attitudinal meaning or evaluative semantic relations as; desirability / inclination; warrantability / probability; normativity / appropriateness; usuality / expectability; importance / significance; comprehensibility / obviousness; and humorousness / seriousness. Apart from these different terminologies and classifications, there are many other linguists exploring evaluative language from their own perspectives. Labov (1981), for instance, emphasizes the comparative nature of evaluation. For him, evaluation may consist of anything that can be compared to or contrasted with certain norms or values, and this comparison / contrasting process provides or shapes evaluation. Labov provides four evaluative categories, namely “intensifiers, comparators, correlatives and explicatives”. Exploring the relationship between stance and Engagement, Hyland (2005) proposes a model for the analysis of stance (hedges, boosters, self-mention, attitude markers) and engagement (directives, questions, shared knowledge, reader reference). For him, stance and engagement are important elements within writers` arguments and they suggest how writers anticipate and understand their readers` background knowledge,