New Glances at the Morphosyntax of Greek

The Greek language has a documented history of nearly 3500 years. This number alone can account for the fact that Greek is one of the most extensively and profoundly studied languages. It has been studied from various perspectives, at all levels, planes, and axes with the welcome outcome that today researchers in Greek linguistics have access to a treasure trove of information, raw data, descriptions, and analyses of nearly all of the language’s phenomena.

Greek, with its long history and typological identity—morphologically classified as a ‘morphological’ or inflecting language—offered an ideal terrain for the new theories to be applied and tested. That said, it is important to note that what we today call morphology and syntax have diachronically constituted the core of language studies. Modern morphology coincided with what was called ‘grammar’ in the past, while syntax constituted a separate field of investigation. Interestingly, the border between grammar and syntax was clear-cut. That changed over time, as is demonstrated in this Special Issue (SI).

In a sense, the SI pays tribute to the ancient tradition of complementarily combining morphology and syntax. However, it also acknowledges the intricate relations between syntax and morphology at the linguistic level, which were not acknowledged by past scholars.

Morphosyntax is a relatively new term, and its exact meaning is largely underspecified; it refers to phenomena that are in principle subject to one, the other, or both a morphological or a syntactic approach, and to the transfer of originally syntactic rules to the morphological component. Scholars have even proposed eliminating the latter as a separate grammar component.

Essentially morphosyntax refers to the interface between morphology and syntax, which, as the literature reveals, ends up defining the division of labor between syntax and morphology and testing their boundaries. Traditionally, syntax manipulates the distribution and the combination(s) of words, while morphology deals with the distribution and the combination of morphemes (or morphs)—the minimal meaningful elements of grammar—to form words. Morphology subsumes three major subcomponents, inflection, derivation, and compounding. Of the three, inflection has always been considered the ‘core’ of morphology. In contrast, derivation and, even more, compounding are the two subfields that are more naturally subject to a morphosyntactic or even a syntactic approach. Greek and its varieties offer data for all the issues relating to morphology and syntax, both from a theoretical and an experimental perspective, as shown by the articles in this SI. Some theoretical questions that have generally occupied researchers are: what are the minimal meaningful units inside a word and what is their exact meaning? How can a word be segmented (if at all)? What is the relation between a particular morpheme and the morph, i.e., its realization? How are allomorphs identified and conditioned? What is/can be an alternative to allomorphy? What role (if any) does diachrony play in the process of identifying the morphemes in a word? From an experimental point of view, questions that have been addressed are the following: what is the relation between gender and case in the comprehension of subject and object relative clauses by Greek-speaking children? Do morphosyntactic characteristics, such as case and word order in Greek, a morphologically rich, free word order language, play a role in the real-time processing of complex structures for Greek-speaking children and adults? Does gender marking affect clitic processing in the performance (production and processing) of simultaneous bilingual and monolingual children? What is the role of cliticization in different varieties, such as in Northern and Pontic Greek, where the accusative is employed on the indirect object? This accusative is claimed to be the descendant of the Ancient Greek dative—its loss is documented in Early Medieval Greek (5th–10th c. AD). What are the language contact effects on the morphosyntactic properties of the DP, such as definiteness marking, directionality of adjectives, and gender marking in Vlach Aromanian? In the VP domain, is the parser sensitive to morphological cues, such as (non)-active voice marking?

The collected papers of the present SI address the above questions both theoretically and empirically. The SI includes articles that use innovative approaches, such as fieldwork methodology, experimental and psycholinguistic perspectives, and corpus studies, to further support the interdisciplinary impact at the level of morphosyntax. Moreover, the papers of the SI are interconnected, as the reader will find out, which is something that reflects the range of the research questions delineating the domain of Greek morphosyntax.

Proceeding to the specific contributions, those with a primarily theoretical status that address particular morphosyntactic phenomena are the articles “Zero-Derived Nouns in Greek” by Artemis Alexiadou and Elena Anagnostopoulou, “Compounding in Greek as Phrasal Syntax” by Dimitrios Ntelitheos, “The Morphotactics of the Cypriot Greek Augment” by Natalia Pavlou, and “Preposition Allomorphy in Calabrian Greek (Greko) and Standard Modern Greek and Its Theoretical Implications” by Georg Höhn.

Since morphosyntax cuts across various branches of linguistics, as well as of levels of analysis, Nikos Angelopoulos, Eleftheria Geronikou, and Arhonto Terzi (“Locality and Intervention in the Acquisition of Greek Relative Clauses”), Kalliopi Katsika, Maria Lialiou, and Shanley E.M. Allen (“The Influence of Case and Word Order in Child and Adult Processing of Relative Clauses in Greek”), Georgia Fotiadou (“Processing of Transitivity Alternations and Frequency-Based Accounts in Greek Adult Language”), and Vasiliki Koukoulioti, Stavroula Stavrakaki, Maria Vomva, and Flavia Adani (“Gender Marking and Clitic Pronoun Resolution in Simultaneous Bilingual Children”) all address morphosyntactic phenomena falling under language acquisition/bilingualism. George Höhn’s article “Preposition Allomorphy in Calabrian Greek (Greko) and Standard Modern Greek and Its Theoretical Implications” tackles an interesting phenomenon involving language contact and dialectology. The field of language contact between Northern Greek and Pontic Greek is chosen by Elena Anagnostopoulou, Dionysios Mertyris, and Christina Sevdali in “High and Low Arguments in Northern and Pontic Greek”, while Ianthi Maria Tsimpli, Alexandra Prentza and Maria Kaltsa in “Bidirectional Language Contact Effects at the DP Domain: The Case of Greek and Vlach Aromanian Speakers” investigate nominal phenomena of Modern Greek compared to those of Vlach Aromanian.

Artemis Alexiadou and Elena Anagnostopoulou in “Zero-Derived Nouns in Greek” answer the question of the locus of the gender feature [fem] that the vast majority of deverbal action/activity nouns bear in Greek. They focus on what they call zero-derived (feminine) nouns based on irregular verbs, namely deverbal nouns that do not contain in their stem the morpheme identified as the perfectiveness/aorist marker (-s-) (called by the authors suffixed derived nominal). The authors center their analysis within the framework of distributed morphology (DM) (Halle and Marantz 1993; Halle 1997; Embick 2010), which crucially holds that what matters most are abstract morphemes that syntax operates on, and which are realized post-syntactically. Following the literature spawned by the theoretical premises of DM, even the major syntactic categories (traditional ‘parts of speech’) are the outcome of syntactic processes operating. For instance, nouns result from pre-categorial roots being inserted in the tree and interacting with various morphemes in their vicinity. Artemis Alexiadou and Elena Anagnostopoulou argue that the feature [fem] relates to the (functional) nominal head n, which in the syntactic tree is adjacent to VoiceP and responsible for the formation of deverbal (activity) nouns. Different approaches to forming deverbal nouns constitute an instructive example of the fuzziness of the relation between syntax and morphology. Deverbal abstract nouns (in their diachrony) were also analyzed by Horrocks and Stavrou (2000), although they did not include the issue of gender in their analysis. Within the general ‘separationist’ morphological framework (Beard 1987; Aronoff 1976; 1994, a.o.), they claimed that the emergence of deverbal nouns was based on the ‘aorist’ stem (regularly, but not uniquely, marked by the -s-) and not simply the affix -s-. In the light of data from the history of Greek, Horrocks and Stavrou (2000) argue that “[…] the aorist stem in -s- and abstract deverbals containing -s- are in fact based on a common ‘morphomic’ entity (the -s- stem) […]” (p. 34).

In the same spirit as Alexiadou and Anagnostopoulou, and within the same ‘morphology-as-syntax’ framework (Koopman 2005; Collins and Kayne 2020; Harley 2009), Dimitrios Ntelitheos in “Compounding in Greek as Phrasal Syntax” accounts for the creation of two types of compounds—‘synthetic’ (one word) and ‘phrasal’ (two words) compounds—both related to agentive nominals which are derived from verbs with an internal argument. A major claim of the paper is that roots project their argument structure before they become categorized. The author argues that both types of nominals are derived in the syntax and that no need for a separate morphological component arises. The domain of compound formation is the lower vP (following Alexiadou 2017, a.o.). The differences observed in the distribution of the two types of compounds are attributed by the author to the properties of the verbal domain and the selectional properties of the nominalizing affixes involved in the word formation process.

Natalia Pavlou in “The Morphotactics of the Cypriot Greek Augment” offers a novel analysis concerning the distributional pattern of the Cypriot Greek augment. Her contribution tackles a phenomenon observed in Cypriot Greek, whereby the past tense augment (prefix) e- can appear in two positions within the verb complex; once at the beginning of the verb complex and once (optionally) left adjacent to the verb root. The analysis postulates an initiality constraint, thus motivating post-syntactic changes in the linear order of the morphemes.

The contribution “Preposition Allomorphy in Calabrian Greek (Greko) and Standard Modern Greek and Its Theoretical Implications” by Georg Höhn, has a theoretical orientation (allomorphy vs. portmanteaux morphemes). At the same time, it is experimentally based on data collected from proficient speakers of Greko, an endangered variety of Greek, spoken by very few speakers in southern Calabria. The focus of inquiry is the alternation in the form of two locality prepositions in Greko (asce and an ‘from’, alongside others) as compared with a similar alternation between the two allomorphic forms of the preposition SE (‘in, at, to’) (nam., se and s-) in both Standard Modern Greek and in Greko. The article claims that in both languages the alternation is an instance of conditioned allomorphic sensitivity to the presence of the linearly adjacent definite article and does not involve portmanteaux morphemes.

The articles below apply experimental methods to shed light on issues concerning the morphosyntax of Greek. Nikos Angelopoulos, Eleftheria Geronikou, and Arhonto Terzi in “Locality and Intervention in the Acquisition of Greek Relative Clauses”, following the theory of relativized minimality (Rizzi 1990) in acquisition, focus on the relation between gender and case in the comprehension of subject and object relative clauses by Greek-speaking children. In Greek, like in Italian, and unlike in Hebrew, gender is not an active morphosyntactic feature and behaves like case, not a syntactically active feature in Greek either. The study shows that (i) although both features are found in Greek nominal morphology, ‘neither gender nor case is responsible for intervention effects in the comprehension of relative clauses by Greek-speaking children’, and (ii) that children comprehend subject relative clauses better than object relative clauses.

The paper “The Influence of Case and Word Order in Child and Adult Processing of Relative Clauses in Greek” by Kalliopi Katsika, Maria Lialiou, and Shanley E.M. Allen presents a study of online sentence processing of subject and object relative clauses in Greek-speaking children and adults. The authors’ goal is to examine (i) whether specific morphosyntactic characteristics, such as case and word order in Greek, a morphologically rich, free word order language, ‘play a role in the real-time processing of complex structures’ for both groups and (ii) whether real-time processing mechanisms of Greek-speaking children differ from that of adults. The data collected concerns word order inside the relative clause (canonical vs. scrambled) and the relativizer (either a relative pronoun or complementizer). The authors conclude that children process complex structures similarly to adults and ‘that, for Greek children, word order is a more prominent cue than morphological case’.

Through L1 acquisition experiments addressed to the monolingual Greek-speaking adult population, Georgia Fotiadou in “Processing of Transitivity Alternations and Frequency-Based Accounts in Greek Adult Language” examines the processing of ‘voice (non)-alternating’ anticausative Greek verbs. For the specific verb alternation, the study is based on empirical data of authentic written speech documented in contemporary Greek written corpora (ILSP sample) and on ‘quasi’-oral speech (Web-based sample). Using a self-paced reading and an offline acceptability judgment task, the author studies the online processing and the acceptability of sentences that include the anticausative reading, considering the acceptability ratings compared to frequency data. Georgia Fotiadou argues that the online self-paced reading processing study ‘indicates that the parser is sensitive to morphological cues (non-active voice)’. At the same time, semantic factors such as the animacy of the syntactic subject and the semantic notion of verb class are important at later stages of processing, thus suggesting priority of syntax over semantics.

An experimental study concerning bilinguals is the focus of the paper “Gender Marking and Clitic Pronoun Resolution in Simultaneous Bilingual Children” presented by Vasiliki Koukoulioti, Stavroula Stavrakaki, Maria Vomva, and Flavia Adani, who address the issue of gender marking on clitic production and comprehension by simultaneous bilingual and monolingual children. The discussion compares the verbal intelligence of the monolingual and the bilingual group, suggesting that ‘gender marking affects clitic processing’ and leading to the conclusion that ‘verbal intelligence has an effect only on the performance of the monolingual’ group in the production task. As far as processing is concerned, the relevant data show that bilingual children are slower than monolingual children in comprehension tasks.

In the absence of a written tradition, the two papers “High and Low Arguments in Northern and Pontic Greek” by Elena Anagnostopoulou, Dionysios Mertyris, and Christina Sevdali, and “Bidirectional Language Contact Effects at the DP Domain: The Case of Greek and Vlach Aromanian Speakers” by Ianthi Maria Tsimpli, Alexandra Prentza, and Maria Kaltsa resort to innovative fieldwork to investigate nominal morphosyntactic phenomena in the varieties of Northern and Pontic Greek and Vlach Aromanian, respectively.

Elena Anagnostopoulou, Dionysios Mertyris, and Christina Sevdali analyze the distribution of the accusative as an indirect object, replacing the Ancient Greek dative, in Northern Greek and Pontic Greek, a dialect group spoken in Asia Minor before 1922. The study is based on data collected in Northern Greece from speakers of both varieties. It offers a theoretical analysis of the morphosyntactic features found in the language of the two groups. More specifically, ‘both varieties use the accusative in all the environments where Standard Modern Greek would use the genitive’. The two varieties described differ in the nature of clitics and clitic doubling. The authors provide evidence showing that (i) ‘Northern Greek employs obligatory clitic doubling’, a mechanism absent from Pontic Greek; in Pontic Greek determiner-like clitics of the Standard Modern Greek and Romance variety are unavailable and (ii) as a result of this, Pontic cannot license “high” arguments, e.g., external possessors and ethical dative constructions.

Ianthi Maria Tsimpli, Alexandra Prentza, and Maria Kaltsa in “Bidirectional Language Contact Effects at the DP Domain: The Case of Greek and Vlach Aromanian Speakers”, appeal to language contact effects of spontaneous language production in Vlach Aromanian (a Romance oral language—the only language spoken in a specific language community) produced by Modern Greek bilinguals across three generations who live in Epirus, Greece, with each language group displaying different levels of proficiency. The study focuses on several morphosyntactic properties in the DP domain that are different in Modern Greek and in Vlach Aromanian, e.g., definiteness marking, positioning of adjectives, and gender marking. The data presented show that ‘language contact effects can be bidirectional’. In particular, Modern Greek influences Vlach Aromanian concerning definiteness marking and adjective placement in younger bilingual groups, while Vlach Aromanian affects Modern Greek as far as gender marking in older bilinguals is concerned.

We are grateful to our colleagues and friends who responded eagerly to our call and generously contributed to the SI with an outcome that by far exceeded our expectations. We are also deeply indebted to our reviewers for their time and their illuminative and constructive comments. Finally, we thank the team of Languages/MDPI for inviting us to edit the SI on the morphosyntax of Greek, for trusting us through the long publication process, and for their substantial help and support at every step of it. Special thanks to Janelle Yang for her unreserved editorial support and help.