Hacettepe University Graduate School of Social Sciences Department of English Linguistics MOTION PREDICATES IN TURKISH: A MORPHO-SYNTACTIC TREATMENT Abdullah TOPRAKSOY Ph.D. Dissertation Ankara, 2022 MOTION PREDICATES IN TURKISH: A MORPHO-SYNTACTIC TREATMENT Abdullah TOPRAKSOY Hacettepe University Graduate School of Social Sciences Department of English Linguistics Ph. D. Dissertation Ankara, 2022 ACCEPTANCE AND APPROVAL The jury finds that Abdullah Topraksoy has on the date of May 25th 2022 successfully passed the defense examination and approves his/her Doctoral Thesis/Ph. D. Dissertation titled “Motion Predicates in Turkish: A Morpho-Syntactic Treatment”. Prof. Dr. Deniz ZEYREK BOZŞAHİN (Jury President) Assoc. Prof. Emine YARAR (Main Adviser) Prof. Dr. Işıl ÖZYILDIRIM (Member) Assoc. Prof. Murat ELMALI (Member) Asst. Prof. Zeynep DOYURAN (Member) I agree that the signatures above belong to the faculty members listed. Prof.Dr. Uğur ÖMÜRGÖNÜLŞEN Graduate School Director YAYIMLAMA VE FİKRİ MÜLKİYET HAKLARI BEYANI Enstitü tarafından onaylanan lisansüstü tezimin/raporumun tamamını veya herhangi bir kısmını, basılı (kağıt) ve elektronik formatta arşivleme ve aşağıda verilen koşullarla kullanıma açma iznini Hacettepe Üniversitesine verdiğimi bildiririm. Bu izinle Üniversiteye verilen kullanım hakları dışındaki tüm fikri mülkiyet haklarım bende kalacak, tezimin tamamının ya da bir bölümünün gelecekteki çalışmalarda (makale, kitap, lisans ve patent vb.) kullanım hakları bana ait olacaktır. Tezin kendi orijinal çalışmam olduğunu, başkalarının haklarını ihlal etmediğimi ve tezimin tek yetkili sahibi olduğumu beyan ve taahhüt ederim. Tezimde yer alan telif hakkı bulunan ve sahiplerinden yazılı izin alınarak kullanılması zorunlu metinlerin yazılı izin alınarak kullandığımı ve istenildiğinde suretlerini Üniversiteye teslim etmeyi taahhüt ederim. Yükseköğretim Kurulu tarafından yayınlanan “Lisansüstü Tezlerin Elektronik Ortamda Toplanması, Düzenlenmesi ve Erişime Açılmasına İlişkin Yönerge” kapsamında tezim aşağıda belirtilen koşullar haricince YÖK Ulusal Tez Merkezi / H.Ü. Kütüphaneleri Açık Erişim Sisteminde erişime açılır. o Enstitü / Fakülte yönetim kurulu kararı ile tezimin erişime açılması mezuniyet tarihimden itibaren 2 yıl ertelenmiştir. (1) o Enstitü / Fakülte yönetim kurulunun gerekçeli kararı ile tezimin erişime açılması mezuniyet tarihimden itibaren ... ay ertelenmiştir. (2) o Tezimle ilgili gizlilik kararı verilmiştir. (3) ……/……/…… Abdullah TOPRAKSOY i 1“Lisansüstü Tezlerin Elektronik Ortamda Toplanması, Düzenlenmesi ve Erişime Açılmasına İlişkin Yönerge” (5) Madde 6. 1. Lisansüstü tezle ilgili patent başvurusu yapılması veya patent alma sürecinin devam etmesi durumunda, tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulu iki yıl süre ile tezin erişime açılmasının ertelenmesine karar verebilir. (6) Madde 6. 2. Yeni teknik, materyal ve metotların kullanıldığı, henüz makaleye dönüşmemiş veya patent gibi yöntemlerle korunmamış ve internetten paylaşılması durumunda 3. şahıslara veya kurumlara haksız kazanç imkanı oluşturabilecek bilgi ve bulguları içeren tezler hakkında tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulunun gerekçeli kararı ile altı ayı aşmamak üzere tezin erişime açılması engellenebilir. (7) Madde 7. 1. Ulusal çıkarları veya güvenliği ilgilendiren, emniyet, istihbarat, savunma ve güvenlik, sağlık vb. konulara ilişkin lisansüstü tezlerle ilgili gizlilik kararı, tezin yapıldığı kurum tarafından verilir *. Kurum ve kuruluşlarla yapılan işbirliği protokolü çerçevesinde hazırlanan lisansüstü tezlere ilişkin gizlilik kararı ise, ilgili kurum ve kuruluşun önerisi ile enstitü veya fakültenin uygun görüşü üzerine üniversite yönetim kurulu tarafından verilir. Gizlilik kararı verilen tezler Yükseköğretim Kuruluna bildirilir. Madde 7.2. Gizlilik kararı verilen tezler gizlilik süresince enstitü veya fakülte tarafından gizlilik kuralları çerçevesinde muhafaza edilir, gizlilik kararının kaldırılması halinde Tez Otomasyon Sistemine yüklenir * Tez danışmanının önerisi ve enstitü anabilim dalının uygun görüşü üzerine enstitü veya fakülte yönetim kurulu tarafından karar verilir. iii ETİK BEYAN Bu çalışmadaki bütün bilgi ve belgeleri akademik kurallar çerçevesinde elde ettiğimi, görsel, işitsel ve yazılı tüm bilgi ve sonuçları bilimsel ahlak kurallarına uygun olarak sunduğumu, kullandığım verilerde herhangi bir tahrifat yapmadığımı, yararlandığım kaynaklara bilimsel normlara uygun olarak atıfta bulunduğumu, tezimin kaynak gösterilen durumlar dışında özgün olduğunu, Doç. Dr. Emine YARAR danışmanlığında tarafımdan üretildiğini ve Hacettepe Üniversitesi Sosyal Bilimler Enstitüsü Tez Yazım Yönergesine göre yazıldığını beyan ederim. Arş. Gör. Abdullah TOPRAKSOY iv ACKNOWLEDGEMENTS No dissertation is a single man’s work. The current study is a product of many helpful colleagues, friends, and an excellent supervisor. I want to express my gratitude to my supervisor Assoc. Prof. Emine YARAR for her valuable guidance and non-stop support with her intellectual personality. I owe special gratitude to thesis committee members Prof. Dr. Deniz ZEYREK BOZŞAHİN and Prof. Dr. Işıl ÖZYILDIRIM who always provided valuable suggestions and enabled me to see the light on the right track until the end of this journey. I also want to thank jury members Assoc. Prof. Murat ELMALI and Asst. Prof. Zeynep DOYURAN for their insightful comments and suggestions during the thesis defense exam. Special thanks to dear Prof. Jaklin KORNFILT who helped me find a track on this dissertation via sharing her valuable time with me during our meetings and cutting my path to meeting dear Prof. Elena BENEDICTO who encouraged me to think about the topic of Motion as my dissertation theme at the very beginning of this journey. I am also grateful to Prof. Len Talmy for his kind interest in reviewing some parts in my research and for setting aside time for his comments on some of the experiment sentences in the fictive part of this study. I owe appreciation to my dear friends and colleagues Zeynep Seza YILMAZ and Yasin TAŞDEMİR for their shares in proofreading some experiments and for their never- ending support and belief in me. This work, as in my master’s thesis, also has the spirit of my parents whom I cannot thank enough for making me as I am. This study has been carried out thanks to TÜBITAK’s (The Scientific and Technological Research Council of Turkey) scholarship program for Ph.D. students (2211-National Scholarship Programme for Ph.D. Students). v ABSTRACT TOPRAKSOY, Abdullah. Motion Predicates in Turkish: A Morpho-Syntactic Treatment, Ph.D. Dissertation, Ankara, 2022. Research on space and language has had fruitful outcomes in the last decades. One of the related domains of study is motion itself, which is central to our experience. Although the work on the semantic analysis of motion events have had fruitful outcome in typological attempts to motion event encoding, a more recent approach to the field support the idea that motion independent properties which govern the morphological, lexical and syntactic resources available to languages may determine the selection or tendency in motion framing of languages. The present study sets off to question this recent approach and focuses on actual motion events in Turkish from a structural point of view and investigates motion expressions in relation to subordination and case marking. It aims to understand what kind of case markings and subordinate expressions are used to encode motion events and to describe the relations, if any, between these structural elements and motion expressions. Apart from the structural investigation of motion expressions, the present study also addresses a preliminary analysis of fictive motion in Turkish, which is a totally different travel from the structural analysis. Two tasks were employed for the analysis of actual motion expressions and one task for the description of fictive motion expressions. Due to the nature of the content, no verbal expressions on it, and easy operability, The Pear film was taken as the first task of the study. A follow-up narrational experiment via a set of animated video clips was organized as the second task of the study. In the final section, a drawing task was administered to participants in order to account for the extent to which fictive motion is observable from the drawings. All the tasks are based on language production in their nature. The tasks were carried out online with individual participants (n=60), who are all native Turkish speakers. The findings of the study taken from both tasks found out that participants made use of certain subordinate constructions to elaborate their narrations of motion expressions. Three subordinate types were described from the frequently used ones to the least. Their relations with motion expressions were explained as encoding mainly the manner of motion; modifying the figure and/or ground elements of motion expressions. In regard to the use of case markings, three types of cases were observed in participants’ descriptions. The functions of these cases were linked to the translocational dynamics of motion expressions. The findings are in line with similar studies of its kind (e.g. Jackendoff 1990, 1996; Croft et al., 2010; Ibarretxe Antuñano, 2009 and Beavers et al., 2010) which suggest a flexible classification or continuum of motion typology vi since languages may exhibit more varied motion constructions than they are expected or proposed to in just two- or three-way typology. In terms of the fictive motion analysis, judging from the differences shown in drawings of fictive and non-fictive pairs, the present study suggests that there may be traces of fictive motion as if there was some form of motion effect, but further analyses are needed to make sure about that. Overall, apart from being the first investigation of fictive motion in Turkish, the present study can be regarded to contribute to the studies within the domain of motion in general and in Turkish in two ways: First, the present study tested the use of framework (by Beavers et al., 2010) which highlights the place of linguistic resources in encoding motion events in a language and as the findings suggest, that framework can be really beneficial in using linguistic resources for the analysis of motion events. Second, using tools rich of motion for the analysis of motion events, the present study can shed light on new insights which emphasize the clausal patterns in description of motion events in Turkish where path and manner verbs are used and even supported via additional uses of subordinate clauses for extended motion events and descriptions via case markings. Keywords Motion events, framing typology, grammar and space, case marking, subordination, Turkish. vii ÖZET TOPRAKSOY, Abdullah. Türkçe’deki Devinim Yüklemleri: Biçim-sözdizimsel bir Yaklaşım, Doktora Tezi, Ankara, 2022. Uzay ve dil üzerine yapılan araştırmaların birkaç on yılda verimli sonuçları olmuştur. İnsanlık olarak deneyimimizin merkezinde yer alan devinim, konuyla ilgili çalışma alanlarından birini oluşturmaktadır. Devinim olaylarının semantik analizi üzerine yapılan çalışmalar, devinimin kodlamasına yönelik tipolojik girişimlerde verimli bir sonuca sahip olsa da, alana yönelik daha yeni bir yaklaşım, diller için mevcut olan morfolojik, sözlüksel ve sözdizimsel kaynakları yöneten, devinimden bağımsız özelliklerin. dillerin devinim çerçevelemedeki seçimini veya eğilimini belirleyebiliceği fikrini desteklemektedir. Bu çalışma, bu yeni yaklaşımı sorgulamak için yola çıkmakta ve yapısal bir bakış açısıyla Türkçedeki somut (reel) devinim olaylarına odaklanmakta ve devinim ifadelerini yantümceleme ve durum belirleme bağlamlarında incelemektedir. Çalışma, devinim olaylarını kodlamak için ne tür durum belirleme ve yantümce ifadelerinin kullanıldığını anlamayı ve varsa bu yapısal elemanlar ile devinim ifadeleri arasındaki ilişkileri tanımlamayı amaçlamaktadır. Bu çalışma, devinim ifadelerinin yapısal incelemesinin yanı sıra, yapısal analizden tamamen farklı bir yolculuk olan Türkçedeki kurgusal devinimin bir ön analizini de ele almaktadır. Çalışmada, gerçek devinim ifadelerinin analizi için iki aşamalı deney ve kurgusal devinim ifadelerinin betimlenmesi için bir deney uygulanmıştır. Herhangi bir sözlü ifade içermemesi ve kolay işlenebilirliği nedeniyle Pear film, çalışmanın ilk deneyi olarak belirlenmiştir. Çalışmadaki ikinci deney olarak, bir dizi animasyonlu video klip aracılığıyla yürütülen bir izleme-öyküleme deneyi düzenlenmiştir. Kurgusal devinimin çizim testlerinden ne ölçüde gözlemlenebildiğini belirlemek için katılımcılara bir çizim testi uygulanmıştır. Çalışmada yer alan her üç deney de, doğası gereği dil üretimine dayanmaktadır. Deneyler, tamamı anadili Türkçe olan katılımcılarla (n=60) çevrimiçi olarak gerçekleştirilmiştir. Her iki deneyden elde edilen bulgular, katılımcıların devinim ifadeleri anlatımlarını detaylandırmak için belirli yantümce yapılarından yararlandıklarını ortaya çıkarmıştır. Kullanım sıklıklarına göre üç yantümce türü tanımlanmıştır. Bu üç türün devinim ifadeleri ile ilişkileri, esas olarak devinim tarzını kodlamak, ve devinim ifadelerinin şekil ve/veya zemin öğelerini değiştirmek olarak açıklanmıştır. Durum belirmeleme kullanımına ilişkin olarak, katılımcıların anlatı betimlemelerinde üç durum tipi gözlemlenmiştir. Bu durumların işlevleri, devinim ifadelerinin yer değiştirme dinamikleriyle bağlantılıdır. Diller, yalnızca ikili ya da üçlü tipolojiyle beklenenden veya önerilenden daha çeşitli devinim yapıları sergileyebileceğinden; çalışmadaki viii bulgular, devinim tipolojisinin esnek bir sınıflandırmasını ya da sürekliliğini öneren kendi türündeki benzer çalışmalarla (örneğin Jackendoff 1990, 1996; Croft vd., 2010; Ibarretxe Antuñano, 2009 ve Beavers vd., 2010) uyumludur. Kurgusal devinim analizi açısından, kurgusal ve kurgusal olmayan çiftlerin çizimlerinde gözlemlenen farklılıklardan yola çıkarak, bu çalışma, katılımcıların çizim örneklerinde kurgusal devinim izleri olabileceğini, ancak bunu daha güçlü savunabilmek için daha fazla analiz yapılması gerektiğini öne sürmektedir. Genel olarak, Türkçedeki ilk kurgusal devinim araştırması olmasının yanı sıra, bu çalışmanın Türkçeye ve genel olarak devinim alanındaki çalışmalara iki şekilde katkıda bulunduğu söylenebilir: Birincisi, bu çalışma dillerdeki devinim olaylarını kodlamada dilsel kaynakların önemini vurgulayan çerçevenin (Beavers vd., 2010 tarafından) kullanımı test etmiştir. Bulgulardan hareketle, bu çerçeve devinim olaylarının analizi için dilsel kaynakların kullanılmasında gerçekten faydalı olabilir. İkincisi, bu çalışma, devinim olaylarının analizi için devinim yönünden zengin içerikli deneyler uygulayarak, Türkçede devinim yapılarının betimlenmesinde yol ve tarz eylemlerinin kullanıldığı ve hatta yan tümcelerin ek kullanımlarıyla ve durum belirleme ekleriyle desteklendiği tümceli yapıları vurgulayan yeni anlayışlara ışık tutabilir. Anahtar Sözcükler Devinim olayları, çerçeveleme tipolojisi, dilbilgisi ve uzay, durum belirleme, yantümce, Türkçe. ix TABLE OF CONTENTS ACCEPTANCE AND APPROVAL……………………………………………….….i YAYIMLAMA VE FİKRİ MÜLKİYET HAKLARI BEYANI………………….…ii ETİK BEYAN………………………………………………………………………….iii ACKNOWLEDGEMENTS …………………………………………………………iv ABSTRACT ..…………………………………………………………………………v ÖZET………………………………………………..………………………………vii TABLE OF CONTENTS……………………………………………………………ix CONVENTIONS……. ………………………………………………………….…xiii LIST OF TABLES …………………………………………………………………xvi LIST OF FIGURES..……………………………………………….………………xviii INTRODUCTION……………………………………………………………………1 CHAPTER 1: THE STUDY…………………………………………………............5 1.1. STATEMENT OF THE PROBLEM…...……………………...............5 1.2. PRESENTATION OF THE STUDY………………………………......7 1.2.1. Aim of the Study and Research Questions…….…………….......7 1.2.2. Boundaries of the Research………………………………….......7 1.3. OUTLINE OF THE DISSERTATION………………………………....8 CHAPTER 2 : LITERATURE REVIEW………….……………………………….9 2.1. THE CONCEPT OF ‘MOTION’………….……………………………9 2.2. MOTION IN RELATION TO LINGUISTICS….……………………..10 2.2.1. Motion Events……….…………………………………….…….11 2.2.2. Acquisition of Motion Events in Early Childhood………………14 x 2.3. ENCODING MOTION EVENTS IN TYPOLOGY ………………….15 2.3.1. Talmyan Framework on Motion Event Encoding………….……17 2.3.2. Further Categorizations Than Binary Typology of Motion Events……...……...……...…………….……...18 2.4. FICTIVE MOTION….…………………………………….…………….21 2.4.1. Traces of Fictive Motion.…………………………………….…....21 2.4.2. Talmy’s Categorization of Fictive Motion………………………23 2.4.3. Fictive Motion in Languages………………………………….....25 2.5. AN OVERVIEW OF MOTION EVENT STUDIES IN TURKISH.......26 2.6. CASE AND SUBORDINATION IN TURKISH………………………..28 2.6.1. Case in Turkish…………………………………………………..28 2.6.1.1. The Accusative case marker……………………………30 2.6.1.2. The Dative case marker………………………………...31 2.6.1.3. The Locative case marker……………………………...32 2.6.1.4. The Ablative case marker………………………………34 2.6.1.5. The Genitive case marker………………………………36 2.6.2. Subordination in Turkish………………………………………...37 2.6.2.1. Noun (Complement) Clauses…………………………..37 2.6.2.2. Relative Clauses………………………………………..39 2.6.2.3. Adverbial Clauses……………………………………...42 2.6.2.3.1. Time………………………………………….44 2.6.2.3.2. Manner……………………………………….44 2.6.2.3.3. Purpose…………………………………….…45 2.6.2.3.4. Causality (Reason) …………………………..46 2.6.2.3.5. Condition……………………….……………46 2.6.2.3.6. Degree…………………….………………….47 2.6.2.3.7. Concessive……………….…………………..47 xi CHAPTER 3: METHODOLOGY………………………………………..................49 3.1. THEORETICAL BACKGROUND………………………..……………49 3.2. PILOT STUDY………………………………………………….………..51 3.2.1. Data Collection Tools.………………………………….….……52 3.2.1.1. First Pilot Study……………………………………….53 3.2.1.2. Second Pilot Study………………………..…………...56 3.2.1.3. Final Pilot Study ……………………………………...57 3.2.2. Data Collection Procedures …………………………….….……58 3.2.2.1. First Pilot Study …………………………...………….58 3.2.2.2. Second Pilot Study ……………………………………59 3.2.2.3. Final Pilot Study ……………………...………………60 3.2.3. Data Analysis and Coding …...…………………...…………….60 3.2.3.1. First Pilot Study …...…………………...…………..…60 3.2.3.2. Second Pilot Study …………………...……………….61 3.2.3.3. Final Pilot Study …………………...………………….62 3.2.4. Contribution of the Pilot Sets……………………………..……..63 3.3. THE MAIN STUDY……………………………………………………...65 3.3.1. Subjects.……………………………………………………. ……65 3.3.2. Data Collection Tools……………………..……………………..65 3.4. PROCEDURE…………………………………………………………….67 3.4.1. For the Analysis of Actual Motion ……………………………...67 3.4.1.1. Transcription of the Narrations…………………….......68 3.4.2. For the Analysis of Fictive Motion…………………………........68 xii 3.4.3. Data Analysis…………………………………………………....69 CHAPTER 4: ANALYSIS OF THE FINDINGS ………………………….…….70 4.1. ANALYSIS OF FINDINGS ON ACTUAL MOTION EXPERIMENTS………………………………………………………………..70 4.1.1. Total Clauses/Clauses with Motion……………………………...71 4.1.2. Main/Subordinate Clauses with Motion……………………........71 4.1.3. Selection of Motion Verbs……………………..……………......73 4.1.4. Selection of the Subordinate Clauses with Motion………….......74 4.1.5. The Use of Case Markings in Motion Expressions……..……….75 4.2. ANALYSIS OF FINDINGS ON FICTIVE MOTION EXPERIMENTS………………………………………………………………..77 4.2.1. Mean Lengths of the Fictive/Non-fictive Sentences……………..78 4.2.2. Comparison of the Fictive/Non-fictive Pairs…………………….78 CHAPTER 5: DISCUSSION OF FINDINGS……………………………………….96 5.1. DISCUSSION OF FINDINGS ON ACTUAL MOTION………....……96 5.2. DISCUSSION OF FINDINGS ON FICTIVE MOTION …………….105 CONCLUSION……………………………………………………...……………….111 BIBLIOGRAPHY…………………………………………………………………....120 APPENDIX 1. ETHİCS BOARD APPROVAL FORM …………………………..136 APPENDIX 2. ORIGINALITY REPORT……...…………………….……………137 APPENDIX 3. TASK SAMPLES IN THE PILOT STUDY………………………138 APPENDIX 4. FINDINGS FROM THE PILOT STUDY………………………...152 xiii CONVENTIONS The following is a group of listing which helps to understand the glosses, where applicable, the abbreviations that may be seen in the whole text, and the format which is followed both within the text and in the examples provided. The following begins with the list of glosses, list of symbols and abbreviations, and finally the typesetting. A. GLOSSES Æ: zero(empty) element 1: First person 3: Third person ABL: Ablative ACC: Accusative ADV: Adverbial ANom: Action nominal AOR: Aorist CAUS: Causative CL: Clause COMP: Complement COP: Copula CVB: Converb DAT: Dative DEF: Definite DEM: Demonstrative DET: Determiner F: Feminine GEN: Genitive IMPF: Imperfective xiv INDF: Indefinite INF: Infinitive LOC: Locative M: Masculine MAIN: Main Clause NUM: Numeral OBJ: Object ObjP: Object participle OBLG: Obligation PART/PTCP: Participle PASS: Passive PL: Plural PERF: Perfective POSS: Possessive PRN: Pronoun PROG: Progressive PRS: Present tense PST: Past tense P.COP: Past copula QUANT: Quantity REFL: Reflexive REL: Relative SbjP: Subject participle SG: Singular SUB: Subordinate VN: Verbal noun marker xv B. LIST OF SYMBOLS AND ABBREVIATIONS -: indicates a mark between morpheme boundaries in a word [ ]: shows the analysed section in clausal elements MANNER: indicates Manner verb PATH: indicates Path Verb ( ):shows the use of case marking C. TYPESETTING In the text: SMALL CAPS : Labels for basic anchoring categories. italics : Terms defined or introduced for the first time, linguistic markers and expressions, variables for semantic contents, or used for emphasis. Initial caps: Descriptive labels for grammatical elements of motion, such as Path, Manner. ‘single quotes’ : Terms used in other frameworks and English translations of the sample sentences from the study. bold: When a marker is in the focus of a discussion, it is emphasized with a bold letter type. In the examples: SMALL CAPS : Gloss items for Turkish examples. bold : Markers being analysed. italics : the English translation of parts that correspond to the analyzed element in the Turkish examples, and sentences with fictive motion is given in italics. xvi LIST OF TABLES Table 1. The number of clauses counted in the short movie and video animations….71 Table 2. Main and subordinate clauses with motion counted in the short movie and video animations……………………………………………………………....73 Table 3. Path and Manner verbs in clauses with motion counted in the short movie and video animations…………………………………………………………………73 Table 4. Path and Manner verbs in main and subordinate clauses……………………..74 Table 5. Distribution of subordinate clauses with motion……………………………...75 Table 6. The selection of case markings in motion expressions…………………….....76 Table 7. Mean length of Non-fictive and Fictive categories total in the drawing task...78 Table 8. Mean length of the pairs with no significant difference (=cm)……………….79 Table 9. Mean length of the pairs with significant difference (=cm)…………………..86 Table 10. The overall frequency of clauses and motion verbs in the study……………96 Table 11. Percentage of participants’ motion verb use in the present study………….97 Table 12. Percentage of adults’ motion verb use from Özçalışkan and Slobin’s study (2003) ………………………………………………………………………98 Table 13. Comparison of the pairs with and without any difference from drawings…105 Table 14. Participants’ judgment scores for Manner Verbs…………………………..153 Table 15. Participants’ judgment scores for Path Verbs……………………………...154 Table 16. Participants’ rating according to the appropriateness of the chosen sentence to the verb………………………………………………………………....155 Table 17. The most frequent 10 verbs in the narrations of participants………………156 Table 18. The distribution of subordinate clauses in narrations……………..………..158 xvii Table 19. Participants’ judgment scores for manner verbs………..………………….160 Table 20. Participants’ judgment scores for path verbs………..………………..……161 Table 21. Distribution of sentence structure based on participants’ ratings(n=18) …..161 Table 22. Drawing comparisons for non-fictive vs fictive by mean average…………162 Table 23. Total distribution of subordinate clauses with motion in both tasks……….165 Table 24. Total distribution of case markings in both tasks…………………………..166 xviii LIST OF FIGURES Figure 1. Case markers in Turkish……………………………………………...…29 Figure 2. The description of the tasks in each pilot study…………………………52 1 INTRODUCTION When we speak, there is actually a process whereby the information is linearized into sequential speech units. It becomes salient to talk about three-dimensional spatial information (Levelt, 1984) via different languages employing different ways to distribute the same spatial information into linguistic units. Changing places in our lives, moving things from one place to another, and seeing things moving in nature are all instances of motion, which comprises a collection of spatial elements. The human experience also includes talking about motion; however, beginning with Talmy’s (1975) seminal work on the typology of the linguistic expression of motion events, ways people talk about motion are said to differ according to the language they speak. The significance of motion in linguistics lies behind the idea that motion is an indispensable part of human thinking, and it is one of the core concepts in our mindset (Goddard, 1998). The representation of motion is not only a fundamental but also a high- level human cognitive ability, which may let scholars carry out deeper analyses into cognition (Landau & Jackendoff, 1993), and motion concepts are acquired quite early in childhood (Choi & Bowerman, 1991). Moreover, as stated above, each and every language has its own way of speaking about motion, and there is variability in the linguistic encoding of motion events. This variation, in turn, can posssibly affect the mental representations, and it can lead to different mental imagery about how one navigates in space. Last but not least, its expression exceeds the lexical level, such that it reaches to sentence and even discourse level (Slobin, 2004). In the last decades, “event” has been identified, with recent research on underlying conceptual organization of language, as one of the building blocks of language and cognition (e.g., Goldberg, 1998; Talmy, 2000). However, the definitions of the notion of ‘event’ differs both within and across different disciplines. For instance, within the field of linguistics and as part of the cognitive theories, events are viewed as conceptual units defined by perceived changes in quality between two breakpoints in the external world (Newtson and Engquist 1976; Zacks and Tversky 2001; Radvansky and Zacks 2014). In more detail, from this viewpoint, it can be assumed that event schemata or frames are 2 stored in the long-term memory and readily accessed during cognitive processing. On the other hand, events are described differently in semantic theory such that an event refers to a specific semantic unit which corresponds to a sentence (e.g., Davidson 1980; Kamp 1979). In this sense, the definition is more linked to the relations of quality and time (Klein 2010; Koenig 2016), as expressed by verbs and their arguments (Levin and Rappaport 2005). Depending on the typological categorization of languages, a wide range of studies on language production, comprehension, and acquisition of this event type (e.g., Bylund et al., 2013; Carroll et al., 2012; Flecken et al., 2015; Levinson 2003; Majid et al., 2004; and Papafragou et al., 2008) take their departure from the peculiarities between verb-framed (V-F) and satellite-framed (S-F) languages in that they differ according to how path and manner are encoded within their framing of events. Slobin (2006) assumes that “this basic contrast leads speakers of satellite-framed languages to pay more attention to manner of motion, since manner is typically expressed in the main verb with “path” expressed by a satellite, while the reverse holds for speakers of verb-framed languages”. However, many crosslinguistic studies demonstrate a high degree of variation even within language types. For instance, languages where path is typically encoded in the verb (e.g., French, Italian, and Spanish) also show relatively high frequencies in the use of manner verbs (Cardini 2012; Kopecka 2009). Likewise, English, categorized as satellite-framed where manner is encoded in the verb slot, displays frequent use of path verbs (Carroll et al., 2012). Thus, questions arise concerning the factors which govern the selection of one pattern (path or manner) over the other in actual language use. There can be different sources which play a role in the variation of these patterns as Pourcel (2004) points out: “(a) the type of entity with respect to the features human/non-human, whereby human figures encourage path over manner in contrast to non-human figures which encourage manner; (b) the type of manner, as with marked forms of motion such as to limp or to dash, which lead to a manner bias, in contrast to default means of moving (e.g., to walk with a bias toward path); and (c) the type of path, whereby a long trajectory often entails a bias toward manner, with a bias toward path given a short trajectory”. In the same vein, Durst-Anderson et al., (2013: 129) compare Russian, English and Danish and they assert that the ‘same’ representation of a motion 3 event can result in different linguistic outlooks regarding mental focus for the languages compared. These language-based outlooks are meant for perceptual and conceptual images in which the grammatical system of a language is the major determining factor. The difference in languages can be based on the selection of the content by the speakers of each language having differing points of departure. Language is very generous in extending motion expressions to the fictive domain as well. Fictive domain includes fictive entities which evoke for the entities and their relationship in between (animate or inanimate) in the real world. Linguistic studies of fictive motion have been attempted by numerous scholars (Bennett 1975; Talmy 1983; Langacker 1990; Levin 1993; Matsumoto 1996; and Matlock 2004 to name a few). Fictivity comes from ‘virtuality’ (Langacker, 1999), through which a lexical noun by itself (e.g., road, mountain) does not specify a type of ‘thing’, namely any specific instance of that type, as with a lexical verb by itself (e.g., go, run) barely specifies a type of event or situation, which Langacker (2005: 170) terms a ‘process’ but not any certain instance of the process. Rather, a type can be essentially described as a fictive entity that “represents an abstraction from actuality which captures the commonality (generic characteristics as a whole) inherent across a set of actual instances” (Langacker, 2005: 170). For example, let’s assume that the sentence “Mountains run parallel to the coast” refers to an actual series of mountains located close to an actual coast. The linguistic reference to motion – i.e., runs – seems to be at the level of instance. However, while the sentence itself is a statement about actuality (both the mountains and the coast refer to actual instances), the process of motion aimed to describe the mountains is fictive because no actual movement occurs from mountains to the coast. Fictive patterns are learned through abstraction, as put by Langacker (2008). Abstraction refers to a means of transcending direct experience and conforms to the structures it is based on but is less detailed. Abstraction is expressed in Langacker’s terms as: Abstraction comes about through the reinforcement of what is common to multiple experiences. And since commonalities are often apparent only in a coarse-grained view, involving lesser precision, abstracted structures are usually schematic relative to these experiences. Though immanent in all of them, an abstracted structure is independent of any instantiation. It represents a 4 generalization with the potential to be evoked in subsequent processing. Without the capacity for abstraction, every experience would be unique and unrelated to every other (Langacker, 2008: 525). Fauconnier and Turner (2002: 349) argue that “fictive motion blends a dynamic scenario of motion with a static situation so that the static situation can be conceived and described as having motion. The dynamic input contributes a moving trajector on a path, which is mapped onto a relevant dimension of the static object in the other input.” There are different ways to analyze fictive motion expressions in languages such as eye- tracking experiments in which participants’ eye movements are followed in a series of tasks by the experimenter; drawing studies in which participants are asked to draw on about the expressions they are showed. For the sake of the scope of the present study, these tools will not be extended here. The origin of variation even within language types are relatively new to the field of linguistic analysis of motion. Few studies (as will be shown in fhe following sections of the present study) have focused on the linguistic elements which play a role in the variation of path and manner encoding in languages. Considering whether and to what extent linguistic elements abovementioned can result in differences in the encoding of path and manner constructions, the present study is one of the first attempts that focuses on the analysis of motion events in Turkish with a view on the role of certain morpho- syntactic elements on encoding actual motion; and also aims to describe fictive motion expressions on the basis of certain drawings derived from fictive and non-fictive pairs of sentences shown to the native speakers. 5 CHAPTER I THE STUDY This section presents the general aims and the research questions of the study. It also briefly introduces the need for carrying out the present study, boundaries and limitations. This chapter concludes with an outline of each following chapter in the present study. 1.1. STATEMENT OF THE PROBLEM Following Talmy’s studies (1985, 1991) on the typology of motion events, various languages1 have been put on test ground for the analysis of motion events. Some of these languages have been categorized as typical of their typological group while some of them may exhibit certain idiosyncrasies. But this typology is mainly based on the lexical meaning of constituents, and do not focus on more general typological parameters such as morphosyntactic complexity, utterance structure constraints and relational information as controlled by verbal predicates (i.e. number of arguments, type of spatial complements, etc.). However, many crosslinguistic studies demonstrate a high degree of variation even within language types. For instance, languages where path is typically encoded in the verb (e.g., French, Italian, and Spanish) also show relatively high frequencies in the use of manner verbs (Cardini 2012; Kopecka 2009). Likewise, English, categorized as satellite- framed where manner is encoded in the verb slot, displays frequent use of path verbs (Carroll et al., 2012). Recent work on encoding patterns underlines the fact that spatial encoding involves a lot more than only lexical meaning (cf. Skopeteas 2008; Beavers et al. 2010). Therefore, the factors which govern the selection of one pattern (path or manner) over the other in actual language use is still open to debate. Turkish motion events have mainly been investigated by Özçalışkan and Slobin (1999- 1 English, Spanish, Hebrew (Aske, 1989; Berman and Slobin, 1994), Danish (Sinha et al., 1994), Russian, Dutch, German, Hindi (Bowerman et al., 2002), Korean (Choi and Bowerman, 1991), Chinese (Peyraube, 2006), Thai (Zlatev and Yangklang, 2004), Japanese (Nakazawa, 2006), Swedish and Icelandic (Ragnardottir and Strömqvist, 2004), Basque (Ibarretxe-Antunano, 2004), French (Kopecka, 2006), Tzeltal (Brown, 2004), Arrente (Wilkins, 2004), etc. 6 2000, 2003); Özyürek and Özçalışkan (2000) in terms of the acquisition of motion verbs and Toplu (2011) has carried out a study on motion events in Turkish with a comparison to those in English and French in terms of the linguistic relativity vs universals discussion. Recenlty Türk (2014) investigated whether there is a relation between accompanying gestures and motion expressions in Turkish. To date, little is known about whether Turkish is to be only regarded in a V-Framed typology or there may be flexibility (degree of variation) in encoding motion events with considerable use of manner verbs as well as ‘default’ path verbs in the verb slot. One outcome of the previous studies within the scope of motion expressions has showed that data comparability and replicability are significant in reaching quantitative and qualitative representativity. Therefore, with this idea in mind, one of the needs to carry out the present study is the methodological approach, and tools in the present study are easy to carry out in any language and open for comparability both within and across languages. Several recent studies (e.g., Beavers et al., (2010); Iacobini et al., (2020)) focus on to emphasize that encoding of motion events is conditioned by a number of motion independent properties governing the morphological, lexical and syntactic resources available to languages. Moreover, in motion events encoding, the focus of the analysis have started shifting from the just the analysis of the semantic components to the parts of speech involved in their expression, with ‘verb’ playing a crucial role. Taking this standpoint, another need to set off this study comes from the main theoretical effort made for the study which consists in the identification of several necessary and sufficient categories for the linguistic analysis of encoding motion expressions. With its framework, the present study is one of the first attempts in analyzing motion events in Turkish with a morpho-syntactic perspective through which linguistic devices of case endings and relations within main and subordinate clauses are taken into consideration. Further, it also covers an analysis of fictive motion descriptions based on a drawing task, which, in turn, makes this dissertation a first attempt to deal with fictive motion in Turkish. Using dynamic elicitation tools and adopting an easy-to-operate and comparable framework into the motion event analysis, the present study will hopefully shed light on fruitful findings into the motion event descriptions in Turkish. 7 1.2. PRESENTATION OF THE STUDY 1.2.1. Aims of the Study and Research Questions The present dissertation aims to provide a detailed description of the relation, if any, between motion verbs and morpho-syntactic elements regarding the uses of case markings and subordination in Turkish and to explain how these elements contribute to the description of motion and to have an outlook of fictive motion in Turkish by analysing drawings of fictive expressions. Accordingly, the research questions to be answered in this study are: 1. What kind of cases and subordinated constructions can go along with motion verbs to elaborate motion events in Turkish? 2. What is the contribution of occurrence for the cases and subordination to expressing motion events in Turkish? 3. Considering that linguistic elements such as case marking and subordination play a role in encoding motion events, in what ways any relation can be linked between subordination and encoding motion events in Turkish? 4. Does case marking play a role in regard to Change of State (CoS) and Change of Location (CoL) situations in participants’ descriptions of motion events? 5. What is the relationship between motion verbs and fictive motion in Turkish? 1.2.2. Boundaries of the Research This study had been planned to be carried out in person earlier, but the current pandemic conditions made this impossible. Therefore, the dataset in the current study was carried out all online, via ZOOM web application. Each session of the tasks with individual participants was recorded on the application. Although the framework of Beavers et al. (2010) followed in this study includes several 8 grammatical devices (e.g., ideophones, subordinate clauses, adverbs, affixes, applicatives, semantic cases, adpositions, particles), the present study analyzes motion events by taking into account two elements from that framework, namely the case markings and subordinate constructions accompanied to motion verbs in Turkish. For the analysis of fictive motion, a drawing task, adapted from Matlock (2006), was used for the descriptions of fictive expressions. The investigation of case marking and subordination was not applied to the analysis of fictive motion in the present study, since it has a different motivation. 1.3. OUTLINE OF THE DISSERTATION The present thesis is further organized into six chapters. The current chapter has outlined the general aims and the research questions of the study. It has also briefly introduced the need for carrying out the present study. Chapter 2 offers an extensive literature review on the background of this dissertation, on the bulk of research on motion covering different analyses and classifications of motion expressions that have been put forward. Chapter 3 consists of detailed information regarding the method and tools administered in this dissertation in a way that covers the beginning of the pilot study to the content of the main study and provides the theoretical framework of the study. Chapter 4 gives place to the analysis of the findings obtained in this dissertation. In addition, the results of the experiments are discussed in detail. Chapter 5 presents the overall results obtained in the main study and is divided into two sections for the discussion of the findings on actual motion and on fictive motion respectively. The discussion of the results is also supported by further samples from the present study. Final section will conclude the dissertation by summarizing the study in the light of re- stated Researh Questions; contributions of the findings in the field of investigation and in Turkish, and providing some suggestions for future studies. 9 CHAPTER 2 LITERATURE REVIEW This chapter offers an extensive literature review regarding the theoretical background of this dissertation, the majority of research on motion covering different analyses and some classifications of motion expressions that have been put forward so far. First, the term ‘motion’ is described in detail with its use in historical periods. Second, the use of motion in the field of linguistics is explained with the emphasis on the works of some prevailing scholars within the field of study. Following this, the core elements of motion events are described and further explained with relevant examples. Later, the acquisition of motion expressions is explained with pointing towards some relevant studies. Then, typlogical considerations on motion events are mentioned with an emphasis on Talmyian typology and on subsequent newly revisioned classifications of languages which can shed light on new possibilities of framing strategies in terms of motion constructions. Moving on to the motion expressions in fictivity, the field of fictive motion is described and followed by defining the field within linguistics. What is more, the subsections include the traces of fictive motion in language, Talmyian categorizations of fictive motion and some studies carried out so far regarding fictive motion in languages. After clearing the typological possibilities and placing the fictive motion, the overall mention is made on the studies of motion expressions in Turkish with some recent studies in the field. Following, the final section covers the case and subordination which comprise the methodological skeleton of the present study are explained in detail with reference to scholars and grammarians in the field. 2.1. THE CONCEPT OF ‘MOTION’ Motion is such an inherent concept in our lives that every time we change places, and we move things from one place to another, we experience motion, be it consciously or unconsciously. It is unsurprising that motion is also central to our language since we have different expressions and ways to describe motion in language. First inquiries on motion date back to Antiquity by pre-Socratic philosophers. Heraclitus of Ephesus (535-475 BC), reported by Plato (388 BC/1997) in Cratylus (401d, 402a), 10 claims that all entities move, and nothing remains still as the universe is constantly undergoing motion and change. For Aristotle, motion (kinesis) is a broad term found in his Physics (350 BC/1995), where he discusses the science of material nature in terms of motion and change and he defines motion as “the actualization of what potentially is, as such”. By ‘actualization’, Aristotle means both energeia, which means being-at-work, and entelechia, which means being-at-an-end, which relates respectively to a process and the result of a process (Waliński, 2018:24). For him, the concepts of energeia and entelechia function as synonyms, though a linguistic analysis of his definition (Kosman, 1969, 1987; Ugaglia, 2016) reveals a subtle complexity included in this definition. Later on his definition was challanged by Descartes, Galileo, Newton and some others in many points2. In sum, from the natural philosophy of Antiquity to present, motion has been seen both as change in position and the energy driving that change. The concept of motion is split in duality: inner and outer, process and result, active and inactive, cause and effect, animate and inanimate, volitional and accidental (Blomberg, 2014: 6), but the present thesis will not go into the details of each duality of motion pairs and will turn to the domain of motion in regard to linguistics. 2.2. MOTION IN RELATION TO LINGUISTICS The duality of motion is not new to linguistics. Tesnière (1959), for instance, proposed a general distinction between movement (mouvement) and displacement (déplacement). The former is “inner motion, the activity involved in motion” whereas displacement is “outer motion concerned with how somebody or something changes its location in space, notably with respect to a given point of reference” (cited in Wälchli, 2001: 298). To exemplify this, the former consists of the movements which are typical of human beings such as to run and walk but should also include the inner motion characteristic of inanimate objects like to oscillate and bounce. Nonetheless, there is an important difference in that the latter, or displacement, requires a reference to a surrounding, objective space: to change location is to be in two different places at two different moments (Blomberg, 2014: 6). 2 For details on views by Descartes and others, see Waliński, J. T. (2018: 24-28) Verbs in fictive motion. 11 It was only after the works of Leonard Talmy (beginning from the mid-1970s) that these stylistic differences and the distinction of Tesnière became the theme for general semantic and typological inquiry in linguistics and motion typology has grown to a research field in its own right. A recent definition of motion proposed by Zlatev, Blomberg, and David (2010: 5) is as follows: “[M]otion … can be defined as the experience of continuous change in the relative position of an object (the figure) against a background”. In the domain of motion, event properties and components such as agency and affectedness, intention and causation, and manner and path of motion are used by speakers. 2.2.1. Motion Events Talmy (2000: 25) describes a motion event basically as the situation that “consists of one object (the Figure) moving or located with respect to another object (the reference object or Ground)3”. A distinction between motion and movement is made by Talmy as the latter being the state of motion at a location (e.g., wriggling at a single spot) rather than change of location, which is the defining feature of motion. In addition, there is another distinction between motion events and motion activities. For Pourcel (2010), a motion event refers to a situation in which the conceptual emphasis is put on directionality and reaching a goal through the path of motion, e.g. “Tom walked to the store”. On the other hand, a motion activity specifies a motion in progress, e.g. “Tom is jogging”. So, the main difference between motion activities and events relates to the presence of directionality or a destination. As put by Pourcel (2010), “on the one hand, motion events refer to directional or goal-oriented motion by entailing a change of location where a manner, if specified, serves merely to follow the course of the path. On the other hand, motion activities do not inherently require overt directionality. They refer to the type of motion performed, which typically describes a specific manner” (Pourcel, 2010: 423). Zlatev, Blomberg, and David (2010) make a point about boundedness of motion. In a motion event, the Path implies bounded motion, whereas the Direction implies unbounded motion. The boundedness of motion refers to a state-transition4, i.e., that the 3 See Talmy, 1975b, 2000 (Vol 1), Ch. 5 for a discussion on Figure and Ground in language. 4 See Pustejovsky, 1991; Vendler, 1967 for more about state transition. 12 Figure departs from the Source, or passes through a Midpoint, or reaches the Goal. On the contrary, the unboundedness of motion means that motion does not reach a definite end point, as in “They marched forward/uphill” (Cappelle & Declerck, 2005). Another distinction between bounded vs. unbounded motion is telicity, which refers to the event completion understood as reaching the goal of motion (from Greek telos5 meaning “end”). In this case, telic and atelic motion can be differentiated (Comrie, 1976: 44–48). Motion activities are generally atelic since they refer to ongoing, uncompleted instances of motion whereas motion events are telic because they involve an endpoint, i.e., a change of location or state. The notions boundary and change represent the essence of our spatial and temporal parameters respectively and they are helpful in determining what situation type a speaker address. Thus, it gives us more about knowing than just the meaning of verbs or tenses used in that situation. This point is highlighted in Langacker (1987: 258), where the distinction between events and states has a “primal character”, because it is linked to a basic cognitive capacity: the ability to perceive change. The human mind is flexible enough to construe the same spaces as both locations and boundaries and it plays an immense role in the lexicalization of spatial domains which is why we are delving into the level between the perceptual and the linguistic, namely that of conceptualization. An event can be conceived of two modules -an internal structure and a degree of complexity-. Thus, as given in Cifuentes Férez (2008: 25), there are complex events, which are composed of a main event or framing event and a subordinate event or co- event (both of which are ‘conceptualized as unitary events’). In sum, there is the relation that the co-event depends on to the framing event. Talmy (1975b) divided motion into several semantic categories such as figure, ground, path and manner. Some additions have been made to the above explanation and some scholars pointed out that languages may show the path of motion, namely, the trajectory of the figure with respect to the ground, as well as the manner of motion, in other words, the rhythm, motor pattern, and rate of motion (Jackendoff, 1990; Slobin, 2004; Talmy 1985, 1991, 2000). Talmyan typology, despite a semantic one, explains that languages exhibit systematic similarities 5 The word entelecheia used by Aristotle in his discussion on motion in the sense of “being-at an- end” comes from the adjective enteles, meaning “complete, perfect”, whose root is ‘telos’. 13 and/or variations on the basis of the way in which they morpho-syntactically express the semantic domain of motion. Two sentences, one in English and the other in Turkish below, exemplify the main frame of motion: (1) The man swam into the cave. (from Skordos & Papafragou, 2014) (2) Adam ev-den koş-arak çık-tı. Man.NOM house-ABL by run-CVB exit-PST.3SG ‘The man exited from the house by running.’ By looking at the example (1) above, it can be said that the man is the figure, the cave is the ground, swam displays the manner of motion, and into the cave encodes the path or direction of the motion event. On the contrary, in the example (2), we can see that adam ‘the man’ is the figure, evden ‘from the house’ is the ground/source, çık ‘exit’ encodes path of motion and koşarak ‘by running’ displays the manner component. As seen from the examples above, languages may encode path and manner elements in one clause as in (1) or in two separate clauses, one to employ path (e.g., to exit) and the other to encode manner (e.g., running), making two as matrix-subordinate construction as in (2). To make those concepts clear, Talmy (2000: 25) explains each four element (as internal elements in motion) below: - FIGURE: It is an object / a person moving or located with respect to another object. - GROUND: It is the spatial reference point, according to which the motion or location of the ‘figure’ is determined. - PATH: It refers to the trajectory followed, or the space occupied by the ‘figure’ during motion. It consists of a source (the starting point), a medium (the intermediate points), and a goal (the end point) (Pourcel and Kopecka, 2006). Aske (1989) broadens the path and divides it into two: One is the ‘telic path’ (e.g. walk to the store from Goodrich and Snyder, 2012) that includes a certain end-point of a motion event described, and the other one is the ‘atelic path’ (e.g. walk in circles from Goodrich and Snyder, 2012) by which no specific end-point but the medium is presented. - MOTION: It is the presence of motion or locatedness itself. Talmy 14 distinguishes motion from movement, the latter being the ‘state’ of motion at a location (e.g., wriggling at a single spot) rather than ‘change’ of location, which is the defining feature of motion (Filipović and Ibarretxe-Antunano, 2015: 527). Apart from the four internal components of motion, Talmy (2000: 26) distinguishes an associated Co-event in the forms of Manner or Cause of a motion event: “a motion event can be associated with an external Co-event that most often bears the relation of Manner or of Cause to it”. Thus, the MANNER component reflects the manner in which the motion takes place, or the way of moving or being located in a motion event. The CAUSE is the reason of the occurrence of a motion event. Talmy (1985: 139) explains that the assessment of whether Manner or Cause is conflated in a verb depends on the verb’s basic reference to what the Figure does or to what the Agent/Instrument does. To clarify, while “He pushed the keg” expresses Cause because the verb refers to what the Agent (he) did; “He rolled the keg” expresses Manner since the verb basically refers to what the Figure (keg) did. 2.2.2. Acquisition of Motion Events in Early Childhood Some researchers have attested that spatial language emerges quite rapidly in young children, and spatial vocabulary seems to be mapped onto prelinguistic space and motion concepts (Piaget & Inhelder, 1956; Bowerman, 1978, 1980, 1996; Baillargeon, 1986, 1987; Clark, 2004; Casasola, 2008; and some others6). Lexicalization biases are found to affect children’s conjectures about the meaning of newly encountered verbs. Maguire et al. (2010) reported that 2-year-olds adopt similar motion verb construals regardless of language typology and they are mainly path- oriented. Moreover, by age 3, children start to diverge in their preferences in accordance with their mother tongue. Other studies have proved variations in the distribution of 6 See Gibson & Spelke, 1983; Hespos & Spelke, 2004; Kellman, 1995; Landau, 1994; Johnston, 1984, 1985; Pruden, Göksun, Roseberry, Hirsh-Pasek, & Golinkoff, 2013; Pulverman, Golinkoff, Hirsh-Pasek, & Sootsman Buresh, 2008 and Quinn, 1994 for further. 15 manner- and path-oriented interpretations of novel verbs in older children based on the languages they acquire (Papafragou & Selimis, 2010; cf. Hohenstein, Naigles, & Eisenberg, 2004). This is also true when ‘transitivity’ is taken into account. In one study, transitivity was tested on participants’ selection of motion verbs where English-speaking and Spanish-speaking adults tended to choose more path-focused interpretations of novel verbs when the verbs appeared in a transitive frame (e.g. She is kradding the tree) compared to an intransitive frame (e.g. She is kradding toward the tree; Naigles & Terrazas, 1998) because transitive frame motivated a relational/path interpretation of the predicate. Even in the domain of Spanish, they selected more path-focused ones because the transitive frame was consistent with Spanish language-specific (focus on path) verb lexicalization biases. On the contrary, in English, where the transitive frame contradicted the (focus on manner) language-specific verb biases, participants were indecisive between manner- and path-based verb generalizations. 2.3. ENCODING MOTION EVENTS IN TYPOLOGY Lexicalization patterns of languages provide insights into how speakers of different languages encode their experiences of events. Although several components of events are lexicalized in all languages, there seems to be significant variation in the way this is reflected in individual languages. Some languages express a group of components more frequent than others and also may omit some other components in particular structures that are habitually used in the lexicalization of a domain (Filipović, 2007:1). There is by no means any doubt that verb meaning is central to any account of motion lexicalization in languages; but there are many other elements such as the roles of verbs, prepositions, prefixes, verb phrases, prepositional phrases, adverbials that are indispensable to the analysis of motion (Filipović, 2007: 2). In regard to the clausal relations and motion events, Hieda (2016) studied Saamia, a Bantu language spoken by the Luhya people of Uganda and Kenya, in terms of event complexity and found that there are two different clausal elements by which a complex motion event is expressed: One is “no linker construction”, the other is “ni- construction”. According to Hieda (2016: 104), events described by a main and a subourdinate clause in the former are integrated into a complex event, while those in the 16 latter are not integrated, but separate. Further, it is concluded that “there is an iconic relationship between the order of events or states and that of clauses in the no-linker construction; the event or state described by the first clause has to happen before the event or state described by the second clause occurs. This could be an example of non- arbitrary relation between meaning and form” (Hieda, 2016: 109). Another study by Chen and Guo (2009) analyzes the place of Mandarin Chinese in motion event typology via an investigation of motion event descriptions in nine Chinese novels. They analyzed motion verbs, constructions and ground elements in Mandarin Chinese. They found that the pattern of motion verb use in Chinese is different from that found in both an S-language like English and a V-language like Turkish by comparing their findings with those in Özçalışkan and Slobin (2003) (Chen and Guo, 2009: 1759) and they had a conclusion that Chinese novelists express motion events in a way that does not clearly match with writers of either V-languages such as Spanish and Turkish or S-languages such as English. Aurnague (2011) deals with French intransitive motion verbs on the basis of change of relation and change of placement. He emphasizes the spatial properties of them with their semantic content. He states that the change of relation and placement introduced by the main verb in these constructions prevents the expression of another change of relation and placement in the infinitive clause (Aurnague, 2011: 28) and adds that some predicates (e.g., intransitive uses of pénétrer and s’infiltrer ‘to infiltrate, to percolate through’) are placed in between the expression of a change of relation and placement and of affectedness, that is, one aspect or the other being chosen according to the construction used (such as par-headed PPs, direct infinitival constructions). Creissels (2009) outlines the topic of spatial cases with several subdomains. First, upon definition of case, Creissels deals with the topics such as simple and complex spatial cases; spatial cases and semantic classes of nouns. Second, he shows some case systems like the ones of unidimensional and bidimensional by exemplifying from various languages. However, since it is beyond the scope of present study, details of these systems in languages will not be mentioned here. 17 2.3.1. Talmyan Framework on Motion Event Encoding Talmy was one of the first scholars to deal with a typology of motion events by investigating how the semantic structure of linguistic representations in different languages reflects the conceptual structure in the domain of motion. His early study on this topic (Talmy 1975) specified the range of surface structures (grammatical categories such as nominal, prepositional, verb constituents and subordination) and their semantic equivalents (Figure, Ground, Path, and Motion). Talmy’s (2000:27) goal was not to provide a comprehensive inventory of every possible codifying structure that could be used to encode motion in a given language, but rather just those that are “characteristic” or typical of the language; that is, those that are prevalent, commonly used by the speakers. The Path of motion is considered to be the fundamental component of a motion event because without Path there is no motion (though there may be movement), as Talmy (1985) states. The explicit presence of other components, such as Manner, though always present in reality, it is not obligatory for the verbalization of a motion event. Theoretical insights and exemplification from numerous genetically varied and geographically distant languages gave Talmy to create a two-way language typology on motion events: − Satellite-framed languages (S-languages): Path is characteristically placed in the satellite. (3) English run out. Satellite-framed languages are most Indo-European (excluding Romance), Finno-Ugric, Chinese, Ojibwa, and Warlpiri. − Verb-framed languages (V-languages): Path is characteristically codified in the verb root. (4) French partir en courant ‘leave running’. (Filipović and Ibarretxe-Antunano, 2015: 528). Languages that are said to be verb-framed are Romance, Semitic, Japanese, Tamil, Polynesian, most Bantu, most Mayan, Nez Perce, and Caddo (Filipović, 2007: 18-19). 18 2.3.2. Further Categorizations After Binary Typology of Motion Events More recent work extends Talmy’s typology to include a third class of ‘Equipollently framed (E-framed) languages’ or Serial Verb Constructions (SVCs) in which ‘path and manner are expressed by equivalent grammatical forms’ such as a series of two or more verbs that seem to be part of a single clause (Slobin 2004: 249; see also Slobin and Hoiting 1994; Zlatev and Yangklang 2004; Ameka and Essegbey 2013). Niger-Congo, Hmong-Mien, Sino-Tibetan, Mon-Khmer, Austronesian languages, Algonquian, Athabaskan, Hokan, Klamath-Takelman and Jaminjungan languages belong to this type. Croft et al. (2010) try to open a revision of Talmy’s two-way typology by saying that this kind of typology is asymmetrical in that only path or manner selection is available, which means that any other option like having both path and manner elements together in motion event constructions are not applicable in any language. So, this point is made in Croft et al. (2010: 201) and they suggest a kind of typology or classification should include such options as well. Another point they made is that Talmy’s typological classification applies to individual complex event types within a language, not to languages as a whole. Further, they give examples from Icelandic, Dutch, Bulgarian, and Japanese of certain widely cited examples in the resultative constructions (ibid: 202) and state that all of these languages use more than one of Talmy’s types to encode complex events. Therefore, it is an important issue for contrastive construction grammar studies: the basic unit of comparison and contrast across languages in regard to motion events is not the language as a whole, but each construction that is used to express an equivalent state of affairs. Fagard et al. (2013) question a two- or three-way typology of motion in languages by asking whether the notion of language types (with respect to motion typology or in general) should not be abandoned, and languages rather be described as conglomerates of constructions and strategies, with complex overlaps (Kopecka 2006; Beavers, Levin & Tham 2010; Croft et al., 2010). They also restate the question of “what exactly should be regarded as MOTION, PATH, and MANNER, since the way in which these concepts are defined, both theoretically and operationally, will inevitably affect the results from empirical investigations” (Zlatev, Blomberg and David 2010; Fortis et al., 2011). They carried out this study with six languages from VF (French and Piedmontese), SF 19 (Swedish and German and Polish) and EF (Thai) typologies. By adopting a holistic spatial semantics (Zlatev, 2003, 2007) framework7, their analysis is based on three elements: FRAMES of REFERENCE (FoR), PATH and DIRECTION. FoR can be sub-grouped as: The VIEWPOINT-CENTERED FoR which refers to deictic points in motion expression like “The woman is coming this way.”(cf Fagard, 2013: 366); The OBJECT-CENTERED FoR which involves a GROUND elements as in “Stand in front of me.” or “He went into the room.”, and finally The GEOCENTRIC FoR that includes an absolute reference points or axes horizontally or the vertically as in “Go west” or “He climbed up the stairs.” One of their findings is that ‘boundary-crossing constraint’ plays a role in the VF/SF distinction due to the claim that manner verbs are highly restricted in VF languages when the FIGURE crosses a boundary, but much less so for SF languages (cf. Aske 1989; Slobin & Hoiting 1994). In short, their stimuli with boundary-crossing element typically elicited utterances with manner verbs in SF languages but did not in VF languages of their sampling. Moreover, the differences they observed in patterns of expression of the PATH are not striking, unexpectedly between VF and SF languages. As a last category in their analysis, DEIXIS was expressed much less frequently than either PATH or MANNER. According to their conclusions, MANNER can be thought as a good distinctive element in framing languages as VF and SF. On the contrary, PATH was not found significantly different between VF and SF languages and the category DEIXIS was the most represented in Thai language. The overall conclusion from their study is that motion event typology should be performed on the basis of separate constructions or strategies, rather than on language as a whole. Fortis and Vittrant (2016)’s study entitled “On the morpho-syntax of path-expressing constructions: toward a typology” deals with discussing Talmyan typology and offers an inventory of constructions used in the encoding of path. They start with the conflation(adjunction) of patterns Figure and static or dynamic move or locatedness by giving the example sentence “The bottle floated into the cave” in which the verb float 7 A theory of the linguistic expression of spatial meaning which attempts to strike a balance between (embodied) universalism and language-specificity. It claims that the minimal unit of spatial analysis is the whole (trans)locative utterance, where the meaning of the parts is dependent on the whole utterance and vice-versa (Fagard et al., 2013:366). 20 results from a conflation of MOVE and AFLOAT. Later on, they cite Talmy (2000[1985]: 53-4) for the components of PATH: the vector, the conformation and the deictic, each of which will be explained right below: - vector: specifies a relation with respect to a ground, and this relation may be either static or dynamic (i.e., evolving through time). There are five types of vectors that correspond to the fundamental relations AT, FROM, VIA, ALONG and TO. - conformation: locates a figure with respect to spatial properties of a ground, such as its front, top, inside etc. For example, the ‘front of a computer’ is a spatial part which may specify the conformation of a figure with the ground in a motion event like: he sat down in front of the computer (cf. Fortis and Vittrant, 2016: 345). - the deictic component: situates the vector with respect to a point of view. Fortis and Vittrant (2016: 356) also exemplify (from Imbertt et al., 2011) that there can be multiclausal elements in some languages and they can serve to express various stages of the same event, and various relations to one or more grounds, like in the example given from Tacanan language (VUILLERMET, field notes in Fortis and Vittrant, 2016: 356): “eta'a-jo neki kwaji-kwaji-jaasowa-ani” [(lit.) ‘He stands at the river, he goes up running’] but turns to motion-reflected meaning as ‘he runs up from the river’. Here, the first clause (i.e. eta'a-jo neki) is functionally equivalent to an indication of the source of the motion event. Thus, they state the possibility that the specification of a path be left entirely to inferences based on sequenced events, with no path encoding form at all. Later in their study, they adopt a model which substitutes VERB with HEAD by showing that sometimes motion-including element is not always verb in languages and there are some cases where two verbs are found in a string of clauses one of which functions as a converb as the verbal satellite of the verb and the other is the verbal head which expresses a change of place away from the speaker as given in Japanese example “Ken- wa gakkoo-ni arui-te it-ta.” ‘Ken went to school walking.’ (Shibatani 2003 in Fortis and Vittrant, 2016: 356) where –ni is relator to satellite arui ‘walk’ verb and it ‘go’ functions as the verbal head of motion expressing a change of place away from the speaker. Finally, they offer a model regarding a series of PATH components as H-framed, SR-framed, HSR-framed and HSRA-framed which aim to cover different possibilities of 21 combinations of each of these frames in a PATH environment in different languages, but the details will not be given further in here. The present study takes its departure with the framework given in the following section. 2.4. FICTIVE MOTION It is not surprising to see that motion event analysis has gone beyond the experiences of actual motion. Thus, many kinds of experiences are so dynamic and tangible that they are thought of, imagined, and spoken of as if including a sense of motion. 2.4.1. Traces of Fictive Motion Early forms of analyses included sentences like “the post office is over the hill” and were explained as involving a reference point through which somebody could get to the location in question (Bennett, 1975: 35). To better understand, the sentences in (5) and (6) below convey a sense of motion but “not actually” in any domain; it is imagined, i.e., motion is layered on a static extended object: (5) The mountain range goes all the way from Mexico to Canada. (Talmy 2000) (6) The path rises steeply near the summit. (Langacker 2006) The sentences like above were descriptively labelled as “directional extent sentences” in Bennett (1975: 42), and “virtual motion” in Talmy (1983: 236). The verbs in those sentences were categorized as “pseudo-motional locatives” in (Dowty, 1979: 67) and “meander verbs” in Levin (1993: 256). In the literature, several different terms have been used to address the issue exemplified above, such as fictive motion (Talmy 2000), subjective motion (Langacker 1990), implied motion (Barsalou 2009), abstract motion (Matlock 2010) and recently non- actual motion (Brandt 2009; Blomberg & Zlatev 2013). Some cognitive researchers and psychologists have argued that the motivation behind using fictive expressions is based on a dynamic attitude on the speaker’s behalf (Langacker 1990; Talmy 2000; Matlock 22 2004, namely as a “mental simulation of motion” (Matlock 2004). (7) a. The balloon rose quite slowly. [objective, actual motion] b. Last year the price of coffee rose steadily. [objective, metaphorical motion] c. The trail rises steeply near the summit. [subjective, fictive motion] Langacker (2006, p. 24) Examples (7a-c) clearly show the different uses of the verb ‘rise’ in terms of motion from actual to fictive. As put by Langacker (2006:25) to clearly explain fictive motion: “This motion by the subject of conception is subjectively construed: the conceptualizer does not think of herself as moving through space, but merely apprehends the scene; the movement is inherent in the very conceptualizing activity, hence offstage and construed subjectively . . . The conceptual element of spatial movement therefore undergoes subjectification when rise is extended from factive to (imperfective) fictive motion.” In addition, the difference between actual and non-actual motion is a matter of construal in Langacker (1990)’s terms to account for linguistic means for signaling possible alternations in the speaker’s perspective. In more detail, actual motion pertains to an objective construal of motion and non-actual motion to a subjective construal; hence the term preferred by Langacker: subjective motion compared to fictive motion of Talmy’s. Matlock (2004: 1390) sees fictive motion expressions on the basis of mental simulation and states that “the conceptualizer (speaker or listener) takes a perspective in the scene and mentally simulates ‘movement’ or ‘visual scanning’ along the figure”. In short, it can be summarized that there is a correspondence between acting and thinking. To exemplify, Blomberg (2014: 166) uses an account of the action verb ‘pick’ referring to a finding in a neurophysics study which shows that the ‘verb’ pick activates the same parts of the brain as when its corresponding action is performed (Pulvermüller 2005). In sum, since sentences having an underlying sense of non-actual motion have their high degree of imaginability and reliance on “creativity”, these motion sentences 23 are clearly figurative (non-literal) and perhaps the only kind deserving to be called for them is “fictive”. In sum, the visualization of motion can be seen as an additional “layer” on top of the two kinds of experiences assumed by the analyses of Langacker and Talmy (Blomberg, 2014: 170). For this reason, the present study will henceforth stick to the terminology of ‘fictive motion’ when dealing with such expressions. 2.4.2. Talmy’s Categorization of Fictive Motion Talmy (1996) proposes a systematic account of fictivity covering the combination of perception and conception in a single continuous cognitive domain associated with visual perception or conception alone. Consequently, he coins the term ‘ception’, which is meant “to cover all the cognitive phenomena, conscious and unconscious, understood by the conjunction of perception and conception” (Talmy, 2000: 139). He, then, categorized the expressions into six types (explained below) based on the following parameters: a. Factive motion of some elements need not/must be present for that fictive effect. b. The fictively moving entity is itself factive/fictive. c. The fictive effect is observer-neutral/observer-based---and, if observer-based: i. The observer is factive/fictive. ii. The observer moves/scans. d. What is conceived as fictively moving is an entity/the observation of an entity. (Talmy, 2000: 105) His classification of fictive motion encompasses six relatively distinct categories: - emanation, which is essentially the fictive motion of an intangible entity emerging from a source. This category comprises a number of relatively distinct types, including orientation paths, i.e. “a continuous linear intangible entity emerging from the front of some object and moving steadily away from it” (Talmy, 2000: 106); radiation paths, i.e. “radiation emanating continuously from an energy source and moving steadily away from it” (Talmy, ibid: 111); shadow paths, i.e. “the linguistic conceptualization . . . that the shadow of some object visible on some surface has actively moved from that object to that surface” (Talmy, ibid: 114); and sensory paths, i.e. “the conceptualization 24 of two entities, the Experiencer and the Experienced, and of something intangible moving in a straight path between the two entities in one direction or the other” (Talmy, ibid: 115); - pattern paths, which involve the fictive conceptualization of some configuration as moving through space. “The literal sense of a sentence depicts the motion of some arrangement of physical substance along a particular path, while we factively believe that this substance is either stationary or moves in some other way than along the depicted path.” (Talmy, ibid: 129); - frame-relative motion, in which the factively stationary surroundings are fictively depicted as moving; - advent paths, which include depictions of a stationary object’s location in terms of its arrival or manifestation at the site it occupies. The two main subtypes include site arrival, i.e. the fictive motion of the object to its site; and site manifestation, i.e. the fictive change in the sense of the object’ manifestation at its site (Talmy, ibid: 135); - access paths, which are depictions of a stationary object’s location in terms of a path that some other entity might follow to the point of encounter with the object. The representation of the object as stationary, without any entity traversing the depicted path, is factive. What is fictive is the representation of some entity traversing the depicted path (Talmy, ibid: 136); - coextension paths, which are depictions of the form, orientation, or location of a spatially extended object in terms of a path over the object’s extent (Talmy, ibid: 138). In what follows, Talmy (2000: 138) also asserts that fictive change can also be expressed vithin fictive motion via some property of path. This is exemplified in the sentences (8) and (9) below: (8) The road disappears for a while by the lake and then reappears toward the border. (9) The fence gets higher as you go down the road. In the sentence (8) above, the spatial arrangement of two road sections towards the lake is construed as a complete single fictive entity. Its fictivity starts to change, from being absent first and then coming back to present again, as our attention scans along the entity. 25 Sentence (9) covers both fictive change and fictive motion, in a way that the fence is construed fictively while extending along the successive states of its different sections via a path through the road. In sum, what is generally meant from fictivity stems from the idea that perception is dynamic in two senses: One is that of a process unfolding together with movement, the other is the perceptual objects which give themselves in the dynamic flow of space. In turn, as can be linked to Husserl’s ideas on kinaestheses, they provide us with the kinesthetic capacity of perceiving static objects as features of the environment that afford movement (see Waliński, 2018: 74). 2.4.3. Fictive Motion in Languages Matsumoto (1996), Amagawa (1997), Takahashi (2001), Rojo and Valenzuela (2004), Stosic and Sarda (2009), and Hoffmann (2011) are among the few studies that cross- linguistically compare Fictive motion expressions often with English since it is one of the first and most studied languages in motion literature. Based on his comparative work of English and Japanese, Matsumoto (1996:194) has proposed two conditions constraining fictive motion expressions which are given as follows: A. The path condition: Some property of the path of motion must be expressed. B. The manner condition: No property of the manner of motion can be expressed unless it is used to represent some correlated property of the path. To exemplify the conditions given above, it is inconvenient for the path information to say “the road began to run” (Matsumoto, 1996: 195) without specifying any path information. We have to include some path elements like “the road began to run along the shore” (ibid). For expressing manner, “the road… zigzagged through…the forest” (Matsumoto, 1996: 196) is more appropriate than “the road … speeds … through the park” (ibid), because zigzag is helpful in depicting the shape of the road. Rojo and Valenzuela express that there is evidence indicating that the two conditions are generally applicable to Spanish (Rojo & Valenzuela, 2004: 141). According to the two conditions proposed by Matsumoto above, fictive motion expressions generally reject manner information but must contain some path information. As a result, when English 26 fictive motion expressions are translated into Spanish, much less path and manner information is removed compared with translations of physical motion expressions (Rojo & Valenzuela, 2004: 134). In a later study by Rojo and Valenzuela (2009: 253), it is found that verbs expressing non-path-related manner information are harder for Spanish speakers to process. Another study compares Serbia and French (Stosic & Sarda, 2009) in terms of the strategies employed in expressing static location in typologically different languages. Serbian is a satellite-framed language where more manner information can be encoded whereas French is a verb-framed language in which manner is less expressible. In expressing locative motion events, Stosic and Sarda (2009: 51) found that Serbian uses more sentences containing posture verbs (which are assumed to be equivalent to manner verbs in translational motion events) and fewer fictive motion expressions compared with French. Two significant outcomes of this study are: a) manner is more salient in satellite-framed languages (Serbian) than in verb-framed languages (French) in the domain of static location and, b) highly manner-salient languages tend to be limited in the use of fictive motion expressions and vice versa (Stosic & Sarda, 2009: 57). All in all, the investigation of fictive motion has been developing since the mid-1980s. From that on, a number of varied tools and methods have been used to study this cognitive-linguistic phenomenon. These means of data collection range from rational linguistic analyses and cross-linguistic comparisons to empirical psycholinguistic experiments, eyetracking studies, and more recently, important insights contributed by neuroimaging (Waliński, 2018: 233). However, although the body of research continues to grow, we are still at an initial step to determine neatly how the conceptual mapping in fictive motion comes out and what the cognitive processes behind this phenomenon are. 2.5. AN OVERVIEW OF MOTION STUDIES IN TURKISH Based on Talmy (2000)’s typology of languages, Turkish is said to belong to the verb- framed languages in which ‘path’ is encoded in the main verb and the ‘manner’ component is given, if necessary, by subordinated means such as the use of adverbials, gerunds or adpositional phrases as seen in the example below: (10) Kadın, ayaklarının ucuna basarak oda-sın-dan çık-tı. 27 Woman.NOM on tiptoes(Advmanner) room-3SG.POSS-ABL exit-PST.3SG(Vmotion+path) ‘The woman exited from her room on tiptoe.’ (from Özçalışkan and Slobin, 2003) Turkish motion events have mainly been investigated by Özçalışkan and Slobin (1999- 2000, 2003); Özyürek and Özçalışkan (2000), Özçalışkan (2009) and recently Toplu (2011) has carried out a study on motion events in Turkish with a comparison to those in English and French in terms of the linguistic relativity vs universals discussion. Another recent work on Turkish motion events comes from Türk (2014) where he dealt with the relationship between the motion events and gestures accompanied to motion event expressions. To briefly explain, Özçalışkan and Slobin (1999-2000) found, through experiments on children and adults of Turkish, English and Spanish, that a) the patterns of Turkish and Spanish were in tune with V- language typology, such as using path verbs frequently and, English patterns were closer to the S-language typology due to the vast use of manner verbs together with path satellites, and b) children also followed language-specific lexicalization patterns of motion beginning at the age of 3. In another study, Özyürek and Özçalışkan (2000) found that gestural expressions of spatial concepts begin nearly at the age of 6 and added, to make sure, that there must be more prospective studies to be carried out with cross-linguistic comparisons of languages in this regard. Özçalışkan and Slobin (2003) analyzed a number of written narratives from selected novels in English and in Turkish and they reached the same typological differentiation in that English novels consisted of more manner verbs than Turkish novels, and Turkish novels included more path verbs. Özçalışkan (2009) carried out another study on how children develop producing spatial motion patterns in a comparison between Turkish and English. She sampled three groups of children cohorts and adults. She found that crosslinguistic difference was evident the manner lexicon of languages in that English has more manner verb types than Turkish and that Turkish speakers do not express manner in the main verb. On the contrary, English has less path of motion lexicon and in terms of path satellites, English speakers tended to use more satellites as path elements added to a single verb of motion but Turkish speakers did not have a tendency to produce path elements other than the verb slot which is by default for Path in Turkish. 28 In her dissertation, Toplu (2011) found that languages under investigation (i.e. English, French and Turkish) did not show language-specific patterns in canonical motion event expressions used in the categorization task of motion events and this shows a universal frame of manner information while describing motion events. She also found similar results with former studies mentioned above in that native speakers of Turkish and French used path sentences nearly in all of their descriptions of motion events while English descriptions included very high manner information and this result is a reflection of V- framed vs S-framed typology as found in previous studies of Özçalışkan and Slobin (1999-2000, 2003); Özyürek and Özçalışkan (2000). Türk (2014) investigated expressions of motion events and the gestures accompanied to them in Turkish discourse from a small gesture and speech annotated corpus. The corpus was obtained from video recorded narrations by the participants who narrated a story from a wordless picture book “Frog where are you?” (Mayer, 1969). The findings of study showed that path gestures were the most used type in the dataset. However, manner information is more frequently gestured when the number of manner expressions and manner gestures are compared. It was also found that the narrators did make gestures for path more than manner information although both types of elements were marked prominent in terms of prosody. Unfortunately, no attempt on fictive motion expressions in Turkish has yet been found in the related literature. 2.6. CASE AND SUBORDINATION IN TURKISH 2.6.1. Case in Turkish Case system in Turkish is represented via suffixation where the case markers are attached at the end of the nouns. Turkish has six8 case suffixes and five of them mark respectively the accusative, dative, locative, ablative and genitive cases. The nominative case is not shown via suffixation but it functions to mark the subjects of the clauses. The 8 The comitative (instrumental) marker -(y)lA/ile have similiar features in common with case suffixes, thus it is discussed in sections 8.1.4 and 17.2.1 in Göksel and Kerslake (2005). 29 function of case marking is to show the link between the noun phrase a case marker is attached to and other constituents of sentences. (Göksel and Kerslake, 2005: 154). Major case markers are given below: Figure 1. Case markers in Turkish Case Categories Marking Suffix Nominative / absolute Æ Accusative (y)I Dative (y)A Locative DA Ablative Dan Genitive (n}In/Im There is an example set of clauses where each one of these cases is used and it is given below: (11) Ahmet Æ [[ Ali -nin gazete -yi Oya -ya büro -da ver -ip ] [ sen -in iş-ten Ahmet(NOM) Ali -GEN newspaper -ACC Oya -DAT office -LOC give -and you -GEN work -ABL konser -e gid -eceg -in] ] -i bil -Iyor concert -DAT go -FNomFUT -3.SG. -ACC know -PR.PROG. "Ahmet knows that Ali will give the newspaper to Oya in the Office and (that) you will go from work to the concert". (c.f. (752) in Kornfilt, 1997:213) As a summary of the markers in (11) above, NOM and GEN markers are attached to the subjects in clauses; person markers are added before case markers are attached. The following examples regarding the details of the cases will not be glossed but only the place of the case markers will be highlighted. 30 2.6.1.1. The accusative case marker The function of the accusative case marker is to mark the direct object of a transitive verb and it is decided through the following ways: (i) The use of the accusative suffix is compulsory where the direct object is definite: (12) Bütün arkadaşlarımız-ı çağıralım. (c.f. (63) in Göksel and Kerslake, 2005: 156) ‘Let’s invite all our friends.’ (ii) Accusative case marking is also required where a non-definite direct object comes before the verb but does not occupy the immediate pre-verbal position: (13) Birçok şey-i şu raflara koyabiliriz. (Indefinite) ‘A lot of things we can put on these shelves.’ (14) Patlıcan-ı her gün yiyebilirim. (Categorial) ‘I could eat aubergines every day.’ (iii) An indefinite direct object which is in the immediately pre-verbal position must still take the accusative suffix in the following crcumstances: (a) If the direct object is marked with a possessive suffix: (15) Bir arkadaşım-ı getireceğim. ‘I’m going to bring a friend of mine. b) If the direct object is an indefinite or plural generic: (16) Ahmet o anda [koşuya hazırlanan] bir atlet-i andırıyordu. (indefinite) ‘At that moment Ahmet looked like {an athlete [preparing for a race]}.’ (c.f. (71) in Göksel and Kerslake, 2005: 333) (17) Ali doktorlar-ı sevmez. (plural generic) ‘Ali doesn’t like doctors’. (c.f. (72) in Göksel and Kerslake, 2005: 333) (c) If the direct object refers to a member or members of a previously mentioned or implied group: 31 (18) Paketin içindekiler eksik çıktı. İki kitab-ı göndermemişler galiba. ‘The contents of the parcel are incomplete. I think they’ve failed to send two [of the] books.’ (19) Salon kalabalıktı. Kapıya yakın duran bir adam-ı tanıdım. ‘The room was crowded. I recognized a man standing near the door.’ 2.6.1.2. The dative case marker A noun phrase marked with the dative case suffix can have following functions as: (i) An adverbial indicating one of the following: (a) The recipient or beneficiary of an action: (20) Çocuğ-a doğru dürüst bakamıyor. ‘S/he can’t look after the child properly.’ (21) Aysel’e anahtar verdim. ‘I’ve given Aysel a key/keys.’ (b) The destination or target of an action: (22) Beni Paris’e gönderdiler. ‘They sent me to Paris.’ (23) Bu koltuk oturma odasın-a konacak. ‘This armchair is to be put in the sitting-room.’ (c) The price at which something is sold or offered for sale: (24) Bu bisikleti iki yüz milyon-a almıştım. ‘I bought this bicycle for 200 million [lira].’ 32 (d) Purpose: This kind of dative-marked noun phrase is almost always a -mAK clause (25) [Seni görmey]-e geldim. ‘I‘ve come to see you.’ (ii) The oblique object of many verbs of emotion, such as sevin- ‘be pleased (about)’, üzül- ‘be sorry (about)’, kız- ‘be angry (with/about)’, can-ı sıkıl- ‘be annoyed (about)’, and certain other verbs, e.g. benze- ‘resemble’, uy- ‘conform (to)’, ‘comply (with)’, inan- ‘believe’, güven- ‘trust’: (26) [Ayşe’nin geleceğin]-e sevindik. ‘We’re glad Ayşe’s going to come.’ (27) [Annemin isteğin]-e uymadım. ‘I didn’t comply with my mother’s wish.’ (iv) The ‘causee’ of a causative construction based on a transitive verb, i.e. the person who is made or allowed to perform the action: (28) Filiz bütün ev işlerini kocasın-a yaptırıyor. ‘Filiz makes her husband do all the housework.’ 2.6.1.3. The locative case marker The locative suffix expresses physical or abstract location. A noun phrase in the locative case can function as one of the following: (i) A time or place adverbial: 33 (29) O günler-de Selim çok sigara içiyordu. ‘At that time Selim was smoking a lot.’ (30) İnsanlar artık komşularını bile tanımıyorlar büyük kentler-de. ‘People don‘t even know their neighbours nowadays in big cities.’ (ii) The oblique object of a small number of verbs, such as karar kıl- ‘decide (on)’, ısrar et- ‘insist (on)’: (31) [Hepsini denemek]-te ısrar etti. ‘She insisted on trying them all.’ (iii) A subject complement: (a) In linking sentences: (32) Anahtar yerin-de değil. ‘The key is not in its place.’ (b) In small clauses: (33) [Onu İstanbulda] sanıyordum. ‘I thought he was in Istanbul.’ (iv) The locational constituent of an existential sentence: (34) Köy-de elektrik var mıydı? ‘Was there electricity in the village?’ (v) Within a larger noun phrase, a compound adjectival modifier expressing metaphorical ‘location’ in some kind of attribute (size, shape, colour, name, age, etc): (35) otuz metre derinliğin-de bir kuyu ‘a well thirty metres deep’ 34 2.6.1.4. The ablative case marker The ablative case marker indicates that a noun phrase is functioning as one of the following: (i) An adverbial associated with concepts such as departure, separation, source, or cause: (36) Ali oda-dan çıktı. ‘Ali left the room.’ (37) Zavallı bunu yorgunluk-tan yapmıştır. ‘The poor thing must have done this out of tiredness.’ In association with the verb geç- ‘pass’ an ablative noun phrase indicates a place or space through which someone/something travels: (38) Gelirken şehir merkezin-den geçtiniz mi? ‘Did you go through the city centre on your way here?’ (ii) The oblique object of certain verbs of emotion, especially those which reflect the concepts of aversion, e.g. kork- ‘be afraid (of)’, iğren- ‘be disgusted (by)’, nefret et- ‘hate’, bık- ‘get fed up (with)’, hoşlan-‘like’: (39) O adam-dan nefret ediyorum. ‘I hate that man.’ Certain other verbs, notably vazgeç- ‘give up’, faydalan-/yararlan- ‘benefit (from)’ and oluş- ‘consist (of)’ also take ablative-marked objects: (40) Zerrin [tenis oynamak]-tan vazgeçti. ‘Zerrin gave up [playing tennis].’ 35 (iii) The complement of: (a) Certain bare postpositions, e.g.  önce ‘before’, sonra ‘after’, başka ‘apart from’, dolayı ‘because of’: (41) Okul-dan sonra genellikle futbol oynuyor. ‘After school he usually plays football.’ (b) Certain adjectives, e.g. memnun ‘pleased (with)’ : (42) Hayatın-dan memnun g rünüyor. ‘She seems content with life.’ (iv) A subject complement with partitive meaning: (a) In nominal sentences: (43) Osman {yakın arkadaşlarımdan (biri)} deği