In this contribution, I offer a summary of my 2015 Ph.D. dissertation from the University of Ghent on the language and metre of Late Medieval Greek πολιτικὸς στίχος poetry as they pertain to information structure.
My dissertation deals with the Late Medieval Greek (LMG) πολιτικὸς στίχος poetry. The language of this type of poetry, whose most representative genre is the romance, is not easily grasped. In the first place, it is renowned as our chief witness of vernacular literature from the 12th to 15th centuries (Beaton 1996:99). However, it does not at all represent a pure vernacular language, but a so-called mixed language, for the poets now and then include ancient “dead” linguistic items (Toufexis 2008:204). Moreover, it adopts an oral-formulaic style (E. Jeffreys 2013). It is presumably this rather curious idiom, together with the prejudice that the metrical constraints of the πολιτικὸς στίχος impose a straitjacket on the language, that prevented linguists from thoroughly investigating the LMG πολιτικὸς στίχος poetry as a whole. Indeed, linguistic studies dealing with the LMG πολιτικὸς στίχος poetry all have a rather specific profile: they typically treat one grammatical aspect or are diachronically oriented. While such “fragmented” studies are of course unquestionably valid, what is lacking so far is thus a more comprehensive approach. It was my aim to change this.
2. The Theory
My dissertation attempts to offer and develop a theoretical framework in which the LMG πολιτικὸς στίχος poetry can as a whole be approached. More concretely, I argue for the benefits of modern linguistic theory, namely the framework of Information Structure (IS). This framework, which pays a lot of importance to pragmatics, has been founded on the basis of contemporary spoken languages. Assuming that there are important parallels (similar cognitive principles of producing and processing information) between orally conceived poetry on the one hand and naturally spoken language on the other, I consider this framework ideal to deal with the poetry’s peculiar features (cf. Bakker 1997). As a result, my analysis, which is thus well embedded in current linguistic theory, constitutes the first attempt to develop a modern linguistic approach with regard to the LMG πολιτικὸς στίχος poetry.
3. The Analysis
More concretely, I have applied three concepts from IS to the LMG πολιτικὸς στίχος poetry: the intonation/information unit, the topic/focus pair and the phenomenon of discourse markers. These three function as leitmotifs throughout my dissertation.
3.1. Intonation/Information Unit
Let me begin with the intonation/information unit (IU), which is the well-established unit of spoken discourse and thus the standard unit of analysis in most IS-based studies (Chafe 1993). An IU is prototypically demarcated by a breathing pause and often corresponds to a syntactic unit. Moreover, it tends to stand in a simple paratactic relation with the surrounding IUs. This actually reflects the different cognitive grounds that underlie spoken discourse as compared to written language: a speaker organises his information less densely and thus imparts the information in smaller chunks rather than in complex grammatical sentences.
With regard to the LMG πολιτικὸς στίχος poetry too, the “traditional” notion of sentence does not make much sense and is best abandoned in favour of the IU. More precisely, I argue that the fixed caesura (#) at the eighth syllable implies an IU boundary and is—beside a metrical boundary—a prosodic boundary separating two IUs. This argument is based on a variety of metrical (e.g. avoidance of elision) and syntactico-semantic arguments (e.g. distribution of Wackernagel P2 particles). From this perspective, the two standard half-verses of the πολιτικὸς στίχος are not merely metrical units, but also cognitive units. Moreover, the fact that IUs coincide with the metrical patterns of the verse make clear that the verse rhythms of the πολιτικὸς στίχος are nothing to be wrestled with, but should actually be considered stylisations of natural speech. To illustrate this, I give two examples:
ἔπαθενἐκ τὰς ἀρχὰς # ἐκείνη καὶ ὁ Ἰμπέρης(Imberios & Margarona 848)1
‘how [she] suffered from the beginning, she and Imberios’
καὶ ἀτοί τους
τὸνἐνδύσασιν # ἐκεῖνον καὶ τὴν κόρην(Achilleïs Byzantina 1502)2
‘and they got him dressed, him and the girl’
The lack of grammatical agreement between the verb (ἔπαθεν; singular) and its subject (ἐκείνη καὶ ὁ Ἰμπέρης; plural) in (1) and between the object clitic pronoun (τόν; singular) and its coreferential object (ἐκεῖνον καὶ τὴν κόρην; plural) in (2) becomes perfectly acceptable if we conceive the language of the πολιτικὸς στίχος poetry as a concatenation of IUs. Thus, ἐκείνη καὶ ὁ Ἰμπέρης and ἐκεῖνον καὶ τὴν κόρην should be analysed as (corrective) afterthoughts added in a new IU.
3.2. Topic/Focus Pair
A second application concerns the topic/focus pair. Albeit oft-used (especially in studies on word order), the notion is often not straightforwardly defined. In simplified words, the topic is that “what the utterance is about” and thus usually presents referentially given or active information, while the focus is that part of the utterance that constitutes the most prominent information and thus usually involves referentially new or inactive information (Gundel & Fretheim 2004). Somewhat confusingly, a distinction should be made between a topic in this strict sense (a so-called sentence topic) and a discourse topic, with which one refers to the central idea of a stretch of connected discourse.
It was my aim to firmly establish the concepts of topic and focus as crucial means to describe the language of the LMG πολιτικὸς στίχος poetry. For this purpose, I have conducted two complementary studies on object clitic pronouns (OCPs). In a first study, I offer more objective evidence for the focus hypothesis, namely that ad hoc focalised constituents can attract OCPs into preverbal position (cf. Mackridge 1993). For the first time, this (disputed) pragmatic principle is dealt with from the perspective of the verb: I concentrate on so-called light verbs (ποιώ & δίδω), which are so low in content that their object with almost absolute certainty constitutes the actual focus of the utterance.
In a second study, I go more deeply into the phenomenon of clitic doubling. This construction has been identified in Standard Modern Greek as a special mechanism for marking objects as topics, yet has never been described in detail for LMG. I demonstrate that both its modern typology and its modern function already hold for clitic doubling in the LMG πολιτικὸς στίχος poetry.
Rather than as a superfluous OCP, την should be interpreted as an instance of clitic doubling:
την τὴν γραφήν, # ἐπῆρα, ἐφίλησά την (Livistros & Rodamni 1747)3
‘I took it, the letter, I held (it), I kissed it’
The next example too should be analysed as clitic doubling, more precisely as an instance of hanging topic left dislocation, for ὁ ἐπίσκοπος τῆς Ὤλενας is not just dangling at the beginning of the utterance. Rather, it is coreferential with the OCP τοῦ and serves the pragmatic purpose of introducing a new topic:
ὁ ἐπίσκοπος τῆς Ὤλενας# τέσσαρα φίε τοῦἐδῶκαν (Chronicle of Morea ms H 1957)4
‘the bishop of Olena, they gave him four fees’
3.3. Discourse Markers
A third and final concept is the phenomenon of discourse markers (DMs). Briefly, DMs are expressions that structure the discourse in some way or another rather than contributing to the conceptual content of an utterance (Brinton 1996). In other words, they have developed procedural meaning(s). Especially adverbs and verbs tend to enter this grammaticalisation cline. As such, the class of DMs includes a wide range of multifunctional items of which some operate as conduits between different segments of a text/discourse (textual DMs) and others mainly help to clarify the relationship between speaker and hearer (interpersonal DMs). Typically, DMs are syntactically independent, which is translated into a preference for initial position (in case of the adverbial DMs) or for a parenthetical position (in case of the verbal DMs).
With regard to Ancient Greek, it has been established that the Wackernagel P2 particles can be compared to DMs (Loudová 2014). However, these gradually die out, so that by the time the LMG πολιτικὸς στίχος poetry is written, they are no longer used in contemporary speech. Nevertheless, the ancient P2 particles are still inserted by the πολιτικὸς στίχος poets. Therefore, I give an overview of their functions in the LMG πολιτικὸς στίχος poetry: a stylistic function (adding a classicizing flavour to a text), a rhythmical function (demarcating IUs), and even still a pragmatic one (exerting an—albeit bleached—discourse role). As such, they are far from being mere verse fillers. On the other hand, though, I looked for expressions which could have adopted the various pragmatic functions of the P2 particles—in other words: I searched for newly (or in any case more recently) developed DMs in my corpus and I identified both adverbial and verbal expressions which show DM-like behavior.
As for the adverbial DMs, ἀπ(’)αὔτου, ἐδά(ρτε), ἐνταῦθα, ἐν τούτῳ, λοιπόν, πάλιν and πλήν all seem eligible candidates. I devote a special section to πάλιν, because this is the only expression which—in its DM-like use—does not occur initially, yet prefers to stand after the first word/constituent of the utterance. I have derived that this positional difference (P1 vs. P2) actually reflects an important functional difference: signalling a switch in discourse topic (P1) vs. one in sentence topic (P2) (although a continuum might constitute a better way to present these two types of topics).
In the next example, for instance, πάλιν structures the discourse rather than contributing to the conceptual content (“again”) of the utterance: it denotes a topic switch (from “I” to “you”) and should thus be analysed as a DM:
Ἐγὼ πατήρ σου εὑρίσκομαι, # ἐσὺ δὲ
πάλινυἱός μου (Bellum Troianum 11309)5
‘I am your father, you (are) my son’
As for the verbal DMs, I have concentrated on the first person singular of the reporting verb λέγω (including λαλῶ) and the second person singular epistemic imperative γνώριζε (including ἐγνώριζε, γνώρισε, ἤξευρε and πρόσεχε). Tellingly, in their DM-like use, these are consistently used parenthetically. Whereas the adverbial DMs can all be considered topic switch markers and thus textual DMs (which does not mean that they are mutually exchangeable, though!), the verbal DMs rather belong to the interpersonal ones. Λέγω/λαλῶ, for instance, is shown to have a clear clarification-signalling function, as can be seen in the next example, in which ἐκεινοῦ τοῦ δεσπότη constitutes an apposition to τοῦ Καλοϊωάννη:
ποῦ ἐλάλησαν καὶ εἴπασιν # ὅτι ἦλθαν τὰ φουσσᾶτα
σὲ λαλῶ, # ἐκεινοῦ τοῦ δεσπότη (Chronicle of Morea ms H 1101–1102)
‘who talked and said that the armies were coming’
‘(the armies) of Kalojohn, I mean, that despot’
In conclusion, it was my aim to offer a theoretical framework in which the LMG πολιτικὸς στίχος poetry can be approached and understood as a whole and which can thus perfectly deal with its peculiar features. This led me to the adoption of a theoretical framework developed on the basis of modern spoken languages, i.e. the framework of Information Structure. Such a linguistic analysis of the texts from a modern pragmatic perspective, which is innovative with regard to the LMG πολιτικὸς στίχος poetry, might result in a better understanding of not only its language, but also of its metre and discourse. Hopefully, it even leads to more reading pleasure with regard to a type of poetry whose language had once been considered a “Graeco-barbaric hotchpotch” ….
Agapitos, Panagiotis, Karin Hult, and Ole Smith. 1999. The Byzantine Achilleid. The Naples version. Vienna: Österreichische Akademie der Wissenschaften.
Bakker, Egbert. 1997. Poetry in speech: Orality and Homeric discourse. Ithaca (N.Y.): Cornell University Press.
Beaton, Roderick. 1996. The medieval Greek romance. Cambridge: Cambridge University Press.
Brinton, Laurel. 1996. Pragmatic markers in English: Grammaticalization and discourse functions. Berlin: Mouton de Gruyter.
Chafe, Wallace. 1993. Prosodic and functional units of language. In Jane A. Edwards and Martin D. Lampert (eds.), Talking data. Transcription and coding in discourse research, 33–43. Hillsdale (N.J.): Erlbaum.
Gundel, Jeanette, and Thorstein Fretheim. 2004. Topic and focus. In Laurence Horn and Gregory Ward (eds.), The Handbook of Pragmatics, 175–196. Oxford: Blackwell.
Jeffreys, Elizabeth. 2013, Byzantine romances: Eastern or western? In Marina Brownlee and Dimitri Gondicas (eds.), Renaissance encounters: Greek East and Latin West, 217–234. Leiden: Brill.
Kriaras, Emmanouel. 1955. Βυζαντινὰ Ἱπποτικὰ Μυθιστορήματα. Athens: Aetos.
Lendari, Tina. 2007. Livistros and Rodamne. The Vatican version. Critical edition with introduction, commentary and index-glossary. Athens: Μορφωτικό Ἵδρυμα Ἐθνικῆς Τραπέζης.
Loudová, Katerina. 2014. On the category of particles in Byzantium. In M. Hinterberger (ed.), The Language of Byzantine Learned Literature, 147–169. Brepols: Turnhout.
Lurier, Harold E. 1964. Crusaders as conquerors: The Chronicle of Morea. New York: Columbia University Press.
Mackridge, Peter. 1993. An editorial problem in Medieval Greek texts. The position of the object clitic pronoun in the Escorial Digenes Akrites. In Nikolaos Panayotakis (ed.), Origini della literatura Neogreca, vol. 1, 325–342. Venice: Instituto Ellenico di Studi Bizantini e Postbizantini di Venezia.
Papathomopoulos, Manolis, and Elizabeth Jeffreys. 1996. Ὁ Πόλεμος τῆς Τρωάδος: Κριτικὴ ἔκδοση μὲ εἰσαγωγὴ καὶ πίνακες. Athens: Μορφωτικό Ἵδρυμα Ἐθνικῆς Τραπέζης.
Toufexis, Notis. 2008. Diglossia and register variation in Medieval Greek, Byzantine and Modern Greek Studies 32 (2): 203–217.
Edition of Kriaras (1955).
Edition of Agapitos, Hult & Smith (1999).
Edition of Lendari (2007).
Edition of Lurier (1964).
Edition of Papathomopoulos & E. Jeffreys (1996).
Edition of Kriaras (1955).
Edition of Agapitos, Hult & Smith (1999).
Edition of Lendari (2007).
Edition of Lurier (1964).