It is with great pleasure that we join in saluting BMGS on the fortieth anniversary of its inception: it is a stalwart, valued presence in the international academic arena. Long may it continue!
A paper we published together in BMGS in 1979Footnote 1 marks our shared fascination with the literature in Greek which arose after the Fourth Crusade, particularly that which broke with Byzantine tradition and allowed the use of elements of the spoken language previously treated as inappropriate for writing. Beside these ‘popular’ forms, the poems include a surprisingly similar number of ‘learned’ details. They are also more repetitious than is normally bearable in writing, and their texts vary substantially from one manuscript to the other. Both of us were trained as classicists at Cambridge, neither having much education in Modern Greek. The ‘mixed’ and unstable language of these texts probably impressed us more than it did others more used to the morphological variation found in medieval texts using spoken registers of Greek.
But other reasons gave us some confidence in approaching the material. EMJ had written an Oxford thesis on two of its traditional motifs, wondering whether the tradition was purely Greek or due to western literary influences in the thirteenth or fourteenth centuries. She then rashly started a project to edit the largest (and virtually unedited) work in this genre, the War of Troy,Footnote 2 a fairly accurate translation into Greek of the Roman de Troie of Benoit de Ste Maure. Thanks to the good offices of Robert Browning, she teamed up with Manolis Papathomopoulos of Ioannina University, at that point a refugee in London from the Greek Colonels. Methodological discussions around this project showed that there were still many open questions to be solved in editing and interpreting these texts.
MJJ too was briefly involved in the project, and family discussions centred on a shared undergraduate experience. They had both attended a lecture course on Homer by Geoffrey Kirk, based on the proofs of his forthcoming Songs of Homer.Footnote 3 This was heady stuff: Homer's hypnotically grandiloquent repetitions had a function! They were metrically convenient, they were the lines’ building blocks – an intriguing new world opened up involving poets whose poems were recomposed as they were sung before an audience. The Homeric language was an amalgam of at least three dialects, usually chronologically distinct, but in Homer often combined in the same lines; these, like the formulas, were useful for flexible oral composition. For Kirk, as for most Homerists of the time, the living oral traditions studied were from various areas of Yugolsavia in the 1930s. Kirk had begun to face the problem of the transfer of fluid recomposition to the massive textuality of the Homeric poems, but interest and excitement were all at the oral end of the theory. We recognized in narrative poems like the Chronicle of the Morea and romances like the War of Troy many symptoms used by Kirk to diagnose the oral background of Homer.
These discussions in the Jeffreys family became the subject of MJJ's London PhD thesis, written under the guidance of Robert Browning.Footnote 4 By this time, oral-formulaic studies had spread like wildfire from Homer through the European Middle Ages, particularly in Old French, Old and Middle English, Spanish and German, with a bibliography running into many hundreds of items.Footnote 5 Most masterpieces of medieval literature were judged to be affected to a greater or lesser extent. Basic research methodology was similar in each language: a definition of formula was established, depending on repetition within certain linguistic and metrical conditions, and a poem or part of a poem was tested statistically for its ‘formulaic density’.Footnote 6 The functional definition possible for Homeric formulas through the complexities of the hexameter was not found in any medieval tradition. Statistics were compared from language to language, and definitions carefully checked to maximize the validity of the comparisons. For comparative purposes it was useful to find other parallel poems with very few repetitions.
However, this straightforward methodology broke down when it came to analysing its results. At first, only two states were recognized, the oral and the written. A. B. Lord insisted that statistics of formulaic density could prove whether or not a poem was orally composed. Above a certain level, composition must have occurred by an oral poet before an audience. Various mechanisms were proposed to suggest how oral creation became text and survived. The most common was the ‘oral dictated text’: an oral poet preserved his poems in writing by singing to a scribe slowly, or else learning letters and dictating to himself.Footnote 7 Below a certain formulaic density the poem was pronounced a literary imitation of oral style. Transitional forms between these extremes were declared impossible.Footnote 8 This binary opposition broke down later, but not before vitiating the results of many early studies. In the last two or three decades it has been established that all texts from the past showing oral influence through formulas will also have been affected by writing, to a greater or lesser extent, in the process of preservation.Footnote 9 Homer's style may suggest an illiterate improviser, but the monumentality of the Iliad and the Odyssey, in organization as well as sheer bulk, in an environment where text was probably rare, makes it hard to deny that a proficient writer was somehow involved in the composition. The transitional text that Lord thought impossible is now taken to be almost a universal rule.Footnote 10
In all the linguistic traditions where formulas were identified, some scholars denied their existence; these objections were generally countered by precise definitions and careful statistics. Other scholars were reluctant to accept that Homer, say, was illiterate, a view that has been vindicated, as mentioned above. There was also an insidious assumption that formulaic influence in a text meant serious downgrading of its poetic quality. Poets would not use formulas, some assumed, if they were good enough to compose without them. This is strange: one would expect that placing a text in the same category as Homer, Beowulf, the Chanson de Roland, the Nibelungenlied and the Cid was a positive evaluation. Of course there were bad formulaic poets as well as good ones. But no poem confirmed as formulaic, by statistical rules sympathetically derived from its genre, has been shown to have been written from scratch, without the influence of oral poetry somewhere in its background. The oral influence might be very distant, involving a huge remove in time and nature from orally composed original to formulaic descendant. Quintus Smyrnaeus, for example, followed the style of Homer and other strands of the epic cycle centuries after they were written down, and even longer after the generally accepted dates of Homer's lifetime and the forging of his style.
Our writing on these issues in Medieval Greek (MJJ's thesis, an article in Dumbarton Oaks PapersFootnote 11 and the 1979 BMGS article under discussion) came late in the use of the theory in the medieval period, and so had benefited from several stages in the international debate. MJJ, when invited by Lord to speak at a Harvard colloquium, experienced first-hand Lord's firm support for the oral/literary dichotomy. Our results were clearly stated: the Chronicle of the Morea was adjudged, by statistical arguments, to be an orally-influenced poem, and in the 1979 article this conclusion was extended to the War of Troy. As we read these studies now, they seem cautious in claiming conclusions which by then in most western European language traditions would have seemed self-evident. We had fought through initial assumptions of binary opposition (primary oral vs. oral imitation), and reached the later belief in the transitional status of all such material. It was in this framework we tried to place the two written poems under analysis. The Chronicle of the Morea seemed to be the written remains of a kind of oral epic on the foundation of the Principate of the Morea, but so far morphed into literate form that it refers more often to legal struggles than the knightly exploits which surface in passages suggesting earlier oral strata of the poem. Even other literary paraphernalia like cross-references took on a formulaic structure. Study by MJJ of the ‘politikos stichos’, the oral metre of Modern Greece, deepened understanding of the cultural context of this oral poetry.Footnote 12
The 1979 joint article in BMGS was personally important for MJJ, a first attempt to survive as a researcher in a new and difficult environment.Footnote 13 EMJ's research chipped away at the huge task of editing the War of Troy with constant methodological problems and difficulties over linguistic minutiae. Later she would organize in 1981 (with the American musicologist Leo Treitler) a large international conference at the Humanities Research Centre in Canberra, entitled ‘Transmission in Oral and Written Traditions’, with a stellar guest-list.Footnote 14 While the 1979 BMGS article showed the narrow work of the editorial process, the broader context of our methodology appeared in the Canberra conference.
The War of Troy posed the problems of medieval orality in a very challenging form. Its five near-complete manuscripts and other fragments for a poem of more than 14,000 lines make up a very large proportion of surviving manuscript evidence for poetry in the non-learned registers of Greek from this period. The War of Troy is a translation of a widely-disseminated French original, itself lightly touched by residual orality and showing a number of repetitions resembling formulas which were not statistically significant. The main characters are heroes from both sides of the Trojan War; but the forms of their names are often severely distorted by passage via Late Antique Greek and Latin into Old French, now rendered back into Greek by a translator largely ignorant of Homer. These often bizarre Greek translated names are repeated with a statistical frequency suggesting strong oral influence: they operate in some way as formulas. The result sometimes suggests a parody of Homer. By contrast, repeated phrases which (statistics suggest) were formulas of action have a more convincing form and were more often found in other similar Greek poems. If, as the theory suggests, the existence of statistically proven formulas demonstrates the existence of oral-formulaic material somewhere in the background, what may be suggested about that material in this case? Might the limited formulas of the French original play a role? More likely, were the name formulas invented by the Greek translator because repetitions were a necessary part of the poetic discourse of the genre, while suitable verbal formulas, as suggested, were already available in other poems? Having put the BMGS article in context, we will not repeat its arguments.
*
The primary issue in the field of Late Byzantine/Early Modern Greek verse in 1979 was the contrasting pressure of ‘oben’ vs. ‘unten’, as in H.-G. Beck's plenary paper at the Byzantine Congress in Bucharest (1971).Footnote 15 Did the new impetus to use the vernacular in literature come from elite and educated strata of Byzantine society, or from attempts by the underprivileged and uneducated to write in their own language? Was this really a peoples’ movement? Nearly fifty years later one may suggest that insufficient allowance was made for the dominance of the elite over the means of recording and preserving text, which inevitably prejudiced evidence in their favour. At other levels there was still a tendency to castigate scribes (and writers) of the texts for appalling orthography, copying, ignorance and general incompetence, particularly over suspicions that a ‘barbarous’ form was not part of the ‘true’ progression from Ancient to Modern Greek. This mind-set derived from the comments of Byzantine grammarians, often reinforced by a traditional education in the classics. The situation was polarized by attempts to read the violent antagonisms of Greece's modern language question back into the different circumstances of the medieval period.Footnote 16 We hoped that our articles on oral formulas would add one wider perspective via comparative, Europe-wide analysis of the period's poetry.
We were disappointed. Few commentators seemed to have read any of the research we adduced on other oral-formulaic traditions, or even our own words. It was even assumed that we claimed all medieval poetry as orally composed, a view rejected in everything we wrote. Some repeated arguments that had been laughed to scorn decades earlier in other linguistic situations. One reviewer, for example, asked if the words ‘Marks and Spencer’, repeated often in a poem, qualified it as orally influenced. The answer, of course, is ‘yes’, provided that such phrases filled at least 25–30% of the whole poem, the statistical limit we proposed in the genre for secure evidence of oral influence.
Other reactions were dominated by the feeling that no poet would use a repetitious language if able to compose without it — a thought rejected above because of the list of masterpieces where formulas have been found. This feeling operated in different ways. The many articles of Giuseppe Spadaro, who knew this genre of poetry well, described its poets as only able to compensate for their lack of talent by repeating their own lines and plagiarizing each other's work.Footnote 17 When Roderick Beaton later wrote a study of the whole genre of Byzantine romance, he asked many useful questions not previously posed. But on this issue, he felt that we seriously undervalued the poets by claiming that their style (especially formulas and mixed language) was not a personal creation but adopted as part of the genre in which they worked. He proposed an alternative, improved approach, which puzzled us. The main difference from our own line seems to be his determination to give the poets full credit for the conscious creation of their styles.Footnote 18 We find his assertion that a large number of poets over some centuries consciously each created a very similar style for their works rather unlikely. The idea of conscious creation seems not to tally with the imitation which Beaton uses elsewhere to explain the similarity of their stylistic choices.Footnote 19 We prefer to believe that — like the best formulaic poets all over Europe — they chose to put in writing a style already widely used at an oral level.
The only support for our proposals came within the amorphous body known as Neograeca Medii Aevi, brain-child of Hans Eideneier and vigorously supported in its early stages by the late Nikos Panagiotakis. Eideneier favoured a different ‘flavour’ of oral analysis from ours, which started from different premises, but ended with similar conclusions. We became energetic supporters of Neograeca Medii Aevi, where we tried out some experimental linguistic extensions to our basic theories, not always meeting agreement.Footnote 20
The most irritating reaction came from a colloquium, held in Oxford in 1989, by SCOMGIU, the Standing Committee on Modern Greek in Universities,Footnote 21 and partially published in BMGS.Footnote 22 The topic was ‘Orality in medieval and modern Greek poetry’. The articles used a variety of subjects and approaches: among those relevant to this paper, the offerings of Peter Mackridge on metre and David Holton on oral residues in Cretan literature are particularly useful. An article by Roderick Beaton is subject to the same points made above about his book.Footnote 23 The keynote paper was given by Ruth Finnegan, entitled ‘What is orality — if anything?’ Despite the discouraging title, its first half praises positive results of the Parry-Lord theories which had extended the canon of literature to include fleeting and marginal items, whilst unlocking access to orally-based forms lying behind much written literature. Finnegan's second half offers balancing negative points — assumptions that the Parry-Lord theories covered all orally performed material everywhere, careless use of the noun ‘orality’ to imply that all oral material belongs to one easily defined category, and the establishment of ‘orality’ as a stage of development preceding ‘literacy’ since this practice, even when praising those composing and performing orally, condemned them as inferior to the writers who followed. The chief figures included in this last criticism were Eric Havelock, Walter Ong and Jack Goody.Footnote 24
The first half of the 1990s brought several studies which, in effect if not intention, pitted the strongly formulaic genre we examined against the dominant oral tradition of Modern Greece, the dimotika tragoudia. Beaton used computer techniques to investigate formulas in manuscript E of Digenis Akritis, publishing a key-word concordance to that text, demonstrating its limited repetitions and connection to the tragoudia.Footnote 25 Grigoris Sifakis questioned the formulaic status of the repetitions we found in the Chronicle of the Morea and the War of Troy.Footnote 26 Bernard Fenik charted the importance of formulas of the Parry-Lord type in Digenis Ms E.Footnote 27 We had early tested our methods on the manuscripts of Digenis, finding that repetitions were statistically insignificant in E and negligible in G. We began to suspect that within a broad framework of formulaic poetry in Medieval Greek, there were two major patterns (at least): an ‘epic’ form on which we were concentrating, and the ballad tradition of tragoudia which would come to dominate the field. This doublet of oral traditions should be no surprise: most of the languages of Europe have something similar.Footnote 28 We return to this later.
With the exception of the reactions listed here, our proposals were ignored. In our disappointment, we consciously turned our backs on the specifics of this controversy, where we seemed to be talking to the deaf. After a couple of attempts by MJJ to set frameworks and rules for discussing this issue, should anyone be interested,Footnote 29 we focussed on other subjects, including Ioannes Malalas and the twelfth-century texts associated with the sevastokratorissa Eirene.
Having invested much time in introducing oral formulaic research into Medieval Greek studies, we continued to observe developments in the field, even though we were no longer contributing to it. In fact, developments were few. Lines of research from an oral perspective had appeared in many poetic genres in different languages, and initially seemed very promising. But, after the first dramatic changes of perspective (which remained valid), the methodology proved somewhat blocked. We remember two personal surprises in our area, showing that some scholars had listened: first, that Ulrich Moennig hunted for formulas in his edition of the late romance Alexander and Semiramis, following the Jeffreys’ flavour of oral research as well as that of Eideneier.Footnote 30 A little later, listening to the sessions of the Neograeca Medii Aevi conference in Ioannina in 2005,Footnote 31 it struck us that the oral background to the Byzantine vernacular material was being assumed by most speakers, even in cases never formally discussed. Private exchanges of views were equally positive.
We were attracted back to research in the subject by three major impulses, two operating mainly on EMJ. The first came from Teresa Shawcross, an excellent and challenging student who chose as the basis for two Oxford theses the Chronicle of the Morea, in all language versions.Footnote 32 Shawcross approached the subject with much greater sophistication than that of MJJ's thesis in the early 1970s, especially over the literary context of the French and other non-Greek versions. In particular, she put flesh on the bones of the conclusion we reached about the nature of the Chronicle, that it derived from a foundation epic for the Principality of the Morea. We had approached the Chronicle as a text from the second half of the fourteenth century (the date of the earliest manuscript), and had not rethought old assumptions of conflict between Greek- and French-speaking elements of the crusader society where it arose. Shawcross demonstrated that this was a Moreot work, showing an almost proto-nationalistic discourse, supporting both French and Greek Moreots against external enemies. She dated the Chronicle in the first half of the fourteenth century, but suggested that any terminus post quem should be fixed well back in the thirteenth century. This work had probably been subjected to rolling updates, reflecting historical events that affected the principality and its mixed populations. Irrespective of the different languages used, the project was an attempt by French rulers to enthuse the entire population, French and Greek, to support the progress of the Morea as a whole.
The second spur to resume research on oral-formulaic poetry was an invitation received by EMJ to a Princeton conference in 2009 on ‘Renaissance encounters, Greek East and Latin West’.Footnote 33 In searching for a subject for her communication in the background to the War of Troy, she discovered that a new element had been added to the history of the ideological reception of the Roman de Troie, the poem's French original. It had long been clear that one motive for its composition was to write the Anglo-Norman Plantagenet dynasty of Henry II of England into the legendary Trojan descent of European rulers.Footnote 34 Recent research has established that this experiment was repeated in the 1260s onwards, at the courts of Louis IX of France and his brother Charles of Anjou in Naples, in relation to the Capetian dynasty. The evidence lies in illustrated manuscripts of Benoit's poem, first made in Capetian court circles, which stress a Trojan genealogy.Footnote 35 Trojan material similar to the Roman de Troie was also used in the first sections of the Grandes chroniques de France, which also attracted illustrations.Footnote 36 This new research gave EMJ answers to two unsolved problems connected with the War of Troy: firstly, why the Greek translator had chosen Benoit's old-fashioned verse text rather than one of the prose rewritings in circulation from the mid-thirteenth century;Footnote 37 secondly, why most manuscripts of the Greek War of Troy had unfilled spaces for a cycle of illustrations, which seems to have gone back to the original translation. This set her searching for further arguments to redate the translation of the War of Troy, which in the 1996 edition was placed in the mid-fourteenth century, for no very good reason: it was around the time when similar romances were written, and in Catherine de Valois there was a rich patron available who might have invested the money needed for this large translation project.
There was a much more likely sponsor in the 1270s. Leonardo da Veroli was Charles of Anjou's chancellor in the Morea, having arranged the marriage in 1267 by which William II Villehardouin transferred the principality to the Angevins. He was a close confidant of both Charles and William, and married into the prominent Moreot Toucy family, known for their command of Greek. He was a cultivated man, who sponsored the copying of manuscripts: his will lists many books, one in Greek.Footnote 38 Though the evidence is circumstantial, it is very likely that the movement to link the Capetian dynasty to the Trojans was understood in Angevin Naples. If this came to Leonardo's notice, he would surely have realized the interesting light it threw on his own position as representative of a ruler of Trojan descent in a land of Greek-speakers.Footnote 39 Nowhere in the world was more appropriate than the Morea for the dynastic politics represented by Benoit's poem: the remaining need was to extend dynastic propaganda to the Morea's Greek inhabitants.
A translation was needed: but in what form? Moreot society was cultured,Footnote 40 but its cultural languages were Latin, Old French and Italian. Some people of Greek background will have known at least one of these languages well, but too few to bridge the linguistic divide. Many of the ruling westerners (like the family of Leonardo's wife) would have known some Greek, 170 years after the conquest. But this will have been more the Greek of daily conversation then the learned or semi-learned written language, which will also have been unattractive or even inaccessible to most Greek Moreots. What was required was a means of oral communication, which could be read aloud to Greek or mixed audiences. Some westerners could understand enough, while the Greeks, if the medium was well chosen, might be informed and entertained in a very attractive way. We believe that this was ‘The traditional style of early demotic Greek verse’ discussed in our BMGS article. Leonardo and his officials probably invested considerable sums in sponsoring the huge translation and several copies, leading to the many manuscripts provable from later centuries. Thus the War of Troy would have been produced some time after the Treaty of Viterbo of 1267, which brought the Morea under Angevin authority, and Leonardo's death in 1281.
This redating of the War of Troy has been accepted by all those expressing an opinion — provisional acceptance, however, for those able to test the new date without extensive research are few. A second, more tentative, proposal, made by EMJ at the same time, has been received with caution. The first step is to consider together the War of Troy and an early form of the Chronicle of the Morea (as it might have been in the 1270s and 1280s), two texts with similar stylistic features and comparable ideological goals: the cementing of unity between western rulers and Greek subjects in a mixed community — arguably the Morea in both cases. The second step follows the War of Troy, with or without the Chronicle, to Constantinople where they would have posed a literary challenge. The War of Troy would have seemed to highjack Greek cultural history (Homer) while the Chronicle rewrote recent Greek history (glorification of the Fourth Crusade and the conquest of the Morea), all expressed in an appropriated medium (an attractive written Greek style referencing oral poetry), making effective Frankish propaganda. Furthermore the War of Troy offered attractive romantic sub-plots – Jason and Medea, Paris and Helen, Briseis with Troilus and Diomedes, Achilles and Polyxene. It is tempting to see in this combination of elements the impetus for the Palaeologan romances in the politikos stichos, with their formulas, mixed language close to the vernacular and anonymity – a Byzantine riposte to intrusive Frankish texts.Footnote 41
MJJ has not returned to searching for formulas. He has been inspired by another Jeffreys family project, the edition of the huge twelfth-century collection of poems attributed to ‘Manganeios Prodromos’, the third of our three impulses to revisit this area of study. Manganeios was a colleague of the well-known litterateur Theodore Prodromos. But his work is much less scholarly than Theodore's; his expertise is in the rhythmical patterns of the politikos stichos, reinforced by alliteration, rhyme, and syntactical parallelism. We have the impression that his work will have been attractive lower down the educational scale than any other surviving twelfth-century text which demoticists would class as ‘learned’. Manganeios claims to be influenced by traditional Byzantine rhetoric. But the politikos stichos had appeared too late to have a place in the rhetorical handbooks. We find the templates of Modern Greek folk-song, as revealed in the structural study of Grigoris Sifakis,Footnote 42 more useful than the old Byzantine handbooks in revealing the generative processes behind Manganeios’ poetry. We wish to make this nexus of problems a central feature of our edition of Manganeios.Footnote 43 The challenge specially presents itself in two different forms: what terminology should we use in the edition — that of Byzantine rhetoric or of Sifakis’ work? And how can one discuss the continuity of folk-song patterns over several centuries, when evidence is largely found at the beginning and the end, and surviving texts from the period between are so few? The problem has been well posed by Sifakis himself.Footnote 44 Perhaps our analysis of Manganeios will help in the quest he has described. It may be that such research will help clarify the link between the two kinds of Greek formulaic poetry, epic and ballad, which seem to have coexisted for much of the last millennium.
We must finally mention Jorie Soltic's recent studies,Footnote 45 not yet properly digested, but offering interesting ways forward. She demonstrates, using more subtle linguistic arguments than are usually found in this area, that the politikos stichos poetry of the thirteenth and fourteenth centuries employs a real language, fit for use by linguistic historians just like the limited prose texts of the time. The evidence derives from a corpus of seven romances (including the War of Troy) and the Chronicle of the Morea. The investigation uses eclectic linguistic methodologies from the spectrum known as Information Structure, based on examination of oral discourse in modern languages (see pp. 39–64). She starts from the premiss that some features of the language of her corpus, notably traditional formulas, have been linked with oral language.Footnote 46 Usage of object clitic pronouns also seems to show close parallels with the likely development of daily speech (pp. 91–124). This is one of several examples of the extreme expressiveness and flexibility of the politikos stichos, and the ways in which the mixed language fits its demands (pp. 81–124). Her main analysis shows that several features of the poems’ style reveal the hasty processing strategies of speech rather than the more considered rhythms of writing. Information is presented in paratactically connected metrical chunks rather than sentences (pp. 125–45), the relations of topic and focus (which we do not understand) are those of oral speech (pp. 146–90), and above all a huge number of different words and phrases are bleached of their regular meanings and have become mere discourse markers (pp. 191–298). Several other syntactical awkwardnesses may be resolved as signs of oral processing (pp. 36–7, 304–5). The evidence for all this is extensive and well presented.
Soltic's conclusions lead to a corollary (pp. 303–4), including the following:
Hence, the adoption of an oral-formulaic style goes beyond mere stylistics, i.e. beyond the ‘simple’ insertion of formulas: the whole underlying conception of the LMG πολιτικὸς στίχος poetry is based on spoken discourse. Reading the LMG πολιτικὸς στίχος poetry with this in mind is truly revealing. This approach does justice to its language, which is orally conceived (rather than barbarically!), and frees us from the preconceptions of written discourse.
The following questions suggest themselves:
1. Greek daily speech at this time did not use politikos stichos with statistically significant formulas. Where then was the ‘real language’ to which Soltic refers?
2. Some features she lists (especially the ‘bleaching’ of discourse markers) demand long periods of usage. Did bleaching happen in daily speech, being then applied as a stylistic feature to the poetry? Or did it somehow occur within the poetry itself?
3. Nearly all the mixed morphological features of the poems were current parts of daily speech at this time somewhere in the Greek-speaking world. But how did they come to be used so effectively, in so similar a way in so many poems?
4. Is it possible that the real language to which Soltic's work points was developed by one class of Greek oral poets (those producing narrative), over many decades, for effective composition and communication with their audiences? Might it then have been transposed into a written style, demanding the use of many features of the Byzantine Volksliteratur, including Soltic's discoveries, to be read on other occasions and with other purposes, but with equal effectiveness, to the same audiences?
5. Will these new linguistic tools, together with the impending Holton-Horrocks Cambridge Grammar of Medieval Greek, remain effective for longer than the Parry-Lord theories in working out the complexities of Late Byzantine language? Or will they even, as Soltic speculates at the end of the thesis (p. 310), show a very long tradition of oral poetry?
