Project Logo Faces of JanusOU logo Classical Receptions in Drama and Poetry in English
from c.1970 to the Present


The Project
About the project

Project Publications
(including Archived Conference papers)

Specialist Bibliography
Masks Workshop Video

Critical Essays

New Voices
Practitioners' Voices

2010 Conference

A Democratic Turn


2009 Democratic Turn Eseminar

1998-2008 Archived topics

Drama Database
Search the DB

Poetry Database

(pilot v. 1)
An Introduction

Case Study 1:
Michael Longley

Case Study 2:
Eavan Boland and
Olga Broumas

Database Pilot Sample:
Eavan Boland
Olga Broumas
Ted Hughes
Michael Longley

Classical historiography, ideas and material culture
Exhibiting Democracy

Classical Reception Studies Network


© Copyright Notice

January Conference 1996


Michael Psellus on Euripides and George of Pisidia[1]

Mary Whitby, Royal Holloway, University of London

Contents      Abstract

Although the stature and appeal of the dramatist Euripides has remained unquestioned and continued to evoke new responses right up to the present day, George of Pisidia, a panegyrical and religious poet who wrote, like Euripides, in iambic trimeters, but worked in Constantinople in the first half of the seventh century AD, is now almost entirely forgotten. And yet just under a thousand years ago, Michael Psellus (1018-?1078/97), a man who was not altogether unjustified in regarding himself as the greatest intellect of his day[2], chose to discuss whether Euripides or George of Pisidia was the better poet, and it is likely that he rated George higher.[3] A question so startling to modern eyes merits further examination. What context gave rise to it and what were Psellus' criteria? How much did he know of his subjects? To what extent does his view reflect the ways in which contemporaries received the poetry of their classical and late antique predecessors?


I begin with a very brief sketch of Psellus' own career and the wider cultural background, paying particular attention to the work of the Patriarch Photius, who lived two centuries before Psellus.

'Much the most versatile man of his generation',[4] Psellus combined in an energetic life a public career as imperial adviser and civil servant, a period of disgrace in a monastery, the direction of the new philosophical school in the imperial university in Constantinople,[5] a reputation as a popular lecturer of international standing,[6] and an extensive and varied literary output which includes an anecdotal history (the Chronographia) of his own time, orations, letters, poetry and a number of literary essays.[7] Our text has survived in only one damaged copy,[8] and the literate and literary élite represented only a very small fraction of the Byzantine population.[9] Yet Psellus was an influential teacher and ideas were disseminated orally: the views of a man who 'dominated the intellectual and sometimes the political life of Constantinople for nearly fifty years'[10] are likely to have been influential in his lifetime, at least among the educated.

Psellus lived in a period of high renaissance. The revival of classical learning is associated with the appearance of a new, easily-written minuscule script, perhaps about 790,[11] and in the ninth century the recovery and preservation of texts from the classical past became a major occupation among the small scholarly elite.[12] The luxury manuscripts of Arethas of Caesarea, which date from the turn of the ninth and tenth centuries, are only the most spectacular manifestation of this activity.[13] An encyclopaedic interest in the classical past was also manifest in the form of handbooks, syntheses and compilations: best known is the tenth-century Suda lexicon, which is both a biographical reference book and a verbal glossary, important to us for its extensive quotation from texts now lost. The earliest Byzantine lexicon, from the second half of the ninth century, listed words useful for prose writers and other obscure vocabulary.[14] It was compiled by Photius, Patriarch of Constantinople, a scholar at least the equal of Psellus in stature and importance.[15]

Photius' major work, the Bibliotheca, although an incomparably more massive undertaking, may be seen as an interesting precedent for the literary essays of Psellus. It is a review, compiled for Photius' brother Tarasius, of 280 rare texts which he had sought out and read.[16] More than half of the texts are Christian,[17] although (at the other end of the literary spectrum) four novels are also included. Photius' reviews vary considerably in format and length, but many comment on style or make other critical assessments which are probably original, although they use the technical terminology of the literary-critical tradition which goes back to the early centuries AD.[18] This range of interests and concern with stylistic questions are paralleled on a much smaller scale in the seven literary essays of Michael Psellus.[19] Three of the latter are concerned with fourth-century Church Fathers (with particular prominence given to John Chrysostom and Gregory Nazianzen), while a fourth is devoted to Symeon Metaphrastes, who in the tenth century rewrote the corpus of saints' lives in a more elaborate style. One essay is devoted to comparison of the novelists Heliodorus and Achilles Tatius.[20]

But in one respect Psellus is distinct from Photius. For Photius, although no doubt familiar with classical poetry from his school reading at least, had included in the Bibliotheca only one poet, the Empress Eudocia, wife of Theodosius II, who wrote Christian poetry in Homeric centos: this suggests that he did not read poetry for pleasure.[21] Psellus, on the other hand, devotes one of his seven essays, to poetry - that on Euripides and George of Pisidia.[22] In addition, he may well be the author of a treatise on tragedy preserved anonymously in a group of texts of which several are certainly to be attributed to Psellus[23] This is a theoretical work based largely on earlier discussions, some of which can be traced to the Chrestomathy of Proclus, and part probably to Hellenistic literary theory. Psellus' interest in poetry is to be explained in terms of his own creative writing, which includes a 540-line poem on rhetoric in Byzantine political verse,[24] and a brief set of versified rules for the composition of iambic poetry.[25] As for his interest in tragedy,[26] it has recently been shown that in his historical writing he sought to present in tragic terms (among other modes) the dramatic events of the coup of April 1042 by which the Emperor Michael V was ousted from the throne, and of which Psellus was a personal witness.[27] Hence Psellus' discussion of tragedy and of poetry in general is generated by its practical utility for himself.[28] This confident and creative approach to classical source material distinguishes him from the more encyclopaedic and conservationist attitude of the ninth- and tenth-century classical revival. Psellus saw himself as a direct successor to the writers of the Hellenic past, an environment in which he felt at ease and able to make his own contribution.[29]


I turn now to the essay on Euripides and George of Pisidia and consider its motivation, themes and arrangement before examining more closely what Psellus has to say about each of the two poets.

According to its heading, the essay was composed in response to a question: 'Who was the better poet, Euripides or Pisides?' This in itself provides important signals about context and content.

Are we to believe that such a question was in fact posed? It evokes an educational environment, where teaching was regularly conducted by means of a series of short questions and answers which served both to aid the teacher in the organisation of material and to test the pupil on how much he had learned. This system was used at various levels from the most elementary to advanced discussions, for example of theological questions.[30] The essay in response to a question may be seen as a development of this technique, and a number of Psellus' minor works take this form.[31] This reflects his role as a teacher and lecturer, interacting with or directing student interest. If the question was not asked, then it ought to have been, and Psellus would not have tackled it had he not thought it worthwhile.

But the length of these responses also associates them with another educational context, that of the public performance as a virtuoso display by a distinguished rhetorician, designed to impress and inspire listeners. The question provides Psellus with the opportunity to show off both his learning and his performance skills. Its literary subject-matter is redolent rather of the lecture-theatre than the public stage, where a less recherché topic might have been chosen, but the display element is an important factor for understanding the work. It helps to account for both the high-flown opening section (lines 12-32) and the climax, the concluding passage on George of Pisidia (lines. 110-32), where Psellus' illustration of the poet's use of figurative expressions turns into a display of his own rhetorical and linguistic expertise.[32] Psellus' reception of the two poets is a creative response, in which he implicitly presents himself as a worthy successor.

The question and answer context and the performance element suggest a further point. The essay is to be seen as an impromptu composition, devised spontaneously in response to a specific demand. As in the case of a modern lecturer, impressive performance is in practice more important than accuracy. This is not to be viewed as a highly-researched, polished, and nicely judged assessment of the two poets. Psellus (as was customary in the ancient world) relies on the information he carries in his memory, and is indeed proud to display his off-the-cuff learning. But his accuracy is not infallible[33] and, more important, reliance on remembered reading sometimes leads him (as it does the modern student) to digression. This is particularly apparent in the discussion of Euripides, where a good deal of space is devoted to comparison and contrast with the other two tragedians, particularly Aeschylus, almost certainly because Psellus recalls such comparisons from his reading. (We know that they began in Euripides' own day, with Aristophanes' Frogs.)[34] This affects the balance of the essay: thanks to Psellus' use of pre-existing analyses, Euripides is accorded twice as much space (lines 33-99) as George (lines 100-32), although Psellus apparently drew less on first-hand knowledge of the texts in discussing the tragedian.[35]

It is not possible to determine to what extent the question-and-answer setting and the pose of spontaneity represent reality as opposed to a convenient literary fiction. We are, however, invited to take the work at face value as a spontaneous response to a genuine student question, and the stance is more important than its veracity.[36]

What aspects of these authors interest Psellus? His agenda is indicated in the opening words, where both poets are said to be 'useful in metrics and the art of poetry' (line 1).[37] After remarking on the difficulty of deciding how the two men differ from one another - not a problem which strikes the modern critic - he continues:

But [if anyone] has a notable knowledge about both [arts], about metres and rhythms, that is, lofty and theoretical knowledge about metres and feet, it is not altogether difficult to discover a certain difference in either man and to bestow a crown of victory on one over the other.[38]

This initial emphasis on metre indicates the major theme of the essay: it leads on to discussion of diction and style, but metrical considerations are Psellus' focal point.

The discussion is arranged in four parts. A preliminary paragraph considers historical changes in metre, concentrating on the hexameter of epic poetry and the iambic of tragedy (lines 12-32). Separate sections on Euripides (lines 33-99) and George (lines 100-32) follow,[39] and a final brief paragraph gave Psellus' verdict. Tantalisingly the text is here too badly damaged for it to be clear what that verdict was.[40] It seems likely, however, that Psellus favoured George, not least because there would seem to be little point in the comparison with Euripides if the Byzantine poet were ultimately regarded as inferior.[41]


I pass over the difficult prefatory discussion of historical changes in metre (whose purpose is in part to show off Psellus' own rhetorical virtuosity, even at the expense of clarity),[42] and turn to his discussion of Euripides.

What interests Psellus about Euripides and why? How much does he know of him? The discussion centres on Euripides' metrical and musical versatility, his ability to imitate every subject and his skilful variation of metre according to character. But incorporation of inherited material from the literary-critical tradition is a prominent feature of this section, although even here personal taste will have influenced the selection.

A number of 'facts' typical of handbooks are adduced: Psellus mentions the name of Euripides' deme (Phlyeia),[43] and attributes to him 'eighty or more' plays.[44] The source of a reference to Euripides' ability to reduce the Athenians to tears (lines 66-7) is more difficult to trace, but it fits with Aristotle's judgement of the third tragedian as 'the most tragic' (tragikotatos).[45] Interestingly it can also be linked with Psellus' description in his History of his own role in the coup which overthrew the Emperor Michael V, where the sight of the refugee emperor and his uncle by the altar caused him to be overwhelmed by the mutability of human fortune and reduced to uncontrollable tears and groaning, like a spectator at a tragedy.[46] Psellus admires Euripides for his ability to evoke a proper response to tragic events, and his inclusion of this point is linked with his own creative writing.[47]

Initial commendation of Euripides' mastery in character-drawing and the delineation of passion leads to the first significant digression from the metrical focus (lines 40-7). Passion is identified as the distinctive characteristic of tragedy as opposed to comedy, and this point is followed by a summary of the different elements of tragedy - stage action, chorus, episodes, etc., patently culled from a standard handbook,[48] before Psellus returns to the importance of varying metre according to character. He soon moves to a comparison with Sophocles and Aeschylus (lines 54-64), [49] who are judged to have 'more profound ideas and a more dignified linguistic arsenal' and, although they sometimes lack grace and rhythmic euphony, they are in general considered more dignified and elegant.[50] Detailed attention is given to the difference in style between the Prometheus Bound and other works by Aeschylus, interesting in view of modern doubts about the authenticity of the Prometheus.[51] Psellus makes clear that, like Aristophanes, he found Aeschylus hard-going:[52] it is perhaps significant that the Prometheus was the first Aeschylean play to be read by Byzantine schoolchildren.[53] Euripides' grace and charm, both in diction and passion, are commended by contrast,[54] while as in Frogs, the less contentious Sophocles gets no more than a couple of passing references (lines 34, 54f.,?84).

The first of Psellus' two references to specific Euripidean passages confirms the derivative nature of his remarks: to illustrate Euripides' subordination of word-accent to music, he cites the opening choral words of the Orestes, with a discussion which draws on a fuller analysis by the first-century AD literary critic Dionysius of Halicarnassus.[55] But in commending Euripides' musical abilities, to which he devoted more attention than he did to plot, Psellus reflects his own interest in this difficult and technical topic, independently attested by a letter and a substantial section in the tragedy treatise plausibly attributed to him.[56] A comment on Euripides' skill in handling barbarian speech, which perhaps alludes to the Phrygian's song in the Orestes (1369-1502), also reflects Psellus' interests, since he elsewhere similarly commends a contemporary orator.[57] Psellus' concluding criticism may be a personal response: Euripides 'deviates from propriety' (line 93)[58] in the debate scene of the Hecuba (like Orestes, part of the 'Byzantine triad' of Euripidean school texts, along with Phoenissae), in giving the captive Hecuba 'the prize of honour' against the eloquent Odysseus.[59]

Although he incorporates second-hand critical material, Psellus' response to Euripides is a personal one. He selects features (metre, music) which interest him as a poet and scholar, and singles out effects which he especially admires or seeks to imitate (reproduction of barbarian speech, tragic pathos). His direct knowledge of the plays appears very limited: he refers only to texts on the school syllabus and in one case at least takes over an earlier discussion. He has no interest in treatment of myth, staging, plot construction, and so on because he does not aim to write a tragedy himself and had no personal experience of living theatre.[60] All this makes it easier to understand why he juxtaposed Euripides with the iambic poet George of Pisidia.


George of Pisidia is much closer than Euripides to Psellus, not only in time, but also in ethos. He lived in the early decades of the seventh century and, like Psellus, is associated with a cultural renaissance, in George's case that fostered by the Emperor Heraclius and the Pariarch Sergius in the 620s, a decade of reviving Byzantine political fortunes. In this period the classicising historian Theophylact Simocatta promoted Herclius' régime by narrating the events of the reign of Maurice, the emperor in whose name Heraclius had ousted his predecessor Phocas, while at a lower literary level the Paschal Chronicle made a new record of world history which culminated with Heraclius' restoration of the True Cross to Jerusalem in spring 630, an event interpreted as a prelude to the Golden Age.[61] George's contribution to this movement was the composition of panegyrical poems (some officially commissioned) in celebration of Heraclius' victorious campaigns against the Persians, together with works on Christian themes which included both propaganda for Heraclius' moves to secure church unity (Against Severus), Christian celebration (the Hexaemeron) and personal introspection (On the Vanity of Life).[62] He himself belonged to ecclesiastical circles, working as an administrator for the Patriarch Sergius who was also his patron.[63]

A good deal of George's poetry is likely to have been available to Psellus. The oldest testimony to a substantial proportion of his extensive surviving oeuvre is an eleventh-century manuscript. It has been suggested that this may reflect renewed interest in George's panegyrical poems in the reign of Basil II (d. 1025), who had ambitions in the east of a similar order to those of Heraclius which might have prompted in him hopes of a panegyrist like George[64] . In addition, extensive quotations from George's poetry are preserved in the ninth-century Chronicle of Theophanes,[65] and in the tenth-century Suda lexicon, the latter at least suggestive of renewed interest in George's work at this period. Psellus' chief focus of interest is in George's religious poetry, notably the Hexaemeron,[66] rather than the panegyric, whose topicality perhaps rendered it less accessible and hence less interesting, just as dramatic aspects of Euripides were ignored.

Psellus' discussion parallels that of Euripides without the digressions occasioned by his dependence on the earlier literary critical tradition. He is our sole source for the fact that George's home town was 'lesser' Antioch in Pisidia (line 101; not the more famous city on the Orontes). After this snippet of biographical information (analogous to the mention of Euripides' deme), he considers in turn metre, diction and style, introducing specific examples only in the last and longest section

In metre, George is commended as a proponent of 'the ancient and uniform type' (line 102), which recalls the very beginning of the essay where George is said to have handled the iambic metre 'with greater precision than many of his successors' (lines 4-6). In contrast to Euripides' metrical versatility,[67] George's fine rhythm, euphony[68] and taste for three-word trimeters are singled out.[69] Brief remarks on diction touch on his naturalistic quality, 'as if he read off whatever came into his head',[70] his exclusive use of Greek vocabulary, euphony and grandeur.[71] Psellus praises similar features at greater length in Gregory of Nazianzus, whom he regarded as a model of stylistic virtue.[72] These qualities may arouse Psellus' admiration for George not so much because the latter was a poet, but almost in spite of it.

The remainder, and by far the bulk of Psellus' discussion (lines 110-32), is concerned with George's style, centring on his use of figurative language, which is indeed a striking feature of his poetry. Psellus commends George's apt use of figures, which do not detract from the point of comparison (lines 112-14), a view not necessarily shared by modern students for whom George's highly metaphorical style presents considerable problems of interpretation.

Psellus illustrates his point with five examples which praise the vividness of George's descriptions respectively of disease, a rope suspended from heaven, the dance of the Seasons, description of the Seasons as a four-horse
carriage,[73] and battle-scenes. Although the allusions are less precise than in the case of the Euripidean examples, all but the last have been connected with specific passages in George's surviving religious poems, three to the Hexaemeron and one to Against Severus[74] . The allusion to battle-descriptions is more problematical, but indicates that Psellus was at least aware of George's military panegyrics.[75] The reference to George's vivid descriptions of disease is corroborated by the modern recognition that George appears to have specialist knowledge of medical vocabulary, a field in which Psellus too had a particular interest.[76] In fact Psellus' examples are perhaps most informative about his own interests. The rope suspended from heaven refers to the opening lines of George's poem Against Severus, where this image is used to describe how the emperor's mind is linked to that of God. But Psellus' description also alludes to a famous passage of the Iliad (8. 18-22) in which Zeus asserts his power by declaring that even if a golden rope were suspended from heaven, all the deities together could not drag him down. Psellus had himself elsewhere attempted to explain the Homeric passage in Neoplatonist and Christian allegorical terms,[77] and what he says here has more to do with Homer, and with a desire to demonstrate his own rhetorical skills, than it does with comment on George. Display of his personal virtuosity is central to Psellus' other examples.[78] On the dance of the Seasons, the relevant passage in George runs as follows: 'and they do this in alternating course / like maidens joining together in dance / and combining together their own fingers, / so that they might weave a dance of well-ordered life.' (Hex. 289-92). Psellus' response may seem overblown: 'If he gathers the Seasons together and urges them on as if in a dance, - good Lord! - the twining of hands, the circle, the whirling, the rotary motion, the strophe, the antistrophe, the epode, the movement!' (lines 118-20, tr. Dyck). This show passage forms the climax of the piece, immediately before the brief verdict.[79] Psellus here uses George as an excuse for display of his own writing, in a typically Byzantine manner.[80]


A final consideration must be George of Pisidia's own reception of Euripides. If Psellus put the two poets side by side, it seems worth investigating whether George's poetry does have a particular affinity with that of Euripides. What did George know of Euripides, and what would he have found interesting in him? But these questions cannot easily be answered, since there is very little evidence to go on. Although he does invoke figures such as Demosthenes, Galen and Proclus,[81] George does not, as far as I know, allude explicitly to Euripides. Hence judgement must be based on his use of the plays.

Textual citation is the most obvious starting-point, but also the least satisfactory in view of the availability from at least the sixth century of lexica, handbooks and anthologies. Just as Psellus incorporated material from the literary critical tradition in his essays, so George certainly used easy guides to the classics where possible, and first-hand knowledge is difficult to prove.[82] The extremely limited evidence available does, however, suggest that George favoured Euripides above the other two tragedians in this respect.[83] But this is in part because more of Euripides survived and he was more commonly cited in anthologies. Indeed the fact that two of the allusions noted come from the 'alphabetical' plays of Euripides rather than from the ten selected for school use may be an indication that George used a handbook such as the anthology of Stobaeus.[84] Much more work remains to be done in this area, but it is unlikely that George could be shown to have direct knowledge of Euripides' plays.

Other stylistic features suggest the limitation of links between George and Euripides. George's use of the three-word trimeter is remote from Euripidean practice,[85] while the most systematic analysis of his diction so far undertaken identifies the inclusion of vocabulary hitherto confined to prose, particularly specialised medical vocabulary, as George's main innovation in this area.[86] This move towards prose usage is perhaps further reflected in the use of rhyming line-ends, in a manner analogous to that employed in some types of rhetorical prose.[87] On the other hand, in one case at least, George apparently expresses a preference for a Euripidean usage,[88] and more extensive research might add to this haul. Even here, however, the proviso about lexical handbooks must apply, quite apart from our own limited knowledge of ancient authors.[89]

In fact in many ways it is preferable to follow Psellus' example and take metre as a starting-point. In making the innovative decision to compose his panegyrics in trimeters rather than the traditional hexameters, George is most likely to have looked to Euripides as a metrical model.[90] Furthermore, metrical characteristics can be analysed and measured.[91]

Modern studies[92] of George's metrical practice uphold Psellus' verdict (lines 4-6) that George handled the iambic metre with greater precision than his successors. He had a good understanding of the metrical rules for the classical quantitative trimeter, even though his verses also reflect care in the placing of accents, to assist the understanding of a post-classical audience whose ear was unattuned to differences in quantitative length between syllables.[93] His trimeters are also more varied than those of his successors, who reduced the rhythm to a monotonous 12-syllable line with predictable accentuation: although the majority of George's lines are dodecasyllabic, he also writes 13- and 14-syllable lines, occasionally 15 syllables.[94] The increase in the number is achieved by the use of resolved (divided) syllables, a practice most common in the classical comic poets and, among the tragedians, in Euripides. Analysis of specific types of resolution has suggested that George is closer to Euripides than he is to the other two classical tragedians.[95] It may be added that, like Euripides, George changed his metrical habits over the course of his career. While Euripides' dramatic increase in resolution in his later plays is well-known[96] George's later panegyrical poems show increased care in the placing of accent at line-end.[97] Although he is unlikely to have been aware of this phenomenon in Euripides, George's own innovative instincts will have attracted him to a poet of Euripides' richness and variety.

Thus George is more closely allied with Euripides than with the other tragedians, especially in his experimentation and refinement of metrical practice: as Euripides responded to new trends in contemporary music,[98] so George was increasingly attuned to the limited sensitivity of his contemporaries to quantitative classical metres and adapted his technique accordingly. Both poets share a readiness to innovate, to experiment with and revitalise a well-tried genre, and thus single themselves out for comment.[99]

But such analogies are too limited to be very instructive. In the present state of knowledge, the verdict on George's reception of Euripides must remain open. It seems that, like Psellus, he found Euripides the most accessible of the tragedians, but there is no evidence that he was directly acquainted with Euripides' plays, as opposed to extracts available in compendia. Technical metrical similarities, although telling, may well be unconscious. But like Psellus, George has no interest in Euripides' plays as drama.


It may be added that Psellus himself was not seriously interested in comparing George's style with that of Euripides. Had he wished to do so, he would have concentrated on Euripides' iambics, rather than the lyrical parts of his plays. Nevertheless his response to the two poets is illuminating, combining as it does the interests of a scholar and a creative writer. His erudition enables him to recall at will learned literary discussions of the classical dramatists, perhaps a passage from the Hecuba which he had found offensive to propriety, to relate an allusion in George of Pisidia to contemporary allegorical interpretation of Homer, as well as to make judgements on technical metrical questions. These latter technical aspects were of interest for his own poetic aspirations. The composition of quantitative iambic trimeters in the eleventh century was likely to have been an exercise analogous to the composition of classical Greek verse today - to be contemplated only by the most able, and with the assistance of all available handbooks. But Psellus is also interested in the effect of poetry on its audience: he admires Euripides' mastery of tragic pathos (and seeks to emulate it in his own historical writing), while George's skilful handling of figurative language appeals to his taste: this he seeks to match and outdo in his own rhetorical prose criticism of the poet. Classical verse composition may have been almost beyond reach, but Psellus saw that classical poetry had lessons also for those who strove to write effective prose.

This one very limited example of reception perhaps encapsulates the best and worst of Byzantium: its degenerative tendency to incorporate, distil and often distort earlier scholarship, but also its sense of direct and confident inheritance of a rich literary tradition to which innovative additions could still be made.

[return to contents]


[1] I am most grateful to Judith Mossman for help on responses to Euripides. In revising this paper for publication, I have benefited greatly from comments generously made by Pat Easterling and by Lorna Hardwick and two anonymous referees. This paper was written with the support of a Leverhulme Special Research Fellowship. [return to text]

[2] Cf. R. Browning, 'Enlightenment and Repression in Byzantium in the eleventh and twelfth centuries', Past and Present 69 (1975), 3-23 at 11 (=Studies on Byzantine History, Literature and Education, Variorum 1977, no. XV): 'sometimes childishly vain about his immense and superficial learning'. [return to text]

[3] On the problem of Psellus' verdict, see further below, section II. [return to text]

[4] L. D. Reynolds and N. G. Wilson, Scribes and Scholars (2nd ed., Oxford 1974) 60. [return to text]

[5] Founded, along with a school of law, by Constantine IX Monomachus in 1045. Psellus claims (Chronographia 6. 37) to have revived philosophy single-handed from decline; cf. C. Mango, Byzantium: the Empire of New Rome (London 1980) 143. For his contribution to the development of philosophical studies in Byzantium, see C. Niarchos, 'The philosophical background of the eleventh-century revival of learning in Byzantium', in (edd.) M. Mullett and R. Scott, Byzantium and the Classical Tradition (Birmingham 1981) 127-35. [return to text]

[6] He boasted of attracting pupils not merely from every region of the Byzantine empire, but from Egypt and India too: Browning, 'Enlightenment' (n. 2) 20. [return to text]

[7] Convenient survey in N. G. Wilson, Scholars of Byzantium (London 1983) 156-66. [return to text]

[8] The thirteenth-century codex Vaticanus Barberinianus graecus 240: details in Andrew R. Dyck, Michael Psellus: The Essays on Euripides and George of Pisidia and on Heliodorus and Achilles Tatius (Byzantina Vindobonensia xvi, Vienna 1986) 25-7. [return to text]

[9] C. Mango, 'Discontinuity with the classical past in Byzantium', in Mullett and Scott (n. 5) 48-57. R. Browning ('Literacy in the Byzantine world', Byzantine and Modern Greek Studies 4, 1978, 39-54) suggests that estimates of literacy tend to be too low, although few would have a sophisticated level of knowledge. [return to text]

[10] Browning, 'Enlightenment' (n. 2) 11. [return to text]

[11] C. Mango, 'The availability of books in the Byzantine empire, A.D. 750-850', in Byzantine Books and Bookmen: a Dumbarton Oaks colloquium (Dumbarton Oaks 1975) 29-45 at 45; id. Byzantium (n. 5) 138. [return to text]

[12] The earliest dated minuscule manuscript is from 835; see Reynolds and Wilson (n. 4) 51-4. [return to text]

[13] See Wilson, Scholars (n. 7) 120-35; R. Browning, 'Byzantine Scholarship', Past and Present 28 (1964) 3-22 at 11f.=Studies (n. 2) XIII. [return to text]

[14] On usefulness as a criterion in scholarship, see further n. 28. [return to text]

[15] See Wilson, Scholars (n. 7) 89-93; Mango, Empire (n. 5) 137-44; Browning, 'Scholarship' (n. 13) 8-11. The classic study of the renaissance here cursorily surveyed is P. Lemerle, Le premier humanisme byzantine (Bibliothèque byzantine, Études 6, Paris 1971) esp. chs. 7-10. Briefer survey (in English): Warren Treadgold, 'The Macedonan Renaissance', in (ed.) Treadgold, Renaissances before the Renaissance (Stanford 1984) 75-98. [return to text]

[16] Wilson, Scholars (n. 7) 93-111, id. Photius, The Bibliotheca (London 1994). Photius does not include familiar school texts. The number of books available to him was extraordinarily large by contemporary standards and there has been much discussion of how he could have had access to so many, see further Mango, 'Availability' (n. 11) 37-43, Warren Treadgold, The nature of the Bibliotheca of Photius (Dumbarton Oaks Studies 18, Dumbarton Oaks 1980); brief statement of problems, Treadgold, 'Photius and the reading public for classical philology in Byzantium', in Mullett and Scott (n. 5) 123-6. [return to text]

[17] Over 40 per cent of these are specifically theological, reflecting the particular preoccupations of a patriarch of Constantinople. Photius' Quaestiones Amphilochiae, composed during a period of exile allegedly in response to questions posed by Amphilochius, metropolitan of Cyzicus, is also predominantly theological, see Wilson, Scholars (n. 7) 114-19. This work deserves mention in connection with Psellus' literary essays, some of which are similarly framed as responses to enquiries, although in Psellus' case no specific interlocutor is identified. Some of Photius' Amphilochian responses are of substantial length (e.g. no 1 has 37 sections). See further below, section II. [return to text]

[18] Wilson, Scholars (n. 7) 103-5. The writings of Hermogenes and Dionysius of Halicarnassus are an important influence on both Photius and Psellus. [return to text]

[19] Surveyed by Wilson, Scholars (n. 7) 166-79. [return to text]

[20] On the continuing popularity of these novelists in the thirteenth century, see Wilson, 'Books and readers in Byzantium', in Byzantine Books (n. 11) 1-15 at 7f.; id. Scholars (n. 7) 26, 186. [return to text]

[21] Wilson (Scholars, n. 7, 100, 111-13) collects evidence that Photius had indeed read some classical poetry at school; B. Baldwin ('Photius and poetry', BMGS 4, 1978, 9-14) argues for deliberate exclusion of poetic texts from the Bibliotheca. Arethas claimed in a letter that 'every schoolboy' knew Sophocles' Ajax, see S. B. Kougeas, ho Kaisareias Arethas kai to ergon autou (1913) 142. (I am indebted to Pat Easterling for this reference.) [return to text]

[22] Ed. Dyck (n. 8). [return to text]

[23] Ed. R. Browning, 'A Byzantine treatise on tragedy' (Acta Universitatis Carolinae, Prague 1963) 67-81 (=Studies, n. 2, XI); cf. Wilson, Scholars (n. 7) 177f. [return to text]

[24] Dyck (n. 8) 33. On this metre, which is specifically associated with works for the imperial family, see Michael Jeffreys, 'The Nature and Origins of Byzantine Political Verse', Dumbarton Oaks Papers 28 (1974) 141-95. [return to text]

[25] Dyck (n. 8) on 16ff. [return to text]

[26] See R. Browning, 'Ignace le diacre et la tragédie classique à Byzance', Revue des études grecques 80 (1968) 401-10 (=Studies, n. 2, XIV) for revival of interest in tragedy in Byzantium in the 9th c., noting the reservations of Wilson, 'Books and readers' (n. 20) 14f. [return to text]

[27] Andrew R. Dyck, 'Psellus tragicus: observations on Chronographia 5. 26ff.' Byzantinsche Forschungen 20 (1994) 269-90. See further below, section III. [return to text]

[28] Usefulness is a regular motive for scribal comment and for selection from texts from the earliest period of classical scholarship, see (recently) Pat Easterling, 'Menander: loss and survival', in Stage Directions: essays in honour of E. W. Handley, Bulletin of the Institute of Classical Studies Suppl. 66 (1995)153-60 at 154-6, 160. Photius' Lexicon was compiled on this principle, and the opening words of Psellus' essay on Euripides and George of Pisidia comment on the usefulness of these authors, see below, section II. [return to text]

[29] Cf. Browning, 'Enlightenment' (n. 2) 10. [return to text]

[30] For the question in elementary education, see Mango, Empire (n. 5) 127. Christian collections are discussed by J. Haldon, 'The works of Anastasius of Sinai: a key source for the history of seventh-century east Mediterranean society and belief', in (edd.) A. Cameron and L. I. Conrad, The Byzantine and early Islamic Near East: I Problems in the literary source material (Studies in Late Antiquity and Early Islam 1, Princeton 1992) 107-47 at 116-25. See also n. 17 on Photius' Quaestiones Amphilochiae. [return to text]

[31] Examples collected by Dyck (n. 8) 52; cf. Browning, 'Treatise' (n. 23) 71, id. 'Enlightenment' (n. 2) 9f. In the case of Psellus' commentary in political verse on the biblical 'Song of Songs', it has been argued that there is no reason to doubt that Psellus was answering an actual question from the emperor, 'What pious conclusions can be drawn from that difficult book?', see Jeffreys (n. 24) 164; a work in political verse, however, is aimed at a rather different audience from the literary essays, see Jeffreys' discussion. [return to text]

[32] So Dyck (n. 8) 71. On the opening section, see n. 42. [return to text]

[33] For example, Dyck (n. 8, 72) notes that Psellus confuses the Elements and the Seasons in referring to George's poetry. [return to text]

[34] Aristophanes was intensely studied almost continuously throughout antiquity as an exponent of the Attic dialect, although Wealth and Clouds were more widely read than Frogs, see K. J. Dover, Aristophanic Comedy (Berkeley 1972) 224-9. The history of competitions and of literary comparisons (synkriseis) between authors, also relevant to the genre of this work, is discussed by Dyck (n. 8) 27-9. [return to text]

[35] A. Pertusi (Giorgio di Pisidia poemi I: panegirici epici, Studia Patristica et Byzantina, Heft 7, Ettal 1960, 30f., 50) doubted whether Psellus had read much either of Euripides or of George. [return to text]

[36] Cod. Vat. Barb. gr. 240 (n. 8) preserves a series of treatises by Psellus: Dyck (n. 8) 26. The text is written on paper in an irregular and abbreviated hand which suggests that the copy was made for practical use rather than as a luxury item. [return to text]

[37] Cf. n. 28. [return to text]

[38] Lines 8-11, tr. Dyck. [return to text]

[39] Although Psellus' citation of material on Aeschylus and Euripides indicates knowledge of comparisons (synkriseis) between poets, he does not himself make direct comparisons either in this essay or the one on Heliodorus and Achilles Tatius, but treats each author separately. [return to text]

[40] The damage probably goes back to the seventeenth century, since Leo Allatius did not include the conclusion in his extracts from the text: Dyck (n. 8) 25. [return to text]

[41] Other arguments are adduced by Dyck (n. 8) 34-6. [return to text]

[42] Cf . Wilson, Scholars (n. 7) 178 'Psellos apparently lapses into florid verbiage, which will not be redeemed even if one day a less damaged text is recovered.'[return to text]

[43] Also paraded in the essay on John Chrysostom, see Dyck (n. 8) on 3-4 and 135. The information may have come from the Suda. [return to text]

[44] Line 64. The biographical tradition gives a total of 92, of which 78, 77 or 68 were said to have been preserved: see Dyck ad loc. [return to text]

[45] Poetics 1453a29f. Plutarch describes how the tyrant Alexander of Pherae was moved to tears by one of Euripides' Trojan tragedies, see J. Mossman, Wild Justice (Oxford 1995) 218f. But he was not, of course, an Athenian. Xenophon (Symposium 3. 11) mentions the ability of the 5th/4th-century actor Kallippides to evoke audience tears, but evidence for audience response is very limited, see A. Pickard-Cambridge The Dramatic Festivals of Athens (2nd ed., rev. J. Gould and D. M. Lewis, Oxford 1968) 274-8. [return to text]

[46] Chron. 5. 40. See Dyck (n. 27) 281-7. [return to text]

[47] For tragic patterning in prose biography, see Judith Mossman, 'Tragedy and epic in Plutarch's Alexander', Journal of Hellenic Studies 108 (1988) 83-93, qualifying the arguments of P. De Lacy, 'Biography and tragedy in Plutarch', American Journal of Philology 73 (1952) 159-71. De Lacy noted the censorious tone of Plutarch's allusions to tragedy, which is criticised, among other things, for its false material, a view associated with Platonic ideas. Psellus, on the other hand, despite his own Platonic leanings, commends Euripides' veracity: '[the Athenians] fancied they beheld the spoken word as living action', lines 67f., tr. Dyck. (I am grateful to Pat Easterling for drawing my attention to these articles.) [return to text]

[48] The anonymous treatise on tragedy (n. 23) also deals with the parts of tragedy (secs. 1, 4). The ultimate source is probably Aritstole's Poetics: Dyck (n. 8) 58. This passage degenerates into banality, and it might be charitable to suspect the intrusion of a scholiast's note. [return to text]

[49] The comparison is already implicit in the suggestion (line 34) that one might prefer Sophocles to Euripides. [return to text]

[50] Lines 54-7. 'Dignity' is already associated with Aeschylean style at Frogs 1004, 1061. [return to text]

[51] Articulated by M. Griffith in The Authenticity of Prometheus Bound (Cambridge 1977), and in his commentary on the play (Cambridge 1983). [return to text]

[52] 'Without being, as it were, an initiate, one would not understand his mysteries' (lines 63-4). Persae is specifically singled out. [return to text]

[53] Wilson, Scholars (n. 7) 112. [return to text]

[54] But at line 37 two apparently contrasting qualities are both associated with Euripides, who is described as 'sometimes taking the lead in other forms of grace and dignity'. [return to text]

[55] On Literary Composition 11; see Dyck (n. 8) 62f. [return to text]

[56] For comparison with the letter, see Dyck (n. 8) on 88-89. Tragedy treatise (n. 23) sec. 5: Euripides uses the chromatic and was the first to use polychords, both marks of the new style in music; in general he shows the greatest metrical variety. See Winnington-Ingram's commentary 74-8. [return to text]

[57] Dyck (n. 8) on 90. Euripides does not, however, attempt to reproduce the Phrygian dialect in the Orestes. [return to text]

[58] 'Propriety' (to prepon) is a standard ancient literary-critical tool: Dionysius of Halicarnassus On Literary Composition is particularly concerned with the topic, e.g. 20 'Propriety is a necessary concomitant to all the rest. Any work which is lacking herein lacks, if not its whole effect, at least the most important part of it.'; propriety is later defined as 'what suits the persons or actions to be handled'. Tr. D. A. Russell in D. A. Russell and M. Winterbottom, Ancient Literay Criticism (Oxford 1972) 334f. [return to text]

[59] Wilson (Scholars, n. 7, 179) suggests that Psellus was genuinely moved by the scene. Judith Mossman has kindly drawn my attention to Theon, Progymnasmata ii. 60. 29-30 (Spengel), 'we censure Euripides because his Hecuba philosophises out of turn' (discussing prosopopoeia): this may allude to much earlier criticism of this scene. [return to text]

[60] Cf. Easterling (n. 28) 160 on rejection of material tied to a particular context in favour of the universally applicable as a criterion for selection from a text (discussing Menandrean maxims). [return to text]

[61] See Michael Whitby and Mary Whitby, Chronicon Paschale 284-628 AD (Translated Texts for Historians vol. 7, Liverpool 1989) ix-xiv. [return to text]

[62] The panegyrical poems are edited by Pertusi (n. 35). Other major poems are to be found in Patrologia Graeca vol. 92. [return to text]

[63] Evidence relating to George's career and writing is collected by Pertusi (n. 35) 11-16. [return to text]

[64] So Pertusi (n. 35) 14f., 49-67. Cf. id. 'Dei poemi perduti di Giorgio di Pisidia', Aevum 30 (1956) 395-427. The 11th-century cod. Paris. suppl. gr. 690 contains George's Hexaemeron, but not the poem Against Severus, to which Psellus alludes: see further below. [return to text]

[65] On the material in Theophanes, see J. Howard-Johnston, 'The official history of Heraclius' Persian campaigns', in (ed.) E. Dabrowa, The Roman and Byzantine Army in the East (Krakow 1994) 57-85, who argues that George himself composed a hybrid prose/verse history of Heraclius' campaigns. [return to text]

[66] This poem has a much more extensive manuscript tradition than George's other works, partly because it was for a time mistakenly attributed to Cyril of Alexandria, but also because its Creation theme gave it a much wider appeal, see Pertusi, 'Poemi perduti' (n. 64) 402-5, 408f. [return to text]

[67] An explicit contrast with the variety required in drama appears to be made in a damaged sentence (lines 103-4), but Psellus is not, it seems, critical of George (a further reason for thinking that he ultimately judged George more highly). Ninety hexameters On the Mortal Life are also attributed to George, but are unknown to or ignored by Psellus, see Dyck (n. 8) on 101-2. [return to text]

[68] George's euphony (of diction) is again commended at line 110, whereas Aeschylus and Sophocles were criticised for lack of euphony at 56f. (The Greek word used is slightly different in each case.) [return to text]

[69] For this phenomenon in George's panegyrical poems, see M. Marcovich, Three-word Trimeter in Greek Tragedy (Beiträge zur klassischen Philologie, Heft 158, Königstein/Ts. 1984) 199. His ratios of 1/90.5 (panegyrics) and 1/118 (Hexaemeron) are much higher than those for Euripides (1/279.4), and closer to Psellus himself (1/44): see Marcovich 160-5. [return to text]

[70] Cf. J. D. C. Frendo, 'The significance of technical terms in the poems of George of Pisidia', Orpheus 21 (1974) 45-55 at 54f.: the most striking single feature about George's poetry is his use of words previously confined to prose. A similar 'artless' style was admired in Menander by the rhetorician Hermogenes, see Easterling (n. 28) 154, and cf. Dyck (n.8) on 108-9. [return to text]

[71] Contrast the comments on Euripides' use of barbarian speech. In the sixth century, the orator Choricius praised his contemporary Procopius of Gaza in similar terms (Or. 7. 8, 112. 10-15 Foerster-Richtsteig): in his lectures to the young, 'he never let pass a word that was not Attic... nor a syllable that spoilt the rhythm, nor a sentence whose construction failed to please the ear.'[return to text]

[72] See Wilson, Scholars (n. 7) 169-71. [return to text]

[73] It is in fact the Elements, not the Seasons, who are likened to a four-horse carriage, cf. n. 33. [return to text]

[74] See Dyck ad locc. [return to text]

[75] Dyck suggests (on 127-30) that Psellus may here be indulging in rhetorical amplification with the aid of a lexicon, since Pertusi (n. 35, 30f.) found that, although George does on more than one occasion describe battle-scenes, the terms chosen by Psellus do not occur in the surviving panegyrics, while several are characteristic of classical or Hellenistic battle-descriptions. In view of Psellus' deviation from the original in the other examples, it is likely that he was amplifying half-remembered material, but it is also possible that he had read panegyrical poems of George which are now lost. [return to text]

[76] J. D. C. Frendo, 'Special aspects of the use of medical vocabulary in the poems of George of Pisidia', Orpheus 22 (1975) 49-56; A. Littlewood, 'The midwifery of Michael Psellos', in Mullett and Scott (n. 5) 136-42. [return to text]

[77] Dyck (n. 8) ad loc., Wilson, Scholars (n. 7) 161f. [return to text]

[78] See Dyck ad locc. [return to text]

[79] And incidentally corroborates the belief that George was to be acclaimed the victor. [return to text]

[80] Cf. Littlewood (n. 76) 136: 'A distinctive feature of the Byzantines' imitation of classical literature is the tension between the desire to exhibit a fluent familiarity with ancient models and the compulsion to demonstrate an ability to create new variations of these same models,...'. Wilson 'Books' (n. 20) 11 comments on Symeon Metaphrastes' verbose elaboration of saints' lives; cf. id. Scholars (n. 7) 167 on Psellus' essay on Symeon. [return to text]

[81] Demosthenes: Exp. Pers. ii. 1, Heraclias i. 93. Galen: Heraclias ii. 41. Proclus: Hex. 60, 75, 78. [return to text]

[82] Pertusi ('Poemi perduti', n. 64, 396) suspected that allusions to the tragedians were derived from gnomological collections. Wilson ('Books', n. 20, 5) is not confident that the sixth-century historian Agathias had read much Euripides. On the limits of Agathias' knowledge even of Herodotus and Thucydides, see Averil Cameron, 'Herodotus and Thucydides in Agathias', Byzantinische Zeitschrift 57 (1964) 33-52=ead. Continuity and change in sixth-century Byzantium (London 1981) II. (I am grateful to a referee for drawing my attention to this article.) [return to text]

[83] Pertusi (n. 35, 38 n.1) noted at least seven allusions to Euripides in the panegyrical poems, as opposed to two to Aeschylus and one to Sophocles. The figure for Euripidean allusion is higher even than that for Homer (5 instances), but in all cases the count is very low. [return to text]

[84] One allusion to Suppl. 508 (Exp. Pers. iii. 52f) and one to Hel. 514 (Exp. Pers. iii. 181): both are of a gnomic character, and the latter is not particularly close. See R. M. Piccione, 'Sulle citazioni euripidee in Stobeo e sulla struttura dell' Anthologion', Rivista di filologica 122 (1994) 175-218. Piccione notes Stobaeus' high proportion of citations from the nine alphabetical plays, and even more (approximately three-quarters) from plays now lost. (I am grateful to Pat Easterling for drawing my attention to this article.) [return to text]

[85] See n. 69. [return to text]

[86] See nn. 70, 76. Pertusi (n. 35, 39-43) noted in addition George's extensive use of rare and unique words, which were influential on later writers. [return to text]

[87] On George's use of rhyme, see Pertusi (n. 35) 45-7, who notes that it is a phenomenon of sixth- seventh-century religious poetry and prose homiletic. The orator Gorgias was the most famous exponent of rhthymic and rhyming prose; see further J. D. Denniston, Greek Prose Style (Oxford 1952) 135f. [return to text]

[88] At Hexaemeron 1891 a relatively rare verb (stomo) is used in a sense known elsewhere only in Euripides ('fence'), see LSJ s.v. IV, and cf. Mary Whitby, 'The Devil in Disguise: the end of George of Pisidia's Hexaemeron reconsidered', JHS 115 (1995) 125 n. 45. [return to text]

[89] Pertusi (n. 35, 39f.) highlights the hazards of this kind of investigation. [return to text]

[90] P. E. Bouvy (quoted by Pertusi, n. 35, 43) suggested that George took Euripides as a metrical model. R. J. H. Jenkins ('The Hellenistic origins of Byzantine literature', DOP 17, 1963, 37-52 at 41) stated bluntly that George's iambics are modelled on the Euripidean rhesis; some clarification is provided by his subsequent comment (42), 'he might borrow his forms from what he understood of Euripides; but...'. [return to text]

[91] Cf. D. Feeney, 'Criticism ancient and modern', in (edd.) D. Innes, H. Hine, C. Pelling, Ethics and Rhetoric (Oxford 1995) 301-12 on the inadequacy of ancient tools of criticism and the problem of finding an adequate critical approach to ancient texts. Metrical and stylistic analysis of late antique hexameter poetry was pioneered by A. Wifstrand, Von Kallimachos zu Nonnos (Publications of the New Society of Letters 16, Lund 1933). [return to text]

[92] Valuable work was done in the metrical field at the beginning of this century by the Pole Leo Sternbach (Rozprawy Akademii Umiejetnosci, Wydzial filologiczny Ser. ii, tom. xv, Krakow 1900, 259-96), who first gave serious scholarly attention to George of Pisidia, and many of his conclusions still stand. See further Paul Maas, 'Der byzantinische Zwölfsilber', BZ 12 (1903) 278-323, esp. 289f. 321 (revised in Kleine Schriften, Munich 1973, 242-88); Pertusi (n. 35) 43-5; M. West, Greek Metre (Oxford 1982) 177-80, 182-5; R. Romano, 'Teoria e prassi della versificazione: il dodecasillabo nei Panegirici epici di Giorgio di Pisidia', BZ 78 (1985) 1-22. [return to text]

[93] Accents are regulated at the line-end and before the caesura, see Pertusi (n. 35) 43-5, West (n. 92) 184, Romano (n. 92). Romano gives a figure of 81.9% paroxytone verse-endings in the panegyrical poems. [return to text]

[94] Pertusi (n. 35) 44 (two instances of 15-syllable lines in the Hexaemeron); Romano (n. 92). [return to text]

[95] Sternbach (n. 92, 289-91) found George's technique in handling tribrachs and dactyls comparable only to Euripides, not to Sophocles or Aeschylus: cf. Dyck (n. 8) on 4-6. [return to text]

[96] Although, of course, our sample for Euripides is much bigger than it is for his predecessors, Aeschylus and Sophocles generally show less variation: see the discussion of West (n. 92) 85-8. [return to text]

[97] Romano (n. 92) 10: increased preference for lines accented on the second syllable from the end (paroxytone). [return to text]

[98] See n. 56. Euripides' new-fangled musical ideas are satirised in Aristophanes' Frogs. [return to text]

[99] George's adoption of the iambic metre for panegyric, as opposed to the traditional hexameter, is his most striking innovation. Euripides is central to Aristophanic and Aristotelian criticism of tragedy, while George became a model for composers of iambics (Wilson, Scholars, n. 7, 187; Dyck, n. 8, 35) and was singled out by Psellus. [return to text]