Στόχος της προκείμενης έρευνάς μου είναι η περιγραφή του φωνηεντικού συστήματος της Παφίτικων λαμβάνοντας υπόψη σημαντικές παραμέτρους που επηρεάζουν την πραγμάτωση των φωνηέντων. Να υπογραμμίσω ότι ως αντικείμενο της εργασίας επιλέχθηκαν τα φωνήεντα (και όχι τα σύμφωνα) διότι αυτά εμφανίζουν κάποιες σημαντικές ιδιότητες για την έρευνά μου. Πρωτίστως, τα φωνήεντα είναι ο ‘πυρήνας’ των συλλαβών, οι οποίες πολλές φορές αποτελούνται από μόνο ένα φωνήεν. Επίσης, χρησιμοποιούνται συχνότερα από τα σύμφωνα και μεταφέρουν περισσότερη ακουστική ενέργεια. Η ομάδα των φωνηέντων είναι πιο ομοιογενής όσον αφορά τα φωνητικά τους χαρακτηριστικά, τα οποία είναι απλά και σταθερά. Συνάμα, έχουν μεγαλύτερη διάρκεια από τα σύμφωνα και είναι φορείς του τονισμού (προσωδία) κατά την ομιλία. Η εργασία βασίζεται στην ανάλυση γλωσσικού υλικού ορισμένου αριθμού πληροφορητών της δεύτερης κυρίως γενιάς Κυπρίων με κοινή καταγωγή από την πόλη και επαρχία της Πάφου. Οι λόγοι που με οδήγησαν στην επιλογή του Παφίτικου ιδιώματος ποικίλουν. Αρχικά, η συγκεκριμένη πόλη είναι μικρή, απομονωμένη από τον υπόλοιπο κορμό του νησιού και με μια κοινωνία σαφώς πιο κλειστή και μια οικογένεια πιο συντηρητική και παραδοσιακή. Σημαντικός επίσης παράγοντας για την επιλογή της συγκεκριμένης περιοχής αποτέλεσε το γεγονός ότι συμπεριλαμβάνεται ανάμεσα στα 18 ιδιώματα που τείνουν να εμφανίζονται στο νησί παρουσιάζοντας φωνολογικές και μορφολογικές διαφορές σε σχέση με άλλες περιοχές και έτσι απώτερος σκοπός ήταν να αναδειχθεί η θέση της στη τοπική γλωσσική κοινότητα. Επιπλέον, παρόλο που οι μελέτες για την Κυπριακή διάλεκτο είναι άφθονες, δεν σημειώνεται στη βιβλιογραφία καμία μελέτη για το Παφίτικο ιδίωμα και έτσι θεώρησα ενδιαφέρον να μελετηθεί η έντονη και βαρετή ντοπιολαλιά που ακούγεται με περιέργεια και προκαλεί πειράγματα. Αναλυτικότερα, στόχος της παρούσας εργασίας είναι: α) να εξετάσει την πραγμάτωση των πέντε φωνηέντων σε τρία είδη λόγου (φυσικός λόγος, συνομιλία κατά τη διάρκεια map task test και αναγνωσμένος λόγος) και να εντοπίσει διαφορές λαμβάνοντας υπόψη γλωσσικές και εξωγλωσσικές παραμέτρους που σχετίζονται με το φωνητικό περιβάλλον, τον τόνο καθώς επίσης και τη θέση της συλλαβής και β) να επισημάνει που εντοπίζονται και που οφείλονται οι διαφοροποιήσεις. Βασική υπόθεση της παρούσας έρευνας είναι η εξής: α) αναμένουμε ότι ανάμεσα στο πρώτο και στο δεύτερο είδος λόγου, δηλαδή στον αυθόρμητο φυσικό λόγο και στο λόγο που εκφωνήθηκε κατά τη διάρκεια map task tests ότι θα εμφανίζονται ομοιότητες με βάση τη διάκριση γλωσσικών και εξωγλωσσικών παραμέτρων, καθώς τα map task tests είναι μια διαδικασία παραγωγής φυσικού λόγου. Τα map task tests σχεδιάζονται με τέτοιο τρόπο ώστε να αποσπάσουν συγκεκριμένα παραδείγματα αυθόρμητης γλωσσικής συμπεριφοράς, σε ελεγχόμενες συνθήκες. Αντιθέτως, έντονες διαφοροποιήσεις αναμένουμε να εντοπίσουμε μεταξύ φυσικού και αναγνωσμένου λόγου. Η παρούσα εργασία θα προσπαθήσει να ανταποκριθεί στα ερωτήματα και στις υποθέσεις που διατυπώθηκαν. / --

This is the first typological study devoted to Medieval Cypriot (MC). The objective of the study is to provide both syntactic and pragmatic factors which are determining for the word order of the language and to open new ways to recording mechanisms of word order change. Cypriot syntax deserves this attention, as it is a language highly interesting for the typologist as for the researcher of other linguistic areas; Modern Cypriot is VOS, and exhibits a series of exceptions to the general rules of V-initial languages. Medieval Cypriot conforms to most of Greenberg�s Universals (1963) which are pertinent to type VSO in that it has V in initial position in all unmarked clauses, in that it is prepositional, that adjectives mostly follow the noun they qualify, and so on. However, the comparison of MC to Greenberg�s Universals is not the aim of this work. Apart form the order of the main constituents, this research mainly focuses on revealing mechanisms of syntactic change not generally known, and on unveiling particular traits of the Cypriot VSO order that are not common to other VSO languages. The analysis can be defined as diachronic for it deals with the language written over a span of many years, as assumed from studying the texts. Some words and structures, used in the beginning of the narrative, seem to decrease in frequency in the end, or vice versa. It is diachronic considering it also allows for comparison with later (colloquial) and earlier (written) constructions of the language. However, it is mostly a synchronic analysis; the patterns observed are from within the same language spoken by the same people living in the same period, more importantly from within the same work. Makhairas is thus the only broad evidence of his period, offered both as a diachronic and a synchronic linguistic testimony of his time. As no language exists in vacuo, my description of MC starts with a historical approach to the language under study; it is almost impossible to realise the problems of colloquial, literary and foreign features without being aware of the earlier history of Greek in general and of Cypriot in particular, in some of its earlier documents. I refrained as far as possible from entering the field of comparative criticism with Medieval Greek. In this way I decided to focus on discussions based exclusively on the Cypriot forms and patterns, as presented and justified by the evidence in Makhairas, and as witnessed by history which, for many centuries, has singled out Cypriot from the rest of the dialects and the Greek language itself. So, alternative views, criticism and discussion of same mechanisms of change recorded within the broader Greek language have been more or less avoided. The exposition of the MC word order patterns is based on my hypotheses that word order, as I understand it, is founded on purposes of communication and that languages with extreme flexibility of order, such as Medieval Cypriot, may adopt patterns that display rigidity of order in a number of their elements. It is within these areas of rigidity that new mechanisms of change may be detected. I also hypothesised that the same syntactic changes within languages of the same branch may be merely coincidental, and that Greek or forms of Greek may well adopt foreign elements, only (but not exclusively) if these acquire the Greek endings, or if they appear as independent affixes, as is the case with the post-medieval referential Cypriot marker �mish� which is from Turkish. Acquiring particular elements from other languages does not mean acquiring their order. However, acquiring patterns that are similar to Greek from a borrowing language which has the same patterns does not exclude syntactic borrowing. Since Modern Cypriot is V-initial, I presumed that this might have also been its order in the Middle Ages. I judge that major mechanisms of syntactic change of the same period may have been triggered by factors internal to Cypriot rather than by the more general, universal mechanisms of change. Moreover, I speculated that MC was a far more marginalised language in the Middle Ages than what history and literature have taught us. Its creative dynamism and potentiality to �juggle� between words and patterns has been its greater forte. Cypriot has not been studied as a dialect, in this work. I avoided having only a partial or a shadowed understanding of its word order patterns. Exhaustive descriptions that show its particularities in the process of completion appear with both rigidity (in some elements) and flexibility of order, and most importantly, they exhibit a long-life endurance. I have also been concerned with forms and /or patterns of Greek such as the future and other periphrastic tenses, although they are already known and have been analysed at length in Greek linguistic studies. I concentrate here on some of these from a Cypriot perspective. Cypriot has never been classified as Balkan Greek or mainland Greek. Following this study, it will be clarified further that any attempt to fit MC into a framework defined along these categorisations will be successful only in some areas of the general Greek syntax. In fact, Cypriot opens the way for a further understanding of Greek syntax with its (almost) boundless flexibility; it is through MC and the unique data of Makhairas that the study of the Greek syntax is being enriched. Areas of fine-grained classificatory criteria result in connecting some MC syntactic traits to those of Greek and accrediting to the language its own word order singularities in what can be righteously called here the Cypriot syntax. Additionally, the study aims to open new areas of investigation on diachronic syntactic issues and to initiate new and revealing answers concerning configurational syntax. To determine the syntactic traits of MC a meticulous work of counting was needed. The counting of the order of the main constituents from both the more general narrative patterns of the Chronicle as well as of those passages thought to be more immediate to the author�s living experience(s) was done manually. The primarily and more difficult task of considering, following and explaining pragmatic word order patterns in the Chronicle has been the stepping stone of this research. Earlier (and forgotten) stages of Greek, and patterns exclusive to Cypriot, assembled in a unique lexicon and with special Cypriot phrasal verbs, have provided answers to explaining the Cypriot structure. In addition to statistics, areas of language contact have also been explored, both in the morphology and in the syntax. More importantly, the extreme word order freedom of MC that illustrates word order processes based entirely on internal structural changes, aims to contribute to discussions regarding morphology and syntax versus morphosyntax. Chapter 1 provides all the background information of the history and language in Cyprus, prior to the Middle Ages. Chapter 2 deals with the description of the data and the methodology used to assess them. Chapter 3 exhibits the MC verbal forms, both finite and non-finite; it examines non-finites more closely, inasmuch as they play an important role in the change of the order of major constituents and uncover and explain the role of V-initial structures. Chapter 4 is the core chapter of this work. It displays Cypriot particularities of word order, reveals data concerned with the word order of the major constituents within the clause and unfolds explanatory accounts of them; lastly, it classifies MC as a V-initial language. Chapter 5 summarises conclusions, adds a further note on the Cypriot morphosyntactic traits while placing the results into the contemporary scholarship on VSO languages, also suggesting additional research areas into the MC patterns. The examples from Makhairas have been written in the monotonic system, where only one accent has been used; other special symbols have been eliminated or modified in the interest of making the text readable in the absence of the right font. However, Ancient Greek words appear with their appropriate accents. Abbreviation C indicates structures or words that remained unchanged in Cypriot over a long period of time, and G means a form or word accepted in both their written and spoken forms over a long period of time in Greek. A morphemic analysis of each form of the glosses has not always been given. I limited myself to glossing some elements only, for the better understanding of some examples.

This is the first typological study devoted to Medieval Cypriot (MC). The objective of the study is to provide both syntactic and pragmatic factors which are determining for the word order of the language and to open new ways to recording mechanisms of word order change. Cypriot syntax deserves this attention, as it is a language highly interesting for the typologist as for the researcher of other linguistic areas; Modern Cypriot is VOS, and exhibits a series of exceptions to the general rules of V-initial languages. Medieval Cypriot conforms to most of Greenberg�s Universals (1963) which are pertinent to type VSO in that it has V in initial position in all unmarked clauses, in that it is prepositional, that adjectives mostly follow the noun they qualify, and so on. However, the comparison of MC to Greenberg�s Universals is not the aim of this work. Apart form the order of the main constituents, this research mainly focuses on revealing mechanisms of syntactic change not generally known, and on unveiling particular traits of the Cypriot VSO order that are not common to other VSO languages. The analysis can be defined as diachronic for it deals with the language written over a span of many years, as assumed from studying the texts. Some words and structures, used in the beginning of the narrative, seem to decrease in frequency in the end, or vice versa. It is diachronic considering it also allows for comparison with later (colloquial) and earlier (written) constructions of the language. However, it is mostly a synchronic analysis; the patterns observed are from within the same language spoken by the same people living in the same period, more importantly from within the same work. Makhairas is thus the only broad evidence of his period, offered both as a diachronic and a synchronic linguistic testimony of his time. As no language exists in vacuo, my description of MC starts with a historical approach to the language under study; it is almost impossible to realise the problems of colloquial, literary and foreign features without being aware of the earlier history of Greek in general and of Cypriot in particular, in some of its earlier documents. I refrained as far as possible from entering the field of comparative criticism with Medieval Greek. In this way I decided to focus on discussions based exclusively on the Cypriot forms and patterns, as presented and justified by the evidence in Makhairas, and as witnessed by history which, for many centuries, has singled out Cypriot from the rest of the dialects and the Greek language itself. So, alternative views, criticism and discussion of same mechanisms of change recorded within the broader Greek language have been more or less avoided. The exposition of the MC word order patterns is based on my hypotheses that word order, as I understand it, is founded on purposes of communication and that languages with extreme flexibility of order, such as Medieval Cypriot, may adopt patterns that display rigidity of order in a number of their elements. It is within these areas of rigidity that new mechanisms of change may be detected. I also hypothesised that the same syntactic changes within languages of the same branch may be merely coincidental, and that Greek or forms of Greek may well adopt foreign elements, only (but not exclusively) if these acquire the Greek endings, or if they appear as independent affixes, as is the case with the post-medieval referential Cypriot marker �mish� which is from Turkish. Acquiring particular elements from other languages does not mean acquiring their order. However, acquiring patterns that are similar to Greek from a borrowing language which has the same patterns does not exclude syntactic borrowing. Since Modern Cypriot is V-initial, I presumed that this might have also been its order in the Middle Ages. I judge that major mechanisms of syntactic change of the same period may have been triggered by factors internal to Cypriot rather than by the more general, universal mechanisms of change. Moreover, I speculated that MC was a far more marginalised language in the Middle Ages than what history and literature have taught us. Its creative dynamism and potentiality to �juggle� between words and patterns has been its greater forte. Cypriot has not been studied as a dialect, in this work. I avoided having only a partial or a shadowed understanding of its word order patterns. Exhaustive descriptions that show its particularities in the process of completion appear with both rigidity (in some elements) and flexibility of order, and most importantly, they exhibit a long-life endurance. I have also been concerned with forms and /or patterns of Greek such as the future and other periphrastic tenses, although they are already known and have been analysed at length in Greek linguistic studies. I concentrate here on some of these from a Cypriot perspective. Cypriot has never been classified as Balkan Greek or mainland Greek. Following this study, it will be clarified further that any attempt to fit MC into a framework defined along these categorisations will be successful only in some areas of the general Greek syntax. In fact, Cypriot opens the way for a further understanding of Greek syntax with its (almost) boundless flexibility; it is through MC and the unique data of Makhairas that the study of the Greek syntax is being enriched. Areas of fine-grained classificatory criteria result in connecting some MC syntactic traits to those of Greek and accrediting to the language its own word order singularities in what can be righteously called here the Cypriot syntax. Additionally, the study aims to open new areas of investigation on diachronic syntactic issues and to initiate new and revealing answers concerning configurational syntax. To determine the syntactic traits of MC a meticulous work of counting was needed. The counting of the order of the main constituents from both the more general narrative patterns of the Chronicle as well as of those passages thought to be more immediate to the author�s living experience(s) was done manually. The primarily and more difficult task of considering, following and explaining pragmatic word order patterns in the Chronicle has been the stepping stone of this research. Earlier (and forgotten) stages of Greek, and patterns exclusive to Cypriot, assembled in a unique lexicon and with special Cypriot phrasal verbs, have provided answers to explaining the Cypriot structure. In addition to statistics, areas of language contact have also been explored, both in the morphology and in the syntax. More importantly, the extreme word order freedom of MC that illustrates word order processes based entirely on internal structural changes, aims to contribute to discussions regarding morphology and syntax versus morphosyntax. Chapter 1 provides all the background information of the history and language in Cyprus, prior to the Middle Ages. Chapter 2 deals with the description of the data and the methodology used to assess them. Chapter 3 exhibits the MC verbal forms, both finite and non-finite; it examines non-finites more closely, inasmuch as they play an important role in the change of the order of major constituents and uncover and explain the role of V-initial structures. Chapter 4 is the core chapter of this work. It displays Cypriot particularities of word order, reveals data concerned with the word order of the major constituents within the clause and unfolds explanatory accounts of them; lastly, it classifies MC as a V-initial language. Chapter 5 summarises conclusions, adds a further note on the Cypriot morphosyntactic traits while placing the results into the contemporary scholarship on VSO languages, also suggesting additional research areas into the MC patterns. The examples from Makhairas have been written in the monotonic system, where only one accent has been used; other special symbols have been eliminated or modified in the interest of making the text readable in the absence of the right font. However, Ancient Greek words appear with their appropriate accents. Abbreviation C indicates structures or words that remained unchanged in Cypriot over a long period of time, and G means a form or word accepted in both their written and spoken forms over a long period of time in Greek. A morphemic analysis of each form of the glosses has not always been given. I limited myself to glossing some elements only, for the better understanding of some examples.

Στη μεταπτυχιακή αυτή διατριβή αμφισβητώ την ευρέως διαδεδομένη άποψη ότι η διάκριση ανάμεσα σε ενδοκεντρικά και εξωκεντρικά σύνθετα είναι κατ’ εξοχήν σημασιολογική (βλ. μεταξύ άλλων Bauer 2009: 350). Αντιθέτως, ακολουθώντας τους Ralli & Andreou (2010), προτείνω ότι η συγκεκριμένη διάκριση είναι δομική. Για να υποστηρίξω τους ισχυρισμούς και τις προτάσεις μου, παραθέτω δεδομένα από την Κυπριακή και την Κοινή Νέα Ελληνική. Καταρχάς, υποστηρίζω ότι η εξωκεντρικότητα δεν είναι ένα περιθωριακό φαινόμενο στη σύνθεση, δεδομένου ότι διάφορες γλωσσικές ποικιλίες, μεταξύ των οποίων και η Κυπριακή, επιδεικνύουν υψηλό βαθμό παραγωγικότητας εξωκεντρικών συνθέτων. Ακολούθως, υποστηρίζω ότι η σημασιολογία δεν μπορεί να μας οδηγήσει σε ασφαλή συμπεράσματα ούτε για την κατηγοριοποίηση των εξωκεντρικών, αλλά ούτε και για τη διάκριση ανάμεσα σε ενδοκεντρικότητα και εξωκεντρικότητα. Επιπρόσθετα, σε αντίθεση με τους Scalise et al. (2009), δείχνω ότι τα μορφολογικά χαρακτηριστικά, όπως για παράδειγμα το γένος και η κλιτική τάξη, δεν μπορούν να ορίσουν την εξωκεντρικότητα. Καταληκτικά, προτείνω ότι η ενδοκεντρικότητα και η εξωκεντρικότητα είναι επιφαινόμενα της σειράς εφαρμογής των διαδικασιών σχηματισμού λέξεων, σύνθεσης και παραγωγής. Με βάση την πρόταση αυτή, ένα σύνθετο είναι εξωκεντρικό όταν η σύνθεση και η παραγωγή οι οποίες συνεμφανίζονται (co-occur) σε ένα μορφολογικά πολύπλοκο πρωτογενή σχηματισμό, αλληλεπιδρούν με τέτοιο τρόπο ώστε η παραγωγή να έπεται της σύνθεσης. Αντίθετα, ένα σύνθετο είναι ενδοκεντρικό όταν περιλαμβάνει μόνο σύνθεση ή όταν συμπεριλαμβάνει παραγωγή και σύνθεση με αυτή τη σειρά. / In this MA thesis, I challenge the widely accepted view that the distinction between endocentric and exocentric compounds is fundamentally semantic (see, among others, Bauer 2009: 350). Following Ralli & Andreou (2010), I propose, instead, that this is a structural distinction. I illustrate my claims and proposals by using data from Greek and Cypriot Greek. First, I challenge the widespread view that exocentric compounds do not belong to the productive word-formation mechanism given that exocentricity in a number of linguistic varieties, Cypriot included, is very productive. In addition, I show that semantics cannot be a safe criterion for either classifying exocentric compounds into various categories or distinguishing between endocentric and exocentric compounds. Moreover, I argue that morphological features, e.g. gender and inflection class, cannot define exocentricity, in the way Scalise et al. (2009) have proposed. Finally, I suggest that exocentricity might be an epiphenomenon of the order of application of the two word-formation processes, according to which, when compounding and derivation co-occur within the same morphologically-complex item, compounding precedes derivation. In contrast, a structure is endocentric, if it contains only compounding, or involves derivation and compounding, in this particular order.