Series Content

Volume I
THE ORAL TRADITIONAL BACKGROUND
OF ANCIENT GREEK LITERATURE

Volume 2
HOMER AND HESIOD AS PROTOTYPES
OF GREEK LITERATURE

Volume 3
GREEK LITERATURE IN THE ARCHAIC PERIOD:
THEEMERGENCEOFAUTHORSHIP

Volume 4
GREEK LITERATURE IN THE CLASSICAL PERIOD:
THE POETICS OF DRAMA IN ATHENS

Volume 5
GREEK LITERATURE IN THE CLASSICAL PERIOD:
THE PROSE OF HISTORIOGRAPHY AND ORATORY

Volume 6
GREEK LITERATURE AND PHILOSOPHY

Volume 7
GREEK LITERATURE IN THE HELLENISTIC PERIOD

Volume 8
GREEK LITERATURE IN THE ROMAN PERIOD
AND IN LATE ANTIQUITY

Volume 9
GREEK LITERATURE IN THE BYZANTINE PERIOD
Acknowledgments

The editor wishes to thank the following scholars for their help and
encouragement: Benjamin Acosta-Hughes, Victor Bers, Emmanuel Bourbouhakis,
Casey Due, Mary Ebbott, David Elmer, Corinne Pache, Jennifer Reilly, Panagiotis
Roilos, David Schur, Roger Travis, T. Temple Wright, Dimitrios Yatromanolakis.
Greek, Literature

Volume 9
Creek Literature
in the Byzantine Period

Edited with introductions by

Gregory Nagy
Harvard University

ROUTLEDGE
New York/London
Published in 2001 by

Routledge
2 Park Square, Milton Park,
Abingdon, Oxon, OX14 4RN

Published in Great Britain by

Routledge
711 Third Avenue,
New York, NY 10017

Routledge is an Imprint of Taylor & Francis Books, Inc.

Copyright C> 2001 by Routledge

All rights reserved. No part of this book may be reprinted or reproduced or utilized in any form or by
any electronic, mechanical, or other means, now known or hereafter invented, including any
photocopying and recording, or in any information storage or retrieval system, without permission in
writing from the publishers.

Library of Congress Cataloging-in-Publication Data

Greek literature I edited with introductions by Gregory Nagy.

p.em.
Includes bibliographical references.
Contents: v.I. The oral traditional background of ancient Greek literature - v. 2.
Homer and Hesiod as prototypes of Greek literature -- v. 3. Greek literature in the
archaic period: the emergence of authorship -- v. 4. Greek literature in the classical
period: the poetics of drama in Athens -- v. 5. Greek literature in the classical period :
the prose of historiography and oratory - v. 6. Greek literature and philosophy - v. 7.
Greek literature in the Hellenistic period -- v. 8. Greek literature in the Roman period
and in late antiquity -- v. 9. Greek literature in the Byzantine period.
ISBN 0-8153-3681-0 (set) -- ISBN 0-8153-3682-9 (v. 1) -- ISBN 0-8153-3683-7 (v. 2)
- ISBN 0-8153-3684-5 (v. 3) -- ISBN 0-8153-3685-3 (v.A) -- ISBN 0-8153-3686-1 (v.
5) -- ISBN 0-8153-3687-X (v. 6) -- ISBN 0-8153-3688-8 (v. 7) -- ISBN 0-415-93770-1
(v. 8) -- ISBN 0-4IS-93771-X (v. 9) --ISBN 0-8153-2-
1. Greek literature--History and criticism.!. Nagy, Greg')ry.

PA3054 .G74 2001

880.9--dc21
2001048490

ISBN 0-8153-3681-0 {set}

ISBN 0-8153-3682-9 {v.l}
ISBN 0-8153-3683-7 (v.2)
ISBN 0-8153-3684-5 (v.3)
ISBN 0-8153-3685-6 (vA)
ISBN 0-8153-3686-1 (v.5)
ISBN 0-8153-3687-X (v.6)
ISBN 0-8153-3688-8 (v.7)
ISBN 0-4159-3770-1 {v.8}
ISBN 0-4159-3771-X (v.9)

Publisher's Note
The publisher has gone to great lengths to ensure the quality of this reprint
but points out that some imperfections in the original may be apparent.
Contents

vii Series Introduction

IX Volume Introduction

Section A. Synchronic and Diachronic Perspectives

1 1. From Ephrem to Romanos
Sebastian Brock
14 2. Wandering Poets: A Literary Movement in Byzantine Egypt
Alan Cameron
55 3. The Sixth Sibylline Oracle as a Literary Hymn
Mark D. Usher
80 4. On the Imitation (Mimesis) of Antiquity in Byzantine Literature
Herbert Hunger

Section B. Registers and Styles

103 5. The Language of Byzantine Literature
Robert Browning
134 6. The Oral Background of Byzantine Popular Poetry
Elizabeth and Michael Jeffreys
179 7. The Function and Evolution of Byzantine Rhetoric
George L. Kustas
199 8. Levels of Style in Byzantine Literature
Ihor Sevcenko

Section C. "The Saint's Life"

223 9. The "Low Level" Saint's Life in the Early Byzantine World
Robert Browning
235 10. Imperial Panegyric: Rhetoric and Reality
George T. Dennis
245 11. Byzantine Hagiography and Sex in the Fifth to Twelfth Centuries
Alexander Kazhdan
258 12. Byzantine Descriptions of Works of Art
Henry Maguire

v
VI Contents

Section D. Literary Renaissance

301 13. The Poverty of Ecriture and the Craft of Writing: Towards a
Reappraisal of the Prodromic Poems
Margaret Alexiou
341 14. Poetic Justice in the Patriarchate: Murder and Cannibalism in the
Provinces
Ruth Macrides
373 15. Honour among Romaioi: The Framework of Social Values in the
World of Digenes Akrites and Kekaumenos
Paul Magdalino
409 16. Aristocracy and Patronage in the Literary Circles of Comnenian
Constantinople
Margaret Mullett
439 17. Amphoteroglossia: The Role of Rhetoric in the Medieval Learned
Novel
Panagiotis Roilos

457 Copyright Acknowledgments

 Series Introduction

This nine-volume set is a collection of writings by experts in ancient Greek

literature. On display here is their thinking, that is, their readings of ancient
writings. Most, though not all, of these experts would call themselves philologists.
For that reason, it is relevant to cite the definition of "philology" offered by
Friedrich Nietzsche. In the preface to Daybreak, he says that philology is the art of
reading slowly:

Philology is that venerable art which demands of its votaries one thing
above all: to go aside, to take time, to become still, to become slow- it is
a goldsmith's art and connoisseurship of the word which has nothing but
delicate cautious work to do and achieves nothing if it does not achieve it
lento. But for precisely this reason it is more necessary than ever today; by
precisely this means does it entice and enchant us the most, in the midst
of an age of "work," that is to say, of hurry, of indecent and perspiring
haste, which wants to "get everything done" at once, including every old
or new book:- this art does not easily get anything done, it teaches to
read well, that is to say, to read slowly, deeply, looking cautiously before
and aft, with reservations, with doors left open, with delicate fingers and
eyes.
(This translation is adapted, with only slight changes, from R. J.
Hollingdale, Friedrich Wilhelm Nietzsche, Daybreak: Thoughts on the
Prejudices ofMorality [Cambridge, 1982]')

Nietzsche's original wording deserves to be quoted in full, since its power

cannot be matched even by the best of translations:

Philologie namlich ist jene ehrwiirdige Kunst, welche von ihrem Verehrer
vor Allem Eins heischt, bei Seite gehn, sich Zeit lassen, still werden,
langsam werden- , als eine Goldschmiedekunst und -kennerschaft des
Wortes, die lauter feine vorsichtige Arbeit abzuthun hat und Nichts
erreicht, wenn sie es nicht lento erreicht. Gerade dam it aber ist sie heute
nothiger als je, gerade dadurch zieht sie und bezaubert sie uns am
starksten, mitten in einem Zeitalter der "Arbeit," will sagen: der Hast, der
unanstandigen und schwitzenden Eilfertigkeit, das mit AHem gleich "fertig
werden" will, auch mit jedem alten und neuen Buche:- sie selbst wird
nicht so leicht irgend worn it fertig, sie lehrt gut lesen, das heisst langsam,

Vll
Vlll Series Introduction

tief, riick- und vorsichtig, mit Hintergedanken, mit offen gelassenen

Thiien, mit zarten Fingern und Augen lesen ...
(F riedrich Nietzsche, Morgenrothe. Nachgelassene Fragmente, Anfang 1880 his
Frnhjahr 1881. Nietzsche Werke V.I, ed. G. Colli andM. Mominari[Berlin,
1971],9.)

This is not to say that the selections in these nine volumes must be ideal
exemplifications of philology as Nietzsche defined it. Faced with the challenge of
describing their own approaches to Greek literature, most authors of these studies
would surely prefer a definition of "philology" that is less demanding. Perhaps
most congenial to most would be the formulation of Rudolf Pfeiffer (History of
Classical Scholarship I [Oxford, 1968]: "Philology is the art of understanding,
explaining and reconstructing literary tradition."
This collection may be viewed as an attempt to demonstrate such an art, in all
its complexity and multiplicity. Such a demonstration, of course, cannot be
completely successful, because perfection is far beyond reach: the subject is vast,
the space is limited, and the learning required is ever incomplete.
Finally, it is important to keep in mind that disagreements persist in the
ongoing study of ancient Greek literature, and thus the articles in these nine
volumes necessarily reflect a diversity of opinions. There is ample room for
disagreement even about the merits of representative articles, let alone the choices
of the articles themselves. It is therefore reasonable for each reader to ask, after
reading an article, whether it has indeed been true to the art of philology. The
editor, a philologist by training, has his own opinions about the relative success or
failure of each of the studies here selected. These opinions, however, must be
subordinated to the single most practical purpose of the collection, which is to
offer a representative set of modern studies that seek the best possible readings of
the ancient writings.
Volume Introduction

By the Roman period, as we saw in volume 8, Greek literature had become so

definitive, so obviously classical, that it appears to be "time-free" (see Feeney 1995,
article 1 in volume 8, especially p. 303). With the onset of the Byzantine period of
Greek literature, such appearances are only reinforced. In fact, the first impression
radiating from the sum total of Byzantine literature is that it speaks for itself
perfectly- that it represents a totally self-explanatory cultural system.
A closer look, however, reveals complex interactions among a variety of
styles and registers of expression. Such variety needs to be examined not only
synchronically, that is, in terms of systems of communication functioning within
their own historical contexts, but also diachronically, that is, in terms of systems
evolving through time. The readings in section A (articles 1 through 4) give a sense
of the vast cultural varieties represented by Byzantine literature and of its
connections to previous phases of Greek literature. Of special importance is the
close link between Byzantine literature and the cultural legacy of Late Antiquity- in
particular, the emergence of Christianity as a dominant worldview. Of general
importance is the intimate connectedness of this same literature with antiquity
itself, viewed as a totality (Hunger 1969/1970, article 4). Equally important is the
pervasive interaction of West with East (as represented especially by Egypt and
Syria).
Questions of style and register necessarily engage various cultural
dichotomies, such as low art and high art, standard and substandard, canonical and
apocryphal, classical and popular, oral and written, East and West, religious and
secular, orthodox and heretical. The rich varieties of such cultural constructs are
analyzed in section B (articles 5 through 8, including important papers by Browning
1978 and Sevcenko 1981, articles 5 and 8).
Section C focuses on the most representative genres of Byzantine
literature, such as hagiography (saints' lives), court rhetoric (especially panegyrics),
and bravura descriptions of art.
Section D rounds out not only volume 9 but also this whole set of nine
volumes centering on premodern Greek literature as a notional totality. The articles
in this section explore the interaction of literary productions with their social and
cultural contexts (especially Magdalino 1989 and Mullett 1984, articles 15 and 16),
the emergence of vernacular literature and its rhetorical subtleties (Alexiou 1986,
article 13), and the generic fluidity and indeterminacy characteristic of many
Byzantine texts (Macrides 1985, article 14). The main topic of this last section in
the nine volumes, "Literary Renaissance," is particularly apt, since it leaves the
IX
x Volume Introduction

reader with a simultaneous sense of closure and open-endedness. The focus here is
on a renaissance, dating from the eleventh and twelfth centuries C.E.- well before
the better-known western European models. Some of the genres analyzed in this
section are clearly identifiable in terms of antiquity- but also in terms of
modernity. It is no accident that the crown jewel of genres in this Greek literary
renaissance is itself the ultimate expression of modernity, the novel (Roilos 2000,
article 17). The "novelty" of the Byzantine novel, as an ongoing notional
rediscovery of all Greek antiquity, is symbolic of the renaissance, the eternal
rebirth, of Greek literature.

Further Readings

Alexiou, M. 1977. "A Critical Reappraisal of Eustathios Makrembolites' Hysmineand

Hysminias." Byzantine and Modern Greek Studies 3:23-43.
Browning, R. 1963. "A Byzantine Treatise on Tragedy." In GERAS: Studies
Presented to G. Thomson on the Occasion o/His 60th Birthday, 67-81. Prague.
Galatariotou, C. 1987. "Structural Oppositions in the GrottaferrataDigenesAkrites."
Byzantine and Modem Greek Studies 11:29-68.
Hunger, H. 1981. "The Importance of Rhetoric in Byzantium." In M. Mullett and
R. Scott, eds., Byzantium and the Classical Tradition, 35-47. Birmingham.
Jeffreys, E. 1980. "The Comnenian Background to the Romans dAntiquite."
Byzantion 10:455-486.
Lord, A. B. 1954. "Notes on Digenis Akritas and Serbocroatian Epic." Harvard Slavic
Studies 2:375-383.
Momigliano, A. 1963. "Pagan and Christian Historiography in the Fourth Century
A.D." In A. Momigliano, The Conflict between Paganism and Christianity in the
Fourth Century, 79-99. Oxford.
Nock, A. D. 1925. "Diatribe Form in the Hermetica." Journal o/EgyptianArchaeology
11:126-137. Reprinted in Z. Stewart, ed., Essays on Religion in theAncient
World, 26-32. Oxford.
Roueche, C. 1988. "Byzantine Writers and Readers: Storytelling in the Eleventh
Century." In R. Beaton, ed., The Greek Novel, A.D. 1-1985, 123-133.
London.
Usher, M. D. 1997. "Prolegomenon to the Homeric Centos." American Journal 0/
Philology 118:305-321.
 From Ephrem to Romanos

S.P. BROCK, Oxford

In late antiquity Syria was an area of extraordinary artistic and literary

creativity: the modern visitor stands breath-taken before the ruins, whether it
be of Palmyra, or of Qal'at Sim'an. The literary remains are equally impres-
sive, though less accessible, being written in both Greek and Syriac. The very
fact, however, that the literary culture of Syria in this period was bilingual
perhaps provides a clue towards answering the question, why should Syria
have witnessed this outburst of energy in the creative arts? As far as literature
is concerned, one might suggest that this was at least in part due to the
meeting of, and interaction between, two great literary cultures, Greek and
Aramaic, the latter revitalized by the adoption of Syriac (the Edessene dialect
of Aramaic) as the literary vehicle for Aramaic-speaking Christianity. The
product of this creative activity proved to be particularly influential in two
areas of Christian literature, liturgy and poetry: the Syrian, or "Antiochian"
origins of the great Byzantine liturgical tradition are well known, while on the
Syriac side, liturgical texts in this language reached India (where they are still
in use) and even China. In the case of poetry I have in mind, as far as Greek
poetry is concerned, primarily that which innovates metrically, abandoning
the old classical metres for the new principles of syllabism and homotony:
here it is the great poets of Syria and Palestine, men like Romanos, Cosmas
of Jerusalem, Sophronios, lohn of Damascus and Andrew of Crete (but
originally also of Damascus), on the Greek side, and Ephrem, Jacob of
Serugh, Narsai on the Syriac, whose names and influence stand out in the
annals of eastern christian poetry and hymnography.
In the title of this paper, where our attention will be focussed on poetry, I
have taken Ephrem and Romanos as representatives of Syria's two great
literary traditions, Syriac and Greek. They are appropriate, not only because
they happen to be the finest poets to have emerged from early christianity,
but also because they provide us with a suitable framework of time within
which to work, the fourth to the first half of the sixth century. At the outset it
should be said that I have no intention of offering value judgments on these
two great poets, let alone of indulging in the reverse of Sozomen's cultural
chauvinism by claiming that anything good in Romanos is ultimately due to
Syriac influence. (It will be recalled that Sozomen, faced with the undoubted
fact of Ephrem's immense reputation in his day as a poet, attempted to show
that the very art of poetry had originally been introduced into barbaric Syriac

1
140 S.P. BROCK

by Bardaisan's (probably legendary) son Harmonios) 1. Rather, my aim is to

look at some of the areas where interaction between the two literary tradi-
tions is likely to have taken place. If examples of influence, one way or
another, do turn up, the existence of these should of course not be seen as
detracting from the creative genius of whatever author in whose writings such
influence can be discerned ~ just as no one in their senses would claim that
the fact that Catullus adapted Greek metres and models detracts from his
genius as a poet.
In studying the Syriac and Greek religious poetry of Syria during our
period it is essential, as William Petersen has recently pointed out in his
valuable book, The Diatessaron and Ephrem Syrus as Sources of Romanos the
Melodist 2, to distinguish between poetic form and literary sources. This is an
obvious enough distinction, but, as Petersen has shown in his survey of earlier
scholarship 3, it is astonishing how often it has been neglected in discussions
of the possibility of a Syriac model for the new Greek literary form of the
kontakion. I myself would prefer to add a third element, and distinguish
between metrical form, literary form, and literary motif4. Since I intend to
devote most of my time to literary motifs, I shall deal in a very summary
fashion with the first two.

METRICAL FORM

Attention is normally focussed on the koritakion, a stanzaic verse form

based on principles of syllabism and homoton)', of which Romanos was the
greatest, but not the earliest, exponent. Was this new metrical form based on
a Syriac model, and if so, on what model? The candidates are three, and in all
three the basic metrical principle is that of syllabism; they are: (1) the
madrasha, or stanzaic poem, often with a highly complex syllabic structure;
(2) the soghitha, which properly belongs to the category of madrasha, but
which is characterized by a simple syllabic structure; and (3) the memra, or
isosyllabic couplet, used for narrative poetry. The fact that all three have their
advocates among scholars who have written on the subject suggests at once
that no exact Syriac model in fact exists (though discussion has not always
been helped by various misapprehensions concerning the nature of these three
verse forms, and of Syriac poetry in general). As Petersen points out, by far

1 Sozomen, Ecclesiastical History. JII.I6. On this see my "Syriac and Greek hymnography:
problems of origin", Studia Patris/iea 16 (1985), pp. 77-81.
2 e.S.e.O. 475, Subsidia 74 (1985). See also his "The dependence of Romanos the Melodist
upon the Syriac Ephrem: its importance for the origin of the kontakion", Vigiliae Chrislianae 39
(1985), pp. 171-87. The same point is also made in my "Syriac and Greek hymnography".
J The Dialessaron and Ephrem Syrus, pp. 1-19.
4 Metrical form and literary form can of course govern each other, but as we shall see there
are good practical reasons for making this threefold distinction, even though the three compo-
nents are not on the same hierarchical level.

2
 From Ephrem to Romanos 141

the most satisfactory discussion of the matter remains that of Paul Maas, in
an article published in 1910 5 ; his conclusion is one with which I would fully
agree: the kontakion must be ultimately inspired by a Syriac model, and this
model is undoubtedly the madrasha; but at the same time the model is
certainly not taken over slavishly: it has been adapted to the Greek language
in a number of different ways, most notably by the introduction of homo-
tony, a feature absent from its Syriac model.
Before leaving the question of poetic form, it is worth mentioning that a
clear case for a more direct borrowing can be made out in the case of the
Greek isosyllabic kala slichon verse. Though there is no evidence for direct
translations of Ephrem's madrashe, or hymns, into Greek, some of his memre,
narrative poems in the 7 + 7 syllable metre, were translated into Greek,
probably at an early date, and for these the translator simply took over the
isosyllabic metre of the original. Many such isosyllabic pieces are to be found
in the tangled corpus of works which go under the name of Ephrem
Graecus 6, and it is evident that in many cases these are original Greek
compositions, or at least free adaptations; we do, however, have one text, on
the prophet Jonah, where the isosyllabic Greek is an exact translation of an
ext.lnt memra whose attribution to Ephrem seems to be reasonably assured 7.

LITERARY FORM

A characteristic feature of much poetry in both Greek and Syriac during

our period is the use of dialogue 8; it is equally a feature of several prose
homilies in the two languages. This dialogue can take a number of different
forms, and the adoption of one particular form may govern the poetic form
employed. Thus, strictly formal dialogue where the speakers are allocated
alternate couplets, is virtually confined to the soghitha. that is, stanzaic poetry
with a simple isosy\labic structure. This particular literary form happens to be
a favourite in Syriac, and a large number of splendid dramatic dialogues
survive, the best of which certainly belong to our period, although they are
usually anonymous and so not exactly datable 9 .
The Syriac dialogue poems normally have a short narrative introduction.
after which the dialogue begins, taking the form of an alphabetic acrostic:

5 P. Maas, "Das Kontakion". By:antinische ZeilschriJl 19 (1910), pp. 285-306, reprinted in his

Kleine SchriJlen (ed. W. Buchwald; Munich. 1973), pp. 368-91.

6 See especially the texts edited by S.G. Mercati, S. Ephraem Syri Opera, I. Sermones in

Abraham el Isaac, in Basilium Magnum, in Elialn (Rome. 1915) .

• Greek text ed. D. Hemmerdinger lliadou. "Saint Ephrem Ie Syrien. Sermon sur Jonas", Le
.\.{useon 80 (1967). pp. 47-74: Syriac text ed. E. Beck. Des heiligen Ephraem des Syrers, Scrmones
II (C.SC.O. 311-2. SeT. Syri 134-5; 1970), no. I.
8 For a more detailed discussion. with a proposed five-fold typology. see my "Dramatic

dialogue poems". Symposium Syriacum IV (O.CA. 2~9. 1987).pp. 135-47,

9 See my "Syriac dialogue poems: marginalia to a recent edition", Le .Huseon 97 (1984),

pp. 29-58,

3
142 s.P. BROCK

each letter of the alphabet has two verses, one for each speaker. A characte-
ristic feature is that the second speaker often picks up the opening words of
the first speaker, but then develops them in a contrastive manner. Dialogue
poems with this strictly formal arrangement are only rarely found in Greek.
Notable examples are an early kontakion on Elijah and the widow of
Sarepta 10, and two pairs of dialogues, between the Virgin Mary and the
Angel, and between Mary and Joseph, which are incorporated into Homily 6
attributed to Proclus, and in another homily attributed to Germanus II. None
of these is a direct translation from Syriac, and it remains to be seen whether
any borrowing of literary motifs from Syriac dialogues on the same topics
exists, or not. But the similarities in literary form here are so great that some
sort of relationship seems inevitable. One consideration in particular suggests
that the literary form has been borrowed from Syriac by Greek writers: the
genre in Syriac is already found in a few of Ephrem's hymns in the fourth
century, and has obvious roots in the ancient Mesopotamian contest poem,
whereas in Greek, while strophomythia per se is well known from classical
Greek drama, there are no good precedents for the manner in which it is
employed in these alphabetic texts.
There are of course other ways in which dialogue can be used in poetry,
and if we compare what happens in the two languages some interesting facts
emerge 12.
Instead of being in alternate stanzas the dialogue can be arranged in
uneven blocks and incorporated into a narrative framework. This manner of
handling dialogue is found in a few of Ephrem's madrashe (dialogues between
Death and Satan), and there is one memra, by Jacob of Serugh, where the
Church and the Synagogue dispute in this way. This type of poetic dialogue,
however, appears to lack any exact correspondences in Greek sources of the
period.
In a third type of dialogue poem the dialogue is incorporated into a bare
narrative skeleton. Most of Ephrem's madrashe containing dialogues between
Death and Satan are of this type, and it also occurs in a few soghyatha; but
no memre exist having this structure. In Greek it is found in a number of
Romanos' kontakia, including that on the Cross, where the theme is that of
the Descent into the Underworld, as in Ephrem's madrashe just mentioned.
A fourth type of dialogue poem consists of narrative into which speeches
and dialogue are inserted. Homiletic material, if present at all, is confined to
the opening. In Syriac this is characteristic of several memre, and there is at
least one prose text; it is never found in madrashe. Turning to Greek, we

10 Ed. C.A. Trypanis, Fourteen Early Byzantine Cantica (Vienna, 1968), no. 8 (cp. also
P. Maas, Fruhbyzantinische Kirchenpoesie (Kleine texte 52/53; 1910), no. 3).
11 C.P.G. 5805, 8009.
12 The following is based Oil my "Dramatic dialogue poems".

4
 From Ephrem to Romanos 143

again find several examples among Romanos' kontakia (e.g. that on the
prophet Elijah), and it also turns up in isosyllabic couplets, in various pieces
concerning biblical characters to be found in the corpus of Ephraem Graecus
- none of which, however, appears to be a direct translation from Syriac.
A fifth type again consists of narrative with interspersed speeches, but this
time a considerable amount of homiletic material is interspersed. In Syriac
this is typical of several memre, or verse homilies, by Narsai and Jacob of
Serugh (i.e. late fifth and early sixth century), and also of a few prose texts. In
Greek, there are again counterparts among Romanos' kontakia (e.g. that on
Adam and Eve), and of course it is characteristic of many prose homilies by
Basil of Seleucia, John Chrysostom and others.
If we set out these observations in tabular form we will discover why, in the
past, scholars have been so uncertain which of the Syriac verse forms to select
as a possible model for the kontakion 13:
- in the case of Greek kontakia of type I, the alphabetic poem with
dialogue in alternating stanzas, the Syriac counterpart is the soghitha;
- in the case of kontakia of type III, it is the madrasha that provides the
Syriac counterpart;
- in the case of kontakia of types IV and V, it is the memra which
provides the Syriac counterpart.
Almost certainly we should see a chronological development here: the
earliest kontakia correspond most closely to the madrasha or soghilha, not
only in the metrical form (as Maas indicated), but also in the way they handle
dialogue. From these beginnings the kontakion then developed and began to
handle dialogue in other ways as well, incorporating it into a narrative
framework - corresponding to our types IV and V, both characteristic of the
Syriac memra. It is hard to say with certainty whether this development in the
kontakion was an independent inner-Greek one, or whether it was (in part at
least) due to the growing popularity of the memra as a vehicle for narrative
poetry in Syriac during the latter half of the fifth century. We would certainly
be more inclined to accept the possibility of the influence of the memra in this
matter of literary form if we could pin-point examples where the author of a
kontakion clearly took over literary motifs from Syriac memre of this kind.
This brings us to our third element, literary motif.

LITERAR Y MOTIF

In a bilingual culture such as must have existed in Syria in late antiquity we

should not be surprised to find literary motifs crossing linguistic boundaries
without necessarily having actual translations to provide the vehicle to convey
the motif from the one language to the other. Thus Ephrem alludes to

13 This shows the importance, for our purposes, of distinguishing between poetic form and
literary form.

5
144 s.P. BROCK

episodes in Greek mythology on a few occasions in his Hymns 14, but his
knowledge of these by no means necessarily implies that he was basing
himself on translations of Greek texts mentioning these episodes. Likewise
the various Jewish haggadic traditions of which he makes use will probably
have reached him at an oral, and not a written, level.
Even a fairly superficial reading of the Greek and Syriac poetic output in
Syria during our period will throw up a large number of literary motifs and
themes handled in very similar ways in the two languages. Often it will be
very difficult, if not impossible, to discern how these have travelled from one
language to the other, and in which direction this has taken place. There
remains scope for much fruitful work in this area, and future investigation
will no doubt clarify the situation considerably: unfortunately most of the
research that has taken place so far on this topic has either been rather
superficial, or limited to only part of the evidence available. An exception is
provided by Petersen's careful collection of 21 instances where Romanos
appears to base himself directly on passages in Ephrem's writings. His basic
case, that Romanos is borrowing from Syriac sources, is not necessarily
weakened by the fact that some of his 21 instances represent commonplaces
in early Syriac poetry and thus need not represent direct borrowings from
Ephrem himself.
Since no one would deny that Syriac authors, both poets and prose writers,
readily made use of Greek literary motifs, I will concentrate here on the
possibility (I would say probability) of the transmission of literary motifs in
the other direction, from Syriac to Greek. A number of general areas seem to
me promising and deserving of detailed investigation; these concern certain
biblical episodes which clearly caught the imagination of early Syriac writers,
and in many cases specifically Syriac developments can fairly readily be
discerned, even though the theme itself may also have been popular with
Greek authors writing in Syria. These general areas are: the episodes of Cain
and Abel (Gen 4), Abraham and Isaac (Gen 22), Elijah and the Widow of
Sarepta (I Kgs 17), Mary and the Angel (Luke I), Mary and Joseph
(Matthew I), the Sinful Woman who anointed Christ (Luke 7), and the
Descent of Christ into Sheol. in several cases some of the Syriac evidence still
awaits publication. Here it will be practicable only to look at one of these
themes in detail and I select the episode of the Aqedah, or Sacrifice of Isaac,
since some new Syriac texts on this episode, in both prose and verse, and
containing relevant material, have recently been published.
The stark, one might almost say apophatic, biblical narrative of Gen 22
invited elaboration, and over the ages writers (and for that matter, artists),
Jewish, Christian and Muslim, have not failed to take up the challenge. From

\4 Hymns on Paradise III.8 alludes to Tantalus, and Carmina Nisibena XXXVI.5 alludes to

Orpheus.

6
 From Ephrem to Romanos 145

Greek and Syriac writers of the fourth to sixth century we have quite an
impressive response, some nine texts in Greek, of which the kontakion by
Romanos is perhaps the most dramatic, and in Syriac six texts, all but one in
verse (in Greek only Ephrem Graecus and Romanos are in verse) 1 5. Roma-
nos' kontakion is probably the latest in time of all these texts, and so it will
usefully serve as our starting point.
The first fifteen stanzas (there are twenty four in all) are devoted to
Abraham's reaction to God's initial command; this takes two forms, hypo-
thetical, and actual. First comes the hypothetical speech, describing what one
might have expected Abraham to have said in reply. Romanos is here making
use of a well-known rhetorical technique, known as ethopoeia, which had
already been applied to this episode by earlier Greek homilists such as Basil
of Seleucia and Ps. Chrysostom; it is a technique also found in some of the
Syriac texts as well - a clear case of borrowing. In the course of this
hypothetical speech of Abraham Romanos introduces a second hypothetical
speech, to illustrate what Sarah might be expected to have said, were he to tell
her what he was going to do with Isaac. This transition from one hypothetical
speech to the other occurs in stanza 7 with the following words put in
Abraham's mouth: "Sarah will get to hear of all your words, Lord, and when
she knows of your purpose she will say ... ". In introducing Sarah into the
episode (the biblical text is entirely silent on this matter) Romanos was again
simply following the precedent of earlier writers - both Greek and Syriac -
who were curious to know how Abraham managed to extract the child Isaac
from his mother's care. The hypothetical speech put into Sarah's mouth runs
on for five stanzas, and her reaction is what one might expect a mother's to
be, and Romanos makes a number of clear allusions to the hypothetical
speech put into Sarah's mouth in the poem on Abraham and Isaac under the
name of Ephrem Graecus (of which no known Syriac original exists, and
indeed it may not be a translation at all). Then, in Romanos' stanza 12, we
encounter an extraordinary transition: what Abraham might have been
imagined to suppose that Sarah might have said, if he had told her of God's
command, suddenly becomes as it were reality, and Abraham starts to rebuke
Sarah:
Do not use words like that, woman, you will anger God:
He is not asking us for anything that does not belong to him,
for he is simply taking what he earlier gave us.
Do not spoil the sacrificial offering with your lamentations;
do not weep, otherwise you will put a blemish on my sacrifice.

Though this dramatic transition from hypothesis to reality is an innovation

on Romanos' part, the content of what he says when he rebukes Sarah is
again basically in harmony with speeches attributed to him in the earlier

15 The texts are listed in Le Museon 99 (1986). pp. 66-7.

7
146 S.P. BROCK

Greek homiletic tradition. The reply that Romanos now puts into Sarah's
mouth, in stanza 14f, is by contrast totally without precedent in the earlier
Greek tradition: not only is Sarah aware of what Abraham is intending to do
with her son (in the earlier homiletic tradition Abraham sometimes hides his
true purpose from her), but she actually gives her assent, thus putting her
faith in, and love for, God on a par with Abraham's. Her words are adressed,
not to Abraham himself, but to Isaac:
If God desires you for life, he will give orders that you live;
he who is the immortal Lord will not kill you.
Now I shall boast: having offered you as a gift
from my womb to him who gave you to me, I shall be blessed.
Go then, my child, and be a sacrifice to God,
go with your father - or rather your slayer.
But I have faith that your father will not become your slayer,
for the Saviour of our souls alone is good.
Romanos' very positive approach to the figure of Sarah in the second half
of the kontakion contrasts dramatically with the way she is treated in the first
half, and those scholars who have made a search for Romanos' sources in
this homily, Nikolopoulos, Moskhos and Grosdidier de Matons 16, have all
found the same thing: in the first half Romanos clearly draws on the earlier
Greek homiletic tradition for his general approach, though he has handled
these sources in his usual creative manner: it is a matter of verbal allusion,
and not of direct borrowing. For the second half of the kontakion, however,
where Sarah shares willingly in the offering up of Isaac, no Greek source at
all provides anything similar: the nearest is a homily attributed to Amphilo-
chius and preserved only in Coptic 1 7 ; here Abraham decides not to tell Sarah
his true purpose (he imagines what she would say), and he leaves her to
suppose that they are going off to make an ordinary sacrifice. Sarah
accordingly sends Isaac off willingly, having first instructed him how to
conduct such a sacrifice; her words to him are heavily loaded with prophetic
irony.
It is of course possible that Romanos, knowing this unusual treatment of
the theme by Amphilochius, himself took it one step further. But this is to
ignore the fact that he was a native of Horns in Syria, and so is very likely to
have benefitted from that region's cultural bilingualism. Now if we turn to
the Syriac literary tradition we find that the second half of Romanos'

.6 N.G. Nikolopoulos, 'Elti "ta~ 1tTlYa~ tOU EI~ t~Y Gllo-iaY tOU A~paall UIIYOIl ProliayOu tOU
MEAqX)OU, 'AG1]vii 56 (1952), pp. 278-85; M. Moskhos, "Romanos' hymn on the Sacrifice of
Abraham: a discussion of the sources and a translation", Byzantion 44 (1974), pp. 310-12;
J. Grosdidier de Matons, Romanos /e Me/ode. Hymnes, I (Sources chretiennes 99; 1964), pp. 129-
35.
17 Ed. L. van Rompay, in C. Datema, Amphilochii Iconiensis Opera (Corpus Christianorum,

series graeca 3; 1978), pp. 274-303.

8
 From Ephrem to Romanos 147

kontakion happens to have a close parallel in an anonymous verse homily

probably of the late fifth century 18. Before turning to this we should note
that the seeds for such a positive handling of Sarah had already been sown in
Ephrem's Commentary on Genesis, where he simply says 19:
Abraham did not reveal the matter to Sarah since he had not been ordered to reveal
it; but had he done so, she would have been beseeching him that she might go and
share in his sacrifice, just as he had made her share in the promise of Isaac's birth.

Building on this, the anonymous Syriac homily provides exactly the same
development that we find in the second half of Romanos' kontakion: Sarah is
not only aware that Isaac is to be sacrificed, but she sends him off willingly.
Not only do we have this general, but very striking, parallel between the two
texts, but, as we shall see, there are also some remarkable similarities in
details of phraseology.
The Syriac homily in question survives in a single manuscript, and like
many verse homilies in the 7 + 7 syllable metre it is wrongly attributed to
Ephrem (whose name is traditionally given to that metre). It cannot possibly
be his, since it draws quite heavily in places on another interesting anony-
mous homily which is itself certainly later than Ephrem 20, Since I shall be
quoting from both homilies, I shall call the earlier one Memra (verse homily)
I, and the later one, with the very favourable treatment of Sarah, Memra II.
Memra II has an added interest in that it is one of the very few pieces of
Syriac literature for which there is some evidence that the author was a
woman; this evidence (unfortunately not entirely unambiguous) hangs on a
grammatical form where the narrator speaks in the first person. The great
prominence given to Sarah in the whole poem might well lend some further
support, for Sarah in fact is made to undergo two trials, to Abraham's one,
and it is she who emerges as the true heroine of the ordeal.
Before we look in detail at the treatment of Sarah in Memra II, a notable
feature of both Memras should be mentioned: Isaac and Abraham bring back
from the sacrifice the ram's fleece. One wonders whether this might not be a
distant reminiscence of the Greek legend of the Golden Fleece, and there
happens to be a curious - but probably purely fortuitous - similarity
between the description of the wonderful fleece in Memra I (line 143) and
that given by Apollonius Rhodius of the Golden Fleece in Argonautica IV,
lines 184-5. The introduction of the motif of the fleece into the episode of
Gen 22 appears to be unique to these two Syriac poems.
How, then, do these two Syriac Memre handle the person of Sarah in their
retelling of the biblical drama? Clearly the question "Where was Sarah" was

18 Memra II in my "Two Syriac verse homilies on the binding of Isaac", Le Museon 99

(1986), pp. 61·129.

19 Ed. R. Tonneau (C.S.C.O. 152·3, Scr. Syri 71·2; 1955), section XX,

20 Memra 1, also edited in my "Two Syriac verse homilies".

9
148 S.P. BROCK

an important one for all the authors. Greek or Syriac, who wrote of the
theme of Gen 2221. In contrast to the majority of Greek homilists, who offer
examples of what Sarah might have said, if Abraham had told her of his true
intention, in Memre I and II we are provided with actual speeches by
Abraham and Sarah. Memra I provides quite a long exchange between
Abraham and Sarah (some forty lines. out of a total of 187 lines in the
poem). Terrified on seeing them about to go off, she exlaims:
Where are you taking my only-begotten ... reveal to me the secret of your
intention ... why are you not revealing your secret to Sarah your faithful wife
who in all the hardships of exile has borne trials along with you 0
To which Abraham evasively replies:
I wish to slaughter a lamb and offer a sacrifice to God.
At the fleece which will come back with us you will givc praise to God all the
more.
Sarah urges her husband to leave Isaac behind:
You are drunk with the love of God - who is your God and my God - and if
he so bids you concerning the child, you would kill him without hesitation.
Abraham tells her not to worry: the fact that he is taking two young men
with him should suffice to allay her fears that he might do something rash.
This persuades Sarah, and so Abraham and Isaac set off.
In Memra II the dialogue between Sarah and Abraham is shorter (just
under 30 lines). When Sarah sees Abraham bringing out a knife to take with
him,
... her heart groaned, and she began to speak to Abraham:
Why are you sharpening your knife0 What do you intend to slaughter with
it? Why have you hidden this secret from me today?
Abraham replies, with characteristic male condescension:
This secret today - women cannot be aware of.
Sarah then starts to remind him of their joint response on an earlier occasion,
when they gave hospitality to the poor - who turned out to be angels (Gen
18). Her speech continues with words borrowed directly from Memra I, but it
develops in a very different way:
You are drunk with the love of God, who is the God of gods, and if he so bids you
concerning the child you will kill him without hesitation: let me go up with you to the
burnt offering, and let me see my only child being sacrificed. If you are going to bury

21 For further details see my "Sarah and the Aqedah", Le Museon 87 (1974), pp. 67-77
(written before I had come across Memre I and II), "Genesis 22: where was Sarah?", Expository
Times 96 (1984), pp. 14-17, and the discussion in "Two Syriac verse homilies", pp. 70-76.

10
 From Ephrem to Romanos 149

him in the ground, I will dig the hole with my own hands, and if you are going to
build up stones, I will carry them on my shoulders; the lock of my white hairs in old
age will I provide for his bonds. But if r may not go up to see my only child being
sacrificed, I will remain at the foot of the mountain until you have sacrificed him and
come back. (Lines 23-30).

Sarah then turns to Isaac and instructs him how to behave when he is being
sacrificed, after which she embraces him with tears, takes him by the hand
and hands him over to Abraham.
If one considers the homiletic tradition on this theme as a whole in this
period, Syriac as well as Greek (and one could add the Jewish tradition as
well) Memra II and Romanos stand out from all the other texts in having two
striking features in common in their handling of the biblical narrative at this
point: in the first place, Sarah sends Isaac off willingly, in full knowledge of
what Abraham is about to do; and secondly Sarah speaks directly to Isaac at
this point, something she is made to do elsewhere only in the homily
attributed to Amphilochius(where, however, as we have seen, she believes
that Abraham and Isaac are going off to sacrifice an ordinary lamb).
These two features alone seem to me to be sufficient evidence to suggest
that there must be some connection, indirect if not direct, between the second
half of Romanos' kontakion and Memra II - and perhaps we should also
add "Amphilochius". Before trying to specify the nature of these connections
any further, however, some chronological uncertainties need to be mentioned.
If the homily attributed to Amphilochius were to be genuine, then it would
date from the late fourth century; the fact that it survives only in Bohairic,
however, introduces an element of doubt, given the large number of mis-
attributions given by Coptic scribes to texts translated from Greek. There are
sufficient links between it and Memra II to make it likely that the author of
whichever proves to be the later text knew the other. Unfortunately we have
no means of giving a precise date to Memra II, and it is only on rather
general grounds that one can suggest a late fifth century date: it could
conceivably be somewhat later, and one should not entirely rule out the
possibility that it is later than Romanos, and in that case theoretically it
could be dependent on him, rather than the reverse relationship. Literary
borrowing by a sixth-century Syriac writer, probably Syrian Orthodox, from
a Melkite Greek writer is, however, intrinsically unlikely, and certainly there
is no other evidence at all for a knowledge of Romanos by Syriac writers of
that period. That Memra II is in fact earlier than, and was directly known to,
Romanos is suggested by a quite different point.
The study of Romanos' Greek sources has shown that we do not ever find
him making any verbal quotations: the most we have is verbal allusions. This
means that if we are to extend our search to Syriac sources, as I believe we
must in the light of Memra II, then we should not expect to find any verbal
quotations there either, only allusions. A possible example of such an allusion

11
150 S.P. BROCK

is provided by another very unusual feature found only, among Christian

writers, in Memra II and in Romanos. In stanza 19 of his kontakion
Romanos has Abraham say just before the sacrifice is to take place, "First I
will bind him, then I will slay him, lest his quivering (skirtema) hinder my
intention". Concern lest some involuntary movement on Isaac's part should
invalidate the sacrifice is not found in any other Greek source; it does,
however, feature in Memra II, though there it is Isaac who brings up the
matter, in a speech to his father at this point:
Draw near, father, and bind me; tie tightly for me my bonds lest my limbs should
shake, and there be a blemish in your sacrifice.

This may seem a small detail, but it is a very telling one. This particular
concern over involuntary movement is purely a Jewish one, and it is surprising
to find it at all in a Christian handling of the theme 22 • Thus in the Palestinian
Targum to Gen 22: 10 Isaac says to his father at exactly this point "Bind my
hands well so that I may not struggle at the time of my pain and disturb you
and so render your offering unfit". Likewise in Bereshith Rabba 56:8 Rabbi
Isaac is recorded as having said:
When Abraham wished to sacrifice his son Isaac, the latter said to him, I am young
and afraid that my body will tremble through fear of the knife and I will grieve you,
for thereby the ritual slaughter may be rendered unfit, and this will not count as a real
sacrifice; therefore bind me very firmly.

There can be no doubt, I think, that in our two Christian texts, Memra II
and Romanos, we have an echo of a concern that was of great importance to
the Rabbis - though of none at all to Christians. At this point it will no
doubt be recalled that Romanos was of Jewish origin, a detail furnished by
the kontakion written in his honour. This information should not, however,
lead us to jump to the conclusion that Memra II must be the borrower, for
one good reason: Memra II represents Jewish tradition more closely than
does Romanos, for in the Jewish texts and in Memra II it is Isaac who asks
for his father to bind him firmly lest he (Isaac) invalidate the sacrifice,
whereas in Romanos the concern about involuntary movement is put into
Abraham's mouth.
These two distinctive features common to Memra II and Romanos -
Sarah's willing sharing in the sacrifice, and the Jewish concern over involuntary
movement - strongly suggest that Romanos either knew Memra II itself, or
failing that, a lost Syriac homily also attesting these two highly atypical
features which set apart these two texts from all the other homiletic treat-
ments, Greek and Syriac, of this episode.
If one accepts the strong likelihood that Romanos is borrowing from a
Syriac source for these two features, then the field is open to the search for

22 Cp "Two Syriac verse homilies", pp. 89,94.

12
 From Ephrem to Romanos 151

other Syriac sources which he may also have known. As far as this particular
kontakion on Abraham and Isaac is concerned, several such cases of the
possible borrowing of motifs can indeed be found, as I have indicated
elsewhere 23 •
By way of conclusion it is worth making a more general observation. There
has perhaps been a tendency on the part of some scholars to think -
consciously or unconsciously - of the Greeks writers of Syria as if they were
emigre Greeks - rather like the British in India, none too observant of the
native literary culture around them 24. This is certainly not a satisfactory
conceptual model: we should instead think of these writers as Syrians of
Greek literary culture, living side by side, and interacting intellectually, with
Syrians of Syriac literary culture. Here one might compare the situation in
modern Syria and Lebanon where there have been authors whose literary
culture and language has been essentially French or English, but who have
nevertheless been fully aware of trends in Arabic literature, and so open to
influence from that quarter. If we adapt this second conceptual model to the
way in which we look at Syria in late antiquity, then we will be far more
likely to accept the possibility that creative writers like Romanos were
conscious of what was going on in contemporary Syriac literature. Further-
more, we need to recall in this connection the great prestige that Syriac poetry
had in the eyes of several fifth-century Greek writers 2 5.
Whether or not the particular examples given here carry conviction,
sufficient has been said to make it clear that one needs to take Syriac
literature - and especially Syriac poetry - into consideration if one is
properly to appreciate the full richness of Syrian literary culture as a whole in
late antiquity.

23 See "Two Syriac verse homilies". pp. 91-6.

24 Thus Grosdidier de Matons in his Romanos Ie },felode el les origines de la poesie religieuse a
Byzance (Paris, 1977) was unduly sceptical about the possible Syriac background to Romanos; on
this point see the review by A. de Halleux, "Hellenisme et syrianite de Romanos Ie Melode", in
Revue d'HislOire Ecc/esiaslique 73 (1978). pp. 632-41.
21 See the eulogy of Ephrem in Sozomen, Eccl. Hisl. 111.16; cpo Theodoret, Eccl. Hisl. IV.29.

13
 WANDERING POETS:
A LITERARY MOVEMENT IN BYZANTINE EGYPT

In the later Roman Empire Egypt, not for the first time in its history,
became the home of Greek poetry. "The Egyptians", wrote Eunapius of
Sardis in about the year 400, "are mad on poetry" ('e:7tt 7tOL'Y)'t'LXn [J.E\I mp6~pcx
[J.CX[\lO\l't'c(1 ), and indeed it is hardly possible to name a single prominent
Greek poet of the fourth and fifth centuries who was not either an Egyp-
tian, or else, like the Lycian Proclus, educated in Alexandria. 2 Yet modern
scholars, while studying in some detail its more important representatives,
have devoted surprisingly little attention to the movement as a whole. 3 The
Dionysiaca of Nonnus is familiar to every student of Greek poetry (even if few
can claim to have read all its 48 books), the hymns of Synesius 4 and the epigrams
of Palladas 5 have been much studied in recent years, and even the lame epics
of Triphiodorus and Colluthus still find an occasional reader. But Nonnus and
Synesius are not really the most typical representatives of the movement, and
Palladas by any reckoning is an isolated figure. Most of their fellow-countrymen
put their talent for poetry to a much more practical use, and though men still

1 Vito Soph. 493.

2 Egyptian domination was spent by the sixth century (below p. 508) when of the
numerous poets who flourished at the court of Justinian, only Julianus Aegyptius seems to
have been an Egyptian - though Agathias received his literary education at Alexandria
(Hist. ii 15, p. 97. I Bonn). P. Waltz (Anth. greeque i [1928J xxiii) suggested that the Cyrus
attested as the father of Paul the Silentiary is none other than Cyrus of Panopolis, and this,
if true, would provide an interesting and important link between the two schools. Unfortu-
nately chronology will hardly permit: Paul was still alive and writing in 567 (date of A P
ix.658), whereas we are told that Cyrus lived up to the reign of Leo (Lite of Daniel the
Stylite § 31), which implies that he did not outlive Leo, who died in 474. Even granted that
Paul was a child of Cyrus' extreme old age he would still have been nearly roo when he
wrote his poem. And furthermore Agathias' reference to Paul's wealthy and distinguish-
ed ;;POYOVOL (v. 9, p. 297. 3) hardly fits an Egyptian adventurer who died without a penny
(below p. 473).
3 It is mentioned in passing, e.g., by l\Iaspero, REG xxiv (19II) 458, Friedlaender.

Hermes xlvii (1912) 56, Bury, Later Roman Empire ii 2 (1923) 431 n. 3. Thraede has virtually
nothing to say on the subject in his article 'Epos' in RA C v (1962) 992-3: Christ-Schmid-
Stahlin, Geseh. d. griech. Literatur ii. 2 6 (1924) 956f. give no more than a list (albeit a
valuable, though not exhaustive, list) of names and references.
4 C. Lacombrade, Synesios de Cyrene (1951) ch. xiv and works there cited, especially the
full edition with commentary by Terzaghi in 1939.
5 See most recently my articles in JHS 1964, JRS 1965 and CQ 1965, and the various
articles of Bowra and Irmscher there cited.

14
 Wandering Poets: A Literary Movement in Byzantine Egypt 47T

continued to call poetry a divine gift, and to invoke the Muses to assist them in
their song, the truth of the matter was that for most poets of the later Empire
poetry was no more and no less than a profession. 6 It is these professional poets
whom I propose to discuss. Not, however, at any rate principally, from the
literary point of view. This would hardly be possible, for the simple reason
that with one notable but slightly atypical exception barely a line of their
numerous and voluminous works has survived. The exception is Claudian of
Alexandria - atypical because his extant works are written in Latin. 7 But
paradoxically enough the fact that the work of these poets has so utterly
perished is of less importance than might at first sight appear; it is not likely
that their literary merit was of a very high order, and their significance, for the
modern scholar at any rate, lies elsewhere, in the light their existence and
activity cast on the social, cultural and even political life of the later
Empire.
From scattered allusions in the correspondence of contemporaries and the
compilations of Byzantine savants of later ages,S it is possible to recon-
struct a fascinating picture of a regular school of poets, born and educated
in Egypt, who spent their lives wandering from city to city·throughout the
Empire in search of fame and fortune. The term 'school' may, I think, be
justified by a number of interesting and important features they have in com-
mon. In a Christian world they were almost all professed pagans, unlike a
Nonnus or Musaeus they all wrote on contemporary subjects, in an age of
regimentation they travelled freely and widely, they were all scholars and many
knew Latin (a rare accomplishment for a Greek of any age), and, not least, so
far from living the usual life of the poet in peaceful seclusion, they were often
shrewd and worldly adventurers, equally proficient in the very different arts of
poetry and politics. I propose to consider each of these features in turn.

I
The fourth century saw the triumph of Christianity and the decline and
eventual proscription of worship of the old gods. In the closing years of the
century Theodosius the Great forbade under the direst penalties worship of

8 Hence the circle of Agathias and Paul the Silentiary, few if any of whom were profes-

sional poets, would fall outside the scope of this study even if they had been Egyptians
(n. 2 above).
7 Of his Greek works we have only a few epigrams dubiously ascribed to him (d. W.
Schmid, RAG iii [1957J 168) and a fragmentary Gigantomachia (see N. Martinelli Misc.
Galbiati ii [1951J 47f., B. Lavagnini, Aegyptus xxxii [1952J 452f.).
8 The references to Photius are all to the Bibliotheca, and I have found it more con-

venient to refer simply to 'Suidas', than to 'the compiler(s) of the Suda'. Most of the
relevant entries in Suidas derive from the biographical dictionary of Hesychius of Miletus,
some from Damascius' Lite of Isidore.

15
 472 ALAN CAMERON

pagan deities in every fonn,9 and fanatical monks joined in bands all over the
East to destroy and loot pagan temples. Io Yet paganism lingered on, especially in
the intellectual and university circles of Alexandria, where the teaching
of the classics and neoplatonism continued in pagan hands until the sixth
century.ll But these poets were not born or even educated in Alexandria alone;
many, indeed most, of them came from the Thebaid, in particular from the
district art;>und Panopolis. Pamprepius, Cyrus, Nonnus and Triphiodorus all
came from Panopolis itself, Horapollon from Phenebith, a village in the nome
of Panopolis, Christodorus from Coptus, Olympiodorus from Egyptian Thebes
and Andronicus from Hennupolis. Of the more important figures only Claudian
and Palladas are actually attested as Alexandrians, though doubtless most of
the others will have spent a tenn or two at least in the lecture rooms there.
Now Panopolis had always been a centre of Hellenic culture: Herodotus
noticed that even in his day it was the only city in Egypt which favoured Hel-
lenic customs,lZ and it continued a stronghold of 'Hellenism' (in the meaning
'paganism' the word bore from the mid-fourth century on) until well into the
fifth century.I3 For the well-to-do landowners of Upper Egypt remained
predominantly pagan until the middle of that century, when they eventually
capitulated before the vigorous evangelism (not to mention strong arm tactics)
of that resolute and uncompromising abbot Schenute of Atripe.u It is precisely
from this class that our poets will have come, for the education necessary to
produce a poet in the later Empire was beyond the means of any but the fairly
comfortably Off.16 Hence it is not surprising that in spite of the fact that Egypt
was at this period a land of monks,!' young men from well-to-do families in the
9 See most recently N. Q. King, The Empel'OY Theodosius and the Establishment of
Chf'istianity (1961).
10 For a convenient table of the terrorism of the monks see Fliche-Martin, Hist. de
l'Eglise iv (1937) 19f.
11 The most important document for paganism in fifth century Alexandria is Zacharias
Scholasticus' Lite 0/ Sevel'us. See also J. Maspero, Bull. de l'Inst. /I'any. d'arch. orient. du
Cail's xi (1914) 176f.
12 ii. 91 (then known as Chemmis).

13 E. Amelineau, Geographie de l'Egypte a l'ipoque copte (1893) 18 and see particularly

Remondon in Bull. de l' inst. jl'an y. d'arch. oyient. du Cail'e Ii (1952) 67f.
14 J. Leipoldt, Schenute von Atripe und die Entstehung des nationaliigyptischen Chris/en-

tums (Texte und Untersuchungen N.F. x [1904]) 27, 175f.

Ii Cf. A. H. M. Jones, Later Roman Empil'e ii (1964) 1001 - though education in Egypt
was by no means confined to the positively wealthy (cf. Sir H. I. Bell, Egypt [1948] 82f.),
and the fact that many of these poets occupied municipal chairs as grammatici (below
p. 49If.) suggests thattheycame from the lower rather than the higher stratum of this class.
On the social origins of the professors of Bordeaux, some of them actually lower than curial,
cf. M. K. Hopkins, CQ n.S. xi (1961) 246--8. The only poet whose social class is actually
attested is Andronicus, who was a curial of Hermupolis (Photius Cod. 279 fin.).
18 There were 30,000 monks and nuns in Oxyrhynchus before even the end of the fourth

century: cf. H. Lietzmann, HistOt'y 0/ the Early Church iv (Eng\. Tr. 1950) 148.

16
 Wandering Poets: A Literary Movement in Byzantine Egypt 473

Thebaid who went to be 'finished off' in Alexandria after a grounding in the

local schools 17 ne,'ertheless remained pagans,
According to Malchus of Philadelphia (probably his contemporary) Pam-
prepius made no secret of his paganism (TO 'EMl]VtxOV T~C; 8p,t;v/.do:c; (J.e:T<X
r:o:ppYjcr[o:C; 7tpoo'~),WC; Oe:tXVUfLe:vov), and was suspected of practising magic;18
he even tried to stir up the pagan circles of Alexandria to rebel against the
Emperor, producing oracles and prophecies that foretold the imminent collapse
and extinction of Christianity. Years later the pious Paralius of Aphrodisias
tried to wean his pagan brothers from their astrology by holding up Pamprepius
to them as the classic example of a sorcerer led to his doom by his own false
oracles,19
Pamprepius' fellow-citizen Cyrus, carried by his talents to the highest
positions in the state (see below, p, 498), was accused of paganism at the
height of his career and deposed from office (he was then consul), Cyrus'
paganism was probably not in fact the chief reason for his disgrace; it is more
likely that it served as a convenient pretext to cover up the real motive, the
jealousy of the Emperor Theodosius II at Cyrus' dangerous popularity with the
people, who chanted all day long in the Circus "Constantine founded the city,
Cyrus rebuilt it" (KwvcrTO:VT~VOC; ~XTtcre:, Kupoc; &Ve:VEwcre:). After his deposition
Cyrus was forced to take holy orders and consigned to the bishopric of Cotyae-
urn in Phrygia. The Emperor was doubtless thinking not so much of the good
that would accrue therefrom to Cyrus' soul, as of the fact that the people of
Cotyaeum had lynched their last four bishops, and might therefore not unreason-
ably be expected to treat their next in the same way. When installed in his see
Cyrus shut himself away in silence for some time, but was eventually forced to
preach a sermon on Christmas Day, to prove that he was in fact a Christian. His
sermon is fortunately preserved in its entirety:20
"Brethren, let the birth of God our Saviour Jesus Christ be honoured in
silence, because the Word of God was conceived in the Holy Virgin through
hearing alone. Amen."
His congregation was evidently too taken aback to lynch him, for he lived on
for another twenty years, and oddly enough became a genuine convert. After
the death of Theodosius he even gave away all his property to the poor, taking
the teachings of Christ a good deal more literally than most churchmen of the
day, and one of the very few specimens of his verse that we possess is the
inscription he wrote for the pillar on top of which St. Daniel the Stylite

17 The fact that so many of these poets came from the Thebaid suggests that there
were some very competent schools and teachers there to foster the tradition; it is most
unlikely that they all received their instruction in Alexandria.
18 fro 20 = Suidas s.v. Pamprepios (IV. 14,9 Adler) (d. n. 93).
19 Zacharias, Life of Severus (Tr. Kugener, Patrol. Orient. ii, 190Y) p. 40.
20 John Malalas, ekron. p. 362 Bonn. For the text of the sermon, Bury,LRE i2 228 n. 6.

17
 474 ALAN CAMERON

performed what Gibbon happily called an 'aerial penance,21 in return for the
saint's kindness in interceding (successfully) with the Deity on behalf of Cyrus'
two daughters, both of whom had been suddenly possessed by evil spiritS. 22
Of the very numerous poet-grammarians in Byzantine Egypt (below, p -49I f. ),
Eudaemon of Pelusium, a friend of Libanius, was a pagan,23 as also was Pal-
ladas. Palladas indeed was bold enough to write a series of ironic poems ad-
dressed to Theophilus, the fanatical patriarch ofAlexandria,24 and a neat distich
mocking the discrepancy between theory and practice in monasticism which
deserves quoting;26
'Et fLovocx,ol, 't'l 't'oerol8e:; 't'oerol8e: 8e:, 1tWC; 1t!XAL fLOUVOL;
<11tA~SUC; fLovocx,wV ljie:uerocfLev'I] fLov!Xiloc.
Likewise Horapollon was a pagan; in fact his family remained pagan for another
two generations, for his grandson, also called Horapollon and a poet and gram-
marian like his grandfather, was a staunch pagan. He used to offer sacrifices
to pagan idols together with his pupils, and when one of them, a Christian
called Paralius, taunted him with believing in a pack of lies, the rest of the
class turned on Paralius and beat him up. The Christian populace of Alexandria
nicknamed him Psychapollon, 'the destroyer of souls'. 26 Late in life, however, he
was suddenly converted to Christianity, much to the disgust of Damascius, who
remarked incredulously that he had had no good reason for this drastic move
(&1t' ouile:fLlocC; .•• &vocyxoclocc; wx''I]C;) - though the arch-pagan philosopher
Heraiscus had acutely foretold some time previously that Horapollon would, as
he put it, "desert to the others" (ocu't'0fLoA1jere:L 1tpOC; htpoUC;).27 Nor should we
forget Helladius and Ammonius of Alexandria, two very militant pagans.
Helladius had been the priest of Zeus in Alexandria, Ammonius priest of an ape-
god (mS-l)xoc;) until they were forced to leave the city in 39I when Theophilus
destroyed the temple of Serapis and made life there unbearable for pagans.
They both repaired to Constantinople, where they set up school and made a name
for themselves; but they both remained quite unrepentant, Ammonius perpetu-
ally grumbling about the particularly ignominious treatment meted out to the
sacred effigy of his ape-god, and Helladius being heard on occasion to boast that
he had slain no fewer than nine Christians with his own hand 128

21 Decline and Fall, ed. Bury iv 73, writing in fact of St. Simeon the Stylite, whom
Daniel aped.
22 Life of Daniel the Sty lite 3I f. (Analeeta Bollandiana xxxii (I9I3) Isof.). On the poem
see H. DeIehaye, REG ix (I896) 2I6f. It is also preserved as An/h. Pal. i. 99.
23 O. Seeck, Briefe des Libanius (I906) I3I.

24 Discussed in JRS 1965; d. also Bowra, Proc. Brit. Aead. xlv (I9S9) 26zf.

25 Anth. Pal. xi. 384. 26 Zacharias, Life of Severus pp. 15, 23, 32.

27 Suidas s.v. Horapollon. His 'desertion' may have been not unconnected with the

fact that his wife·t;'an off with a foreign lover; d. Pap. Cairo Masp. iii. 67295. and the
article of Maspero cited in n. I I.
28 Socrates, Hist. Eccl. v. I6.

18
 'Vandering Poets: A Literary Movement in Byzantine Egypt 475

Seeck identified Andronicus of Hermupolis with a poet whom Libanius

mentions as swearing by the gods of Egypt,29 but though he.may be right it is a
mere guess. But perhaps his involvement in a treason case accused of dabbling
in magic together with the fact that the only men we know him to have
associated with are Libanius, Themistius and Q. Aurelius Symmachus, three of
the staunchest pagans of the day, are sufficient to lay the onus probandi on
those who would have him a Christian. That Soterichus of Oasis, author of an
enormous number of epics, both historical and mythological; was a pagan is
rendered extremely probable by the fact that he wrote one on the life of the
miracle-monger Apollonius of Tyana, the pagan counterpart of Christ. 30 The
only clue as to the religious affiliations of Christodorus of Coptus is that he wrote
a poem "On the pupils of the great Proclus" (m:pt T(;)V eXXPOOCT(;)V TOU [LEycXAOU
IlpoxAou) ;31 it is hard to see what motive he could have had for writing such a
work unless he was sympathetic to the teachings of Proclus, the foremost pagan
teacher of his day and the author of a work "Against the Christians" in no fewer
than I8 books. 32 If it was anything like Damascius' Life of Isidore, it will have
provided a sketch of the careers of all the leading pagans of the day; indeed, it
is hard to see how it can have been otherwise, for Proclus' pupils became, like
him, very prominent pagans, Heliodorus, Ammonius son of Hermeias, Severia-
nus, Marinus (who wrote a life of Proclus in both prose and verse) and many
others. Perhaps then we may conclude that, if not an out and out pagan - his
parents were presumably Christians, or they would hardly have called him
Christodorus 33 - he nevertheless had pagan sympathies. We have the unim-
peachable evidence of no less a person than St. Augustine, his exact contempo-
rary, that Claudian was a pagan, and Augustine's disciple Orosius goes so far as
to call him 'paganus pervicacissimus'.34 Curiously enough many scholars,
partly on the ground that there is nothing specifically anti-Christian in his
poems, partly because he was the court poet of a pious Christian emperor, have
contended that he was a Christian, or at least paid lip-service to Christianity.
Neither argument is worth much. No court poet would have been so tactless or
misguided as to attack the faith of his emperor, unless he wanted to cut short

29 Briefe des Libanius 70.

30 Suidas s. v. Soterichos; for Apollonius of Tyana cf. P. de Labriolle, La reaction pa'ienne
(I934) I75 f., and references collected in n. 35 of my article in JRS I965.
31 John the Lydian, De Magg. iii. 26. 82 Suidas s.v. Proklos.

33 Baumgarten, in P-W iii. 245I, expresses surprise at the pagan character of the works

of a man with such an obviously Christian name, and toys with the idea of identifying him
with Christodorus illustris from Thebes, who wrote an epic on the miracles of Kosmas and
Damian (Suidas s.v.). There are no grounds at all for accepting this identification, and no
need to suppose that Christodorus was such a devout Christian as the parents who named
him.
34 Augustine, Civ. Dei v. 26, Orosius, Adv. Pag. vii. 35: I shall be discussing the matter

in detail in a forthcoming study of Claudian.

19
 ALAN CAMERON

his livelihood and probably also his life: and one has only to think of a man like
Themistius, who won the confidence of a succession of Christian emperors
despite his professed paganism and to whom Theodosius himself actually
entrusted the education of his eldest son, to realise that it would have been
perfectly possible for Claudian to have remained a pagan at the court of
Honorius. Augustine and Orosius doubtless had information other than that
provided by the poems on which to base their verdict, and I see no good reason
to dispute it. The poet Theotimus, who wrote a panegyric in Anthemius, the
chief minister of Honorius' brother Arcadius, was probably a pagan as well. 35
As for the versatile Olympiodorus of Thebes, we have the express testimony of
Photius that he was a pagan, an opinion that seems to have been well grounded.
The surviving fragments of his history reveal what Photius evidently considered
an unhealthy interest in neoplatonism,36 and the staunch pagan Hierocles dedi-
cated to him his m;pL 7tpovo[OCC; XOCL df.l.OCpf.l.EV1JC;, naming him at once the critic
and inspirer of his work. 37
It is worth remarking in conclusion that the mythological poets as well, the
'school of Nonnus' seem to have been for the most part pagans. Authorities on
Nonnus himself are now generally agreed that he must have been a pagan when
he wrote the Dionysiaca 38 - though he must presumably have been converted
by the time he wrote his verse paraphrase of St. John's Gospel. The same
probably holds of Triphiodorus (also from Panopolis) and Colluthus; nothing in
their surviving works suggests that they were Christians. It has been argued
that Musaeus was a Christian on the grounds that he imitates Nonnus' para-
36 Synesius ep. 49; d. Bury, LRE i 2 213 n. 8.
36 Photius Cod. 80 init.; cf. E. A. Thompson, CQ xxxviii (1944) 43. It is worth adding
that Photius was also of the opinion that the sixth century poet and antiquary John the
Lydian was a pagan (Cod. 180).
a? Photius Cod. 314 init. This passage, of great importance for our knowledge of

Olympiodorus, was unfortunately missed in the otherwise excellent accounts of him by

Thompson (n. 36) and Haedicke in P-W. There can be no doubt that the Olympiodorus
there mentioned is the famous poet and historian, as realised by Elter, Rh. Mus. lxv (1910)
176 and P. Henry, in his Bude ed. of Photius, iii (1962) 243 n. We should no longer
make the mistake of regarding the teachings of Neoplatonism as diametrically opposed to
those of Christianity at every point - the surviving writings of Hierocles in fact reveal
strong traces of the influence of Christianity (Praechter, Byz. Zeitschr. xxi [1912] 1 f.) and
St. Ambrose was strongly influenced by Plotinus (Courcelle, Recherches sur les Confessions
de S. Augustin [1950] 106f.) - but Hierocles himself was a militant pagan; when teaching
in Constantinople he was publicly flogged and expelled from the city "dripping with blood"
(Suidas s.v. Hierokles). Bury, LRE i2 376 is surely wrong to say that this treatment was not
connected with his paganism: see de Labriolle, La reaction paienne 483.
38 Bogner, Philologus 89 (1934) 320f.; Keydell (P-W xvii. 1. 915) is probably correct
in maintaining that the sarcastic scepticism displayed at xl. 834 towards the idea of
parthenogenesis is "eine versteckte Polemik gegen das Christentum". A recent exception
is E:. Thraede, RAG v (1962) 1001, who still contends that the Dion),siaca could have been
written by a Christian.

20
 Wandering Poets: .-\ Literary Movement in Byzantine Egypt 477

phrase of St. John,39 but it would have been only natural for a devoted admirer
of Nonnus like ~Iusaeus to have read Nonnus' other works besides the Diony-
siaca even if they were Christian and he was a pagan. And it is worth adding
that Synesius only became a Christian late in life. The important and significant
fact remains that whether or not the mythological poets of Egypt were actually
professed pagans, their work is entirely secular (and therefore pagan in inspi-
ration), conceived and excuted in a tradition that can be traced back through
the Alexandrians to Homer himself.
For all that Christianity made more rapid strides in Egypt than practically
any other land of th(; Empire,40 and was the centre whence monasticism spread
all over the Eastern and then the Western provinces, these distant epigonoi of
Horner and the Alexandrians remained pagans almost to a man, some, like
Pamprepius, Helladius or Ammonius, unrepentant and militant adherents of
the old gods, all at any rate sufficiently attached to the pagan past to derive all
their inspiration therefrom, and not to betray the slightest hint that they lived
in a Christian world. This last flowering of Greek poetry on Egyptian soil was
backward-looking, "the last manifestation of the pagan spirit in literature".41

II

The author of the latest comprehensive history of Greek literature an-

nounced his intention of omitting the names of 'verlorenen und zum gr6Bten
Teil hiichst unbedeutenden Werken' in treating of the Greek literature of the
Roman Empire. 42 A perfectly understandable decision in view of the enormous
bulk of Imperial literature that is actually extant, but nevertheless unfortunate
in that the impression thereby conveyed is that the poets of the fifth century
concerned themselves almost exclusively with the trivial and outworn mythology
of an age long past, and wrote on subjects entirely remote from the world in
which they lived. In fact at scarcely any time was more contemporary poetry
produced, and nowhere more so than in Egypt. Now these contemporary poets
were, as already remarked, professionals. Ample proof of this is to hand. Libanius
advises his friend Reraclianus to get the poet Diphilus to sing his praises - but
warns him that it will not be afLtcr6L 43 Synesius compares the relationship of his
friend Theotimus to Anthemius with that of Simonides to Riero of Syracuse -
adding that even Simonides wfLoMy€t 1tpOC; aprUptov 3tCXAEy€cr6CXt. 44 That
perfect Byzantine civil servant John the Lydian composed a panegyric on the
39 KeydeU, P-W xvi.!. 767: the argument does not seem very strong when one re-
members that the pagan Libanius read and imitated Eusebius' Lile 01 Constantine (A. F.
Norman, Rh. Mus. I07 [I964] I73).
'0 As remarked by Eusebius, Dem. Evang. 297b, p. 286 Heikel.
n C. M. Bowra, Oxl. Class. Dict. (I949) 32I.
'2 Albin Lesky, Gesch. d. gyiech. Literaturl (I963) 862.
'3 cp. 969. " cp. 49.

21
 ALAN CAMERON

praetorian prefect Zoticus for which he was paid at the rate of a solidus per line
from public funds 45 - and with a rich wife, a reward likewise earned by both
Claudian and Pamprepius. 46 Palladas, however, strongly deprecated the sordid
financial motives of his fellow-poets, vigorously denouncing one of them 47 for
TIWAWV tlXfL~OU':; W':; EAIXLOV EfLTIOPO':;. What would he have thought of Dioscorus
of Aphrodito, who actually ends his wretched unmetrical effusions with a request
for money:48
't"ej} aej} opc;~ov obdTIJ 5A~O\) X.C;~plX.
Naturally enough a professional poet reckoned to make his living principally
by writing panegyrics on local dignitaries, or military commanders who had
inflicted a defeat on some troublesome frontier tribe. It is attested of most of
the Egyptian poets of the period that they wrote such eyxwfL~1X and we
may well assume that even when it is not explicitly attested most of the
others did as well. In addition to the more famous poets, and those of whom we
know at least the name, the sands of Egypt have turned up a number of
fragments of anonymous eyxwfL~1X from the pens of much humbler exponents of
the genre, sometimes actually in the autograph copy of the poet himself.49 The
eyxwfL~ov was fairly easy to write. All that was required was to adapt the
virtues of the subject to be addressed to the elaborate rules formulated by
rhetoricians like Aphthonius and Menander, with their numerous compulsory
subdivisions, TIpoo(fL~OV, 'YEvo,:;, 'YEvc;a~.:;, &VIX't"pOCP~, emTIjad)fLIXTIX, TIpOC~C;~C;,
eTI(Aoy0C;, with optional extras such as the auyxp~cnc;. Thus the very minimum
of imagination was required - indeed it was hardly possible to go wrong.
Careful analysis of Claudian's panegyrics has shown that they conform to the
rules of the rhetoricians with astounding fidelity, and his panegyric on the
fourth consulate of Honorius has been described as an almost perfect example
of the rhetorically constructed enkomion. 50
.. De Magg. iii. 27. Oppian was paid at the same rate (Sozomen. Hist. Eccl. praef.)
as also was Alexander's court poet Choerilus (Porphyrio on Horace epp. ii. 1. 234).
4. Claudian ep. ad Serenam (c.m. xxxi); Suidas s.v. Pamprepios and Rhetorius (n. 95) I.e.
47 Anth. Pal. xi. 291: a certain Nicander, if any reliance can be placed in the lemmatist

of the Palatinus - which is not likely (cf. CQ 1965).

48 Heitsch (next note) XLII, 5.62 = 9.20; d. 16.6. It might seem tempting to mend
at least one of the holes in the metre (which is intended to be iambic) by writing 'If.pC!.
for Xei:PC!. - did we not have the authority of Dioscorus' own handwriting for Xe:LpC!.! Com-
pare also Corippus Laud. Just. praef. 48, "pro munere carmina porto", Pan. Anast. 48,
l'fessae miserere senectae'J.
49 The fragments are conveniently assembled by D. L. Page, Greek Literary Papyri i

(1948) Nos. 132f. (omitting Dioscorus) and in the fuller and more recent edition by E.
Heitsch, Die griechischen Dichterfragmente der romischen Kaiserzeit, Abh. Gott. Akad. 1961
(2nd. ed. 1963, 1964) Nos. XXVlIf. (on some omissions and shortcomings of this edition d.
R. Browning in CR n.s. xiii [1963] 158 f. and M. L. West in GGA 215 [1963] 164f.) .
• 0 Cf. L. B. Struthers, 'The Rhetorical Structure of the Encomia of Claudius Claudianus',

HSCPh xxx (1919) 49f. (citing the parallels from the Greek rhetoricians in extenso), and
P. Fargues, Claudien (1933) 191 f.

22
 Wandering Pacts: A Literary ?lIovemcnt in Byzantine Egypt 479

In addition to the panegyric the professional will have been equally adept
in the complementary art of the invective ('~ayoc;), though t he opportunity
for making use of it will not normally have arisen so often. The yayoe; was
composed according to the same rules as the e:yxwfLwv, only in reverse (that is
to say with each of the subdivisions serving as an opening for vituperation
instead of praise). The Egyptians, especially the Alexandrians, notoriously had
a flair for invective. According to Seneca Egypt was "loquax [= saucy] et in
contumelias praefectorum ingeniosa provincia ... cui etiam periculosi sales
placent",51 and Herodian was of the opinion that the Alexandrians were by
nature 52 qnAoO'xwfLfLovEe; ... xcx.~ ,x,tOppmTouVTEe; e:e; TOUe; lJ1tEpEZ0VTcx.e; 7tOAA!X
Zcx.PLEVTcx. fLE:V cx.,jTo~e;, ).U7t"f)p!X DE: TO~e; O'xW!fl{k~O'L - a characteristic perfectly
exemplified in the person of Palladas, author of many a blistering lampoon on
<XPXOVTE':;, including one on the great Themistius, a series of eight on the un-
fortunate rhetor Gessius, and several on Theophilus the patriarch of Alexandria
himself. 53 He vowed a thousand times, he tells us, not to write any more lam-
poons,54 but simply could not contain the '100'0':;, openly admitting that 55
, AL'I(~m fLE:V apLO'Tov, 6 DE: cj;oyoe; ~X8EO':; c1.px.~·
c1.McX xaxw.:; d7tE~'I ' AT'nxQ'I e:O'TL (LEAL.
Claudian wrote full scale invectives in two books each against Arcadius'
ambitious ministers Rufinus and Eutropius, as well as many lampoons, and
his ready command of the genre 56 implies that he had left many a tattered
reputation by the wayside on the road to such mastery.
Lastly the professional will have taken care to master the secrets of the
e:m8cx.ArXfLwv, a very necessary item in his repertoire, for which the rhetorical
handbooks prescribed a similarly elaborate series of rules. 57 Naturally such
works were of more than usually ephemeral interest and few have survived,
but we still have the em8aArXfLwv Claudian wrote on the occasion of the
marriage of the Emperor Honorius 58 together with some traditionally
licentious Fescennine verses, another on the wedding of a notary called
Palladius,59 and we possess no fewer than five e:m8ocAafLLa from the prolific
pen of Dioscorus of Aphrodito - on the wedding of the magnificent count

51 Dial. xii. 19. 6.

fi2 Hist. iv. 9. 2; d. also SHA Vita Firmi vii. 4.
63 Gessius:Anth. Pal. vii 681-8 (ei. Byz. Zeitschr. 57 [1964J 279f.). Themistius: ib. xi.

292 (ei. CQ 1965). Theophilus: ix. 393, x. 90, 91 (ef. ] RS 1965).

64 ib. 340. 66 ib. 341.

S6W. Suess, Ethos (1910) 264f., (on the In Eutropium), H. L. Levy, TAPhA Ixxvii

(1946) 57f. (on the In Rufinum).

57 A. L. Wheeler, 'Tradition in the Epithalamium', AJPh Ii (1930) 205f., R. Keydell,

'Epithalamium' in RA C v (1962) 927f.

6B p. 87f. Koch, II9f. Birt; See, for a detailed study of the epithalamia of Latin poets
of the late empire and early middle ages C. Morelli, Stud. ltal. xviii (I9ID) 319-432.
69 C.m. xxv, p. 226f. Koch, 301 f. Birt.

23
 ALAN CAMERON

Callinicus, the most splendid Isacius and various other distinguished Aphro-
ditopolitan bridegrooms. 60
Better, or at any rate more ambitious, poets would also try their hand
at an epic,6o' if the Emperor or one of his generals won a suitably
glorious victory. Properly speaking, the epic - "res gestae regumque
ducumque et tristia bella61 " - should be a very different thing from
the panegyric. But when the subject of the epic was the campaign of a
living Emperor or general, and was intended, moreover, to be recited in front of
him, it was only natural that it should take on many of the aspects of the
panegyric. Indeed, as Wilamowitz remarked when publishing one of the
papyrus fragments mentioned above, it is virtually impossible to tell which of
the two the whole poem was. 62 Nevertheless, though the line between the two
might often be hard to draw, it was always felt to exist, and was marked by the
poet himself in the title he gave his poem. Claudian's numerous poems, for
example, on the successive consulates of Honorius are entitled" Panegyricus de
tertio (quarto etc.) consulatu Honorii Augusti", whereas those on Honorius'
wars against Gildo and Alaric are called "de Bello Gildonico" and "de Bello
Getico": the former are as the title implies panegyrics, the latter definitely
epics.
The epic was more of a challenge to the poet than the panegyric, for he
had a free hand in the construction of his poem and more opportunity to
demonstrate his skill, or lack of it. Claudian on the whole acquits himself
better in his epics than in his panegyrics, which, confined by the hampering
rules of the rhetoricians, seldom rise above the competent but tedious level
of those of lesser writers. The de Bello Gildonico is particularly effective, and
Corippus' Johannis, an epic on the campaign of Justinian's general John
Troglita against the Moors, is a very creditable production for a sixth century
small-town African schoolmaster. Of the epics of their fellow-poets we have
only the titles, which permit a guess at their subject matter, but no more. We
learn from Socrates of two epics on Arcadius' victory over the Goth Gainas
in 400, by Eusebius Scholastic us and Ammonius, both otherwise unknown. 63

60 Heitsch XLII, 21, 22, 23, 24, 25. Several are in acrostics. For an anonymous fourth

century epithalamium see Heitsch XXV, Page 139. Cf. also Suidass.v. Panolbios fin.
60. Following a Hellenistic precedent; numerous epics were written on the deeds of
Alexander, Antiochus Soter, Perseus, AttaIus and Eumenes of Pergamum - but the
temptation was repeatedly (and wisely) resisted by Horace, Virgil and Propertius. Cf.
Kroll, 'Das historische Epos', Socrates N.F. iv (1916) 1-14 (who completely ignores the
historical epics of the later empire), and Pertusi, Giorgio di Pisidia: poemi I (1959) 32 f.
61 Horace, AI'S Poet. 73.

62 Berliner Klassikertexte v. 1 (1907) 107; for some good remarks on this topic, with

special reference to George Pisides and Corippus, see Nissen, Hermes !xxix (1940) 298 f.
63 Hist Eccl. vi.6. Ammonius is probably the pagan grammarian mentioned ib. v. 16, in

which case he was an Alexandrian.

24
 \Vandering Poets: A Literary Movement in Byzantine Egypt 481

From Suidas we hear of the 'IO"o(UpLxli64 of Pamprepius, presumably on the

restoration of the Isaurian Zeno in 476 or the supremacy of his Isaurian
general Iilus, another 'IcrcxupLxli in six books by Christodorus, on the victory of
Anastasius against the Isaurians in 497, and a Ile:pO"Lxli by Colluthus, on the
Persian victory of the same Emperor in 506.
Not all the epic poets of the day were Egyptians. There is no evidence,
for example, that Eusebius was Egyptian (though likewise no evidence
that he was not) nor the Callistus who wrote an epic on the Persian
expedition of the Emperor Julian. 65 The Empress Eudocia, better known
perhaps as Athenais, who wrote an epic on the Persian wars of her
husband Theodosius II in 421, was an Athenian. Nevertheless Egyptian
domination of the field is manifest. Ammonius, Pamprepius, Christodorus
Colluthus and Claudian were all certainly Egyptian, as also were the authors of
various epics on the wars of lesser persons, numerous fragments of which have
turned up on papyrus. We have part of a poem on the Persian war of Dioc1etian
and Galerius,66 perhaps the work of the prolific Soterichus of Oasis,67 another
on an Egyptian campaign of Heraclius, a general of the Emperor Zeno,68
another on the campaign of an otherwise unknown general called Germanus
against the Blemmyes. 69 • That all these and others by unidentifiable persons on
unidentifiable events were written by Egyptians is sufficiently obvious from the
frequent references in them to Thebes, the Nile, the Blemmyes and Egyptians,
some of them being plainly written from an Egyptian point of view. 70 Further-

M Suidas appears to say that the 'IacxupLxcX was in prose, but his entry is confused, for
he also implies that Pamprepius' book on etymology was an epic poem (it being natural to
assume that works ascribed to an epic poet are epic poems unless there is an indication to
the contrary). As Bernhardy long ago realised, the word xcx'rlXAoYcXlll]v, 'in prose', ·is mis-
placed, and should come before, not after, 'IacxupLxcX. Suidas is saying that the book on
etymology was in prose. Whether the transposition is Suidas' own or his scribes' is anybody's
guess. For what may be some fragments of Pamprep ius' 'IacxupLxcX see Heitsch XXXV, I & 2;
granted that XXXV, 4, a panegyric on Theagenes, is by Pamprepius (d. n. 97), then it is
possible that these fragments come from the 'IacxupLxcX, though it is known that Pamprepius
wrote many other eyxw(LLcx (n. IS3), and we cannot be certain that all the poems on the
papyrus are by the same poet.
66 Socrates ib. iii. 21; possibly also the IIap6LxcX of Seleucus of Emesa mentioned by
Suidas s.v. Seleukos was an epic on Julian's Persian expedition: d. Seeck, Briefe des
Libanius 272, Norman, CQ li (1957) 130 n. 1.
18 Heitsch XXII, Page 135. 87 Cf. Suidas s.v. and Heitsch ad 10c.

18 Heitsch XXXIV, Page 144. Cf. Keydell, Byz. neugriech. Jahrb. xii (1936) Sf., and

P-W viii. 503 no. 15 for Heraclius.

89 Heitsch XXXII, Page 142.
70 E.g. Heitsch XXXVI (Page 143) was written by an inhabitant of Thebes. It is, of

course, only to be expected that papyri should give us poems by Egyptians, so to this
extent their evidence is biased. There is no need to assume that Egyptians had a complete
monopoly in the field (fairly competent epigrams on government officials have been found
all over the Greek East in this period: d. L. Robert, Hellenica iv [1948] passim). though
31 Historia XIV/4

25
 ALAN CAMERON

more it is worth mentioning that though Eudocia was an Athenian, she was
taught by the famous Egyptian grammarians Orion of Thebes (also the teacher
of Proclus) and Hyperechius of Alexandria. 71
In one respect these epics and eYX~fLLoc may claim the attention of even the
literary scholar. As is well known, the character of the Greek language changed
during the early centuries of the Christian era, and by the fourth century was
pronounced no longer according to pitch accent, but with a stress accent so
heavy that the quantities were totally obscured. When a Byzantine read
hexameter verses, therefore, (a purely quantitative metre) they did not sound
like verse at all.?1 The iambic trimeter on the other hand could be easily
adjusted to suit Byzantine pronunciation by regulating the number of syllables
and making the stress accents occupy the position of the original long syllables
- some quantitative iambic lines indeed automatically fulfilled these require-
ments - and it is not surprising that a metre so convenient to the new pronunci-
ation should eventually begin to supplant the hexameter as the metre of elevated
poetry.73 The activity of Marianus of Eleutheropolis, a high official at the court
of Anastasius, is often cited as a clear example of this usurpation by the
trimeter of the place formerly held by the hexameter. He wrote paraphrases
(fL&t"oc'PpOCa&LC;) in iambics of the hexameter works of all the leading Hellenistic
poets, Theocritus, Apollonius, Callirnachus, Aratus, Nicander 'and many
others',74 presumably with the aim of making them accessible to a wider public
than those who were sufficiently educated to enjoy them in the original hex-
ameters. And against all precedent, classical and Hellenistic, Helladius of
Antinupolis wrote a didactic poem in iambics, Cyrus of Antinupolis and
Andronicus of Hermupolis iambic enkomia75 • But for all that the uneducated

even on the evidence of the literary sources they clearly dominated it. And in view of the
fact that many of these poets were grammatici (belowpp. 492 f.) who taught all over Greece
and the Levant (p. 485). it is probable enough (though naturally not susceptible of proof)
that other non-Egyptian poets of the period besides Eudocia and Proclus received their
instruction from Egyptians, for it was with the grammaticus that the poets were studied.
We know of at least three Egyptian poet-grammarians who taught at Constantinople at
the end of the fourth century, Helladius, Ammonius and Horapollon the elder, and it
would hardly be surprising if a young poet like Theotimus who recited his Gainea there in
400 while still a student had fallen in some measure under the influence of one or other of
them. One further Egyptian poet not mentioned anywhere in my text is Pancratius of
Alexandria, about whom absolutely nothing is known except that he was the recipient of
two letters from Procopius of Gaza (epp. 31. 133).
71 John Tzetzes. Chilo X. 58f.
72 "Fiir das Ohr des Byzantiners existierten diese Verse nicht", P.Maas, Byz. Zeitschr. xii

(1903) 302.
73 l\Iaas, Greek Metre, Trans. H. Lloyd-Jones (1962) 20 § 25. and art. cit.

74 Suidas S.V. Marianos. Eugenius of Augustopolis, a grammarian, also wrote a number

of iambic poems under Anastasius (Suidas s.v.).
76 Photius Cod. 279, p. 536 Bekker.

26
 Wandering Poets: A Literary Movement in Byzantine Egypt 483

would no longer have been able to appreciate their work, almost all these
contemporary poets continued, in obedience to Aristotle,76 to write their epics in
hexameters,77 like the mythological poets of the period, both before and after
the metre was refined and adapted to the new accentual pronunciation by
Nonnus. Moreover the anonymous papyrus fragments of such works are all in
hexameters.78 But by the end of the sixth century the enkomia of Dioscorus are
sometimes written in hexameters, sometimes in iambics, both of which metres
he handled with equal and abysmal incompetence. 79 And by the seventh
century, when the monk George of Pisidiaso wrote his epics on the campaigns
of Heracliu~ in iambics and from an openly Christian point of view, we realise
that we have finally passed from the ancient to the Byzantine world.
It is perhaps hardly surprising that, with the exception of two lines dubious-
ly ascribed to Ammonius 81 and the fragments preserved by chance on papyrus,
this mass of contemporary poetry should have perished without trace. Their
disappearance is probably to be attributed not so much to their poor quality as
literature (if this were a sufficient reason for a work to perish, then very little
Byzantine literature would survive) as to the fact that their appeal was
strictly limited to audiences who had some knowledge of and interest in the
persons and events described. Only two of these epics are mentioned by anyone
other than a lexicographer who had probably never read, perhaps never even
seen, a copy of the works whose titles he faithfully transcribed. The ecclesi-
astical historian Socrates refers to the poems of both Eusebius and Ammonius
in a way that indicates that he was familiar with them: but he had lived
through the revolt of Gainas himself, and was very probably present in the
audience at Constantinople when they were recited. When a century or two
had passed men ceased to read these Gelegenheitsgedichte celebrating the

76 Poetics 1459b. 34.

77 It is expressly attested of practicaUy every poet cited in the course of this study
(except those mentioned above as writing in iambics) that they wrote a~' brii:J'J or ijpum<iji
fLz't'Ptp or err1) or were brw\I 7tO~'>1't'IX[ - aU of which terms mean that they wrote hexameter
poetry.
78 Though sometimes with an iambic preface, on which convention d. Friedlaender,
Johannes von Gaza (1912) 119f. Claudian's prefaces, however, are always in elegiacs (cf.
Parravicini, Athenaeum ii [1914] 183f.), as also are those to Sidonius Apollinaris' Panegy-
rics, and Corippus' Johannis; on this variation between iambic and elegiac prefaces see
especially Schissel, Berl. Phil. Woch. 49 (1929) 1073f.
79 See the excellent analysis of his 'style', metre &c. by Maspero, REG xxiv (19II)

46gf.
80 Significantly enough George also wrote an iambic Hexaemeron. On his epics, or rather
panegyrics, see Nissen, art. cit. (n. 62), and Pertusi's edition of 1959.
81 Preserved in Etym. lYlag n. s.v.M[fLlX\I't'O~, but the name Ammonius there has been
variously emended; cf. Reitzenstein, Gesch. d. griech. Etymologica (1897) 287. On the
strength of the metrical technique of just these two lines Friedlaender, Hermes xlvii (1912)
43f., argues that Ammonius wrote before Nonnus.

27
 ALAN CAMERON

forgotten deeds of forgotten men, and accordingly they ceased to be copied and
rapidly perished, only the names of even the most famous being known to us
from chance references in the works of contemporaries, and entries in bio-
graphical dictionaries. By sharp contrast the mythological poets of the day
continued to be read by succeeding generations, and works by even such
mediocre poets as Triphiodorus survive; the Dionysiaca is preserved in its
entirety - some 20,000 lines. In the East contemporary epics and panegyrics
were ten a penny, but in the West, where they were a rarity, Claudian was read,
enjoyed and copied long after men had ceased to know or care who Stilicho
was,82 and while his Greek works perished, every one of his Latin poems has
survived, as also has Corippus' epic on the able but not especially distinguished
general John Troglita and panegyric on the mad Emperor Justin II. On the
whole it is probably safe to say that the loss of their Greek counterparts is more
deeply felt by the historian than the student of literature.

III
Perhaps the most characteristic feature of this Egyptian school is mobility.
It is the traditional (and largely unfounded) view that the later Roman Empire
was a sort of prison, a state in which it was rigorously ordained in a series of
iron edicts that every man should follow his father's profession, that anyone
who tried to seek improvement of his status or make for pastures new was
promptly dragged back to the place and calling of his origo. "Immobility",
writes P. Charanis, "was the principal feature of the social structure of the
later Roman Empire".83 But there is another side to the picture, and in certain
professions (notably the bar, the army and the civil service)B4 there were
excellent opportunities for a promising young man to rise from the bottom right
to the top of the ladder and travel widely. Attention has recently been drawn to
the exceptional mobility, social and geographical, of professors and teachers of
rhetoric. 85 But it has not been remarked that professional poets travelled from
city to city every bit as much as an ambitious professor of rhetoric like Libanius
or the young st. Augustine.
In the heyday of the cities of the Hellenistic East it would have been natural
to assume that they moved about in order to compete at the great festivals held
82 The Repertorium of Conrad von Mure (t 1281) has an entry for Stilicho (or' Stillicio' as

he calls him) as follows: "proprium nomen cuiusdam, de quo satis dicit Claudianus".
83 Byzantion xvii (1944/5) 39; d. Alf6ldi, Confiict 0/ Ideas in the Laic Roman Empire

(1952) 28.
B( Jones LRE ii 549f. There were also good prospects for an able eunuch, though this

should not perhaps be described as a career in the ordinary sense of the word; see M. I\:.
Hopkins, Proc. Camb.Phil. Soc. 189 (1963) 62f.
8~ M. K. Hopkins, CQ n.s. xi (1961) 239f. On social mobility in the Later Empire as a
whole see especially now R. MacMullen in JRS liv (1964) 49f.

28
 Wandering Poets: A Literary Movement in Byzantine Egypt 485

in most of the leading cities in Greece and Asia Minor, where prizes were offered
for the recitation of epic and dramatic poetry.86 But the majority of such
festh'als had long since declined or disappeared by Byzantine times, and the
proud lists of victors drawn from all over the Greek world in the &UfLe:A~xot
&.yW\lE~ so common in Hellenistic and early Roman times,87 peter out completely
in the course of the third century.88 There was little demand in a Christian
Empire for the traditional epic hymn to the god in whose honour the festivals
had been founded. Not that such contests died out completely. Libanius refers
once to O'''t'[XW\I &.fL~AAoct on a festival dayBD and even to an&1t[8Et1;t~ &m7)\1 in
honour of Apollo, one of the patron gods of Antioch: 90 poetic contests were still
flourishing in late third century Oxyrhynchus, successful contestants being
rewarded by exemption from taxes. 90a As a young man St. Augustine entered
for a 'theatrici carrninis certamen' in Carthage - and to such good purpose that
the proconsul "coronam illam agonisticam imposuerat non sano capiti meo".91
Poets will doubtless have continued to recite at any festivals where such &.yw\le:~
were still held, but it is unlikely that either the prizes offered or the prospect of
the glory to be won would any longer have been sufficiently tempting to attract
for their own sake the professional poet of the later Empire, who normally
looked to private patrons as a far more lucrative source of income than the now
impoverished city councils.
It was in search of these patrons that our poets moved from city to city,
exploiting in tum each centre of learning and fashion. If Athens did not suit
their particular talents - or proved too hot to hold them - they would move on
to Antioch or Constantinople or (if like Claudian they could compose in Latin)
to Rome. It was generally recognised that some talents were better suited to
one city than another. Eurtapius, for example, tells us of an Alexandrian named
Eusebius who was preparing to go to Rome: "he seemed to be particularly
well suited to Rome, because he knew how to flatter and fawn on the great"
(XOAOCXe:UEt\l "t'E XOCL O'OC[\lEW "t'O \l1tEPEX,0\l), whereas in Athens he had been con-
86 On the festivals and contests in general see Jones, G1'eek City (1940) 233.
87 For lists of victors at the Museia at Thespiae see P. Jamot, BCH xix [1895] 321 f.;
see pp. 365-6 for a list of the ethnics of the victors, as an indication of how far men travelled
to compete at the Museia. Many other inscriptions record ttrw\I 1tOL'I)T(l( winning prizes at
such &y(J)\lE~; add to the references collected by Jones, c"eek City 355 n. 44, SIG3 447, 448,
451, 452, 704 E 20, 7II L 30, 728 K 25 (dating from 3rd - 1St. c. B.C.), OGIS 51. 35f. and
many others.
88 J. Frei, De cerlaminibus thymelicis (Diss. Basle 1900) 29.
89 P1'ogymnasmata ed. Foerster Vol. viii. 540. 18.

90 0,.. 54.37. These recitations can hardly have been proper 6UfJoEALXOl &y(J)ve:~; The
Antiochenes of Libanius' day, much to his regret, preferred less intellectually demanding
entertainments; see ch. III of P. Petit's Libanius et la vie municipale a Antioche (1955) -
who does not cite even the two references given above.
80a Pap. Oxy. 2338, with C. H. Robert's commentary.
91 Confessions iv. If.

29
 ALAN CAMERON

sidered a troublemaker ((j't'IX(j~W~1J~92). The wily Pamprepius93 evoked the same

reaction in Athens; after practising his poetic skill at home in Egypt, he set out
for Athens, where it stood him in good stead for a while, but he was qlLA6't'~f.I'o~.
He contracted a rich marriage and then fell foul of his patron, a powerful
Athenian grandee called Theagenes,94 and was obliged to leave Athens in some
haste. He had evidently spent a while in several other cities - and perhaps left
them just as hurriedly - for we learn, from a rather unusual source, his horo-
scope, that he rpuYOt~ &V 7tOAAoi:~ 't'67t0~~ &yevE't'095 before reaching Athens. After
Athens he made for Constantinople. Here he soon established himself in the
favour of the powerful Illus (below, p. 499 f.), but soon made himself so unpopular
with the Emperor that he was banished from the city and lived for a while in
Pergamum (of his subsequent movements more below). A poem that seems to
have been written to be recited in Athens and has been plausibly ascribed to
Pamprepius, ends with an announcement that the poet is leaving for Cyrene,
summoned by the Muses of Libya, and an exhortation to his audience to wish
him bon voyage :96
, AAAOt flo~ EUflEVeoLTE xlXt &~ "E[AA1JVO~ &:pouP1J~
7tefl7tE't'e flE (j7tEL(jIXV't'E~ &rpL(j't'OCflEV[ OV vee:> E:pYe:>·
Kup~V1l xlXAeEL flE, ~LIX~6flEVO~ [8£ flE <l>oi:~o~
~AY.EL 61lporp6voto rpLA1J~ &7tt yOUVIX't'1X [vUfl ]rp1J~.
8[Ei:i't'E], rpLAOL, 7tpO~ ~8E6AOV &:PE~flIXV£O~ Ih[OA ]EfLIXLOU
~v[61X flE [y.~Y. ]A~[ (j]y'ou(j~ AL~U(j't'L8E~ &~(jh~ [Mo ]U(jIX~.
(Some critics contest the ascription to Pamprepius,97 but the authorship is im-
material for my present purpose; if it is not by Pamprepius, then we simply
have here yet another Egyptian wandering poet who tried his fortune in Athens.)
The end of the poem has the appearance of being a variation on a regular
formula, to be adjusted according to where the poet was and where he was
intending to go next.
The Alexandrian Claudian, like his fellow citizen Eusebius admirably suited
to 'flatter and fawn on the great' (as his panegyrics amply attest), not
92 Vito Sopko 493.
93 On Pamprepius see Asmus, Byz. Zeitsch,. xxii (1913) 327f., Gregoire, Bull. de l'Ass. C.
Budi 24 (1929) 22f., Keydell, P-W xviii. 3. 408f. and n. 95 below.
94 For a fragments of what may be Pamprepius' panegyric on Theagenes, see Heitsch

XXXV, 4, Page 140 b.

95 A. Delatte's realisation that the horoscope analysed in detail by the sixth C. Egyptian
astrologer Rhetorius in CCAC viii. 4. 222f. is none other than that of Pamprepius has
enabled a number of events in his life to be dated with a precision that can be scarcely be
paralleled for anyone in the ancient world (Bull. de la Classe des Let/res de l'Acad. de
Belgique ix (1923) 58f.). We now know for example that he was born on 29th. September
440, at 3. 48 p.m. local time (1. 52 GMT).
96 Heitsch XXXV, 3. 193f., Page 140 a 151 f .

• 7 Cf. Page ad loco p. 565, Keydell P-W xviii. 3.414 f. cf. n. 64. Schissel argues that the

reference to Cyrene shows that the poet was a Cyrenean, but this is hardly compelling.

30
 Wandering Poets: A Literary t.lovement in Byzantine Egypt 487

unnaturally made for Rome, where he quickly established himself as

court poet to the young Emperor Honorius and his powerful general Stilicho. It
is not recorded that he performed elsewhere than at Alexandria,98 Rome and
:'IIilan (where most of his Latin poems were recited), but more than one critic
has drawn attention to the frequency in his works of allusions to the sea and
sailing,99 a fact which suggests that he had, like so many of his compatriots,
travelled widely, and it seems likely that one or two of his poems were composed
in Sicily. 99a
Fairly typical would seem to be the career of Andronicus of Hermupolis,
whose fame as a poet is attested by three contemporaries, Libanius, Ammianus
and Symmachus, and reached through five centuries to Photius - though
unfortunately not much further, since not a line of his poetry survives. We
learn from Photius that he started on the road to fame and fortune by reciting
a panegyric on his fellow-townsman the comes Phoibammon. loo While still quite
a young man he was unwise enough to become involved in a treason trial in
Scythopolis in 359,lO1 but, helped, it seems, by the good offices of his friend
Libanius, was acquitted. After this narrow escape (not many were acquitted in
the treason trials under Valens) Libanius advised him to accept some official
post offered him that would allow him to continue with his poetry as well (w~
e:vov &pXe:w -re: ofLot) XlXt ~~e:~v), but Andronicus replied (Libanius professes to
be quoting his exact words) &)). . ' e:yw e:tfL~ ~wawv e:fLIXU-rov 0e:fL~a-r[cp fLIX61JTIJv,
XcXAAWV ~youfLe:vo~ -rot) TtOMWV &pXE~V.1°2 So Libanius duly wrote him an
introductory letter to the philosopher-rhetorician Thernistius, at this time in
very high favour under the Emperor Constantius II, who had recently had a
bronze statue erected in Themistius' honour in Constantinople. lo3 In his oratio
xxix Themistius refers to an AlyuTt-r~o~ vElXv(axo~ who had lately stayed for a
while in Constantinople (I!.vIXYXOC, e:ma1JfL-fJalXc,), proficient at ~uV"t"~6evlX~ -rPlXycpi)[lXv
XlXt 1!.1t1) XlXt ~~6uPcXfL~OU~.104 Now Photius records that in addition to writing
panegyrics Andronicus ~PlXfLcX-rwv &a-r~ TtO~1JTIJc" ~~1X'P6po~c; fLhpo~~ -rou~ Myou~
e:vn[vwv; it has often, and not unreasonably, been assumed that Themistius was
speaking of Andronicus, on the grounds that there are hardly likely to have
been many Egyptian 'dramatic poets' in an age when the legitimate theatre
hardly existed. IDS Naturally however the identification depends on the date of

98 Where his Greek Gigantomachia was recited. There is no need to assume, with Mar-

tinelli, Misc. Galbiati ii 48, that it was recited at a poetical &yoov. It was probably recited
publicly, but this was the regular practice with almost all literary works in Hellenistic and
Roman times, d. Rohde, Griech. Roman2 (1900) 327 n. I.
99 E.g. Rolfe, T A PhA L (1919) 146, and d. particularly the preface to his Greek

Gigantomachia.
99a Cf. P. Fabbri, Raccolta ... Ramorino (1921) 9If. d. also n. II9.
100 Cod. 279 fin. 101 Ammianus Marcellinus xix. 12. 1 I. 102 Ep. 77.
103 Seeck, Briete des Libanius 296. 104 Or. xxix 347a.

105 Christ-Schmid-Stahlin, ii. 28958£., Keydell, Btlrsians Jahresb. 272 (1941) 69f.

31
 ALAN CAMERON

the speech in question. Seeck, on the strength of the identification, dated it to

c. 35g-Oo,106 but Schenkl and Scholze date it on other grounds some twenty
years later, to 377,107 which would rule out the possibility that Andronicus was
the VEtIVLcrxOt;, for he would no longer then have answered to such a description.
In fact whatever the date of the speech it is quite unnecessary, even hazardous,
to assume that Andronicus was the VEtIVLcrxOt; in question, for there are many
other possible candidates to hand. For example, in 358, one year before
Andronicus set out for Constantinople, another Egyptian poet, a certain
Harpocration, set out for the same destination, armed, like Andronicus, with
an introductory letter from Libanius to Themistius. Harpocration had evidently
made his name in Antioch, but his nationality can be deduced from the letter
Libanius wrote for him before he left for the capital, warning him to be careful
of the climate there: O"WfLtI 3e: At)'o 7tTLOV 7ttIptI300t; 0pcpdoLt; 7tVEOfLtIO"LV 8ptI TL
3pcXO"ELt;.108 And we hear from Photius of one Serenus, an Egyptian grammarian
who wrote 3pOCfLtITtI 3LcXCPOPOC 3LtICPOPOLt; fLETpoLt;, and also that his fellow-gram-
marian Horapollon O"UVTW1jO"L 3e: XtIt tIu't"Ot; 3pcXfLtITtI. Horapollon, according to
Suidas,l09 taught in Constantinople in the reign of Theodosius (379-395),
and if the later date for the speech is accepted it is possible that Horapollon is
Themistius' VEtIVLcrxOt;. But he might equally well be anyone of a number of
such versatile Egyptian poets, of whom our scanty sources leave us in ignorance.
We only know that Andronicus, Serenus and Horapollon wrote 3pOCfLtITtI because
the works of all three happened to be bound up together with Photius' copy of
the Chrestomathia of Helladius of Antinupolis. However the identification of
Themistius' VEtIVLcrxOt; is not of itself of the slightest importance; what is
important is to realise what a surprisingly large number of Egyptian poets with
qualifications similar to those of Andronicus made the trip to Constantinople
during this period.
To resume the story of Andronicus himself, by 378 he was back in
Antioch, where he refused to write a panegyric on a very distinguished
Antiochene called Hypatius (on which see below, p. 500). He seems at sometime
to have either visited or contemplated visiting Rome, for he could count Q.
Aurelius Symmachus as a friend and correspondent. no Libanius knew Sym-
machus' father and corresponded with Symmachus,lll and it may well be that
he gave Andronicus an introduction to Symmachus as he had earlier to
Themistius. In reply to a letter from Andronicus Symmachus thanks him for a
complimentary copy of his poems and promises to direct the attention of his
friends to them ("tradam publicae admirationi ... bonum saeculi publicabo");

106 Briefe des Libanius 300.

107 See P-W. VA 2. 1665 and also Bouchery. Themistius in Libanius' Brieven (1936) 148
n·3·
108 Ep. 368 • 109 Suidas s.v .. and d. Maspero op. cil. (n. II).
110 Syrnmachus ep. viii. 22. 111 Ep. 1004.

32
 Wandering Poets: A Literary ~Iovement in Byzantine Egypt 489

this is presumably what Andronicus had asked him to do. No more is known of
him, but already in 359 his reputation, according to Libanius, had reached the
Ethiopians - though he adds rather mysteriously that the 'misfortune of his
mother and city' had so far prevented him from 'revealing all the treasures of
his soul' (r.cl.v'roc at:~~OCl 'r0: &ycl.)'fLOC'rOC -r-~C; ~'JX1iC;).l12
Another such poet, about whom less is known, is Eudaimon of Pelusium,
who travelled widely in at any rate the Eastern provinces of the Empire. He
studied rhetoric at Elusa, and then his movements can be traced by various
allusions in the letters of his friend Libanius back to Egypt, thence to Antioch,
Elusa again, and then to Constantinople and finally back to EgypU13 He was
sufficiently well known to merit an entry two centuries later from Hesychius of
Miletus, whence he passed into Suidas. From a solitary reference in another of
Libanius' introductory letters 114 we learn of another much travelled poet, called
Dorotheus, a man 1tOAA~V ., . la6v'roc yiiv 'rt: xoct 6cl.AOC'r'rOCV, 1tocv'rocxou ai: ad~ocV'roc
~v fLocv[ocv. Mention also might be made of Helladius and Ammonius who
both moved from Alexandria to Constantinople, and Palladas may have done
so as well, though the evidence usually adduced to prove it is hardly convincing.ll5
Christodorus of Coptus wrote an ekphrasis on the statues in the gymnasium of
Zeuxippus in Constantinople (still extant),l16 a poem on the Isaurian wars of
Anastasius and another on the history and antiquities of Constantinople in
twelve books; all of which works suggest that he had himself lived there for a
time. He also wrote similar poems on the antiquities (1tcl.'r(noc) of Thessalonica,
Nacle, Miletus, Tralles and Aphrodisias,l17 and whereas he might easily have
written them all without stirring from his library, it is natural to assume, in an
age of professional poets, that he visited all the cities in question and was paid
by them to celebrate their past history.l18 His poem on the pupils of Proclus
suggests that he had himself studied at Athens under Proclus. A certain Claudian
(probably not the famous Claudian) likewise wrote a number of 1tcl.'rPlOC (now
112 Ep. 77. I. Had this 'misfortune' perhaps something to do with the reasons that
prompted him to get involved in treasonable activities?
113 Seeck, Briefe des Libanius 13I.
114 Ep. 1517. 5; such introductory letters were Libanius' speciality (cf. W. Liebeschuetz,
His/aria xii (1963) 228 f..
116 Cf. iHS 1964 pp. 56 f.. 116 Anth. Pal. ii.
117 Suidas s. v. Christodoros. These rrchpLC% are obviously the successors of the XT(cre:L<;
performed by the Hellenistic poets at the various 6ufLe:ALxot o<ywve:<; (Christ-Schmid-
Stll.hlin ii. 1 6 141, and below p. 492).
118 When Agathias was in Tralles half a century later he read a rr,hpLo<; TOG O<GTe:O<; [crTOp(CX
there (Hist. ii. 17, p. 102. 4) - perhaps, as Biicheler acutely suggested (Rh. Mus. xxxvii
[1882] 330f.), Christodorus' rrctTpLC%. This need not, of course, entail that Agathias could
not get hold of a copy except by going to Tralles, but it would be natural to assume that the
Trallians would have had several copies made of Christodorus' work and that, being of
purely local interest, very few copies circulated outside TralleS. It is to the (30\)1.1) of Ancyra
that Libanius commends the poet Dorotheus (ep. 1517).

33
 ALAN CAMERON
490
lost), on Tarsus, Anazarbus, Berytus and Nicaea1l9 - a fact which suggests that
he too had travelled widely in Asia Minor. Even Dioscorus of Aphrodito paid
two visits to Constantinople, though principally on business :120 at least one hopes
that he did not recite any of his poems there, or he will have found the circle of
Agathias and Paul the Silentiary rather more discerning than the Duke of the
Thebaid. Mention might also be made of Corippus, though he was not an
Egyptian, nor even a Greek. Emboldened by the success of his Johannis in
Carthage he set sail for Constantinople, and provided for his old age with
panegyrics on the quaestor Anastasius and the Emperor Justin n.l2l
Lastly we may consider Olympiodorus of Thebes. Best known to us as the
author of a copious and accurate history of the period 407-425, unfortunately
preserved only in an abridgement by Photius,122 he was by profession a poet
(7tOL1jTIje;, we; IXUTOe; 'P1jO'L, TO e7tLTIjaEUfJ.IX), and it was evidently his skill as a poet
which launched him on his distinguished career (below, p. 497) as a diplomat. He
was deemed a suitably qualified person to send as ambassador to the Hunnic
king Donatus in 412,123 and in 415 visited Athens, where he was able to exert
sufficient influence to secure a chair of rhetoric for a man called Leontius (who
apparently did not want it!),1 24 He then returned to his native city - of which
he was extremely proud; Homer himself, he claimed, drew his stock from
Egyptian Thebes. 125 As a trusted representative of the Emperor he ....-ill obvious-
ly have resided for periods in Constantinople, and probably lived also in Rome
or Ravenna for a while, since his history shows an intimate knowledge of
Western affairs such as would have been difficult to acquire in the East. 126 In
Rome he was so impressed by the huge size of the mansions of the enormously
wealthy Roman aristocrats that when describing them in his history he actually
burst out into verse: 127 de; MfJ.oe; &O'TU 7tEAEL, 7tOALC; &O'TEIX fJ.OpLIX y.Eo6O:L.

no Attested by the lemma on Anth. Pal. i. 19, an epigram of Nonnan technique, and
therefore not by the Claud ian (Wifstrand, Von Kallimachos zu Nonnos [1933J 159). But we
cannot be sure that the erratic and careless lemmatist of the Palatinus (d. CQ 1965) was
correct in identifying this Claudian with the author of the IIoc't'pLIX he mentions, and the
possibility remains that the author of the IIoc7p'lX is the famous Claudian. Reitzenstein (Das
iran. Erlosungsmysterium [1921J 181 f.) made the attractive suggestion that Nonnus'
detailed section on the K7tm<; of Berytus in xli was based on the IIoc7p'lX of Berytus here
attested; but see KeydeU, Bursians Jahresb. 230 (1931) I I 1-2.
120 Cf. R. G. Salomon, JEA xxxiv (1948) 98t., Jones, LRE i 407-8.
121 The preface to Partsch's ed. of 1878 is still invaluable for study of Corippus; d. also

Stein, Bas-Empire ii (1949) 692 f., and KrestanfWinkler, RA C iii (1957) 424 f.
122 Cod. 80; P. Henry's new Bude ed. of Photius, while providing a more securely based
text, unfortunately omits the conventional numbering of the fragments entirely, and does
not therefore supersede Muller's FHG and Dindorf's HGA! for reference purposes.
123 fr. 18. 124 fr. 28 init. 125 fr. 33 fin. 126 Haedicke, P-W xviii, 1. 201.

127 fr. 43. Thompson (CQ 1944 p. 45) is wrong to claim that Olympiodorus is the only

ancient historian' to break out in verse in a history. Menander Protector includes no

fewer than three whole elegiac couplets of his own composition (fr. 35, HGM ii. 71. 3f.).

34
 Wandering Poets: A Literary Movement in Byzantine Egypt 491

Ironically enough this is the only line of his poetry to survive, but in his lifetime
his reputation stood so high that the chieftains of the Blemmyes, marauding
barbarians of the Egyptian desert and a perpetual threat to the Thebaid
frontier, in their eagerness to meet him (or so he assures us) gave him an
honourable safeconduct well into barbarian territory, to within 30 miles of Abu
SimbeP28 Having voyaged not only round the Mediterranean but far beyond
the Imperial frontiers to both North and South, to the lands of the Huns
beyond Lake Maeotis in one direction and Ethiopia in the other - always ac-
companied, so he tells us, by a faithful parrot which could mimic anything
-rwv,xv(JP<D7t(p 1tPIXTTOfLEVWV and even say 'Olympiodorus'129 - he must surely
have been the most travelled of all the 'wandering poets' of the later Empire.

IV
In the best tradition of the great Alexandrian school of the second and third
centuries before Christ, these latter-day Egyptian poets were not mere pane-
gyrists or Gelegenheitsdichter: they were scholar poets. To complete the remark
of Eunapius already quoted, "The Egyptia."'1s are mad on poetry - but quite
unfitted for serious study".I30 But whereas the first half of his statement is
true enough, the second was inspired by that mixture of jealousy and contempt
that other nations so often felt for the talented but unlikeable Egyptians. Greek
philosophy continued to flourish in Alexandria right down to the Arab conquest
and after, and it was still a centre of learning after Justinian had closed the
schools of Athens. 131 And almost all the 'wandering poets' were scholars and men
of letters, many of them highly regarded in their day.
Pamprepius, for example, after securing himself T<1: aVIXYXIXLIX Tij:J ~(cp in
Athens by means of his poetic talents, became a grammaticus there, and after
a short time €~o1;e )tIXL 1tOAUfLlX(JEO"-rIX-rOC; -rWV lXuT6e~ 1t1X~/)e;[IXC; fLe-re~A1)x6-rwv,
Proclus alone excepted. And after he had moved to Constantinople and secured
the favour of IIIus, he was appointed to a public chair there as well. Nor was
this a mere nominal honour, for he confirmed his scholarly pretensions by
publishing a work on etymology in Athens, and delivering a remarkable
discourse on the soul in Constantinople. 132 Both Andronicus and Harpocration
were evidently keenly interested in philosophy, if they were prepared to go to
128 fr. 37 init.; in fact to Prima, a town long since inhabited by barbarians - as Olympia
dorus himself observes.
129 fr. 36 - recording also a good example of the sort of 'traveller's yarn' that such a
wanderer might be expected to tell.
130 Vito Sopko 493.
131 Simon, Hist. de Neole d'Alexandrie ii (I845) 586f., H.-D. Saffrey REG lxvii
(1954) 396f. Eunapius' prejudice against Alexandrians in particular comes out very clearly
from fr. 83 of his history.
132 Damascius ap. Suidas S.V. Pamprepios.

35
 ALAN CAMERON
49 2
Constantinople specially to study under Themistius, Andronicus actually turning
down a post in the civil service to do so. And Harpocration, according to
Libanius, was a first rate teacher (1tort)T"YJ~ &yod}o~ XIXL 1tIXL~e:\)T"YJ~ &!1e:\\I(u\I) who
was always buried in his books (~e:~L(uXW~ ~E: cr\)\le:x.w~ t\l ~L~A(OL~) - though by
no means a pedant, a thoroughly decent and upright fellow (&.1tAOUC; TL~ W\I X(xL
ye:\I\IIXLOC;).133 Serenus and Horapollon, whose ~pa{LIXTIX have already been
mentioned, were both grammatici. Horapollon in fact was a most distinguished
scholar, who wrote a work on temples (TE{LE\lLxa) and commentaries on Sopho-
cles, Alcaeus and Homer; a mhpLIX of Alexandria and a work on hieroglyphics
ascribed to Horapollon are probably by his homonymous grandson.l 34 He was
considered to be the match of any of the grammarians of old (TW\l1taAIXL AOyL{L-
(uTaT(U\I yplX{L{LIXTLXW\I OUaE\I TL (Le:LO\l X),EOC; &1te:\le:yxa{LE\lO~), 135 and his grandson's
prowess in the same field is highly praised even by the Christian Zacharias
Scholasticus. 136 The influence of the Alexandrian tradition of the Lehrgedicht
is evident in the popularity of 1t1X-rpLIX, versified histories of the founding and
antiquities of cities after the pattern of the Hellenistic XT(cre:LC;.137 Christodorus,
as we have seen, wrote many such 1t1X-rPLIX, that on Constantinople in I2 books,
that on Thessalonica in no fewer than 23; an enormous amount of antiquarian
knowledge must have been required to fill out poems of this length - truly
'Alexandrian' erudition. And as the pupils of Produs he wrote about were
philosophers, as, of course, was Produs himself, he probably dabbled in
philosophy as well.
There was a particularly large number of poet-grammarians in fourth
and fifth century Egypt. Eudaemon of Pelusium wrote a Te:X\I-y) YPIXf.lf.lCXTLX1j
and {l\IO{LIXTLX-Y) opEloypcxrptlX as well as his various poems - and even practised
as a barrister at Elusa for a while.1 38 (Eusebius Scholasticus, author of
the epic on Gainas, was also, to judge by his title, a lawyer by profession, as
later was Agathias). Ammonius, author of another poem on Gainas was a
grammaticus, as also was his colleague Helladius, who compiled a lexicon on
which that of Suidas is largely basedp9 Yet another grammaticus, Helladius of
Antinupolis, had the bright idea of casting a whole encyclopaedia into an epic - a
Lehrgedicht with a vengeance (mercifully it has not survived).140 Triphiodorus
also taught grammar when not turning out his dreary epics on Troy, Odysseus,
Hippodameia XIXL &MIX (the list is in Suidas). A rather better poet was
133 ep. 364. 5. 1M Maspero, op. cit. (n. I I.) 135 Suidas s.v. Horapollon a.
136 Zacharias, Lite of Severus p. 15. 137 See n. 117 above.

138 Seeck, Briete des Libanius 131.

139 For Helladius' poems cf. Suidas s.v.; Suidas does not say that any of the works
he lists are poems, but the titles, including an enkomion of Theodosius II, seem more ap-
propriate to poetry than prose. Scholars often complain of omissions in P-W, but Helladius
has earned by some oversight two separate entries by different scholars on the same page!
Gudeman (viii. 102 no. 3), and Seeck (ib. 103 no. 8).
140 Except in a fairly full prose paraphrase by Photius, Cod. 279.

36
 \Yandering Pacts: A Literary :'Iovement in Byzantine Egypt 493

l\Iusaeus, whose Hera and Leander still finds many admirers; he too was a
grammaticus. H1 So was Palladas, but to judge by his poems, he was not a great
success in that profession, and certainly did not enjoy it. In many a biting
epigram (usually, in typical schoolmasterly fashion, laden with puns) he
laments the woes his profession has brought upon him, the tricks his pupils
play on him to avoid paying their fees (palming him off with lead coins instead
of silver), and it seems that in the end his chair was taken away from him with
the result that he was reduced to penury and forced to sell his books :142
KO()),[fLO(ZOV 1tWAW xcd rHvoO(pov, ~oz xO(t o(t)'t'cXC;

1t't'WO'E~C; YPO(fLfLO('t'LX'~C;, 1t't'WO'LV ~Zwv 1tEV[·~C;.

One thinks also of Bentley's verdict on Nonnus: "he had a great variety of
Learning and may pass for an able Grammarian, though a very ordinary poet. "143
As for Claudian, if his distinguished eighteenth century editor Matthias Gesner
was exaggerating a little when he wrote "Scivit, credo, Claudianus quicquid
tum sciebant homines, et multo plus quam vulgo sciebant etiam qui docti
habebantur",144 his poems reveal that he was thoroughly imbued with the
literature, both prose and verse, and the history, of both Greece and Rome, and
even conversant with what in his day passed for science and philosophy.145 He
was regarded by his contemporaries as a most distinguished man of letters, and
the inscription on the base of the statue erected in his honour by the emperor
and senate describes him as "inter ceteras decentes artes praegloriosissimo
poetarum. "146 Even Dioscorus was a scholar of sorts, probably the most learned
man in sixth century Aphrodito; among other quasi-scholarly activities he
compiled a Greek-Coptic glossary, and, ironically enough in view of the metrical
shortcomings of his verses, a little treatise on weights and measures. I47 But let
Olympiodorus close the list once more. His history reveals that he was an
intelligent and well read man.1 48 Like Herodotus he travelled up the Nile as
far as he could go to'1"Op(O(C; ~VEXO(149 (in fact he reached a good deal further than
Herodotus), recording what he saw for his history, which has a valuable section
on the oases of the Egyptian desert based on his own investigations. 1OO Olympio-
dorns was not, of course, the only Byzantine historian to model himself on
Herodotus, but whereas his fellow-historians limited their imitation to aping his

1U Attested by the manuscripts of the poem. U2 Anth. Pal. ix. 175; d. JRS 1965.
143 Dissertation upon the Epistles of Phalaris (ed. Wagner 1883) 90.

144 Preface to his edition of Claudian (1759) vi.

145 P. Courcelle, Les lettres grecques en occident 2 (1948) 120£., J. Steiner, Das geographische
Weltbild des Claudius Claudianus, Diss. Graz 1950 (uncritical); I hope to discuss the subject
in my forthcoming book on Claudian.
146 CIL vi 1710, ILS 2949. 147 H. I. Bell, jHS !xiv (1944) 30, with refs.
148 Cf. passages collected by Thompson, CQ xxxviii (1944) 45.
149 fr. 37 init.

150 Cf. Thompson, op. cit. 44 n. 2.

37
 ALAN CAMERON
494

style and vocabulary, Olympiodorus eschewed this absurd purist affectation;161

alone among the historians of the age he imitated rather Herodotus' inquiring
and independent mind and actually put facts before literary style - a scholarly
attitude that won him the contempt of Photius (7tpOe; TI)v 7tE7tIl'"lfLEV"f)V XIl"rEV-
"f)VEYfLEVOe; XU8IlLOAOYLIlV, <han fL"f)8' &~~oe; de; O"unpllcp~v eXvllypcicpEcr6IlL b 1.6-
yoe; .... ), but the gratitude of the modem historian. Nor were his scholarly
activities confined to exploration and the writing of history. Whether or not
he wrote the work on alchemy that was until recently ascribed to him,162 he was
keenly interested in philosophy (exov"rll "rou cpLAoaocpEi:v epw"roo), and according
to Hierocles it was discussion with Olympiodorus that sparked off his own work
7t'EPL 7tpOVOLIle;.163
Another aspect of the claim of the Egyptian poets of the age to the title
doctus pocta is that a surprisingly large number of them seem to have had some
acquaintance with Latin literature. It had always been extremely rare for
Greeks to take the trouble to learn what most of them had never ceased to
regard as a barbarian tongue, and in the Later Empire there were few indeed
who took the trouble to learn it for any but strictly practical reasons, with a
view to one of the careers where at least a smattering of Latin was still essential,
the law, the army or the civil service. If Gibbon was exaggerating when he
remarked that there is not one allusion to Virgil or Horace in the whole of
Greek literature from Dionysius of Halicarnassus to Libanius, he was not
exaggerating much. Libanius affected complete and ostentatious ignorance of
Latin, and even required an interpreter to make out a letter written in that
language.l 54 The emperor Julian was able to make himself understood in
Latin ("Latine... disserendi sufficiens sermo", as Ammianus patronisingly
observed) ,155 but" seems guiltless of the most rudimentary acquaintance with
Latin literature" .106 But in sharp contrast to this conventional attitude, as a
result of the widespread Latinisation of Egypt following the reforms of Dio-
cletian, the study of Latin began to flourish, comparatively speaking, in fourth
and fifth century Egypt. Papyri have revealed the unexpected fact that many
Egyptians began in the fourth century to acquire not merely the necessary
smattering of legal jargon in Latin that was required for the bar, but read a
modest range of classical authors, and not only in the traditional centres of
learning. The schools of Alexandria must have had quite a reputation for Latin
in the fifth century, for Severns, the future patriarch of Antioch, was sent there
from Sozopolis as a young man in order to study grammar and rhetoric in both
Greek and Latin. 167 Pupils battled their way through Cicero's Catilinarians

151 Cf. CQ 1964 pp. 312 L, and Averil Cameron, Byz. Zeitschr. Ivii (1964) 33 f.
162 Haedicke, P-W xviii. 1. 202
153 Photius Cod. 314 init. 1M epp. 1004, 1036.
166 xvi. 5. 7. 156 T. R. Glover, Life and Leiters in the Fourth Cmtll'y (1901) 33 n. I.
157 Zacharias, Life of Severus p. II.

38
 Wandering Poets: A Literary Movement in Byzantine Egypt 495

with a word-for-word Greek 'crib' (fragmeljlts of which have survived), and

in addition to these and other prose works, read the poets as welP58 The most
popular poet, as might have been expected, was Virgil, who is represented by
more papyri than any other Latin literary writer; several juxtalinear texts of
the Aeneid have come to light in recent years, some marked with accents and
symbols denoting quantity.u9 Even the 'Silver' poets were read; Lucan was
studied,l60 and even so hard a poet as Juvenal was read ""ith an elaborate
commentary in fifth century Antinupolis. l61 One Virgil papyrus has been found
dating from as late as the sixth century, but by the second half of that century
Dioscorus seems to have known no more than the odd bit of Latin jargon he had
picked up in his legal studies (e.g. a~a't"LvlX't"(6)) and by its close Theophylact
Simocatta thought that rex was Gothic !183
In view of this remarkable interest shown in Latin poetry in Byzantine
Egypt, it is not so surprising as is usually thought that several Egyptian poets
of the age were familiar with a number of Latin poets. Christodorus, for example
alludes in his only extant poem to the Aeneid,183 and Triphiodorus seems
likewise to have read some Virgil.184 It has been plausibly argued that Olympio-
dorus read and modelled his own history on that of Ammianus,185 and if he
studied a Latin history before embarking on his own, it seems only reasonable
to assume that, being a poet by profession, he had read a certain amount of
Latin poetry as well, though none of his own poetry has survived to confirm the
conjecture. He was certainly perfectly familiar with Latin and sometimes
quotes whole sentences of it in his history (he even knew a little Vandal as well).
Of course it is unlikely that more than a fairly small number of particularly
bright pupils attained to any real familiarity with Latin literature, and very
unlikely that more than a handful could write Latin with any degree of fluency,
but in view of the attitude of such as Libanius and Julian, it is significant
enough that there were any. Nonnus read and imitated Ovid, and perhaps the
Latin works of Claudian as well.1 66 And in Claudian we have a poet who wrote
158 P. Collart, 'Les papyrus litteraires latins', Rev. de Phil. xv3 (1941) 112 f., and R.
Cavenaile, Corp. Pap. Lat. (1958) pp. 7-II7; for a more recent list of Latin didactic texts
see G. Zalateo, Aegyptus xl (1961) 196£.
159 None of the nine juxtalinear Virgil papyri so far found dates from before the fourth

century; d. Remondon, Jl. Jur. Pap. iv (1950) 244f. For the numerous Virgil papyri in
general see Cavenaile, op. cit. pp. 1-70. 160 Cavenaile, op. cit. p. II4.
181 C. H. Roberts, JEA xxi (1935) 199f.., Aegyptus xv (1935) 297f.
162 Hist. vi. 9. I; That Theophylact was "t"</l YEVEt AlyU1tno~ is attested by Photius
Cod. 65 init.
183 Anth. Pal. ii. 414; d. Austin, Aeneis ii (1964) p. 289.
1M Funaioli, Rh. Mus. 98 (1939) If., Keydell, Bursians Jahresbericht 230 (1931) 13of.

and 272 (1941) 46.

lSi Thompson, op. cit. (n. 148).
166 J. Braune, Nonnus und Ovid (1935) and in Maia i (1948) 176f.; Keydell Jah,esbericht
272 p. 39 for several further references and see also now G. D'Ippolito, Riv. di. Fil. xc

39
 ALAN CAMERON

with equal fluency in both Latin and Greek, and could boast an intimate
knowledge of the whole of Latin poetry from Lucilius and Ennius down to
Silius, and even the Christian Juvencus as well. 187 On the evidence of the
papyri we might conjecture that Claudian was not alone in actually being able
to compose in Latin as well as read the literature. One particularly interesting
find has given us a school exercise representing a pupil's attempt to rewrite in
his own words a passage from the first book of the Aeneid - a task which he
accomplished in fairly tolerable if uninspiring hexameters. lea Perhaps the
Eusebius whom Eunapius mentions as about to set off to flatter and fawn on
the great of Rome wrote panegyrics on them in Latin like Claudian. It remains
astonishing, but not, perhaps, entirely inexplicable and unparalleled that "in
the decline of arts and Empire a native of Egypt, who had received the edu-
cation of a Greek, assumed, in a mature age, the familiar use and absolute
command of the Latin language, soared above the heads of his feeble con-
temporaries, and placed himself, after an interval of three hundred years, among
the poets of ancient Rome."I69
The fact that so many of these poets were scholars and frequently practised
as professional schoolmasters is of no little importance for a proper appreciation
of their careers. Forit means that they were not wholly dependent on patronage;
they could always earn a respectable living by their professional skill as teachers
when they had exhausted (or failed to arouse) the generosity of the local
notables, and those who knew Latin would find it easy to gain a post in the
civil service, and indeed often did so (below p. 501). With so much competition,
especially in the Eastern provinces, it was probably only rarely that a man could
secure himself a comfortable position as court poet over a period of years, like
Claudian, and it is likely that a large number of our poets reckoned to earn their
daily bread by their teaching, and looked to their poetic skills as a useful though
irregular means of supplementing their income. Conversely the patrons they
obtained by means of their poems were likely to prove of great value in
promoting their teaching careers. For success as a teacher in the later Empire
depended to a large extent on the cultivation of influential connections who
would recommend teachers to well-to-do parents of their acquaintance. 1Sta

(191)2) 300. Similarly many scholars have argued that Quintus of Smyrna was familiar with
a number of Latin poets. F. Vian denies this in his recent Rec1lerc1les sur Ie Postlwmericrs de
Q. de S. (1959) 95f., but see also Keydell's review in Gnom01l xxxiii (1961) 278.
m See the lists of imitations collected beneath the text in Birt's edition and also his
preface, pp. lxiv & cei, and G. B. A. Fletcher, Mnemos. il (1933) 196f.
118 C. H. Moore, 'Latin exercises from a Greek schoolroom', Class. Phil. 19 (1924) 322 and
R. Cavenaile, Les Etudes class. xviii (1950) 285f.
lit Gibbon, Decline and Fall ch. xxx {in. If I have perhaps exaggerated the Latin
scholarship of Byzantine Egypt, I am inclined to think that C. H. Roberts unduly minimized
it in Mus. Helv. x (1953) 276f.
lIta See P. Petit's ad mirable analysis in his Etudiants de Libanius (1956) ch. iii.

40
 Wandering Poets: A Literary Movement in Byzantine Egypt 497

Thus it is not surprising and hardly accidental that so many poets of the
period combined the complementary and mutually advantageous roles of poet
and schoolmaster.
v
Perhaps the most important feature of this school from the point of view of
the historian is that few of them seem to have been content just to recite their
epics and panegyrics or deliver their lectures to an admiring audience. Many
took an active part in public life and even tried their hand at the dangerous
game of politics. Some were successful, and won high office in the state, some
less so.
Among the most successful we may turn once more to that remarkable
character Olympiodorus. He explicitly states himself that he was a poet by
profession, and it is therefore reasonable to assume that he drew attention to
himself in the first instance by means of a few well-aimed panegyrics at carefully
chosen people in high places. He was soon recognised as a man of tact and
eloquence, and became one of the most distinguished diplomats of his day,
representing the Emperor on numerous important embassies to barbarian
peoples, who held him, we are told,170 in the highest esteem. He tells us himself
of one particularly hazardous embassy to the Hunnic king Donatus. On his
arrival at Donatus' camp after a very dangerous sea-voyage, Donatus was at
once foully murdered and promptly succeeded by a certain Charaton. Olympio-
dorus was providentially supplied with gifts from the Emperor with which he
could conciliate Charaton, and Charaton in turn agreed not to cause trouble to
the Romans.l 71 It looks suspiciously as though the intrepid Olympiodorus had
been entrusted with the delicate mission of bringing precisely this chain of
circumstances to pass; if this is a typical example of how he handled Rome's
relations with barbarian tribes, then his reputation as a diplomat was well
earned. But it is interesting to note that his statement that he was a poet by
profession occurs in his history, written when he was probably quite advanced
in years; this implies that for all his success in other walks of life he remained a
poet by profession, and did not just use his poetry as a means to a lucrative
career, thereafter abandoning it.
Another highly successful poet was Cyrus of Panopolis. After catch-
ing the fancy of the literature-loving Empress Eudocia (herself a poetess
of some distinction in her day)17Z he was launched on a meteoric career

170By Hierocles, in the preface to his IItpl IIpo'JO(<X~ (Photius Cod. 314 init.).
171fro 18; cf. Thompson, A History of Altila and the Huns (1948) ch. iii.
17. All that is known of her and her works is collected by Ludwich in his useful Teubner
ed. of 1897. The modern reader is hardly likely to agree with the favourable verdict on her
poetic talent (as exemplified in her one extant poem, on the martyrdom of S. Cyprian)
expressed by Gregorovius in his Athenais (ct. Ludwich p. Sf.).
3" Hisloria XIV/4

41
 ALAN CAMERON

which bore him to the city prefecture in 435 and again in 439, when he
was praetorian prefect as well (the first man ever to hold both offices simultane-
ously),173 and eventually the consulship in 44I. He made himself immensely
popular with the people and was very active in building and in beautifying the
city; he even made provisions for its illumination by night.174 But at the height
of his fortune he was deposed and exiled to Phrygia under the circumstances
narrated above; we still possess the poem he wrote and recited on the occasion
of his expulsion from the city.176 It is again worthy of note that for all that he
was the most popular and distinguished statesman of his day, his primary
claim to fame among his countrymen remained his skill and attainments as a
poet. John the Lydian, writing a century later, remarks that his poetry was
still being read with admiration (he adds that it was the only thing Cyrus was
any good at, but this is just sour grapes),176 shortly after this Evagrius refers
to him as Kupoc; 0 1tOL1jTI)C;,177 and the tenth century lemmatist to the Codex
Palatinus of the Greek Anthology calls him 0 (.Leyo:c; 1tOL1jTI)C;.l18
Another shrewd young man who put his talents to good use was Claudian of
Alexandria. After arriving in Rome in 394, without wasting any time on lesser
personages he wrote a panegyric on the two consuls of 395, brothers, and
members of the Anicii, the most illustrious noble family in Rome. This attracted
the attention of Stilicho, generalissimo of all the armies of the Empire, as
demonstrated by the fact that Claudian's next poem, recited the following
January, while ostensibly a panegyric'on Honorius' third consulship is in fact
skilful propagandal711 for Stilicho. And for the next eight years Claudian
continued to publicise Stilicho's claims, put across his propaganda and answer
his critics. Nor had Stilicho any lack of critics, both in Italy and in Constanti-
nople, where a series of powerful and ambitious ministers dominated the feeble
Arcadiusl80 and steadfastly refused to recognise Stilicho's claim to rule both
halves of the Empire. Claudian must have realised from the start the enormous
risk he was running in binding himself irrevocably to a man such as Stilicho, a
178 John the Lydian says explicitly that he held both offices at once (De Magg. iii. 42).
When Sozomen remarked in the preface to his Eccl. Hist., published in the Hos,that "most
of the prefects" ({mcipxwv o! 7tAdow;) were poets, he was obviously thinking of Cyrus.
174 Priscus ap. Chron. Pasch. p. 588 Bonn.
176 Anlh. Pal. ix. 136 - perhaps only a fragment of a full-length poem; likewise Anth.
Pal. xv. 9 is probably only part of a panegyric he wrote on Theodosius II.
171 De Magg. iii. 42. Sour grapes because when Cyrus was praetorian prefect he discon-
tinued the practice of issuing decrees in Latin, John's speciality (cf. Jones, Studies in
Roman Government and Law [1960] 172£.), and furthermore after a lifetime's loyal service
John's own poetic attainments had taken him no higher in the prefecture than the post
of cornicularius.
177 Hisl. Eccl. i. 19. 178 Lemma to Anlh. Pal. ix. 136.
1'111 On the propaganda aspect of Claudian's work see below p. 503 f.
180 Eutropius, according to Zosimus (v. 12. I), lorded it over Arcadius as though he was
an ox (X\)P~UWV 'AFx:x8£ou x:x6ci7tcp [joax-ljfL:X'O~).

42
 Wandering Poets: A Literary Movement in Byzantine Egypt 499

military dictator of barbarian stock with all but the army against him. He must
have foreseen that if Stilicho should fall he would inevitably fall ...-ith him; yet
he threw himself into Stilicho's cause ...-ith a zest and vigour which must have
made Stilicho's enemies his own, and which they could never have forgiven. In
the event, when Stilicho did finally fall, beheaded at Ravenna in 408, Claudian
was, it seems, already dead,l8l 'felix opportunitate mortis'. But while his patron
lived he had been rewarded by the high and decorative post of tribune and
notary, which admitted him to membership of the Senate, had acquired a rich
wife, and had seen a bronze statue erected in his honour in the forum of Trajan.
Stilicho evidently valued highly the propaganda enshrined in the sparkling
series of panegyrics, epics and invectives Claudian wrote for him from 395 to
404 (ten poems, many of them in several books), for he had them all republished
in an 'omnibus' edition after the poet's death.1 82
Pamprepius too made use of his skill as a poet to serve a barbarian general
trying to dominate a weak Emperor. After his various earlier adventures, he
arrived in Constantinople in spring 476, then aged thirty five. As was his custom
he won Illus' favour by public recitation of a panegyric ,183 and was rewarded at
once with a chair in Constantinople, by 479 was quaestor of the Sacred Palace
and honorary consul, shortly afterwards Patrician. Illus evidently found
Pamprepius very useful, for he continued to protect him in spite of the fact
that his open profession of paganism made him extremely unpopular, especially
with the Emperor, who even had him exiled once when Illus was absent on a
campaign; Illus recalled him on his return. In 482 Illus sent him to Alexandria
to try and win support in the still strong pagan circles of that city for his
treasonable plans, but the Alexandrians, probably wisely, mistrusted him and
would not listen to his proposal. Eventually in 484 Illus revolted, declaring
Pamprepius his master of offices, but the revolt was crushed and the ringleaders
executed; Pamprepius preceded his master to the block, after being caught by
him playing a double game and trying to save his own skin. l84 Thus ended
another adventurer from Egypt, characterised by Damascius as a brilliant poet
and by Malchus as a shrewd politician (7tOAL1"LX'ij~ O"\)'IEO"tWt;; tll7tJ,EWt;;), but
physically ugly (IlD,Il(<; "t"7J'I Xpo[Il('I, dk..(6~t;; 1",xt;; /)ljItL~). arrogant, unscrupulous
and treacherous to the last (&:mO"1"61"Il("1"o<; ~[AOLt;;). It is plain enough from

181 His stream of poems stops abruptly in 404; numerous theories have been propounded
to explain his silence, but the most likely explanation is quite simply that he died in or
about 404 (cf. E. Demougeot, De l'unite a la division de l'empire 1'omain (1951) 288f., E.
Merone, Gio1'n. it. di. fil. vii (1954) 309f.).
182 Birt's ed. of Claudian, praet. lxxviii.

183 Explicitly stated by :\falchus, ap. SUlidas s.v. Pamprepius (b). Rhetorius (d. n. 95)

says Pamprepius wrote many E·(/.(;'[l-~:I( on important people. For the details of what follows
see the studies cited in n. 93.
1M Illus had Pamprepius beheaded in November 484 and threw his body over ilie
ramparts of the fortress in which he himself held out till 488 (cf. Stein, Bas-Empire ii 30-1).

43
500 ALAN CAMERO;';

Damascius' remark OU7t(u yr1.p 'rL~ &~Ot~ 'rWV VUV ~wv'rwv &\l6pw7twV otoc; {:, IIOCfL-
7tptmo~ &yty6\1tL. 185 that he made a very deep impression on his contemporaries.
A good example of the sort of wit that aroused such antipathy against him has
corne down to us. The Syrian neoplatonist Sallustius once asked him the
question l86 'r[ 6tol 7tpO~ &\l6pw7touC;; to which Pamprepius replied 'r[~ ... OUK
ol8t\l w~ oU'r' &yw 7tW7to'rt 6to~ &yt\l6fJ."tJ\I, oU'rt 01J oc\l6pw7to~; Even the astrologer
Rhetorius. a fellow pagan and fellow Egyptian. condemned him as a charlatan
and libertine. His career was in many respects similar to that of Claudian. though
more spectacular; both poets earned by their varied skills eightl 87 years of
fame and power as the right-hand man of a would-be dictator.
Lastly we may turn again to Andronicus. He also. though in a more modest
way. was apparently unable to confine his interests exclusively to his poetry.
Whether or not he was guilty of the charges laid against him in 359. it is signifi-
cant that he was deemed the sort of person who might have been involved in
such an affair. Evidently the authorities did not look upon him as just a
harmless poet. Twenty years later in Antioch after the disastrous battle of
Adrianople in 378. he was asked to write a,panegyric on Hypatius. recently
appointed city prefect of Rome. Oddly enough he refused and left the city. in
spite of the fact that a panegyric on so exalted a personage would have been
quite a lucrative feather in his cap. But shortly afterwards Hypatius got into
trouble for assuming his office in Antioch instead of Rome. and was later
actually accused of treason. though he Was in fact acquitted of the charge. It
looks as though Andronicus had his ear very close to the ground and anticipated
these developments. If so. then he was shrewder than Libanius (usually very
wary in such matters) who was eager to step in and fill the breach at the time.
but later congratulated himself on not having done SO.188 Mindful of his narrow
escape in 359. Andronicus took good care not to make the same mistake twice
and resisted the temptation that lured so many of his fellow-poets to venture
into the dangerous world of politics. It is not recorded that he held any public
posts. Seeck's claim that he was made governor of Cappodocia was based on a
letter of Libanius that has been shown to be a forgeryl883: when offered such a
post in 359 he preferred to go and study philosophy instead. and it would seem
that he was as politically awake as a Claudian or Parnprepius. but without
their ambition.

How is it that poets could corne to play such an important role in public
life? The answer is twofold. In the first place men of letters were very highly

1& Suidas s.v. Sarapion b. fin. IN Suidas s.v. Salustios c. fin.

187Claudian: 396-404. Pamprepius: 476-484.
188 Libanius or. i. 180-1; this is deduced by Dr. A. F. Norman in his forthcoming com·

mentary on the speech ad loc .• which he was kind enough to show to me.
lB8a Bouchery, cp. cit. (n. 107) ISO n. 10.

44
 Wandering Poets: A Literary Movement in Byzantine Egypt 501

thought of and frequently rewarded with posts in the Imperial administration,

in the second poetry served, or could be made to serve, a more practical purpose
in the later Empire than is generally realised.
Cyrus, for example, was promoted to the responsible posts of both the city
and praetorian prefectures (in the West Ausonius also became praetorian
prefect), and Olympiodorus was entrusted with Rome's relations with barbarian
tribes. The poet Icarius received the post of comes Orientis for a panegyric on
Theodosius; TIjv OCPXT,v a6AOvdX&v htwv remarked Libanius. l89 Aeneas of Gaza
wrote a letter to a friend commending a poet called Paul: l90 e:t {Lh vEoe; ~v 0
XP"y)G"t"oc; notuAoC;, he said, 7tO~"I)TIje; 8E,"t" ~{L~e; ~;~OC; ~v. He goes on to explain that
Paul's case is rather different, for he is an old man; evidently both Aeneas and
his friend took it for granted that a bright young poet could reckon on walking
straight into his "t"~{L~ (= Jwnos, magistratus) without any difficulty. Corippus
gained admission to the sacra scrinia by means of his poetry,l9l and John the
Lydian's panegyric on the praetorian prefect is not likely to have damaged his
immediate prospects in the officium of that prefecture (it was not John's fault
that the prefect soon changed, and with him John's luck).192 Even Dioscorus'
humble efforts secured him a comfortable position as notary in Antinupolis.
It may seem strange that men with such apparently unpractical qualifications
as poets should be promoted to responsible administrative posts.l93 But, as A.
H. M. Jones has recently observed, "the traditional Roman view that adminis-
tration was something which any man of normal ability could undertake
whether it involved finance or jurisdiction, still prevailed ... and the Emperors
who made the appointments regarded themselves as distributing prizes as much
as choosing suitable persons to carry out administrative taskS".194 Where such
a system prevailed and where literary culture was so highly prized,l9S it seemed
only natural to reward in this way men who had distinguished themselves in
so noble a field as poetry, and though probably only few poets turned out to be
such able administrators as Cyrus or such skilful diplomats as Olympiodorus
there is no reason for supposing that poets in general were significantly less able
to carry out the duties required of them than the next man.ID6 Their success in

1811 Or. i. 225.

19<1 ep. compare also Symmachus' remark that "iter ad capessendos magistratus
10;
saepe literis promovetur" ('ep. i. 20).
III Partsch's ed., praef. xlvi-ii. m H. 1. Bell, JHS lxiv (1944) 34.
m For vigorous criticism of the practice cf. Alf5ldi, Conflict of Ideas in the Late Roman
Empire ch. v, and for a more sympathetic approach, Marrou, History 0/ Education in
Antiquity (Tr. Lamb 1956) 3II f.
1JN LRE i 386, 383.
lit On the high premium set on literary culture see Alfoldi, op. cit. ch. v passim.
1M For a very low level of technical competence and virtually no specialist qualifications
were required; cf. the analysis of Jones, LRE i 386f. Latin, however, was always at a
premium - which means that several of our poets will have had a positive qualification for

45
 502 ALAN CAMERON

public life would not have seemed so surprising to contemporaries as it does

to us.
More interestivg and important is the practical purpose to which poetry
could be put.l96a It has long been realised that the purpose of the panegyric was
not solely to gratify the vanity of the recipient, but to represent him as he wished
to be thought of by others, to emphasise certain characteristics (or alleged
characteristics) and to publicise certain policies which he claimed to be fol-
lowing. Few if any would believe all of it, and intelligent men would believe
none of it, but a fairly large number of people could probably be reckoned on to
believe some of it. The poet, with his panegyrics, invectives and epics, filled a
position in society somewhat analogous to that of the journalist today. To
capture a clever and able poet like Claudian was like gaining control of a leading
newspaper; to antagonise him like incurring the hostility of the press. Libanius
recommends a certain Dorotheus197 ~8ov..oc (.LtV 00Y. &vw Moua(;)v •.. t7tOCLvtaocvroc
(.Ltv 't"ou~ 8(lVTOC~, 't"ou~ 8t 00 86v .. oc~ 00 !J.t(.Lljia(.Ltvov. The implication is that
a poet who was not paid what he asked for his poems might lampoon his erstwhile
patron, and if he was an Egyptian with a flair for invective (above, p. 479), he
might be able to cause a good deal of pain and embarrassment, if not actually
damage the man's reputation and social standing. Claudian did not for a
moment expect, or even intend, his invectives against Rufinus and Eutropius
to convince his listeners that Rufinus and Eutropius were really anything like
as black as he painted them l98 - but he knew that some of the mud would
stick; that for ever more men would think instinctively, whenever the name of
Rufinus' or Eutropius came up in conversation, of Claudian's savage caricatures
rather than the far less interesting and spectacular truth. Copies of the latest
work of a poet like Cyrus or Claudian would be recited in literary circles all over
the Empire. They would not, of course, reach a very wide public, for the masses
would have neither the opportunity nor the education to appreciate such
productions. But the favour of the masses and the army could be won by less
subtle means. It was precisely the educated classes, those who would attend
the recitations and read the published works of such poets, that an Emperor or
general or politician would want to conciliate. Pamprepius' patron Theagenes

office (above p. 494 f.). Even today in our own civil service, accordingtoC. Norlhcote Parkin-
son, "we should do well to remember that administrative ability is far less specialised than
most people suppose" (The Law and the Profits [1960] 178).
UI& Not that this was anything new in the Greek world: compare the pages in which
Eduard Fraenkel characterises the practical & often political ends to which the poets of
the so caJled 'Lyric Age' had put their gifts 1000 years before (Horace [1957] 36f.), and ct.
also G. L. Huxley, Early Sparta (1962) 54; one thinks also of the Augustan poets.
197 ep. 1517.

1118 Nevertheless we find many of Claudian's slanders repeated by several ancient (and
still too many modem) writers who, one would have thought, should have known better.
I shall be discussing Claud ian's success as a propagandist in my study of him.

46
 Wandering Poets: A Literary Movement in Byzantine Egypt 50 3

was a man of no importance at all outside Athens, but Pamprepius' panegyric on

him caused his purely local reputation to spread at least as far as Egypt, where
a fragment of it has turned up on papyrus.1" Palladas' poems were read and
even translated into Latin during his lifetime in far off Gaul. 200 and copies of
Claudian's In Rufinum and In EutTopium reached St. Jerome's cell in Bethle-
hem within months of their first performance in Milan.201 We may be sure that
Stilicho took care to see that Claudian's work was rapidly disseminated in
Constantinople, and there is indeed evidence it was actually read in the circles
hostile to Stilicho there.20z It was worth enlisting the services, or at any rate
staying on the right side, of poets who could command a circulation like that.
And we can see from the case of Claudian that an intelligent poet could employ
a far more subtle technique than merely to heap praises on his patron and abuse
on his patron's enemies. He could present a systematically misleading and
doctored version of the facts that in default of other and more accurate versions
might often carry some sort of conviction with a large section of those who
mattered. This is precisely what Claudian did. When he arrived in Rome
Stilicho was engaged in trying to persuade the world that Theodosius the Great
had a few months earlier on his death-bed entrusted to Stilicho the regency of
both his sons, and de facto therefore the government of the whole Empire - a
contention made no more plausible by the fact that there had been no witnesses
but Stilicho to this momentous pronouncement of Theodosius,203 and that
Arcadius, Theodosius' elder son, was already nineteen, quite old enough by
Roman standards to rule in his own right. Claudian's poem contains a speech
put in the mouth of Theodosius saying precisely what Stilicho had claimed
that he said. Theodosius is addressing Stilicho:104
"Tu curis succede meis, tu pignora solus
Nostra fove; geminos dextra tu protege frat res
.. '. iamiam securus ad astra
Te custode ferar .....
No-one could wish for a clearer presentation of the claim that Stilicho was to
be regarded as the de facto heir of Theodosius. A few years later, in his poem on
the Gildonic war (recited only a matter of weeks after news of the victory
reached Milan), he played down the part of Arcadius' ministers in fomenting
111 See n. 64 above.
100 L. A. Stella, Cinque poeti dell' antologia palatina (1949) 382 f. His popularity there has
been confirmed by the recent discovery of the Epigr. Bobiensia; cf. CQ 1965.
101 H. L. Levy, AJ Ph lxix (1948) 62f., and my article in Vigiliae Christianae xix
(1965). - C. Lacombrade, Pallas v (1956) 15f.
103 As Claudian himself admits at iii Cons. Hon. 142f. I shall be discussing these
questions in detail elsewhere, but for the moment see Fargues, Claudien (1933) 57f., Maz-
zarino, Stilicone (1942) 91 f., Demougeot, op. cit. (n. 181) passim, and Straub, La Nouvelle
Clio iv (1952) 94f. 1M iii Cons. Hon. 152f.

47
 ALAN CAMERON

Gildo's revolt against Honorius, and gave the impression that it was entirely the
result of Gildo's criminal ambition. We can see from the letters of Symmachus
that Western aristocrats knew very little about Eastern affairs, and it may well
be that, even if they thoroughly mistrusted Claudian, they were taken in by at
least some of his propaganda just because there was no other version by
which to expose and confute him. In a period when communications were slow
and inefficient it will not have been difficult for Stilicho to censor such reliable
information as did reach Rome and Milan. Such poems as Claudian's might
also serve as a convenient sounding board for testing the public reaction to a
proposed course of action. For instance, in celebrating the marriage of Stilicho's
daughter Maria to Honorius Claudian expressed the wish that Maria's sister
might find a similar match: a 'similar match' could only mean that she should
marry Honorius' brother Arcadius. Reaction to the suggestion that Stilicho
should marry his other daughter to the other Emperor was evidently quite
negative (the first match was unpopular enough) and Claudian has no more to
say about it. Another example of Claudian's systematic distortion of the truth
is his insistence that Arcadius' ministers acted without the authority and
against the wishes of their Emperor; and in all ~s vilification of them he is
careful not to say a word in criticism of Arcadius himself. In the first place
it would have jeopardised the plausibility of Stilicho's claim to be acting in
Arcadius' interests if his propagandist attacked Arcadius, and in the second it
would not do Stilicho's standing in Italy any good if it became realised that
Arcadius was in sympathy with the hostility his ministers displayed towards
Stilicho's ambition. To answer Stilicho's critics in Italy all Claudian's ingenuity
was called for. It is amusing to watch him lean over backwards to defend
Stilicho from the charge of intriguing with the Visigothic chieftain Alaric. It
was alleged that Stilicho had deliberately let Alaric go when he had him in his
power, and plotted with him instead of crushing him - an accusation not
wholly without foundation.204a Claudian stoutly (if lamely) protests that:
"ignoscere pulchrum
iam misero, poenaeque genus vidisse precantem",
and puts into Alaric's mouth a plaintive lament to the effect that he would
infinitely have preferred to perish, army and all, than be left alive, disgraced
and friendless 205 (but with enough friends to sack Rome itself eight years
later!). All this is exactly what one would expect of a partisan journalist, and
it may be conjectured that Claudian's propaganda attained that modest degree
of success with at least the less politically alert of his audience that a controlled
press will always encounter.
But if Stilicho was the first to find a poet to serve his interests in
In this way, his rivals in Constantinople were quick to follow suit. \Ve

204& See the opposing views of Bury, LRE i 173 and Mazzarino, Slilicone 273f.
0()6 Bell. Gel. 81 f., vi Cons. Ron. 305 f.

48
 Wandering Poets: A Literary Movement in Byzantine Egypt 505
hear from Synesius of the poet Theotimus, who "'Tote panegyrics in honour
of Anthemius, a successor of Rufinus and Eutropius as the power behind
Arcadius' throne, and Eusebius Scholasticus wrote an epic on the revolt and
death of Gainas, an old ally of Stilicho whom Stilicho had in fact used to rid
himself of Rufinus. Eusebius' poem is lost, but Socrates based his version of the
revolt on it, and it is easy to see that Eusebius completely misrepresented the
affair. 206 It is only natural to assume that Theotimus and Eusebius presented
the inhabitants of Constantinople with as carefully doctored a version of the
facts as Claudian concocted (from the opposite point of view) for his audiences
in Rome and Milan. It is not known whether Theotimus or Eusebius were
Egyptians but Ammonius, who wrote another poem on Gainas a few years later
certainly was.:I07 So also was Synesius of Cyrene,208 who wrote a number of
highly tendentious propaganda pieces for Anthemius' predecessor Aurelian, in
both prose and verse. Only the prose works, 7te:pt ~(£crtAe:((£C; and ' Aty\m'not
have survived, but he alludes to the poems at 7te:pt ~(£(jLAe:l(£.; § 18, where he
says that "many others" were writing such poems as well. Aurelian was the
leader of a 'nationalist' movement which sought to wrest control of the armies
and public affairs from barbarians (Gainas was a barbarian and Stilicho the son
of a barbarian), and Synesius' pamphlets (and doubtless his poems as well)
glorify the 'nationalists' and denounce the barbarians and those who trusted
and relied on them.
Combination of several neglected entries in Suidas can perhaps cast some
light on another example of such 'journalistic warfare', this time at the court
of Zeno.
Suidas S.v. rr(£VOA~W';. rrQ(VOA~tO';, E7tWV 7tr:.t1j'rIj.;, e:YPlXye: OtaqJoplX, xd 7tpOC;
,At8tpwv fLe:'!,x TIjv VOGOV o[ E7tWV, X'XL 7tpOC; 'Epu6pwv, XlXt 7tpOC; ~wpo8e:ov
~yefLOV(£ XlXt XOfL1),!IX, XIXL de;' A(jl6ovwv XOfL1j,!IX, XIXL Em'!&.(jlwv 'Y7tIX'![IXe; 6uYIX'!poe;
'Epu6p(ou.
The addressees of Panolbius' poems were evidently people of some distinction.
Dorotheus was a dux and comes (a military man), Aphthonius a comes, and
Eruthrius, who is known from other sources, was praetorian prefect no fewer
than three times. We learn from Malchus that when he was appointed prefect
for the third time by Zeno he discovered that the finances were in such a low
state that they could only be brought up to the required level by harsh taxation.
Being of a philanthropic disposition he refused to be a party to such measures
and asked Zeno to be relieved of his office, an act which caused public dismay
(AU1t1)V '" Tfi 7tOAe:t 7tlXpecrx.e:v) for he was regarded as the only politician of

206 I hope to discuss this question elsewhere.

207 See n. 63 above; the name Ammonius is extremely common in Byzantine Egypt.
:ro8 Cyrene is in the province of Pentapolis,. which is technically distinct from Egypt, but

came under the jurisdiction of the Patriarch of Alexandria. Synesius was educated in
Alexandria.

49
506 ALAN CAMERON

the day who really had the interests of the people at heart (fLOVOC; ,\
e:1t"\ -
T<ilV
mx.VT<ilV &:y0l:6ii> E1t"e:<:pUXe:\).:Iffl Now Aetherius, I would su~est, is the poet
Aetherius 210 mentioned s.v. by Suidas: .
'A\6tp\oe;, btw'l 1t"Ot'1j7t,C;, ~Y?O('fe: a tcJ.<:po pOI: , xoct bt\6ocM:.I.~ov a~ 'f:.1t"WV
de; 1:~tJ.7tA[Xtov TO'l LatO'l &:ae:).?6v.
Now it seems fair to assume that Suidas copied the titles of all these various
poems accurately from his source, Hesychius of Miletus (who only \\Tote a
generation after Panolbius and Aetherius.) If so, then it may at once be
remarked that whereas Panolbius \\Tote poems 7tpOC; Aetherius, Dorotheus and
Eruthrius, he wrote de; Aphthonius. The distinction between the use of 7tpOC;
and de; in the titles of books and poems is quite clear and universal: de; is used
of a book, poem, hymn etc. written in someone's honour, 1t"p6e; of a work written
to refute, answer or discredit someone. It seems therefore legitimate to deduce
that Panolbius wrote invectives on or answers to Aetherius, Dorotheus and
Eruthrius, but a panegyric on Aphthonius. Let us turn back to Ma1chus' account
of Eruthrius' third prefecture. It is likely that he was not writing sine ira et
studio in his eulogy of the popular and philanthropic Eruthrius and condem-
nation of the grasping and unpopular Zeno. Zeno was very unpopular in
senatorial circles; indeed his reign was one long series of conspiracies.211
Ma1chus was \\Titing from the point of view of one of the factions hostile to Zeno.
If we turn once more to Suidas, this time s.v. 'HpOl:tcrxoe;, we find that:
o
Te yilp 'AfLfLw'ItOe; xoct 'Epu6ptoe; 0 ' AtyU7t'noe; ate:fLcJ.XO'lTO 7tpOC; &:M~AOUC;
EV BU~0I:'17(<P XOl:l aU;:"r€Ae:\ 7tp0<il6w'l &d 0 ~Te:pOe; TO'l ~'t'e:pO'l de; 't'ooe; ~O'x.cJ.'t'oue;
x\'1Mvoue;.
Chronology and location fit exactly, and there is no reason to doubt that this
is the same Eruthrius.212 We do not know and it would be pointless to try and
guess what he and Ammonius (presumably the son of Hermeias) were so
bitterly quarrelling about, but we have at any rate positive evidence that
Eruthrius had enemies in Constantinople who openly attacked him, as well as
an historian who sung his praises. It may be that Panolbius and Aetherius were
spokesmen for the opposing parties, poet-journalists like Claudian and Eusebius.
But not even now have we exhausted the list of poets who became involved
in politics at the court of Zeno. In 490,213 shortly after the revolt of Illus and
Pamprepius had finally been crushed, Zeno had a certain Pelagius put to death,
009 /1'.
6 = Suidas s.v. Eruthrios.
210 Christ-Schmid-Stlihlin, ii. 2" 1079 identify Suidas' Aetherius with a sixth century
grammarian of that name, but his entries often cross-refer.
m Cf. Bury, LRE i ch. xii, Stein, Bas-Empi1'e ii ch. i.
m The passage was missed by Seeck in his P-W article on Eruthrius (vi. 1. 602).
213 All our sources date this event to 490, but Stein, Bas-Empire ii 75 n. 3, argues that it

must have happened in 486. If he is right, then Zeno's action is perhaps more understanda-
ble, for Illus would still have been holding out against him, and he could not have risked
another plot at home.

50
 Wandering Poets: A Literary }\Iovement in Byzantine Egypt 50 7

a distinguished epic poet (OIXUflIXO"':"oV Xo:L 7tc;pl 7to["fjO"w E7tW'1 &:;~6),oyov). 214
Pelagius, like Olympiodorus, was a diplomat, and had represented both Zeno
and his predecessor Leo on important embassies to barbarian kings,215 being
rewarded with the title of patrician. 216 We are told by Malalas 217 that Zeno
asked a clairvoyant who was going to succeed him, and was told that it would
be one of the silentiaries, whereupon he promptly had Pelagius, who had once
been a silentiary, arrested and shortly aftenvards strangled. It may be, of
course, that Pelagius was the innocent victim of Zeno's fear and credulity, as
Malalas evidently believed. On the other hand the fact that Zeno's successor
Anastasius was one of Zeno's silentiaries strongly suggests that the whole story
about the 'prophecy' is a post evenium invention of Pelagius' adherents to
whitewash his memory and blacken that of the hated Zeno, which suggests in
tum that he may not have been so innocent after all. Significantly enough
Zonaras, who knows nothing of the prophecy, says that Pelagius was accused
of paganism. 218 Whatever the truth of the matter, it is interesting that of all the
silentiaries and ex-silentiaries he might have picked upon as the man most
likely to plot his overthrow, Zeno singled out a poet.

VI
It is not hard to see why poetry became such a popular profession in the
later Empire. A good poet could always reckon on his services being in demand.
If he was content to remain a poet he could rely on a steady income from the
writing of panegyrics and epithalamia for high officials, epics on the campaigns
of generals, and 7tlh'P~1X of the conveniently ancient cities of Asia Minor. But a
few years residence in one place would normally exhaust the subject matter it
could offer in the way of both local history and dignitaries, and he might then
set up school as a grammarian, or if he had ambitions in other directions, seek a
post in the Imperial administration, or cast in his lot with a politician and
become a propagandist.But it would seem that few were content to stay in one
place for long, and we find them wandering from city to city in search of fresh
interests and new patrons; doubtless, like Andronicus and Harpocration,
armed with letters of introduction from the last patron where possible, though
on occasion it will have been the displeasure of a patron that obliged them to
leave (as in the case of Pamprepius), often the result of one of those injudicious
lampoons (Seneca's "periculosi sales") that the Egyptian found so hard to

Ii( Theophanes Chyon. p. 135. 30 de BOOT. Pelagius' activity as a poet has escaped the
notice of Christ-Schmid-Sta.hJ.in.
215 Malchus f1'. 19, HGM i. 420. 6, fY. 2, ib. 386. 27 (restoring IIoAtl·(lOV for the TZAOYlO'1
of the manuscript).
2U Malalas p. '390, Chron. Pasch. p. 606, Theophanes I. c.

217 p. 390. 218 xiv. 2. P n. 53. p. 257. 19 Dindorf.

51
508 ALAN CAMERON

resist. But the chief motive was probably always quite simply the love of travel
and adventure. 219
In Ptolemaic and early Imperial times Egypt (that is to say the hinterland
of native Egypt as distinct from the independent and hellenised Alexandria)
had been notorious as a cultural backwater; all educated Greeks (and especially
those of Alexandria) would have echoed Theocritus' description of its in-
habitants as &Mci),OL~ O!lIXAOl, XIXXO: 1tIXlZ'lLIX, 1tci'l't'e:c; &piiLOL.220 It was not until as
late as the third century A.D. that Upper Egypt made its first contribution to
the annals of Greek literature, in the person of Plotinus, but by the fifth
century the Thebaid was the most productive source of Greek poets in the whole
Empire. Modern scholars tend to deplore the sterility and oppression of the
'Byzantine servile state' in Egypt,221 but the fact remains that it was not till
Byzantine times that Egypt could boast that she was in the literary van. 222 But
her domination was short-lived; after not much more than a century or so the
'wandering poets' were no more. It is not always profitable to ask why literary
movements are born and fade away, but the reasons for the collapse of Greek
poetry in Egypt are not far to seek. The poets, as we have seen, were almost
all pagans, and when paganism died in that breeding ground of poets, the
Thebaid, Greek poetry died as well. The well-to-do pagan landowners of the
Thebaid were converted around the middle of the fifth century. Pamprepius,
the last of the 'wandering poets', and the last pagan poet known,223 was born
in 448. Furthermore the language of Egyptian Christians had for some time
been largely Coptic, and with the dominance of Schenute, who knew Greek but
wrote only in Coptic,2M it began to supplant Greek, and did so almost entirely
after 45I, when the fourth Ecurnenical Council atChalcedon condemned Mono-
physitism, the religion of the Egyptian Church, which signalled its severance
from the Greek Church by abandoning Greek as its official language. After this
not only theological literature but secular works and even poetry were written
in CoptiC. 226 It is hardly surprising that Greek poetry could not survive the
change. Not that the attempt was not made in some quarters to keep the

21t Similarly one suspects that it was not always only a burning desire to see the holy

places that led so many pilgrims (often via rather circuitous routes) to Palestine and Egypt
in the fourth and fifth centuries. Cf. Glover. Life and Letters 125 f.
!20 xv. 50: for the last word see Gow ad loco

221 The phrase is H. I Bell's, JEA iv (1917) 86f.

222 In his Greek City (1940) 283-4 Jones comments on the fact that the names and

metropoleis of Egypt were intellectually quite barren until after proper cities had been
established there by Diocletian, v.ith curiae after the Hellenistic model. remarking that "the
coincidence may be fortuitous, but it would seem that city life was a real stimulus to
literature". See also his review of Bell's Egypt (1948) in J RS xxxix (1949) 172.
223 "Der letzte heidnische Dichter. von dem wir wissen" (Keydell, poW xviii. 3. 414).

224 J. Leipoldt, Schenute von Atripe (n. 14) 71 f.

226 Cf. Maspero, REG xxiv (19Il) 481.

52
 !vliszclle 50 9
torch alight; witness Dioscorus. But Dioscorus was a Christian, and sadly
deficient as his poems are in respect of grammar, metre, style and content alike,
they are in fact the clearest proof that the flame had already gone out. 226
Bedford College, London Alan CA~IERO~

m The fact that he compiled a Greek-Coptic glossary is a clear enough sign of the
changing times.
I am grateful to Professor Rudolf Key-dell, Professor Hugh Lloyd- Jones, Mr. Keith
Hopkins and :';lr. Peter Parsons for reading a draft of this paper and making valuable
comments.

53
This page intentionally left blank
 The Sixth Sibylline Oracle
as a Literary Hymn
M. D. Usher

I N 1856 C. ALEXANDRE! remarked that the twenty-eight-line

poem that comprises Book 6 of the Dracula Sibyl/ina is "not
so much a prophecy as a hymn, and, apart from the meter,
nearly a lyric."2 Subsequent commentators have used the term
"hymn" to characterize Book 6, but no one has provided a
formal literary analysis of the book as such, or tried to classify it
within a larger hymnic tradition. 3 The following study is offered
toward that end.

1 The following abbreviations will be used: Excursus=C. Alexandre, Excur-

sus ad Sibylla (Paris 1856); OracAp=H. W. Parke, The Oracles of Apollo in
Asia Minor (London 1985); OS=J. Geffcken, Dracula Sibyllina (Leipzig 1902);
SibProph=H. W. Parke, Sibyls and Sibylline Prophecy in Classical Antiquity
(London 1988); SW=A. Kurfess, Sibyllinische Weissagungen (Berlin 1951);
TU =J. Geffcken, "Komposition and Entstehungszeit der Oracula Siby Hina, "
Tate und Untersuchungen 23 (Leirzig 1902); VisDor=A. B. M. Kessels and
P. W. van der Horst, "The Vision 0 Dorotheus (Pap. Bodmer 29)," VigChr 41
(1987) 313-59.
The following will be referred to by authors' names alone: J. J. COLLINS,
"The Sibylline Oracles," in J. Charlesworth, ed., Old Testament Pseud-
epigrapha I (New York 1983) 317-472; E. BErTSCH, Die griechische DichteT-
fragmente deT romischen Kaiserzeit (Gottingen 1961); R. JANKO, "The
Structure of the Homeric Hymns: A Study in Genre," Hermes 109 (1981)
9-24; L. KAPPEL, Paian: Studien zur Geschichte einer Gattung (Berlin 1992); K.
KEYSSNER, Gottesvorstellung und Lebensauffassung in griechischen Hymnus
(Stuttgart 1932); W. H. RACE, "Aspects of Rhetoric and Form in Greek
Hymns," GRRS 23 (1982) 5-14; D. F. SUTTON, ·P. Ross. Georg. I.11," in
Papyrological Studies in Dionysiac Literature (Oak Park, II. 1987); J. J.
THIERRY, Christ in Early Christian Greek Poetry (Leiden 1972).
2 Excursus 615: "non vaticino, non historiae sed potius hymno similis ...
paene Iyricus."
3 Collins (406) does not address the poem's hymnic or poetic qualities.
Thierry but not Heitsch includes it in his collection. M. Lattke, Hymnus:
Materialien zu einer Geschichte der antiken Hymnologie (Gottingen 1991) 25,
has an up-to-date bibliography but no new contribution.

55
26 THE SIXTH SIBYLLINE ORACLE

The poem is best described as an "oracular hymn" that

celebrates, as a vaticinium ex eventu, the career and apotheosis
of Jesus of Nazareth. An unusually large number of bucolic
diaereses and a relatively high poetic quality set Book 6 apart,
statistically and stylistically, from the rest of the Sibylline
corpus. 4 The poem's chief importance, however, derives from
its status as our first extant piece of Christian hexametrical
poetry. Lactantius' citation (Div. Inst. 14.15-22, published ca
311) provides the decisive terminus ante quem and makes Book
6 at least as early as P.Bodm. 29 ("The Vision of Dorotheus"),
hailed by Kessels and van der Horst as "the earliest known
specimen of Christian hexametric poetry." Their dating of the
Dorotheus poem to the late third century, however, is nearly a
century too early and has been convincingly refuted by J. N.
Bremmer. s Book 6, then, written at least a generation before
Lactantius and-as I shall argue-in Syria-Palestine by a Chris-
tian Middle Platonist during the Severan period (193-235), is the
real beginning of Christian poetry.
Part I of this study presents a text, translation, and line-by-line
commentary on the poem with a new interpretation of some
troublesome lines and stylistic and semantic parallels from
hymns, oracles, and other relevant literature. Part II addresses
the problem of the poem's mixed genre and its literary and
historical contexts. This two-part format best facilitates
discussion of this little known work within the larger context of
Greek hymns. Impatient readers are invited to read the hymn
and proceed directly to Part II.

4 A 71 % incidence (20128 lines) as compared with 44% for the Orac. Sib. as
a whole. I base these numbers on the text I produce below. In his study of the
Sibylline hexameter based on Geffcken's edition, J.-M. Nieto Ibanez, El
Hexametro de los Oraculos sibilinos (Amsterdam 1992) 156-60, reckons
64.2% for Book 6 (18/28), as compared with 60% for Theocritus, 58% for
Nonnus, and 56% for I Iomer.
5 In J. den Boeft and A. I Iilhorst, edd., Dei Laudes: Essays on Early Chris-
tian Poetry (Leiden 1993) 253-62. Bremmer also shows that the identity of
Dorotheus as the son of Quintus Smyrnaeus is unfounded. The uther Dor-
otheus poem from the Bodmer collection, an acrostic about Abraham and
Isaac, is dated by E. Livrea, ·Un poema ineditio di Dorotheus: Ad Abramo,"
2PE 100 (1994) 175, to the late fourth century.

56
 M.D. USHER 27

I
Book 6 owes its current place in printed editions of the Oracula
Sibyl/ina to Alexandre, the first to separate it from Book 7. Most
Mss. of the major <l> and 'I' families include Book 6 with the sub-
scriptions Myo~ EK'O~, (J'tlXOl K'rl.6 In establishing a text, given
the complicated textual history of the Orac. Sib., I agree with
Geffcken (and reproduce here with abbreviated apparatus) the
Mischtext he concedes is inevitable. Where I have differed
from him I have done so, following his own advice and prece-
dent/ using parallels drawn from hymns and related literature.

a8ava'tOu IlEyav uiov aOl5llloV EK <j)PEVO~ a'Mil>,

C!> 8povov U"'l(HO~ YEve'tTl~ 7tape5WKE Aa~eo8at
OU7tW YEvv118evn' E7td Ka.a oapKa .0 OlOOOV
Ttyep811, 7tpoxoa'i~ U7tOAOUOaIlEvo~ 7tOWIl0'io
5 '!opoavou, o~ <j)epEWl yAauKcp 7to51. KUllaw oupwv.
EK 7tUPO~ h:<j)Eu~a~ 7tpGHO~ 8EOV lhVEWt 'h5U
7tVEUIl' E7tlYLVOIlEVOV, AEuKa'i~ 7t.Epuyeaol 7tEA.ell1~.
uv8i)on 0' av8o~ Ka8apov, ~pUOOUOl 5£ 7tl1yaL
5et~n 5' UV8POl7tOlOlV 08ou~, oet~El o£ KEAEU8ou~
10 oupavla~' 7taV'ta~ O£ O'O<j)Ol~ IlU80l0l 5l5a~El.
a~n 5' €~ .E OlK1)V Kat 7tetOEl Aaov U7tEl8ii
aivEtov auxi)oa~ 7ta'tpo~ yevo~ oupaVloao'
KUIla'ta 7tE~£UOEl, VOOOV uv8pW7tWV U7tOAUOn,
o.i)on 'E8vl1Gyta~, U7tWO£Wl aAyea 7tOAAa'
15 fK O£ Illll~ 7ti)Pll<; apwu KOpO<; €OOnal uvopil>v,

3 'to ~itooov cp 'If / OOSEt(Jav Q 6-7 iK '¥ fl> / 0<; Q SEOU OIjlE'tal Tjou
7tVt:l*' fmywollEvOV Fabricius /9EOV OIjlE'tal Tjouv 7tVEUllan YEvvTJ9tv-
'ta Q / 9EO<; 0<; 'tE Kat TjOuv 7tVEUllan YWOIlEVOV fl> / 9EO<; O'tE Tjouv
7tVEUllavn YWOIlEVOV '¥ 12 ~pu(Joum codd. aut ~pUOU(JL / ~pl9oum
LactA.13 (Migne) 11 a.~n ~ '¥ I' T\~n Q a,1tnSll Alexandre / a,7tExOll
codd. 13-15 secundum Lact. 4.15, 25 15 mlPTJ cf Lact. 4.15 16/(J7trlPTJ<;
Q / Pl~TJ<; fl> '¥

6 Unobserved by Parke, SibProph 168, 171 n.S, who believed that "Book 6,
the most explicitly Christian of the Oracula Sibyllina ... as transmitted ... is
very short and probably defective."
as
7 xxvi: "Es bleibt aber immerhin von Wert, einmal einen Blick auf die
Parallelliteratur zu werfen." A full discussion and complete apparatus at as
ix-Iii, DOff.

57
28 THE SIXTH SIBYLLINE ORACLE

OtKO<; otav ~aUtO cpun CPU'tov' EV XEpt 0' au'tO\>

KOOJ.l.O<; OA.o<; Kat yaia Kat oupavo<; tiOE 8aAaooa.
ao'tpa'l'n 0' £1tt rfIv, otov 1tO'tE 1tp&ta cpavEV'ta
doov a1t' aAATtAWv 1tAEUp&V Mo YEvvT\8€v'tE<;.
20 fOOEtat, TjVtKa yata xapTtoEtat (A1ttOt 1taloo<;.
OOt OE J.l.ovn, LoooJ.l.i'tt yatn, KaKa 1tTtJ.l.a'ta KEt'tav
au'tit yap ouocppwv 'tOY OOV 8EOV OUK EvoT\oa<;
1tal~Ovta 8VT\'tOlOt voTtJ.l.aotv· aA-A' a1t' aKav8T\<;
£o'tE\Va<; O'tEcpavcp, CPOPEPTtV OE XOATtV £KEpaooa<;
25 ri<; ~p&OtV Kal1t&~w.· 'to OOt KaKa 1tTtJ.l.a'ta 'tEu~n.
C1 ~UAOV C1 J.l.aKapLo'tov, £cp' o{) 8EO<; £~E'tavu08T\,
OUX E~n Of X8wv, aAA' oupavo\> OtKOV £oo'l'n,
TjVtK' a1taO'tpa\jlnE 'to oov, 8EO<;, EJ.l.1tUPOV OJ.l.J.l.a.

19 YEvv1']9tvtEC; Alex. / YEvv1']9tvtE Hase / YEvv1']9tvta codd.23

xa{~ovta. 9V1']tOlOt VOTlIl<X.OtV Lact. 4.18, 20;
cf August. De civ. D. 18.23:
ipsa enim insipiens tuum Deum non intel/existi, ludentem mor-
talium mentibus, sed et spin is coronasti et horridum Jel miscuisti 25
EiC; pproOtv mtxrolla . to OOt Kurfess / dC; Upptv Kat x rolla . to OOt
Alex. / EiC; UpptV Kat xrolla ti OOt (xVEullatoc; OOt aut oi) codd. 27
oupavou Kurfess /oupavov codd. 28 UOtpa\j1Et(E} Geffcken /0'
uotpa\j1Et aut uotpa\j1EtC; aut uotpa\j1TI codd.

Translation

A great son, the subject of song, I sing from the heart

immortal,
To whom a begetter most high gave a throne to possess
Before he was born. After he, in the flesh, has been a second
time
Raised, having bathed in the streams of the river
5 Jordan, which moves with bright foot sweeping the waves,
And having fled fire, he shall be the first to see God,
A sweet spirit, alighting on the white wings of a dove.
And a pure flower will blossom, and springs will surge.
He will show people the ways and the paths,
10 Heavenly ones, and will teach all with words of wisdom;
And to justice will bring, and persuade, unpersuadable
folk,
Boasting praiseworthy descent from his father, son of
Ouranos.
He shall tread the waves, and cure human disease;
He shall raise those who have died and banish much grief.
15 And from one purse shall be surfeit of bread for mankind,

58
 M. D. USHER 29

When David's house shoots forth a sprout. In his hand

The whole world rests, earth and heaven and sea.
He will hurl lightning to earth-just as once at his first
manifestation
They saw him, those two born from each other's sides.
20 There will be a time when the earth shall rejoice in the hope
of a child.
For you alone, land of Sod om, is destined calamity.
For you were malicious, and did not recognize your own
God
When he was here playing with mortal perceptions. With
acanthus
You crowned him, and terrible gall you mixed
25 Into his food and his drink. That will cause you calamity.
o wood, happy wood, upon whieh God was
stretched,
Earth shall not hold you, but you shall see a home in
heaven
When your fiery eye, 0 God, flashes forth like lightning.

Commentary

1. ci9avcX'tou ... £lC cpp£v6~: aOcXvu'to<; is possibly an adjective used

absolutely as a noun (sc. O£o\» as Collins and Kurfess translate, but
this leaves an awkward repetition of Jesus' father in the genitive then
in the nominative (y£vi't-rl<;). It is better taken as an adjective with
'Pp£vo<;. Medea prophesies from an a9uva'tou <l't(>j.lCX'tO<; at Pind. Pyth.
4.11 and in a Hellenistic epigram Apollo, the Sibyl's patron deity,
possesses a 'Prov~v aOU[Vajulv (H. Lloyd-Jones, "The Seal of Posidip-
pus,» JHS 83 [1963] 89). Because the Sibyl's own 'immortality' lacked
agelessness (cf Ov. Met. 14.130), she was imagined as a perennially
aged woman if not, with the conflation of yet another myth, a shriv-
elled-up cicada wishing very much to die (Hellanicus, FGrHist 4F140;
Petron. Sat. 15.48; SibProph 57£), a motif first attested in the Hymn.
Hom. Ven. in the myth of Eos and Tithonus.
£lC cpp£v6~: The phrase amplifies the emotion with which the hym-
nist sings, i.e., "from the heart- as at Aesch. Cho. 107, Sept. 919; cf. IL
9.343, h: 9u~o\> 'PlA,tov. For the resultant hyperbaton with aOuva'tou
cf [Longinus] who says that this figure is "the truest mark of
strenuous emotion- (Subl. 22.3f). Here the emotional intensity
inherent in the idiom EK 'PP£vo<; is heightened by the hyperbaton.
Note also the prophetic 'PPllV, which "sees something more than is
apparent- that Prometheus offers as proof for the veracity of his
prophecies to 10: Ol1~£la OOt 'taO' Eo'tl 't1l<; E~ll<; 'PP£vo<; 6><; OEPK£'tUt
1tA,EOV 'tt 'to\> 1t£'PUO~£vov (PV 842f; for this usage cf. Nonnus Par. Jo.

59
30 TI IE SIX'!'I I SIBYLLINE ORACLE

1.53). The Sibyl is, in these opening lines, laboring under the weight
of time and inspiration.
aoUhJLov: "a subject of song." This, the only epithet of the anony-
mous son in this hymn, occurs only twice in the Orac. Sib., here and
at 14.310 (a late 7lh c. Jewish oracle) of an unnamed diabolical prince.
Other than at Hymn. Orph. 72.5, where Tyche is hailed as aOlSt~O~, I
find no example of it predicated of a Greek divinity. In Greek litera-
ture this somewhat rare word is variously used to describe prophets,
poets, heroes, emperors, and villains. Predicated of Helen and Hector
in Homer (II. 6.358), it means "notorious," a passage imitated/
parodied by Callimachus (Ap. 121), who uses the word of the ~av'tt~
Teiresias. Homer's aOlSl~o~ in bucolic diaeresis was the model for its
use in later prosody, though it has a more positive meaning at
Hymn. Hom. Ap. 299, where it refers to the god's oracular sanctuary
(vaov; ([ Pind. Pyth. 8.59).
With the meaning "famous for song" Pindar refers to himself as an
UOlOt~OV nlfPlOWV npoq>umv (Paean 6.1-6), as does the Christian Do-
rotheus (VisDor 272). We also find the adjective in a Delphic oracle
(ca yd c. B.C.) predicated of the poet Archilochus (see further W. Peek,
"Neues von Archilochus," Philologus 99 [1955] 4-50):
uOa.vat6~ OOL lta'i<; Kat UO(OlIlO~ C1> TfA£aIKAn<;.
fOtaL tv UVOpO)ltOlOLV. 0<; UJlltp&n6<; Of ltpoodltn
Vll0<; UltOOPWlOKovm q>IAllV d<; ltatploa raiav.
Note the syntactical parallels here with the hymn: a9ava:t6<; and
aOlOt~o<; occupy the same position in the line; both texts use the fu-
ture tense and preserve the anonymity of the son in question. The
occurrence of this Archilochus oracle at Amh. Pal. 14.113.1 suggests
that it once circulated in a florilegium. The author probably modeled
his first line on oracles available to him via such a source-perhaps
this one. Eusebius (Pracp. fvang. 5.32-33.1) quotes these lines (as far
as uvOpwnoLOLV) and knows that the unnamed nu'i<; of the oracle is
Archilochus. From this he concludes that Apollo is a false rrophet
because the god sanctions so shameful a poet, the likes 0 which
-even the noble Plato banished from his Republic." This vehement
disavowal suggests that someone (Christians?) used the oracle,
divorced from its original context, as a prophetic proof-text.
Herodotus' use of thNldjective (2.79) to describe the mytho-his-
torical figure Linus is also apt: "called by various names," this Linus
was UOiSl~O<; throughout the Near East, the only-begotten son
(nu'iou ~OUVOYfV£U) of Egypt's first king, who died young and is
since honored by laments. This same" Linus," the legendary tutor of
Orpheus (Diod. 3.67; D.L. proem. 3), later becomes a pseudonymous
authority for Greek hexameters composed by Jews and remained
current in Christian circles as well (Aristobulus apud Eus. Praep.
Evang. 13.12.13, 13.34; see further H. Attridge in Charlesworth 823f).

60
 M. D. US[[ER 31

IV Maccabees, a Hellenistic Jewish encomium, speaks of a martyr's

life (10.15) and death (10.1) as aoiOl~o<; and is largely responsible for
the meaning "of famous memory" or "renowned" in Christian
hagiographies and sermons, where the martyr is e.g. Ev06Kl~o<; ~ev
1tapa eEC!>, aoiOl~o<; OE 1tapa aVepOO1tOl<; (Basil Caes. PC XXXII 1272).
Eulogy and encomium combine in two epigrams from the Anth. Pal.:
in a sepulchral epitaph the poem itself is a Ilvfilla ... aOlollloV (669.1); in
another, "Lord Christ gives the nod of assent" (i:vEuOEv ... ava~
XPlO'tO<;) to a Nicomedes who has entered into a ~lOV ... aOlol~oV
(8.141.1). In both poems aOlOlllov is in the first line and in the same
sedes as our hymn. From these uses of aoloq.lO<; it is clear that Jesus is
being analogized to a Greek hero, martyred in his prime.
avSi.O: This denominative verb from the substantive avo~ (which
can mean "song" or "ode," e.g. Pind. Nem. 9.10f) is related to adow,
"to sing," the verb par excellence for hymnic preludes Uanko 9f; Race
5ff) and is used in this way by e.g. Pindar (01. 1.7). In the hymnic
prologue to the Orphic Hymns avow at line-end is the single verb
used to invoke a pantheon of twenty-five deities and daimones (line
39). In oracles Apollo uses it, also at line-end ('tl<; OOq>l~ 1tpww<;
mxv'twv, w{mp 'tp i1too ' avOw, Diod. 9.3.2=L. Andersen, Studies in
Oracular Verses: Concordance to Delphic Responses in Hexameter
[=Historisk-filosfiske M eddelelser 53 (Copenhagen 1987)] no. 72).
avoi.O sets this first line, indeed the whole poem, squarely in both the
hymnic and oracular traditions.
2.q,:The relative pronoun (or adverb) whose antecedent is the god in
the accusative case is a well-known hymnic feature Uanko 9f); cf.
Hymn. Hom. Cer. 1.£, Mere. Iff, Bacch. 7.1£.
9p6vov: Cf Orac. Sib. 7.32: 'tC!> yap 't' a{no<; EOWKE 9Eo<; 9p6vov .... The
throne or "seat'" occupied by a divinity is a topos in Greek literature:
generally, cf. IL 5.360; specifically, epovo<; is used of Apollo's oracular
seat at Delphi (Eur. IT 1254, 1282). Theocritus' encomium to
Ptolemy uses language similar to that here: 'tllvov Kal ~aKaPEOOl
1ta'tiw OIlO'tlIlOV EeUICEV/ aeaVa'tOl~, Kal oi. XPUOEO<; ~ EV dlO<; ~
oto~ll't(ll (Id. 17.16f). The NT concern with Jesus' accession to the
royal "Throne of David" (Lk 1:32, Heb 12:2, etc.), however, is
paramount. In Peter's speech at Pentecost (Acts 2:30f) the possession
of Christ's epov6<; is said to have been "foreseen" (1tpOlOWV) by King
David, a prophetic singer of atoll-a't(l (viz. Pss 132:11, 89:4, 16:10),
and his foresight is interpreted as referring to the resurrection (ava-
o'taOl<;; cf ~'Ytpell below) of Christ. Here the Sibyl foresees Christ's
kingship from the pagan quarter-·teste David cum Sibyl/a, as it were.
;5'1'\o'toC; 'YEV£'tllC;: The word UIjIl.o'to<;, theological Gemeingut in the
Greco-Roman world, is very common in the Orac. Sib. In VisDor it
is the favorite appellation for God (lines 100, 134,231). In the "Mag-
nificat" the Holy Spirit is the ouva~l<; UIjIlO'tOU descending upon Mary

61
32 THE SIXTH SIBYLLINE ORACLE

(a NT prose hymn at lk 1:35); in the "Benedictus" John the Baptist

is lauded as the 1tpoQ)'rl1:llS UljIlO1:0U (lk 1:76; el Pind. Nem. 1.90-93 of
Teiresias; also Nem. 11.2; Aesch. Eum. 28). In Aristonous, Apollo,
the son of Zeus, is praised as orllvov ayuAllu ... UljIlOtOU (Kappel 385
no. 42 line 7). The noun YEVEtllS ("father") is largely post-classical
and relatively rare: only here and at o rae. Sib. fro 5.3. Nonnus (Dian.
7.80) uses YEVEtllS to express the father-son relationship between
Zeus and Dionysus (7.80) in an oracle about his son, not yet born,
that Zeus utters in/to primordial time (Aion), wherein he prophesies
Dionysus' future exploits, beginning at his nativity. Zeus' promise
to his son-Zllv\. OUVUOtPU1ttOVta OEOt~Etat utoAoS ui9T]p (99)-
recalls the closing scene of our hymn (see ad line 28).
3. OU7tCD 'Y£vv"Otvn: A fine paradox/oxymoron with YEVEtllS prob-
ably referring to the pre-existence of Christ, a theology best known
from the prologue to the Gospel of John. Nonnus' paraphrase of
John's first verse is a poetic elaboration that also picks up the pre-
existence, sonship, and enthronement of Christ in one stroke: the
logos is YEVEtllPOS OIlT]At1(OS UlOS UIlT]tWP and Utrpllovl ouv9povos fOPTI.
E1td ICUta. OUpICU ... ICtA.: Beginning with End and continuing
to line 7 these lines describing the baptism of Jesus (Mk 1:9-11 pars.)
present problems of interpretation on a number of levels. The
language shows considerable (perhaps considered) ambiguity, and the
text may be corrupt. An "embodiment" of Christ's pre-existent
spirit at baptism seems to be envisioned. unoAououllEVOS and h-
<PEU~US continue the temporal clause introduced by End, which
answers to OIjlEtUl in line 6. The aorist after Enri, however, is unusual
with a main clause in the future. 'Iopouvou in line 5 may have been
originally a marginal gloss on 1tOtUIlOlO later filled out into a hex-
ameter line (without caesura), and the EIC of Mss. 'II and <l> corrupted
to n's os to accommodate it. For a similar explanatory interpolation
involving a river note II. 21.157f: E~ 'A~lOU Eupi> ptovtosl'A~lOU, os
ICuAAlO'tOV Dnwp [n\. "'(U\UV 'l110lV. ["'(EtPW (in the passive) is the pre-
ferred NT verb for Christ's resurrection (e.g. Mt 28:6; Mk 16:6; 1
Cor. 15: 12f), but is not used of his birth as it seems to be used here.
ICUta. (JUPICU is a NT idiom used by Paul (in a neutral sense) for
Christ's human lineage (e.g. Rom. 1 :3; 4:1; 9:3, 5). The first begetting
that this second one (to OlO(JOV) presupposes must be the heavenly,
non-corporeal one implied atlines 2f.
4-5. 7tpoxon'ic; ... oupwv: The personification of the Jordan River, in
fact the hyperbole of the whole description, is strangely but pleas-
antly pastoral. The reading YAuuICronlol KUllun of n is possibly cor-
rect (el Hes. Th. 587). YAUUICip noot is a bold metaphor making the
Jordan a divinity (el pede, Verg. Aen. 9.125, of the Tiber). Although
(as the text now stands) the first reference in this relative clause is to
the river Jordan, the context and vocabulary here suggest that the
foot in question may also be Christ's. A god's gait is especially

62
 M.D.USHER 33

important in epiphanies. At the epiphany of Apollo in Callimachus

the god "raps upon the doors with beautiful foot·: 'tu eUpE'tp<x
I(<XAC!> 1t001. ... apaooEl (Ap. 3), I(<XAC!> 1toSI. occupying the same sedes as
'YA.aUI(C!> 1toSI. in bucolic diaresis. Plutarch preserves an anapestic hymn
to Dionysus in which the god is invoked to come POEP'll 1toS( (see W.
R. Halliday, The Greek Questions of Plutarch [Oxford 1928] 128).
Nonnus (Dion. 1.104f) also puts 1toO( in the same sedes but as here
with the aquatic metaphor: the goddess Demeter u'YP01tOP'll O£
'YA<XUI(U Ol<XOX(~£l pOE'll 1toOI. VC1nu eUAaooll~. Note Nonnus' con-
flation of two images in our hymn: the personified water, here with
"sea-green back" (cf the Homeric EupEa VW'tCl eUAaooll~), and a
divinity (Demeter) "who parts the waves with cow foot leaving a
watery trail." With reference to Jesus I(UAC!> 1tOSl is brilliant hypallage
for the color of the water and a superb word play on 'YAUUI(O~, which
can also mean "bright" or "gleaming· (for the ordinary meaning of
"sea-green" cf Nonnus Dion. 20.353; 2.14). For oupwv used personally
cf Limenius' Paean to Apollo, where it describes that god's "tracing"
or "spreading· the foundations of his oracular shrine "with immortal
hand" (a1t[AEtOU~ eEj.1EAiou~ 't'] aIlPpo't~ XElPI. oupwv, Kappel 390 no.
46 line 24). Note also a hymn from Epidaurus in which the Mother
of the Gods moves oupouou f>U'tU[V] I(0IlUV (M. L. West, "The Epi-
daurian Hymn to the Mother of the Gods: CQ N.S. 20 [1970] 212f).
If, in similar fashion, oupwv describes the motion (or its effect) of
Jesus stepping into or out of the river, it may be meant to suggest
his walking on water (Mk 6:45-52 pars.)-an intimation at baptism
of that mark of his divine character described in line 13.
6-7. itt: 1tUpOC; h:CPEU~<XC; ... KEA.el,lC;: Note the similarities in
Orac. Sib. 7.6M: OUI( EYV~ 'tOY qov eEOV, QV 1to't' EAOUQEV' 'lopOOv% EV
1tpOXOnqt KUI. E1t'ta.'tO 1tVEUl1U 1tEAdn (see further Part II infra). fJSu is
printed by Kurfess and Alexandre, but it should be noted that this
line has many variants in the Mss,., most of them attempts to make
the fJSu masculine or put the Spirit in the dative. If fJSu is correct,
then it modifies 1tVEUIlU; the phrase 1tVEllIlU ... E1tt'YlVOIlEVOV is in
apposition to 9EOV: thus "God, a sweet spirit alighting," etc.
AtUIC<xiC; 7t'tfpuytO'O't 7tEAt1tlC;: The color of the baptismal dove is
not specified in the Gospels. It is probable that Lactantius (Div. Inst.
4.15) gets the color white from Orac. Sib. 6.7, for three other quota-
tions of Book 6 (not this verse) occur in his same chapter.
7ttA,thl: At Orac. Sib. 1.2471to..Etu is used of Noah's dove, but
otherwise only here and at 7.67. The NT uses 1tEplo'tEpa at the bap-
tism, the same for Noah's dove in the LXX (Gen. 8:8). 1tEAEill may
be simply metri gratia, for 1t£Plo't£pa is not well suited to the hex-
ameter (though metrical correption is frequent in this poem).
nEA,£lUl at Paus. 7.21.2, 10.12.10 are the doves or dove-priestesses of
the oracular shrine of Zeus at Dodona. Pausanias' excursus on Sibyls
and Sibylline prophecy highlights these Peleiai (also "Peleiades") who

63
34 THE SIXTH SIBYLLINE ORACLE

"though not called Sibyls by men give prophecies from God.- To

them, Pausanias reports.; is ascribed the venerable oracle ZEU<; ~V, ZEU<;
ECJttV, ZEU<; (oonat" w flqUAE ZEu-a theology consonant with
Christian beliefs about deity (cf Rev. 1:4, 8: <> rov Kal <> ~v Kal <>
EPXOflEVO<;). The appearance of an oracular bird adds a nice touch to the
poem.
8. eXv9ito£\ 5' av90c; lCa9ap6v. PPUOO\)Ot 5e 7tTl'ya{: Cf line 20:
yu"ia xaPTtOnal EAntOl 1tulOO<; and also 8.475 (of Christ's nativity).
Such sympathy of nature with the events recounted in the narrative,
what Keyssner calls "the hyperbolic style: abounds in Greek hymns,
and often accompanies the epiphany of the god (Keyssner 30-34), as
at Apollo's nativity in the Hymn. Hom. Ap.-flelO"OE O£ ya"i' i.l7t£-
vEp9EV (116f; cf also 135, 139)-or Limenius' paean (Kappel 390 no.
46 lines 7-10), where at the same event Ou[S o£ 'Y]qa9E 1toAo<;
OUpUVLO<;. For the collocation oupavoS-'Ya"ia-9uAaoou cf Hom.
Hymn. Cer. 13f. ~puw, "a word belonging to the language of poetry
and religion '" [for] ritual cries or the natural expression of religious
exaltation- (Dodds ad Eur. Bacch. 107), occurs frequently in com-
pound epithets in the Hymn. Orph. As an expression of hope and
happiness, or of a return of a Golden Age-primal restoration is
hinted at in lines 18f (Adam and Eve's glimpse of Christ in the
Garden of Paradise)-the sentiment is not unlike the cosmic
sympathy attending the birth of the child in Vergil's fourth Eclogue,
(esp. lines 19f, 23; cf E. Norden, Die Geburt des Kindes [Leipzig
1924]) 46-50, 58).
iiv90<;: Cf. <p\nov below (16). In the Hymn. Hom. Mere. Apollo is
described as having uv9o<; ... <plAOl(UOEO<; T\~,,<; and in paeans Apollo is
often an (PVOS (e.g. Macedonius [Kappel 383 no. 41 line 4). In Hymn.
Orph. 50.6 to Lysios/Lenaeus (=Dionysus) the god is praised as 'lEpov
uv90v (i.e., wine). For Lactantius the floral imagery of the hymn
stemmed from Isaiah 11:1, which he quotes in Latin: Exiet virga de
radice Jesse et flos de radice ascendit (where flos=LXX uv90<;).
Lactantius explains that Jesse autem fuit pater David ex cuius radice
accensurum esse flo rem praelocutus est, and quotes line 8 of Book 6
as Sibylline support for his interpretation (Div. In st. 4.13= Migne,
PL XXXVII 485f, with the variant ~p19ouOl O£ 7tUVta unreported by
Geffcken and Thierry. For ~p~w in an epiphanic context cf Hom.
Hymn. Cer. 473). Isaiah was probably foremost in the hymnist's
mind as well.
9-14. 5£{~£\ 5' eXv9pc.01tOtoW 050uC; .. , aA'Y£a 1toA.A.a: These
lines are the hymnist's efficient epitome of Jesus' ministry. The
whole section is akin to an aretalogy, and the healing mentioned at
13f is the stuff for a paean to Asclepius-like Isyllus' where the god
is'tov voowv 7tUUOtOPU, SWtllPu u'YlelUS (Kappel 382 no. 40 E lines
56£). In the same author's hcxametrical aretalogy the Spartans uuoito-
uvtoS UKOUOUV, orotElPUV <PTtflUV, 'AOKAU7tl£, Kui OCPE oawouc;

64
 M.D.USHER 35

(Kappel 383 no. 40F lines 79f; for UUS,,(HXV'tO<; cf. line 1 above). Lactan-
tius cites lines 13ff as caelestis indicia virtu tis. Interestingly L. but-
resses his argument with an oracle from "Milesian Apollo," which
Parke regarded "as an authentic attempt of the prophets of Apollo to
define their position with regard to the claims of Christianity"
(OracAp 104). Apollo too acknowledges Jesus' miracles and cruci-
fixion, going so far as to use the Christian phrase Kuta OUPKU (sec
line 3) in his exposition (PL XXXVII 484):
9vTltoe; ETlV Kata oapKa, OOq>Oe;, t£patwO£olV EPYOle;
aU' uno xaAoalwv KPltWV OnA.olC; ouvaAr09Elc;
yo,.UpOle; Kal. OKOA.On£OOl1tUCpT]V uvitATlO£ t£A£Ut';v.
10. Sd;£t ... oupav{ac;: Note the anaphoric repetition of Sd~El and
the assonance olSu~El/a~El. oupuvla<; here in enjambement emphasizes
the quality of Christ's teaching. For "heavenly paths" cf. Clement's
Paedagogus hymn: lXVlUXPlotOV!OOOC;OUpUVlU[SC. Eotlv] (Thierry 10
no. 7.34f). Cf. also Christ's words in the Gnostic (anapestic) ·Psalm
of the Naasenes·: ,.1Op<pac; oe 9EWV E1tlo£i~(!) 'ta KEKpUjljlEVU t11C; aYluc;
~ (Thierry 13 no.8.21 f).
OOcpo'il; ~u90\c;: Keyssner 55 notes" Auch die oo<pla des Gottes be-
greift gelegentlich seine Macht und seine Fahigkeiten in sich" and cites
Macedonius' paean (Kappel 384 no. 41 line 17; cf. the self-proclaimed
oracular wisdom of Apollo cited above ad line 1).
11. 7t£{0£\ Aabv aKu9f\ ... lCd,.: Presumably the Jews are the
·unpersuadable" or "disobedient people" (same collocation, same
sedes at Orac. Sib. 1.204, 3.668; cf. Is. 30:9). AUOC; is the usual, positive
term for "Israelites" in LXX as opposed to Eev" who are "Gentiles."
12. aiuov ... oupav{Sao: Christ here boasts of his "praiseworthy
descent from his father, son of Ouranos." The rare patronymic oupu-
viO"C; is very peculiar. It occurs only here in Orac. Sib. and in Greek
epic only once in the singular in Hesiod (never in Homer or
Nonnus) of Cronos (Th. 486; imitated by Oppian, Cyneg. 3.12). The
word belongs to the language of theogonic myth-an idea very
popular in the Orac. Sib. (Collins 334; Kurfess, "Homer and Hesiod
in 1. Buch der Oracula SibyIlina," Philologus 100 [1956J 147-53). In a
related passage, Orac. Sib. 7.69ff (see Part II infra), ·Ouranos has built
three towers· for Christ who, as Logos, "donned flesh and swiftly
flew to his father's home" (OUpKU tE OUOUjlEVOC; 'tUxuc; 'l1ttU'tO 1tu'tpoc;
E<; OlK01)<; tPElC; S' uutcp 1t1Jpy01)<; IlEyuC; OUpuvOc; EOt"Pl~EV). Ouranos-
Hyperion is a mythological figure known from the Gnosticism of
Valentinus whom Epiphanius accuses of plagiarizing the theogonies
of Hesiod and Stesichorus, "changing only the names" in his
elaborate system of thirty aeons and spiritual pleroma (Adv. Haer.
31.2.4, 3.2). Thus, perhaps, oupav iOllC;=Hclios, the son of H yperion
(Th. 374, 1011; Hom. Hymn. 31.4f}=Yahweh. This would explain the
sun imagery at lines 26ff (q. v.). Though the language of Hesiod is

65
36 THE SIXTH SIBYLLINE ORACLE

evident, identifying a Gnostic source for oupuv(81l<; is impossible.

Unfortunately nothing in the only Valentinian hymn extant
(Heitsch 155 no. 43) helps here.
Alternatively YEVO<; may=Christians, a "praised race [belonging to]
the father, son of Ouranos" of whom, out of the larger stock of
Auot U1tEl8El<; who reject him, Christ has" boasted." This use of YEVO<;
originates in 1 Peter 2:9, where the author accommodates the phrase
YEVO<; (A[nOV from Is 43:20 (of Israel). Thus, in apologetic literature
Christians call themselves a tpttov YEVO<;, a third people alongside
both jews and pagans (Keryg. Petro 2) or elsewhere a 1CUIVOV ... YEVO<;
(Ep. Diog. 1).
15. ile 5£ j.I.\i\c; 7tTtP'I1C; ... le'tA.,: This must be an allusion to the
feeding miracles of jesus (e.g. Mk 6:35-44 pars.), for which Lactantius
reports "the disciples said they had five loaves and two fishes in pera"
(Div. Inst. 4.15.16). The 1tT,PU was a characteristic feature of Cynic
costume (D.L. 6.22, 33). Thus the "one wallet" from which will come
"surfeit of bread for men" (cf Orac. Sib. 1.357) also evokes the
disciples' vagabond lifestyle during jesus' three-year ministry, for
which period they are said to have shared a common purse (jo.12:6,
13:29; cf also Acts 4 :34 f).

17. 1COOj.l.O<; OAO<; ... lCtA.: A characteristic example of the hyperbolic

style (see above line 8). For the sentiment cf Hymn. Orph. 34.14-17
to Apollo.
18f. cxotpaljfu .... : "He will hurl lightning to earth." For the image
note esp. Lk 17:24: Wo1tEP yap ;, UOtpU1t~ UOtPU1ttouou (1C tll<; U1tO
tOY oupUVOV Et<; t~V U1[' oupuvov AUIl1tEl, outw<; (O't<X1 0 uio<; uv8po)1tou
(V til ;'IlEP~ UUto\>. In· this hymn the verb is used absolutely and
personally. I find no OT or NT precedent for such vividness. Even
Satan gets a smile when he "falls like lightning from the sky" (Lk
10:18). The lightning image is often associated with Yahweh, e.g. the
theophany at Sinai (Ex 19:16), in a military prayer (Ps 143:6 LXX),
an apocalyptic vision (Ezk 1:13; cf Ap 4:5) and, of course, with
Zeus, too: Orph. Hymn. 20, a hymn specifically to Astrapaios Zeus.
Lightning is associated with Dionysus in P. Ross. Georg. 16ff
(Sutton), a hymn about king Lycurgus' reluctant recognition of that
god and his subsequent punishm~t. Thunder and lightning are signs
attending him, given by his father Zeus because Zeus "honors him
greatly":
crYu<; iowv [Sc. Lycurgus) cA)e6v'tCl f!E't' ao'tEpoltal<; ~t6vuoov
a't ltUlCVal. (J l£MlY\~oV UltO ~[pov hl]Ol 9af!cla u;
\)i£o<; (py' ai]ol]Aa ~\O<; f!£ya 1(uoaivov'to<;.
So too with jesus' lightning (see line 28 below). Nonnus uses
uotpa1ttw in his paraphrase of that definitive NT indictment of dis-
belief, jo. 3:19 (=par. Jo. 3:97-103). "The judgment upon an irre-
ligious world is this: From heaven came the light (q>EyYoC;) to earth,

66
 M. D. USHER 37

but the race of unstable men, though the light it up the sky ('9£11£0C;
ao'tpax'tov'toc;), loved more gloom.·
otov '/ton Kpm'ta cpavtv'ta ... ax' aAAitAcov '/tA£upmv Buo
y£vvT)9iv't£c;: ·The two born from each other's sides· is an unparal-
leled circumlocution for Adam and Eve, who, the text implies, saw
Christ in the Garden of Eden. The Christian apologist Justin Martyr
explains the anthropomorphism of the Jewish God in the OT, in-
cluding the one in Eden (Dial. 131), as Christophanies, i.e., appear-
ances of a philosophic Logos personified. Though not called "Logos·
in this poem, we find that term predicated of him in the related
passage in at 7.83 (Part II, infra). In powerful hexameters Gregory
Nazianzus hymns ·the celebrated spark and aerial torch of the Logos
(oniv9T)p 5£ I\Oyou Kal nupo~ a£pB£lc; ... aoi~huoc; [ef. line 1 above]),
which pervaded the whole world· at Adam's fall from paradise (PG
XXXVII 1231). The words £~xupov and a7tao'tpa7t'tco juxtaposed have
an astrological significance in a fantastic passage from Orae. Chald. (fr.
146.7 Des Places).
21. 001. BE JoLOvn. Io8oJoL\n yaiT) ... K'tA.: A city may be
·spiritually· (7tV£U~a'tlKroc;) stylized a ·Sodom· (Rev 11:8). The
reference here as in Rev 11:8 is Judaea or Jerusalem. The Hebrew
prophets speak of Judah in its worst days as Sodom (e.g. Is 1:9f; if.
Ezk 16:46, 55). For xtl~a'ta K£i'talCf. an oracle from Hdt. 1.67: nfi~'
Ent ntl~a'tl K£\'tat (same sedes).
22. 8UOCPPCDV ... 'tOY oov 9tov OUK ivoT)oac;: A heavily spondaic
line for emphasis. Recognition of Christ is also important in the
hymn-like piece to the Daughter of Sion (Orae. Sib. 8.335), recog-
nition that the song fosters. At 7.53 the inhabitants of I1ias 8£ov OUK
EVOT)OaV (same sedes, as also at VisDor 74).
23. KaiCov'ta OVT)'toiot vOtlJoLaow: This is the oldest reading ap.
Lactantius, probably followed by Augustine. Of all editors only
Thierry prints it. Geffcken rejected the reading on no certain grounds
(OS xxviii n.1), though he accepted Lactantius' 7ttlP'l1C; in line 15 over
on£ip'l1c; in nand piC'I1C; in <l> and 'I' (OS xxix). The vividness of the
image is the stuff we have come to expect of this poet. Thierry
adroitly compares Ps. 2:4: b Ka'tOtKrov EV oupavoic; EK'Y£A.ac:J'tEIat
au'touc; Kat b ciptoc; E""UKtnptE'i au'touc;-that in response to the
previous verse where ·kings and rulers have allied themselves against
the Lord and His Annointed (XPto'toC;).- Christ's ·playing with
mortal perceptions· is also much in the Dionysiac tradition, e.g., the
hallucinations of Pentheus and Agave in Eur. Bacch. At P. Ross.
Georg. (lines 29f) the god ·pours an illusion- (lv5aA~ov EX£UV) over
Lycurgus such that he, mistaking his children's consoling arms for
snakes, slaughters them. For the ·playing· ef Hymn. Hom. Bacch.
7.14f, where bonds fall miraculously from Dionysus' hands and the
god ·smiles· (~£taUxCl». There, unlike here, a lone helmsman ·recog-

67
38 THE SIXTH SIBYLLINE ORACLE

nized" or "understood" (v011oac;, 15) Dionysus' godhead. At Bacch.

497-518 Dionysus faces danger with sweet abandon. Christ displays
the same attitude in the Acts of John 94ff, where, before going off to
die, he sings a lengthy hymn and dances with his disciples. A
Dionysiac recognition motif is crucial for the Acts of John hymn: cf
esp. 67-70: 0 XOPEUWV VOEl 0 npaoow, Otl oov fOtlV to\ltO to av9pronou
na9oc; 0 !lEAAW naOXElv (Thierry 36).
23ff. ax' aleav9"e; ... otEcpavcp ... XOATtV ... Ie"tA.: A drastic
change of mood from "playing" to the vile elements and mockery of
the Passion. Is it in the vein of mock-pastoral that we are to read the
odd use of the singular &1Cav9" (Vergil's acantho? Eel. 4.20)? The NT
and early Christian fathers use the plural for the crown of thorns.
The detail of the poison/gall, found also at Orac. Sib. 8.303, 1.367;
Mk 15:18,23 (pars.), is ultimately dependent on LXX Ps. 69:21 (as in
Kurfess' emendation PproCHv for UPPlV): i:ow1Cav de; to ppro!la !lOU
XOA~V l((lL Eic; t~V Olljlav !lou f7tOtloav !l£ o~OC;. For the Synoptic
writers the sop given Jesus on the Cross was the fulfillment of
prophecy; here it is that, but it also fills out the pathos of the scene.
For Lactantius, line 24 (which he cites) brought Jer. 11:19 to mind,
where that prophet describes himself as "slaughtered like a lamb with-
out blemish" whose enemies propose to put "wood in his bread"
(lignum [=~UAOV LXX] in panem [=aptov LXX] eius) and "eradicate"
him from the earth. Lactantius' allegorization of the Jeremiah passage
is indicative of the metaphorical potential of bread (15) and Cross
(26) in our hymn: Lignum autem crucem significat, et panis corpus
eius [sc. jesus], quia ipse est cibus, et vita omnium qui credunt in
carnem quam portavit, et in crucem qua pependit (Div. Inst. 4.18.28).
26. Zl ;uAov, Zl llalCapl(J'tOV ... lC'tA.: The word ~uAov ("Cross")
is a favorite in kerygmatic passages in Acts (5:30, 10:39, 13:29).
According to the apostle Paul (Gal. 3:13, quoting LXX Dt. 21:33)
Christ's "hanging on the Cross" made him a "curse," but a beneficial
one, for the Cross was believed to release Gentiles and Jews from the
commands of Mosaic law, making them fellow sons of God, and
·setting them free" to "serve each other in love." Some such para-
doxical view of the Crucifixion informs our author's qualification
!la1CaplOtOV of what otherwise was an instrument of torture and not
an object of Christian veneration \.VItil the early Byzantine period (G.
Snyder, Ante Pacem: Archaeological Evidence of Church Life before
Constantine [Macon (Ga.) 1985] 27). The ~UAOV is apostrophized here
metonymically for Christ. Most commentators have taken the
hymnist to mean that the Cross literally will ascend to heaven. This
literalist reading is based largely on a later vague tradition to this
effect: examples in A. Rzach, "Sibyllinische Orakcl," RE 2A.2 (1923)
2141. The legend, however, does not seem to antedate our hymn and
was demonstrably influenced by it. Sozomen certainly was (HE 1.1.6,
2.1.10). We do best not to foist their literalism upon the hymnist.

68
 M. D. USHER 39

~UAOV is perfectly understandable by metonymy as Christ-a view

supported by the vast majority of Mss., in which at 8.31 'to ~UAoV as
Christ in the famous Sibylline acrostic takes a masculine participle
(<pO>'ti~o>v). In the Wisdom of Solomon (14.7) ~UAOV rounds off a bril-
liant literary conceit in a sententia where, initially used for a pagan's
"crooked'" idol (~UAOV OKOAtOV, 13.13), it becomes, by catachresis,
metonymy, and synecdoche, the saving agent of Noah's ark: EUAO-
"f11tUt yup ~UAOV St' of> YlvEtUt StKUtOOUVTl. Similarly, an interpolation
at Orac. Sib. 5.261 plays on ~UAOV used in a periphasis for Jesus'
profession as a carpenter, but certainly also used ironically for "cruci-
fixion'" (see line 26 below; cf also Orac. Sib. 1.291,7.49). The hymnist,
under the spell of theological paradox and poetic conceit, has
abruptly reversed the lurid details of the Passion with this unex-
pected benediction to the device of death.
28. a.7tcxo'tpa'V£u~ ... E~7tUpOV o~~u: a.1tuotpa1tto>="flash like light-
ning forth· (cf line 18). Although at Rev. 1:14 the heavenly Christ
has fiery eyes (cf 19: 12) and could be represented as a Helios figure
with horse and chariot, as he was in a mausoleum under St Peter's
(Snyder 62 pI. 31), "God the Father" seems to be the recipient of this
second apostrophe. Cf Synes. Hymn. 1.368: oov S' 0IlIlU, 1tat£p. 9£0<;
here is nominative for vocative as often. The indefinite temporal
clause states when the action of t.he main verb (£oo'VEl) will be ful-
filled: Jesus (=~UAOV) will see his "home in heaven" (objective or
appositive genitive; Smyth § 1322, 1332-33) when God, as it were,
gives the signal. For lightning as a divine sign cf fl. 9.237f: ZEi><; ...
OTtllutu <pUlO>V ucrtpazttEl. Cf the prophecy in Nonnus Dion. 7.99
(cited ad line 2), where father Zeus and son Dionysus are to be
reunited in heaven amidst flashes of lightning. At par. Jo. 3:69f the
Son of Man inhabits a "starry home (aOt£pOEVtt IlEAa9pcr) in the
ethereal land of his Father (ztatplOV o\)Su<; ... ui9EpU). Justin Martyr
describes Christ's relation to the Father as that of light emanating
from the sun (Dial. 128), a metaphor that finds its muse in Synesius
(Hymn 5; Thierry 23 no. 13): uuto<; <p&<; d 1tuyu"iov oUAAa~'Vu<;
UKtl.<; ztutpt.
As for OIlIlU, Hclios is 1tU~<PUE<; ui9Epo<; OIlIlU in an Orphic-style
hymn embedded at Nonnus Dion. 40.379 (see F. Braun, Hymnen bei
Nonnos von Panopolis [Konigsberg 1915] 8f). In hymns from the
Magical Papyri, Helios is in fact identified with "lao" (i.e., Yahweh;
Heitsch 183 no. 5=PG M III 198-228 line 14). The Orphic Hymns
laud the Ollila dlKTl<; (62.1, 69.15; cf Procl. Hymn 1.38) and the dto<;
o~~a 'tEAElOV (59.13). An oracle at Macrob. Sat. 1.20.13 calls Serapis
Ollila tEtTlA.<XUyE<; Aall1tPOV <pao><; ilEAlOtO, and Apollo Didymus extols
his "swift eye· (OracAp 78). Gregory Nazianzus, in hymning
Christ's cosmic rule, describes the sun as the 1(1)J(AOV ... £1l1tUPOV and
the moon as the Ollila to VUKtO<;. Perhaps due to the solar religion of
the Severans and, later, the emperor Julian, Ollila was a popular word

69
40 THE SIXTH SIBYLLINE ORACLE

in late antique oracles. The Christian editor of the Tubingen Theos-

ophy shows a special interest in oracles mentioning the o~~a of Zeus,
notably Zl1VO~ 7taVOfpKEO~ aq>9ltov o~~a (Wolff no. 7 239; see also
Orac.Ap 207). These astrological, apocalyptic, and majestic connota-
tions of the author's choice of words in this last verse create a grand
pyrotechnic finale.

II

The hymnic structure of Book 6 may be outlined as follows: 8

I. Invocation in the er-Stil (1)
II. Pars Epica/Narrative and argument (2-20)
A. Heavenly preparations for Christ's birth and kingship
(2-3a)
B. Epiphany at baptism (3b-7)
C. Career of Jesus (8-17)
D. Christophany in Eden and Christ's advent compared
( 18H)
III. Apostrophe I: Rejection/failed recognition of Christ by
mortals (21-25)
IV. Apostrophe II: Makarismos on the Cross and Ascension
(26ff)
As outlined, this formal structure, unique among pre-Constan-
tinian Christian hymns, doxologies, and prayers,') resembles the
shorter Homeric Hymns. Some peculiar features, however, not-
ably the double apostrophe-one of admonition (21-25), one of
blessing (26ff)-bring the hymn to a close. A similar juxtapo-
sitioning of admonition and blessing (in that order) occurs at
Hom. Hymn. Cer. 480ff, referring to initiation into the mys-
teries at Eleusis. The first apostrophe of our hymn entails the
condemnation of a city. Although this is not characteristic of
the hymn, the condemnation of cities is very Sibylline and gives
the hymn its distinctively oracular quality. The second apos-
trophe, the macarism of lines 2y.ff, is a common feature, usually

8 "Er-Stil" is E. Norden' term for third-person, as opposed to second-per-

son, invocations: Agnostos rheos (Leipzig 1923) 143-76. For general overviews
of hyrnnic form and style and further explanation of the terminology used
here, see Janko, Race, and Keyssner and J. Bremmer, "Greek Hymns," in H.
S. Vernel, ed., Faith, Hope and Worship (Leiden 1981) 193-214.
9 The most comprehensive collection of early Christian material is still W.
Christ and M. Paranikas, edd., Anthologia Graeca Carminum Christianorum
(Leipzig 1871); note also C. Del Grande, Liturgiae preces hymni Christi-
anorum e papyris collecti (=Biblioteca Filologica Loffredo 3 [Naples 1934]).

70
 M.D.USHER 41

found near the end of the Greek hymn. 10 "0 happy the race of
those mortals ... which builds a temple to lord Apollo" sings
Philodamus, for example, in his Paean to Dionysus. 11 The final
element in a hymn, however, is most often a request or some
expression of xapt~, addressed to the god directly in the imper-
ative or optative mood. The request is based on an argument
consisting in the accumulation of divine attributes and epithets
(e.g. in the Orphic Hymns) or, as in this hymn, implied in the
pars epica (a narrative of the rast glories of the deity). In our
hymn the fictive futurity 0 the events described and the
oracular persona assumed by the hymnist render the argument
a priori: Christ merits praise because the Sibyl has prophetically
intuited his divine paternity, career, death, and ascension.
Accordingly, the notion of Xapu;, in addition to coming earlier
than usual in our hymn (before both apostrophes in line 20), is
presented as a fact, not couched as a request. Thus, though
clearly working within the traditions of classical and Hellenistic
hymnody, the author seems to have adapted the genre.
The Hymn as an Oracle-The Sibyl as a Hymnist. Lactantius
cites this poem four times, each as if it were a Sibylline
prophecy. Why was this hymn read in antiquity as a Sibylline
oracle? The intrusion of prayers and hymns into other collec-
tions of 'popular' religious literature such as the Greek Magical
Papyri and the Corpus Hermeticum is well known. 12 Thus the
early identification of the poem as a Sibylline prophecy and its
inclusion in the collection are most easily explained as accidents
of manuscript transmission or the caprice of ancient editors. As
I have tried to show in the commentary, however, the oracular
quality of Book 6 is also a literary affectation of its author. We

10 E.g. Hom. Hymn. Cer. 486£; Hom. Hymn. nos. 30.7 to Ge, 25.4f to
Apollo and the Muses=Hes. Th. 94-97.
II·U )lUKap oA.~ia 'tt Kdvv(J)v y£[vdxj ~po'twv ... a K'tlO'T]l vao[v] uvaK[nj
cl>o$(1)l, Kappel no. 39 lines 118-21.
12 For the papyri see E. Heitsch, ·Zu den Zauberhymnen," Philologus 103
(1959) 216-20, 223f; for the Hermetic corpus see G. Fowden, The Egyptian
Hermes (Princeton 1993) 84ff. In the Sibylline corpus the didactic poetry of
[Phocylides] is inserted between 2.55 and 149 (Mss. \jJ only). See further P. W.
van der Horst, The Sentences of Pseudo-Phocylides (= Studia in Veteris Testa-
menti Pseudepigrapha [Leiden 1978]). The longer narrative poems about
Christ at Orac Sib. 1.323-86, 8.251-336, which may also be intrusions, are
closer generically to paraphrases of biblical passages, on which see in general
M. Roberts, Biblical Epic and Rhetorical Paraphrase in Late Antiquity
(Liverpool 1985).

71
42 THE SIXT! I SIBYLLINE ORACLE

find precisely this type of affect at Vito Plot. 128, where Por-
phyry (or his source) passes off a hymn in praise of Plotinus as
an oracle of Apollo. He likens the hymn to the Delphic oracle
about the wisdom of Socrates. As Harder notes, "'es ist hier auf
sehr besondere Weise platonischen Philosophie in den Formen
einer zeitgenossischen Kunstepik oder besser Kunsthymnik
(und nicht im iiblichen Orakelstil) ausgesprochen. "13 Por-
phyry's 'oracle' begins, in true hymnic fashion, with an invoca-
tion of the Muses and ends with the apotheosis of Plotinus. Our
author has done much the same for Jesus with the Sibyl as his
authority.
The Sibyl did in fact have a reputation for hymnody in
antiquity. Plutarch for example can, without special emphasis,
use the verb UJlVfUW to cite the Sibyl (Mor. 388F). Pausanius,
who prided himself on a firsthand acquaintance with oracular
literature (10.12.11), believed her to be an authoress of hymns as
well (10.12.2; cf SibProph 40; 49 n.31). Pausanias reports that
the Delians in his time claimed that a hymn to Apollo used at
Delos was written by a Sibyl named Herophile, whom he
places in the Troad and the local historian Apollodorus of
Erythrae identifies as Erythraean. 14 That Herophile's alleged
hymn is none other than the famous "Homeric" Hymn to
De/ian Apollo (Sib Proph 44) can be seen from what is agreed to
be a genuine fragment of this Erythraean Sibyl,15 in which she
'predicts' the literary career of Homer, alluding to his blindiness
and calling him the Chian (Orac. Sib. 3.419-22; c{. Thuc. 3.104)
-details of Homer's identity that stem ultimately, perhaps ex-
clusively, from the Hymn to Delian Apollo (172).16 The Ery-

R. Harder, ed., Plot ins Schriften 5c (Hamburg 1958) 116.

Il
FGrHist 422; cf Paus. 10.12.7. Parke, I think confusedly, says Pausanias
14
co-identifies Herophile of Erythrae with a Delphic Sibyl (SibProph 38). But
Pausanias speaks (10.12.5) only of her oGINIsionai «)!ton: BE acpl1wl'to) presence
at Delphi in order to chant her oracles from a rock. She is nowhere said to be
resident at Delphi and was apparently itinerant, having spent time in Samos,
Claros, and Delos as well as the Troad (viz. Marpessus) where, as a temple
attendant (VE(J)1(OpO~) to Apollo Smintheus, she died.
15 Collins 359, following Geffcken, TV 13; also Parke, SibProph 44. The cor-
respondence of the subject matter in Orac. Sib. 3.401-88, corroborated by a
report from Apollodorus of Erythrae in Yarro (ap. Lact. Div. Inst. 1.6=
FGrllist 422), is the basis for identifying this section as ultimately Erythraean.
16 Semonides of Amorgos (fr. 29 Diehl) places him in Chios without
mentioning his blindness.

72
 M. D. USHER 43

thraean Sibyl's implicit indictment of Homer for plagiarism 17 is

very telling if she is in fact Pausanias' Herophile, reputed com-
poser of the Homeric Hymn, which that blind man from Chios
in the poem itself claims as his own. IS It is less pertinent here
whether an historical Sibyl named Herophile actually composed
the Hymn to De/ian Apollo, nor even if she ever was in fact
Erythraean. Rather, the allusion of Orac. Sib. 3 to the Delian
hymn she is thought to have composed shows that readers in
Pausanias' time did not distinguish sharply between authors of
oracles and authors of hymns. After all, 'Orpheus' bridged the
two genres. Of course the hexameter and epic diction used in
both oracles and hymns encouraged intertextuality, even
verbatim borrowing. 19
Date and Provenance. The date and provenance of this hymn
depend upon the relationship of Book 6 to Book 7. The close
connection of fire, water, and the dove at Christ's baptism
(Orac. Sib. 6.3-7) has a puzzling parallel in the ritual prescription
at Orac. Sib. 7.76-86, where a fire is sprinkled with water and a
white dove (1tEA-fiT!) is released to heaven. 2o The ri te, apparently
Christian, seems to involve the commemoration of the partici-
pant's baptism, a re-enactment of Jesus' own, and is accom-
panied by a ritual cry (poiJcrac;):
~ (1I: A.6yov yiVVTlOE 7t<lt~p, atap OpVlV a<PlllCa
O~l)V a7taYYEAtllpa A.6Y(J)v A.6yov, UOaolV ayvo'ic;
patv(J)v oov ~U7ttlOlla, Ol' ot 7tUpOC; E~E<pauVeTlC;.21
It is clear from 7.76-86 that "the begetting of Christ is closely
associated, if not identified, with his baptism" but to interpret it

17 Orac. Sib. 3.423ff: cn£wv yap CJlWV Jl£tPWV lCP<Xtlj(Hl. npWto~ yap
lCEpaUxlOlV [Jlat; ~l~AD\l~ UVUnNOOEl.
18 As Alexandre (Excursus 12) notes, however, the charge of plagiarism
against Homer was common: cf. Diod. 4.66.
19 E.g. the famous Homeric line (from the description of Achilles' shield) on
the -tireless sun and the moon waxing into her fullness" (~EAl6v 'tUlC<XJl<XV't<X
OEAtlVtlV tE 1tAlj9o\lo<xv, II. 18.484), which turns up once at Orac. Sib. 3.21 and
in the entirely unrelated Hymn to all the Gods from Epidaurus (line 9).
20 Geffcken emended 7.79 to uypillv ou n£AEl<Xv on rather questionable
grounds: -die wilde Taube f1iegt schneller davon, in aile Welt, man kann sie
nicht verfolgen, gerade so wie der A6yot; sich zum Himmel aufschwingt" (TV
34 n.4). Wilamowitz's upyTttu (-white") dove is better and reflects the con-
nection of this passage with Bk 6.
21 -Just as the Father begat the Logos, r have released a bird, the Logos, a
swift messenger of words, sprinkling your baptism with holy waters, through
which you appeared from fire."

73
44 THE SIXTH SIBYLLINE ORACLE

as a "theophany by fire, "as do Collins and Geffcken, is, strictly

speaking, inaccurate.u The participant in this ritual is said to
have "emerged" from fire much as Christ "escaped" in Book 6.
I suggest rather that Christ's "flight from fire" in Book 6 is best
understood as his descent from the empyrean realm of the
YEvEtTle; u\jItatoe; at line 2 to become "embodied" at baptism in
accordance with the cosmology of Middle and Neoplatonism, a
cosmology promulgated in the (hexametrical) Chaldaean
Oracles, through which (we might imagine) these ideas reached
our author. In the Neoplatonic universe fire was the dwelling
place of the supreme, transcendent 'father' 2J_an empyrean
region created by a second god, the father's "intellectual
power"24 who was identified as his son. 25 This cosmology per-
meated popular culture. It underlies a theological oracle of
Clarian Apollo (3 rd c.) in Lact. Div. Inst. 1.71.1,8 2 (, and is found
in hymns from the Magical PapyriY
Proponents of Middle and Neoplationism, the more promin-
ent of whom hailed from Syria-Palestine, made much of the
"rational man passing through the ... sublunary universe (YEV-
Eate;) and returning to his celestial home." 28 In particular, Nu-

22 Collins 409. Geffcken overemphasized the epiphanic nature of the fire at

baptism in Book 6 because he saw in it an echo of the (non-extant) Gospel of
the Ebionites, which supposedly began with the baptism and included a
Feuererscheinung: EUe\><; rr£pIO..aIlIj/E 'tov 'torrov Ipw<; Il£ya (ap. Epiph. H aero
30.13.7). ]. G. Gager, "Some Attempts to Label the Dracula Sibyl/ina, Book
7," HTR 65 (1972) 94, however, has rightly pointed out that light, not fire, is
mentioned and the description in Bk 6 is actually closer to Justin Martyr's
report of the event: rrup uvTiq>911 tv 'til> 'lopMvn (Dial 88.3).
2) Fr. 3: 0 rrmTjp l1p1tuc}Orv euu'tov ouS' £v en SuvallEI VOEp<;t lCA.£lO(x(; iS10V
rrup (E. Des Places, cd., Oracles Chaldafques: avec un choix de commentaires
anciens [Paris 1971] 13).
24 Fr. 5.3f: vou yap v60<; co't!V OlCOOIlOU 'tEXV1'tf\<; rruplou (Des Places 124).
25 A.-J. Festugicre, La revelation d'l/ermes Trismegiste (Paris 1949-54) III 54
n.2.
26 G. Wolff, Porphyrii de philosophia ex-oraculis haurienda librorum reli-

quiae (Berlin 1856) 234 no. 2: £cr9' i)ltl:p oupavlou rrupO<; uq>9hou aieOIl£Vll 1pA6~; a
shorter version in the Tubingen Theosophy (Wolff no.l line 15) speaks of
God tv rrEp! valCilv. A. S. Hall, ZPE 32 (1978) 263, cited in Parke, OracAp 66,
has shown that yet another version of the Clarian oracle, found inscribed in
the city wall of Oenoanda and datable to the late third century, actually
formed part of a shrine to the "Most High God" ("Ylj/lo'to<;9EO<;).
27 One of four versions of a hymn to the sun from PC M reads: "HAlE ...
lhc1tCilV IpA.oyo<; u1(ulla'tov rrup (Heitsch 181 no. 4 line 2).
28 R. LAMBERTON, Homer the Theologian (Berkeley 1986: hereafter
'Lamberton') 53.

74
 M. D. USIIER 45

menius of Aramea developed his allegorization of the Odys-

sean Cave 0 the Nymphs along these lines,29 in which refer-
ence to water and wave in Homer and Plato's myths were inter-
preted as symbols for the world of YEV£<H<; (cl Lamberton 71).
Numenius loosely cites "the prophet" of Gen. 1:2 ("the Spirit of
God hovered over the waters") as additional ancient support
that "souls settle upon God-inspired water. "30 The resonances
with our hymn are striking: a pre-existent Christ (as the 1tvE\)~a
... E1ttYtv6~EVOV in line 7) descends from the empyrean world of
the Most High God (1-4), is materialized upon the water (5-8)
as the human Jesus, and returns to his fiery home (oh:ov ...
E~1tUpOV o~~a, lines 27f). It is worth noting that when the
author of o rae. Sib. 7 speaks of a "baptism through which you
appeared from fire," he uses the verb E~E<paaVel1<; at line end,
the same metrical position as at Od. 12.441-the book of sea
adventures culminating in that 'rational man's' escape from
Charybdis (i.e., the world of yEVECH<;).31 Numenius may well
have been familiar with the life of Jesus. Origen reports that he
"published a story about Jesus, not mentioning his name, which
he allegorized" (Fr. lOa Des Places=Orig. c. eels. 4.51). Numen-
ius, if not our anonymous hymnist, is thus the model from
which to sketch a profile: a Syrian religious philosopher conver-
sant in pagan, Jewish, and Christian literatures. Celsus, a percep-
tive critic of early Christianity, actually mentions a Christian
group called the Sibyllistae. But who these Sibyllistae were,
whence they came, and whether Sibyllistae was their own label
or Celsus' is unclear.
The relationship of the ritual passage in Book 7 to Book 6 is
closer than Geffcken and Collins have allowed. Alexandre and
Kurfess rightly insist that much else in Book 7 stems from
Book 6, especially lines 64-95, which are introduced by a woe
pronounced upon Cocle-Syria for "not recognizing your God
when the Jordan washed him in its streams and the Spirit soared
in (the form of) a dove" (OUK EYVW<; tOY OOV eEOV, QV 1tOt'
EAOUOEV 'I6p8avo<; 1tpOXOnOt Kat E1ttatO 1tVEU~a m~Adn). Book
6 and 7.64-95 also show especially strong (non-formulaic)

29 Od. D.llOff; Numenius frr. 30f, 33 Des Places.

30 Fr. 30.3-6 Des Places: lm\ 'tOY 1tooqn]tnv dPTPcEVUl E~q>EPEoeUl E1t(XVIJ) 'tou
1J&l'tO~ aEQU 1tVEU~(l.
31 Same position at II. 4.468, 13.278; Sutton line 4.

75
46 THE SIXTH SIBYLLINE ORACLE

verbal affinities. 32 To the arguments of Kurfess and Alexandre

in favor of the relationship bctween the two books, I would add
two furthcr observations: (1) the Orac. Sib.'s only two refer-
ences to the Jordan River by name occur in Books 6 and 7, and
(2) the mythological figure OUpavo~ at 7.71 parallels the odd use
of the rare patronymic oupavloao at 6.12. In both passages Ou-
ranos is linked contextually to each book's description of the
baptism of Jesus, which involves the unusual (non-NT) manifes-
tation of fire. Furthermore, in the n Mss. Books 6 and 7 often
form one continuous book; this suggests that ancient editors,
among whose ranks may well be our hymnist, felt the
similarities. Alexandre thought that the same author did in fact
have a hand in both books;33 if so, the verses against Tyre, Coele-
Syria, and Berytus (7.64ff), coupled with warnings of the time
"when another warlike tribe of Persians will come to rule"
(7.40ff), referring to the establishment of the Sassanid dynasty in
224, point to a Syro-Palestinian provenance J4 and a date roughly
in the Severan pcriod. 35 Alcxandre's suggcstion (Excursus 384ff)
of 233-235, compensating for the ex eventu nature of Sibylline
prophecy, is eminently sound.
In further support of a Syro-Palestinian provenance I offer
the case of the Christian apologist Justin Martyr, a Greek from
Samaria and one of only two other authors to refer to fire at

32 Alexandre persuasively demonstrated this point but subsequent commen-

tators have neglected his discussion: Excursus 382f; cf esp. Or",c. Sib. 6.11 with
7.24ff, 6.16 with 7.3J.
JJ Cf Excursus 380: ·doctrina, sententiis ipsisque aliquando verbis ita inter
se conveniant ut plane ab eadem manu profecti videantur." Kurfess, in E.
Hennecke, W. Schneemelcher, M. Wilson, edd., New Testament Apocrypha
(London 1963-65) II 708. Gager's discussion of Book 7 (supra n.22: 91-97)
does not treat evidence from Book 6.
H Geffcken (TV 31, following Mendelsohn) speculated a Syrian provenance,
seeing an Ebionite influence in the conjunction of fire and baptism. Collins
and Kurfess (SW 313) rightly object that nothing in Bk 6 necessarily indicates
a sectarian much less Ebionite origin. ColliM (406), however, who like Geff-
cken minimizes the relationship of Books 6 and 7, dismisses the hypothesis of
a Syrian locale.
J5 In 194 Septimius Severns divided Roman Syria into two provinces, ·Syria
Coele" stretching north and east of Mount Lebanon, and ·Syria Phoenice" to
the south, including Palestine. But, as F. Millar notes (The Rom",n Ne",r £",st:
31 BC-AD 337 [Cambridge (Mass.) 1993] 21, 423f), Septimius' designation of
the northern reaches as • Syria Coele" was • against all previous usage," for
epigraphic and numismatic evidence shows that Greek cities of the Decapolis
(trans-Jordan) described themselves as belonging to Coele Syria decades
before and after Septimius officially divided the province.

76
 M. D. USHER 47

Jesus' baptism (Dial. Tryph. 88.3)-a detail for which he is

probably dependent upon local Levantine traditions. The details
of the hymnic argument too are remarkably close to Justin's,
who in a compressed passage (Apol. 1.30) cites Christ's ful-
fillment of Old Testament prophecy and his miracles as
evidence for his divinity and the truth of Christian teaching-a
teaching that he believed was also adumbrated by Greek
philosophers and poets including the Sibyl.36 With Justin our
author shares an interest in pre-existence, miracles, and
Christophany (see ad Iff, 13ff, 18f). With a deft touch, an apolo-
gist's criteria have been carefully interwoven to form the fabric
of our hymn. Justin also anticipates our author's developed
symbolic, metaphorical interest in the Cross (lines 2Mf), seeing
visual reminders of it in the shape of the human body, a ship's
mast, even the standards of the Roman army (Apol. 1.55).31
Two Parallels. Finally and briefly, the relation of this poem to
other ancient literature. We have seen how in Book 6 the
author's use of prosopopoeia evokes the authenticating power
of the Sibyl for early Christian apologetic: the hymnist sings as a
Sibyl and by so doing wields the great authority of pagan
antiquity. This prophetic impersonation has an intriguing
prototype in Lycophron's Alexandra, which Parke aptly called
"'a Sibylline oracle transmuted into high literature." The first-
person persona and the predominant use of the future tense for
narrative action in Lycophron's iambic poem certainly
anticipate the technique employed in our hymn, also charac-
terized by the future tense, which in the pars epica is a feature
virtually unknown in other hymns.
It will be more readily observed that Book 6 shares many sty-
listic and thematic features with Verg. Eel. 4. Kurfess has dis-
cussed the possible influence of Sibylline oracles on Eel. 4 and
(more persuasively) the influence of Eel. 4 on the Orac. Sib.
(notably at 7.146-49),38 though he does not mention Book 6 in

)6 Apol. 1.30, alluding to Orac. Sib. 3.42f. The Sibyl is proffered as pagan

testimony to a final eschatological conflagration for which Justin also cites the
Stoic doctrine of ekpyrosis.
)7 Cf H. Rahner, in 1. Campbell, ed., Pagan and Christian Mysteries: Papers

[rom the Eranos Yearbooks, trr. R. Manheim and R. F. C. Hull (New York
1963) 186.
)8 ·Virgil's vierte Ekloge und die Oracula Sibyllina," HI 73 (1954) 120-27;

·Virgil's 4. Ekloge und die christlichen Sibyllinen," Gymnasium 62 (1955)

11 Off.

77
48 THE SIXTH SIBYLLINE ORACLE

either study. Nonetheless, Vergil's Sibylline posture,39 his use of

the future tense to describe narrative action,40 the nativity/
epiphany of an ominous child,41 his cosmic desti ny42-all of it
clothed in natural imagery, rooted in agricultural metaphors-is
present in our hymn as well. 43 It is unlikely that our author read
Lycophron. As for influence from Vergil, as Norden says of in-
fluence in the opposite direction, this is "schwer zu antwor-
ten."H Greek translations of Vergil did exist in the second and
third centuries,45 and a Greek translation of virtually the entire
Eel. 4 is cited by the Emperor Constantine in his speech Ad
sanctorum coetum (delivered at Antioch in 325), where he
offers the earliest surviving Messianic interpretation of the
poem and cites the Sibyl to support it.46 I leave open whether or
not we have in Book 6 a rare case of Greek dependence on
Latin poetry. If nothing else, clearly the Eel. 4 illustrates how
hymnic (or encomiastic) and Sibylline elements may co-exist in
the same poem.
Whatever its classical and Hellenistic prototypes, Book 6 is
surely a welcome exception to Norden's claim that "Die
Sibyllistik war kiinstlerisch so wertlos, dass sie kaum Poesie
heissen darf." This oracular hymn, probably written by a
Christian Middle Platonist in Syria-Palestine in the Severan
period, is also without question the first bloom of a poetic
tradition of late antiquity that will reach full flower in an author
like Nonnus of Panopolis, whose hexameter paraphrase of
John's Gospel is a bold attempt to lift the Gospel story to the
level of epic drama. In Nonnus, Jesus virtually carries the

19 Cumaei ... carminis (4); non me carminibus vincet nec Thracius Orpheus
(55).
40 solvent (14), accipiet ... videbit (15), videbitur (16), reget (17), [undet (20),
etc.
41 nO'lia progenies caelo demiuilur alto (7); cara deum suboles, magnum 10000is
incrementum (49).
42 pacatumque reget patriis virtutibus orbeJIJ (17).

4} For pastoral description in Orac. Sib. 6, cf lines 5, 8, 16-20, (23f), and

comm. ad. locc.

44 Die Geburt des Kindes (Leipzig 1924) 145.

4S B. Baldwin, ·Vergilius Graecus," AlP 97 (1976) 361-68; E. A. Fischer,

-Greek Translations of Latin Literature in the Fourth Century .... 0 .. " YCS 27
(1982) 173-215.
44 R. Lane Fox, Pagans and Christians (New York 1986) 647-52, attributes
the translation and Messianic interpretation to Lactantius-an intriguing
suggestion given Lactantius' interest in Bk 6; see above comm. ad 9-14, 23ff.

78
 M.D. USHER 49

thyrsus,47 and the depiction of Christ as a Dionysus is no less

bold, or perhaps intentional in our hymn. The apostrophe to
the Cross in particular, found first (so far as I know) in this
hymn, has had a long literary life. Fortunatus develops this
motif into a conceit in his famous hymn Pange lingua (Misc.
2.2), to which Robert Herrick adds an Alexandrian touch,
rounding off his H esperides with His Anthem, to Christ on
the Crosse, a cross-shaped poem to "the sacred tree."48

UNIVERSITY OF CHICAGO
July, 1995

47 E.g. Par. Jo. 7.185=Jo. 7:49. Conversely in Nonnus' Dion. a Christie

Dionysus appears now and again, displaying behavior inconceivable without
a literary/theological influence from Christianity. For this view of Nonnus see
G. W. Bowersock, lIellenism in Lale Antiqutiy (Ann Arbor 1990) 41-54. For
the syneresis of Christ and Dionysus on late antique sarcophagi see E. C.
Olsen, Dionysiac Sarcophagi in Baltimore (Baltimore 1941).
48 I would like to thank C. A. Faraone and Peter White for carefully reading
earlier drafts of this paper and making excellent suggestions for its
improvement.

79
ON THE IMITATION (MIMHLIL)
OF ANTIQUITY
IN BYZANTINE LITERATURE

HERBERT HUNGER

80
 C ONTRARY to the opinion prevailing in modern theories of art and
modern poetics, which places the original work of the artist far above
every imitation, no matter how good this imitation may be, Greek antiq-
uity and the Byzantine Middle Ages cared very little for "original genius."
Even the anonymous author of the work DEpl v,+,ovs (Longinus on the Sublime),
of whom A. Lesky says, "er ist in seinem den Modernen unmittelbar anspre-
chenden Verstandnis fur die Werte grosser Dichtung als genialer Einzelganger
seiner Zeit weit vorausgeeilt" (as a solitary genius he by far passed beyond
his own time through his understanding of the values of great poetry which
directly appeals to modern man),l even he knows and appreciates "die eifrige
Nachahm ung der alten grossen Schriftsteller und Dichter" (the zealous imi ta-
tion of the great ancient writers and poets) as a"zweiten vVeg zu den Hohen"
(a second way to the summit [of creative writingJ).2 He believes in the possibil-
ity of inspiration through this imitation, and that such inspiration may result
in literary works-in analogy to the insp!ration of the Delphic Pythia and the
oracular decrees resulting from it. Like the priestess of Apollo, those affected
by the divine spirit would also partake of the divine creative power, even
though they had not shown any signs of originality before. 3
Yet, about two generations before this work was written, the two critics and
grammarians Caecilius of Caleacte and Dionysius of Halicarnassus had al-
ready developed a theory of imitation. DEpl 1J11Ji]crEWS by the latter, it is true,
has been handed down to us in scanty fragments only,4 but it is clear that the
imitation of Attic culture, which demanded a close study of the classical
models, would also call forth practical instructions concerning it. The young
rhetorician wanted to know which authors he had to imitate, and in which
way, in order to be successful. Quintilian is convinced of the high value
of imitation and at the same time emphasizes the importance of exploiting
several models simultaneously and in an eclectic way (Inst. or., X.2.1): neque
enim dubitari potest, quin artis pars magna contineatur imitatione, and (X.2.26) :
plurium bona ponamus ante oculos, ut aliud ex alia haereat et quo quidque loco
conveniat aptemus. This sentence by Quintilian already contains the recipe
which has again and again proved to be effective with the numerous rhetori-
cians, epistolographers, and poets of the Empire, of late antiquity, and of
the Byzantine Age. The aliud ex alio haeret was the magic formula for the
admirably retentive memories of the authors who wrote those much abused
patchwork poems, the centos, of which we shall have to speak in greater detail
later on.

1 A. Lesky, Geschichte der griechischen Litera/uy, 2nd ed. (Berlin, 1963), 886.
• nEpi iAyo\JS. 13.2: ... Kai Mi\T] Tts .•. 656s hri 'ra V\jIT]ACx TEl VEt. "Troia 5f Kai TiS aVTT]; ti TWV E\l"TrpOaSEV
\lEY"""'V avyypaq>E"'v Kai "TrOtT]TWV \l1\lT]aIS TE Kai [1\i\",atS.
3 Ibid.: Kai 01 llti i\lav q>Otl3aO"TtKoi TC;; hopwv avvEvSovcnwal \lEyoSel.
• Cf. W. Schmid and O. Stahlin, Griechische Literaturgeschichie, II (i\lunich, 1920). 470.

81
 18 HERBERT HUNGER

Who has read much can considerably improve his style: o:vcXyvcucns TpOcpi) AE-
~Ecus5-this catchword was certainly followed very often.
Dion of Prusa recommends to the adept of good style the reading of the works
of Menander and Euripides, not by himself but by having them recited to him
because in that way he could better concentrate. 6 This short speech (No. 18)
by Dion is a particularly important testimony for the theory of imitation in
the early period of the Empire. What is said here, for example, on the signif-
icance of Homer was still valid throughout the Byzantine millennium: ('Homer
is the first poet every child meets, the grown-up man meets him in the middle
of his life, and the old man as the last one, and from his wealth he gives to
each of them as much as he can possibly hold. "7 In his judgement of other
authors and literary genres Dion usually proceeds from the standpoint of a
scarcely veiled utilitarianism.
From the centuries of late antiquity and of the Byzantine Age, by the way,
we have only a trifling number of theoretical remarks on the subject of imita-
tion. s The reason for this is probably that the tradition of imitating rhetorical
models had for a long time become a customary practice. When customs and
institutions have become firmly rooted in a continuous and stable convention,
explanations and justifications of their existence are seldom wanted. Of the
practical application of imitation, however, we have evidence from the various
centuries.
From the Christian sphere we learn, for instance, that the sermons of St.
Gregory of Nazianzus were not only written down officially by stenographers-
which may have been of importance for the later publication of the text-but
that also other churchgoers took notes of these sermons. Evidently, people
endeavored to collect beautiful figures of speech, similes, and images used by
the famous preacher, in order to use them occasionally for themselves. 9 Gregory
of Nyssa in his fourteenth epistle tells how a letter was passed around in his
literary circle, which was repeatedly read and even learned by heart by some,
whereas others copied down excerpts in their notebooks. This situation of
publicity in regard to literary production is characteristic of the entire By-
zantine Age: You met in a rather large circle interested in literary matters-
it is frequently called .9EClTPOV, presumably because in its functions it had to
compensate for the old stage theatre which was already dead at that timelO

5 Ibid., 457, note 11.

6 Dian Chrys., Or. 18.6 (ed. J. de Arnim, II [Berlin, 1896J, 252): 'IT!.EiVlV yap f) cxierS'1cr1S CmcxAAcxyeV'Tt
T~S mpl TO aVCXYlyvwerKEIV a"Xo!.!cxs.
7 Or. 18.8: ·O~'1pOS 01 Kcxl 'ITpGlTOS Kat ~Eeros Kcxl VcrTCXTOS 'ITaVTl 'ITalol Kcxt Qvopt Kat yepoVTI, 'roeroVrov QCP'
cxVTOVOIOOVS oerov EKCXcrTOS 00VCXTCXI !.cxJ3dv.
• E. g., Syrianus, ed. H. Rabe, I (Leipzig, 1892), 98, 1. 20f1., 104,1. 17ff. John Siceliotes, in Ch.
Walz, Rhetores Graeei, 8 vols. (Stuttgart-Tiibingen-London-Paris, 1832-36), VI, 71£. A. Brinkmann,
"Phoibammon nEpl ~1~f)erEws," Rheinisches ]'v!useum j. Phiioiogie, 61 (1906), 117-34, gives the text of
Phoibammon, who tries to invalidate the various objections against the practical possibilities of
successful imitation. It is always the imitation of style, of the three XCXpCXKT~PES (l"Xv6S, .,'eros, aop6s),
that is discussed here. The excerpt of John Doxopatres is printed below the text.
• Gregor. Kaz., Or. 42.26 (PG, 36, col. 492 A).
I. For examples for this use of the word, see H. Hunger, Reich del' Neuen 111ilte (Graz-Vienna-
Cologne, 1965), 341.
82
 ON THE IMITATION (MIMH~I~) OF ANTIQUITY 19
-being an aesthete you enjoyed the recital of rhetorical pieces of some kind
(official speeches, practice speeches, descriptions, letters, etc.)' you were de-
lighted with the contest of the authors, you participated if possible in the
acclamation concerning the value of the recited pieces-and you took notes
of everything that seemed useful for your own employment. Thus Quintilian's
aliud ex alio haeret was turned into practice. There is evidence for the continua-
tion of this institution from the fourteenth and fifteenth centuries. We also
have references to the contest, the passing around of letters, and the acclama-
tions from the later part of the period. We may assume that the excerpting
and jotting down of interesting phrases and expressions were still practiced.
For the rest, the questions as to the nature and the extent of imitation will
have to be judged from the texts of the Byzantine Age that have been pre-
served.
Before we tum to the discussion of the different forms in which classical
models were imitated in Byzantine literature we shall consider to what extent
ancient subject matter was received into Byzantine literature. First of all, it
becomes evident that the introduction of ancient historical or mythological
persons and events into epic poems, plays, or novels, which occurs so frequent-
ly in modem European literature ,11 remained relatively rare in the Byzantine
literary productions in the high language. The mythological epic poems of
Nonnus, Quintus Smyrnaeus, Tryphiodorus, Colluthus, Musaeus, and Maria-
nus12 belong entirely to the fifth and sixth centuries, that is, to the early
Byzantine Age. Later pieces of this genre are only the Carmina Iliaca, the
Theogony, and the verse chronicle of John Tzetzes (twelfth century).l3 Apart
from the very common didactic epic poems, the Byzantines on the whole pre-
ferred contemporary sUbjects. The drama, which continued to exist merely
in the form of a few scanty productions intended for reading, as well as the
novels in the high language from the time of the Comneni, which adopted
very freely the background and the motifs prevailing in the novels that were
written in the time of the Empire, are of no moment here. It is only the lit-
erature in the demotic language that shows signs of a certain tradition in
respect to historical and mythological themes throughout the centuries. One
has only to think of the Alexander prose romance or verse narratives, the
poem on Belisarius, the Achilleid, or the Vita Aesopi. To what extent the
original historical facts were blurred in these works or preserved only in a sug-
gestion of classical atmosphere is not our concern here.
In Byzantine rhetorical writing we find classical subjects in practice speeches
(I-iCAFrOI), descriptions (B<q>pO:aEIS), narratives (01T1YTJIlOTO), and character-drawings
(iJ$o1Tolial) that were continuously transmitted by the schools; with more or
11 Many hundreds of such instances of the effects of ancient mythology on even modern literature
are collected in my Lexikon der griechischen und romischen Myth%gie, 6th ed. (Vienna, 1969).
12 Presumably from his circle are 211 trimeters on the Twelve Labors of Heracles, ed. B. Knos in
Byzant. Zeitschrijt, 17 (1908), 406-21.
13 Carmina Iliaca, ed. I. Bekker (Berlin, 1816); Theogonie, ed. I. Bekker (Berlin, 1840), and epilogue,
ed. H. Hunger, in Byzant. Zeilsclirijt, 46 (1953), 302-307; Verschronik, ed. H. Hunger, in Jahrbuch d.
Oesterreich. Byzant. Gesellschajl, 4 (1955). 13-49. Although written in verse, the Horner allegories of
Tzetzes belong to the category of the commentaries, and therefore to nonfictional literature.
83
 20 HERBERT HUNGER

less extensive changes they belonged to the standard material of rhetorical

handbooks and instruction periods. Just as in the second century Aristides
had defended the old politicians of Athens, Miltiades, Cimon, Themistocles,
and Pericles in his work nepi TETT6:pWV, historical figures like Epaminondas
and Pelopidas or Alcibiades appeared in the Miscellanea philosophica et histo-
rica by Theodore Metochites. Similarly, his contemporary Thomas Magister,
treats the old motif of the contention between the fathers of the two heroes
killed in battle near Marathon in imitation of the rhetorician Polemon (ca.
A.D. 130).14 The same Thomas wrote a plea and a counterplea on the topic of
exemption from taxes (mEAElo), relying completely on Demosthenes' speech
against Leptines as far as the contents are concerned. 15
In order to offer a contemporary example of a crVI.I~OVAeVT1K6s MyoS, in his
Outline oj Rhetoric Joseph Rhacendytes the Philosopher (first half of the four-
teenth century) feigns the following situation: Somewhere in the Rhomaean
empire an isthmus (1o-.91.16s) is to be cut through; in accordance with the ancient
rhetorical technique he then lists the individual points on which the orator
had to give his opinion. At that time, however, such an enterprise would or
would not have been carried out only in accordance with the authoritative
decision of the emperor. 16 An attitude analogous to that of Thomas Magister
and other Byzantines toward Demosthenes we find in John Chortasmenus
toward Libanius, when, in the first half of the fifteenth century, he answers
four letters of the famous sophist in his correspondenceY
Of the seven recently published model specimens of rhetoric by Procopius
of Gaza (sixth century)18 two are character-drawings of Aphrodite and of
Phoenix as he appears in the ninth book of the Iliad; but in the other five
pieces too (description of the spring, of a meadow, etc.) we find ourselves in
the world of the ancient Greek gods and heroes. This is the rule with the great
mass of such productions throughout the Byzantine centuries. One might, for
instance, point to the "narratives" (5111yiJI.IO-rO) of Nicholas of Myra,19 Severus,20
or Nicephorus Basilaces,21 which have been transmitted together with the
Progymnasmata; one might also refer to the character-drawings of Aphthonius22
and his imitators, which are collected in the first volume of the Rhetores
Graeci by Walz. The short allegories on Tantalus, Sphinx, and Hephaestus by

14 :t\ew edition by F. W. Lenz, Funt Reden Thomas Magisters (Leiden, 1963). On p. VIII, Lenz very
characteristically comments on the nature of the imitation: " ... in der Weise, dass wir auf den ersten
Blick liberal! glauben, Polemon zu Ie sen, und doch nur ganz selten w6rtliche Entlehnungen oder
Zitate finden." ( ... in such a way that at first sight we think we are reading Polemon, and yet find
verbatim adoptions or quotations very rarely.) A fundamental feature of Byzantine imitation!
1> Also edited by Lenz, op. cit., who, on the basis of manuscript studies, could deny to Aristides

the authorship of these two speeches and attribute it to Thomas. The still unpublished speech
"For the Olynthians" certainly belongs to the same category of speeches with ancient historical
contents.
10 Walz, op. cit., III, 516-21.
17 My edition of the letters of John Chortasmenus appeared in 1969.
18 Procopii Gazaei epistolae et declamationes, ed. A. Garzya and R.- J. Loenertz (Ettal, 1963), 83-98.
19 Walz, op. cit., I, 269-72.
20 Ibid., 537-39.
21 Ibid., 428-42.
'" Aph thonji Progymnasmata, ed. H. Rabe (Leipzig, 1926), 34f1.
84
 ON THE IMITATION (MIMHLIL) OF ANTIQUITY 21
Michael Psellus belong here as well.:!3 In the case of the ethopoeia, Christian or
contemporary topics were exceptions . Yet, of the approximately two dozen
character-drawings of Nicephorus Basilaces (twelfth century) in Walz's col-
lection, about twelve are concerned with subjects taken from the Old or the
New Testaments, one dealing even with a "modern" topic. 24 In addition to
these, we should like to mention one ethopoeia by Nicephorus Chrysoberges
(twelfth-thirteenth century), namely, "What might a Christian philologist
have said when Julian the Apostate forbade [the Christians] to read Hellenic
books?" and one on the Virgin by Nilus Diassorenus (fourteenth century).25
Historiography and official rhetoric (speeches to the emperor and official
orations of all kinds) were thoroughly rooted in the classical tradition, yet the
main emphasis was never laid on classical contents. Most of the Byzantine
historians were interested in a more detailed account of the time they them-
selves were living in, that is, of contemporary history. Ancient history they
treated, if at all, as conscientiously as did the chroniclers, who always began
their often rather sketchy outlines with the creation of the world.
In conclusion we can say that only a relatively small part of Byzantine
literature is determined by the reproduction of classical contents and subject
matter. Much greater is the number of those works that are in some other
respect or even in a number of ways-and to a varying extent-characterized
by the imitation of the ancients. This long-known fact which, as far as I know,
has never found a specific and coherent treatment, must hardly be understood
as if the Byzantines had consciously conceived the hundredfold application
of ancient motifs, figures, and quotations as imitation. The fact is rather
that the Eastern Empire had not experienced a break in its historical and
cultural development as had the ,,vest, where such an interruption had been
caused by the establishment of Germanic empires on formerly Roman soil.
Again and again one discovers from remarkable details in the literature, art,
and architecture of Byzantium that the cultural continuity had been preserved
since antiquity.
Much of this situation was due to that intellectual development in the
course of which highly gifted as well as learned Christian Fathers-I am
referring to Clement of Alexandria, Origen, Eusebius, and above all the Cap-
padocians-succeeded in introducing the new Christian values into the long
approved literary-stylistic forms of pagan antiquity, and could eliminate
every thought of the discrepancy between these two elements in the course
of the development. Thus they created that intellectual attitude of Christian
humanism which at all times claimed the loyalty of the most noble minds and
the most outstanding writers of Byzantium. 26 A particularly original testimony
23 Ed. H. Flach, Classen und Scholien zur Hesiodiscken Tkeogonie (Leipzig, 1876), 424-28 .
.. Walz, op. cit., I, 466-525. The girl of Edessa who is deceived by a Goth (ibid., 519-22) is
borrowed from the Passio SS. Curiae, Samonae et Abibi by Symeon Metaphrastes (PG, 116, col.
145 D-161; Bibliotheca HagiograPkica Craeca, no. 736) .
• 5 K. Krumbacher, Cesckichte der byz. Litl., 2nd ed. (Munich, 1897), 470 and 560. The elhopoeia of
John Geometres on the Emperor Nicephorus II Phocas (PG, 106, col. 932) comprises only a dozen
lines.
" Hunger, Reich dey Neuen Mitle, 355-69.
85
 22 HERBERT HUNGER

of this intellectual attitude, which reaches from the literature of the educated
far into the realm of folklore, are those supposed prophecies of pagan philos-
ophers on the Trinity, the Incarnation, and other Christian dogmas, which
not only occur in the form of short passages in numerous manuscripts, but also
are known in the form of characteristic presentations in large fresco cycles of
many late- and meta-Byzantine monasteries and churches of monasteries. 27
In an essay on the "Vorbildqualitat und Lehrfunktion der byzantinischen
Kunst," O. Demus recently emphasized the naturalness, the universality, and
the representativeness of this art which he rightly calls a living Christian art. 28
Thus we usually feel that the imitation of this "intellektuellen Kunst, die ihre
Mittel der analytischen Bewaltigung des antiken Erbes mit bewusster Meister-
schaft anwendete" (intellectual art that applied its means of analytically
mastering the classical inheritance with conscious skill)29 is a survival rather
than a revival. Analogous conditions have to be kept in mind in regard to the
literature of the Byzantines. Since one knew one's Homer, the tragedians,
Herodotus, Thucydides, Plato and Demosthenes, Plutarch and Lucian so well,
one used quotations and allusions and adopted motifs and various associ-
ations very freely, and was quite unaware of utilizing foreign property or of
even committing plagiarism. 3o The mutual penetration of the old Hellenic-
pagan and Christian traditions in accordance with the principles of Christian
humanism usually enabled authors to introduce mythological or historical
examples of antiquity without giving offense to anybody by doing so. Those
cases where personal enmity or extreme zealotry did bring about attacks on
the "Humanists" shall be disregarded in this context.31
Some examples of this naive imitation of classical models are found in those
cases where Christian or contemporary Byzantine personalities are replaced by
mythological figures. The "as-if-by-chance" element of this pagan-Christian
mixture gives a particular charm to the works of art we here refer to. The
substitution of mythological figures tanked from circumstantially drawn com-
parisons to mere references or allusions.
In his sixth hymn (counting according to Terzaghi), Synesius presents
Christ as a second Heracles, without mentioning the name of the Greek na-
tional hero. Like Heracles, Christ "cleaned up" the earth, the sea, the air (the
07 Among the "prophets" in the refectory of the Lavra on lIit. Athos are depicted Socrates, Pytha-
goras, Hypatia, Solon, Cleanthes, Philon, Homer, Aristotle, Galen, Sibyl, Plato, Plutarch. Cf. Hunger,
Reich der Neuen Mitte, 303. N. A. Bees, "Darstellungen altheidnischer Denker und Autoren in der
Kirchenmalerei der Griechen," Byzantinisch-Neugriechische Jahrbucher, 4 (1923), 107-28; K. Spet-
sieres, Ehc6vES 'EA7Ii}vwv q>lAocr6cpwv Els bcKAT]cr1as, in 'ElTIcrrTJ~. 'E-1TFTT]p1S Tiis <!lIAOS. ~xoAfis TOO nallnnCTT.
'A.9T]V6iIl, Ser. 2, Vol. 14 (1963-64), 386-458; I. Dujcev, "Die Begleitinschriften der Abbildungen heidni-
scher Denker und Schriftsteller in Backovo und Arbanasi," Jahrbuch d. Oesterreich. Byzant. Gesell-
schalt, 16 (1967), 203-209. Of great interest would also be a study of the abundant material offered
by the frescoes of the Romanian monasteries of Moldavia.
28 O. Demus, "Vorbildqualitat und Lehrfunktion der byzantinischen Kunst," Aktelt ties 21. Intern.
K01lgresses luI' Kunstgeschichte in Bonlt, 1964, Vol. I: Epoche" europaischer Kunst (Berlin, 1967), 92-98 .
.. Ibid., 98.
30 That imitation is not to be understood as kAO'ITi} is emphasized already by the author of nepi V'l'OVS,
13.4. On writings 'ITEpi KAo'1Tfis, d. E. Stemplinger, Das P/agiat in del' griecllischen Litel'atur (Leipzig-
Berlin, 1912), 33-80.
31 Cf. Hunger, Reich de,' Neuen Mitte, 359ff.

86
 ON THE IMITATION (MIMHl:Il:) OF ANTIQUITY 23

domain of the demons), and the netherworld and eventually descended to

Hades as the "helper of the departed" and as a god. The descent to the nether-
world and the victory over hell are also portrayed as the deeds of a Christian
Heracles: Christ defeated Hades and Thanatus just as Heracles overcame
Cerberus and rescued Alcestis from death. Some years ago K. Weitzmann
pointed to the iconographic connection between the figure of Heracles with
Cerberus on a second-century sarcophagus in the British Museum and the
Resurrection picture in the lectionary preserved in the treasury of the Lavra
on Mt. Athos.32 One may regard as the missing link between these two ancient
monuments the reverse side of coins that depict the emperor dragging a captive
after him, a representation which was popular on coins in the fourth and fifth
centuries. 33
Georgius Pisides is known to have written a number of epic poems glorify-
ing his imperial lord Heraclius (610-641). In the first book of his Heraclias he
works out a a\iyKPIO'IS (comparison) between the Emperor and the Greek hero
Heracles, which is suggested by the similarity of the names. Homer should not
have addressed Heracles as a god, since the merit of having slain a boar and a
lion were not as important for mankind a~ was the salvation of the world through
the KOO'iJOpVO'TI1S (savior) Heraclius. 34 The poet then juxtaposes the individual
labors of the 6w6EKa.9Aos and the achievements of his Emperor. In analogy to
the Cerberus adventure, Heraclius descends to Hades, overcomes the mad
monster (the Sassanid Chosroes II), and raises Alcestis, i.e., the oikoumene
(the Christian world) from the dead. Like Heracles, the Emperor has killed a
dragon (of the Hesperides) and a hydra (Chosroes). Just as Heracles cleaned
up the Augean Stables, he has cleansed life which before had been covered
with dirt (pVTrwvra TOV TIpiv eset<OTIpwO'EV ~iov)-here we have a reminiscence
of the crusade motif of the war against the Persians !35 The Emperor has
strangled the lion that destroys the world (KOO'iJOCP.9opov = Chosroes); he has
secured the golden apples of the Hesperides, i.e., he has reconquered the
Byzantine cities (Tas TIOAEIS OAas) that had been occupied by the Persians. The
darkness (Chosroes) has vanished and the light (the imperial sun) has risen;
a new life, another cosmos, a new creation have begun: Kai KOO'iJOS aAAOS Kai VEW-
TEpa KTiO'IS (I.83). As in the writings of ancient Greek philosophy, the word
"cosmos" has here the meaning of world, but also of the great, tremendous
order.
As is so frequently found in Byzantine literature-a tradition that goes
back even to Clement of Alexandria-the mythological example is followed
by a Christian one from the Holy Scriptures. Pisides compares his imperial
lord to the patriarch Noah (6 Nwe TfjS yeas oIKOV\.lEVT'\S). Just as Noah had en-
32 Cf. K. Weitzmann, Geistige Grundlagen und Wesen der Makedonischen Renaissance (Cologne-
Opladen, 1963). 39f.• pIs. 37.38.
·33 A reference from a lecture of O. Demus. given at the Oesterreichische Byzantinische Gesellschaft.
Cf. H. Cohen. Description historique des monnaies frappees sous l'Empire Romain, I-VIII (Paris. 1880-
1892): Constantine I (C 237). Constans (C 133). Julian (C 75ft). etc. R. Ratto. Monnaies byzantines
(Lugano. 1930; reprint. Amsterdam. 1959). 161ft
.. Heraclias. I.65-70. ed. A. Pertusi in Giorgio di Pisidia. Poem;, I (Ettal. 1960).
35 Ibid .• I.76.

87
 24 HERBERT HUNGER

trusted himself and his family to the ark, Heraclius entrusted his existence to
his army and was therefore secure from the 1<OT01<AVUI-IOS Xoupoov. As the dove
brought Noah an olive branch when the Flood was over, the Emperor could
now gain the olive branch of peace. 36
The Byzantine poet, however, does not simply compare a Christian example
with a pagan one. Repeatedly he stresses-quite in accordance with imperial
propaganda and the prevailing Byzantine attitude-that this war had the
character of a crusade: A Christian basileus fought here against the heathen
fire-worshipperS' and astrologer Chosroes. Ironically Pisides asks: "Where
are [nowJ the investigations into the secrets of the stars? Who drew up the
horoscope of Chosroes' fall ?"38 Heraclius, hO'wever, was victorious with the
support of Christ, who had once again proved himself to be the "cornerstone."39
Through Christ as the "door" (iTlJATl),40 the Emperor found the right way and
became the good shepherd of his people. The aim of the crusade, however,
was the destruction of the Persian gods, not out of hatred against the element
of fire but in order to offer the creation, free and rescued, unto the Lord. 41
Following a basic thought of Byzantine imperial ideology, the poet has God
act as "collaborator" of the Emperor: Heraclius appoints God commander-in-
chief and thus secures the victory.42 The role of the Holy Ghost is to overcome
the language difficulties between the various peoples composing the army.43
The victorious struggle of Heraclius with his predecessor on the imperial
throne, Phocas, is also presented in a mythological garb by Pisides. The
usurper (T\ipavvos) Phocas is portrayed as the "tyrannical sea-monster on
land".44 The poet deliberately does not give the name of the girl threatened
by the monster, so as to leave it to the educated reader to choose between
Andromeda and Hesione, both of whom were chained to a rock as sacrifices
to such a monster and afterward set free, the one by Perseus, the other by
Heracles. Again Pisides combines this substitution of contemporary persons
by ancient mythological figures with the Christian tendency of his work and
with an entirely topical allusion: Heraclius is said to have confronted the
"debaucher of virgins" Phocas with the picture of the Immaculate Virgin
(a pun upon the word nopSevos)-a tale which evidently refers to the fact that
the fleet of the two Heracliuses, father and son, sailing from Carthage to
Constantinople for the overthrow of Phocas carried with it a picture of the
'Virgin and Child as a palladium. 45

3. Ibid., 1. 84-92.
37 TIVpaoACrrp'l" ibid, 1.14, 22ff., 181.
38 Ibid., 1.61 f.
39 Epbes. 2:20. He,'., 1.184.
4. John 10:7,9. Her., 1.193.
41 Her., 11.213-30.
"Exped. Pel's., IL118£., ed. Pertusi. Cf, H. Hunger, Prooil11,ion (Vienna, 1964),88, Idem, "Kaiser
Justinian 1.," Anzeiger d. Oesterreich. Akademie d. Wissensclzajten, Phil.-hist. Klasse, 102 (1965), 346.
43 Exped, Pers., lI,170-76 .
.. Tfis yfjs ,.6 KfjTOS, He,'" 11.11; Kfj,.OS ,.vpavvov, ibid., 11.22.
"Ibid" lI.B f.: ahA' av-rl1"Cl~as "0 q>SOpEl ,.wv 1TapS/vwv / TO q>plKTOV £150, TfjS axpc'xvToV napS/vov.
Cf. Tbeopban., ed. C. de Boor, I (Leipzig, 1883), 298, I. 15ff.; Pisid" Exped. Pers., I.139ff., with the
commentary of Pertusi, p, 142 f.
88
 ON THE I l\HTA TI ON (l\IIM H ~ I~) OF A:.JT IQU I TY :25

To the historian Agathias the relating of a historical event of his time pro-
vides an opportunity for including a mythological example in his account. King
Chosroes I has his defeated general Nachoragan flayed and the empty skin
put up on a pole. Agathias then tells the story of ApoUo and Marsyas, not
without rationalistic criticism of the myth and particularly of the dispro-
portionate cruelty of the god (emavSpc,mia).46 With a relevant quotation he
refers to the Dionysiaca by Nonnus (1.42£,), adding casually that he cannot
remember the context and the immediately preceding lines in Nonnus' work 47
In his attempt to describe the character of the Emperor Andronicus 1,
Eustathius of Thessalonica compares him to Proteus and Empusa. 48 Nicetas
Choniates likens the Empress Euphrosyne, the wife of Alexius III, to Penelope
because she succeeded in "undoing" the plot contrived by Contostephanus,49
whereas Isaac II Angelus, who superseded the old Andronicus I on the throne.
seems to him like Heracles who saved Andromeda (Constantinople) from the
monster (Andronicus I) threatening her. 50 Ducas compares Andronicus IV
who imprisoned his father and brothers, to Zeus who kept his father Cronus
and his brothers Pluto and Poseidon within bounds in order to secure the
powers of ruler for himself. 51
The mythological example may be regarded as a stock element of Byzantine
epistolography. With Libanius, for instance, the examples are legion. Usually
mythological figures are introduced as models for the author's contemporaries,
particularly for his addressees. Thus, as a model to an addressee, Libanius
presents the Agamemnon who did not rebuke Odysseus for having failed in
his mission to Achilles (Il., IX.676£.).52 In the eighteenth letter, the Athena
of the first book of the Iliad, who appeases Achilles' wrath by order of Hera
(Il., I.194ff.), serves as a model for Libanius' relationship to Tatianus. 53 On
another occasion Libanius challenges his addressee to imitate Achilles, who
first inflicted the wound upon Telephus and then healed it again: "lVIay you
become an Achilles unto Telephus and heal the consequences of wrath through
clemency."54 The wording shows-here as well as in other instances-that in
the mind of the writer the mythological figure has taken the place of his con-
temporary. Thus Libanius says on one occasion: "We beg you, Ajax, to lend
him [SalviusJ your shield and help him in a seemly way."55 Odysseus' relation-
ship to his protectress Athena, which is characterized by confidence, serves
as a model for the relations between contemporaries of Libanius, namely
.. Agatbias, 1\'.23 (ed. R. Keydell [Berlin, 1967], 151£.).
47 Ibid.: ou yap 8~ ,.c.,v 'TfpoT1Yovl'ivwv E'TfWV Eml'i~V'1~al; by doing so, might he not want to create
the impression that he has been quoting from memory I
.. La espugnaziane di Tessalanica, ed. S. Kyriakides (Palermo, 1961). 14, l. 31-16. l. 1.
.. Nicetas Choniates, Bonn ed., 687.
50 Nic. Chon., Speech No.9, trans. F. GrableI', in Byzantinische Geschichtsschreiber, VoL 11 (Graz,
1966), 156.
51 Ducas, 12.3 (ed. V. Grecu [Bucharest, 1958J, 73, II. 8-12).
32 Liban., Ep. 432.4 (ed. R. Forster, X [Leipzig, 1921J, 422, 1I. 10-14).
;a Idem, Ep. 18 (ed. Forster, X, 10, 1. 6ft.).
5-1 Idem, Ep. 1105 (ed. Forster, XI [Leipzig, 1'922J, 212, l. 13f.). Cf. the same motif in Ep. 754.7
(ed. Forster, X, 680, 1. 4fI.).
;5 Idem, Ep. 1433.4 (ed. Forster, XI, 472, 1. 6ff.): oeo\.l€$a 8iJ O'ov "00 Aiavros llE'l'aoouva[ ,.e a\J'1'('ij "iis
clO''TfiooS Ked j3oT1$iiaQI ,." dKO,.a.
89
 26 HERBERT HUNGER

Tatianus and Hesychius,56 and Theodore and Eusebius. 57 As a model for the
similarity between children and their parents, the story of Odysseus and
Telemachus in the fourth book of the Odyssey is used. 58 In all these examples,
the number of which could be enlarged at will, mythological figures in charac-
teristic and representative situations take-at least for an instant-the place
of the addressees. The modern reader thus receives the impression of an ancient
atmosphere.
Something similar can also be found in epigrams; here two examples from
the Palatine A nthology may serve our purpose. Paul the Silentiary dedicates
a poem to the myth of Danae, to which he gives this rationalistic interpreta-
tion: Basically, every woman can be bought with gold. In fact the epigram-
matist wants to confirm a general experience (the venality of love) by means
of the story of Danae. 59 In the poem immediately following, Agathias Schola-
sticus tells (or feigns?) that a rival has alienated the affections of his girl from
him. To illustrate the situation he uses well-known titles of comedies by Me-
nander: the girl corresponds to Perikeiromene (The Shorn Girl), the rival to
~Iisoumenos (The Odious One). the author himself to Dyskolos (The Malcon-
tent) .
In a second group of examples the imitation of antiquity pertains to whole
scenes, to individual motifs, and finally to mere associations and slight allu-
sions. All these may, but need not, be accompanied by quotations from classical
authors.
A particular problem in this connection are various passages from Byzantine
historians who provably borrowed also from the contents of some classical
record which they may originally have wished to imitate in its formal aspects
only. In such cases there is, of course, the danger that the historical truth
might be falsified, a problem to which G. Moravcsik recently gave his atten-
tion. 60 Our distinguished colleague and friend was able to demonstrate by
means of several examples that the identification of a classical model which was
used by a Byzantine author for his report even in respect to its contents, does
not necessarily discredit the Byzantine historian. Thus, for example, Priscus
relates the conquest of Naissos (Nis) by the Huns in 441 in the form of a
narrative 61 which in many respects reminds us, in its phraseology and coloring,
of Thucydides' description of the conquest of Plataea (I1.75f.), and which is
reminiscent too of the conquest of Philippopolis described by Dexippus. 62
Nevertheless, the remark that the Huns used siege engines on the occasion
need not too readily be doubted, for Procopius, for instance, gives the same
account of the Sabirs, who were related to the Huns. 63
"Idem, Ep. 855.1 (eeL Forster; XI, 13,1. 15ff.).
"Idem, Ep. 905.1 (ed. Forster, XI, 54, I. 5f.).
'" Od., IV.141ff. Liban., Ep. 93.1 (ed. Forster, X, 92, L 15ff.).
59 A. P., V.217 (ed. H. Beckby, I [Munich, 1957], 234).
60 Gy. Moravcsik, "1(lassizislTIns in der byzantinischen Geschichtsschreibung," Polychronion, 10",/-
schl'i/t F. Dolgel' zum 75. Geb"I'tstag (Heidelberg, 1966),366-77.
Gl Hist01:iei Gmeei Minol'es, ed. L. Dindorf, I (Leipzig, 1870),278-80: ff. 1"-

"Ibid., 184-86: fr. 19.

63 Procop .. Dc bel/is, YIIl.11 (cu. ]. Hann', [Leipzig, 1963J, II, 539ff.) lIIoravcsik, op. cit., 370.

90
 ON THE Il'vIITATION (jIIM H ~I~) OF ANTI QUITY :27
Classical reminiscences are very likely to occur in the historical episodes
presented by authors who based their imitations on Herodotus and Thucydi-
des. \"hen the same Priscus, referring to the conclusion of peace between
Per6z, king of Persia, and Kunchan, chief of the Kidarite Huns, tells that the
Persian deceived the Hun by giving him not his sister-as had been agreed-
but a servant as a wife,54 the anecdote in Herodotus (UI.l), where Amasis
gives not his daughter but some other woman to Cambyses, suggests itself as a
model. Nevertheless, the fights between Persians and Huns as well as the
marriage which followed are not to be doubted, since they can be verified by
means of Oriental sources. Priscus, however, adopted the form of a short tale
from Herodotus in order to give his presentation a classical touch and offer
his audience associations with the famous father of historiography.s5
Procopius, in telling about a stratagem of the Hephthalites who built well-
camouflaged pitfalls against hostile horseman,66 has obviously in mind the
report by Herodotus (VIII.28) on a similar stratagem of the Phocians;67 but.
again, his account can be verified by Oriental sources, as Haury was able to
show. 68 In such cases the historical credibility should not be prematurely
questioned on the grounds that the imitation of the Byzantine author pertains
even to the contents of his model.
In order to present certain qualities of their heroes more strikingly, histor-
ians and rhetoricians introduce mythological and historical figures for the sake
of comparison. In presenting the heir to the throne Constantine (X) Ducas
and the nobility of his birth, Psellus goes back as far as Achilles, the latter's
father Peleus, and his grandfather Aeacus. 69 For the reigning emperor, famous
rulers of antiquity are preferably used as patterns, for example Alexander the
Great, Caesar, Augustus, Pyrrhus, Epaminondas, and Agesilaus for Constan-
tine IX,70 Alexander the Great for Alexius I Comnenus ;71 Cato the Stoic.
whose equanimity could not be disturbed by his fever, appears as a counter-
part to Isaac Comnenus, who ceaselessly tossed himself from side to side in
his fever. 72 To impress more effectively upon the reader's mind the fascinating
beauty of her mother Irene, Anna Comnena even invokes the petrifying effect
of the Gorgon's head,73 and in describing the battles of her father Alexius she
introduces Typhon and the struggle between the Giants and the gods by way
of comparison!4
As a result of the tradition which had remained unbroken since antiquity .

.. Exc. de legat., ed. C. de Boor, I (Berlin, 1903), 153, 1. 25-154, L 32.

" R. Benedicty, "Die historische Authentizitat eines Berichtes des Priskos. Zur Frage der histono-
graphischen Novelli5ierung in der friihbyzantinischen Geschichtsliteratur," ] aizrbucil d. Oesterreich.
Byzant. Gesellschatt, 13 (1964), 1-S .
.. Procop, , De bellis, 1.4 (ed, Haury, I, 15; II. 6-20).
"Cf. also Polyaen., Strateg., VI.1S.2 (ed, E. Woelffiin and I. ::'lelber [Leipzig, 1887:,297).
58 Cf. Moravcsik, op. cit., 37sf.
W Psell., Chron., ed. E. RenauId, II (Paris, 1928), 134.
7. Ibid., slf.
71 Anna Comn., XV.7.S (Bud" CoIL).
" PseII., Chron., ed. RenauId, II, p. 130.
13 Anna Comn., III.2.4.
"Idem, 1.7.3.
91
28 HERBERT HUNGER

a great number of mythological and historical situations and motifs had so

deeply penetrated the minds of all educated Byzantines that we find them
again and again in the works of the historians and rhetoricians. Some ex-
amples may suffice. First, those from mythology: ideal friendship is repre-
sented by the pairs Orestes-Pylades and Theseus-Pirithous, quarrelsomeness
by the apple of Eris, a time of peaceful tranquillity by the "halcyon days";
the union of Alpheus and Arethusa stands for all-conquering love, Ixion and
Hera for unfulfilled-today we should say frustrated-love, the Heliads' tears
that were turned into amber stand for never-ceasing grief, the "helmet of
Hades" for invisibility; cunning changeableness is represented by the versatile
Proteus, rapid transitoriness by the" Gardens of Adonis", etc. Second, motifs
from history: for the wisdom of the emperor and legislator stand Numa Pom-
pilius or Lycurgus, for the murder of the tyrant, Brutus or Harmodius and
Aristogiton, for fear of God, Publius Cornelius Scipio, for presumptuousness,
Xerxes, for security of the walls, Semiramis, for a capricious disposition, the
changing current of the Euripus, etc.
A limited number of mythological motifs became particularly popular in
epistolography; in these cases one can speak of specifically epistolographic
topoi. There is, first of all, Hermes, who, being the god of rhetoricians, was
at the same time regarded as the divine friend of letter-writers. Libanius, for
instance, writes on one occasion: "If Hermes and the other gods permit it,
we shall meet within this month,"75 and another time: "By Hermes, do remove
that prooimion from your letter to me .... "76 'Ep~o0 Aoyiov 7VlTOS one used to
say of correspondents to whom one wished to pay a compliment."
To the stock themes in Byzantine letter-writing belongs the unquenchable
desire to visit the addressee. To reduce the sometimes great distances one
frequently longed for the winged sandals of Perseus. 78 The wish to visit his
friend by way of flight is already expressed by the writer of a papyrus letter. 79
Libanius speaks several times of the desired wings, but only once of Perseus. so
Procopius of Gaza, on the other hand, presents the motif in a clearly pro-
nounced way: "Could I but become a Perseus and sail through the air on
wings and be carried across the sea, so that I might-as soon as I wish-be
with you and enjoy your 10ve."Bl Later, we come across this motif in Nice-
phorus Uranus82 and Michael Psel1us. 83 Basil the Great uses the words of the
Psalm (54:7): "Oh, that I had wings like a dove!" in the same connection. s.l

" Liban., Ep. 894.3 (ed. Forster, XI, 44, 1. 18H.).

" Idem, Ep. 1497.1 (ed. Forster, XI, 524, I. 21f.); d. also Epp. 199, 269, 338, 884, 1145, 1400.
" Synes., Epp. 101 and 159 (ed. R. Hercher, EPistolographi Graeei [Paris, 1873J, 699, 739); Thorn.
;lIag., Ep. 1 (PG, 145, col. 405.2).
,. On this motif, d. G. Karlsson, I dlologie et cerlmonial dans l'epistolographie byzantine, 2nd ed.
(Uppsala, 1962), 57f.
'" U. Wilcken, Chrestomathie de. Papyruskltnde (Leipzig-Berlin, 1912), No. 481.
.0 Liban., Ep. 44.2.
81 Procop., Ep. 58.6-8 (ed. Garzya and Loenertz); d. also Epp. 29, 90, 123 .
• 2 J. Darrouzes, EPistoliers byzantins du X, sieele (Paris, 1960), V, 47, I. 55f.
.3 PselI., Ep. 14 (ed. E. Kurtz and F. Drexl, II ([Milan, 1941J, 17, I. 15) .
•• Basil., Epp. 47 and 140 (ed. Y. Courtonne, I, II [Paris, 1957, 1961J); d. also Niceph. Basilaces,
Ep. 3, I. 4 (ed. A. Garzya, in Byzant. Zeitsehrift, 56 [19631, 232).
92
 ON THE IMITATION (MIM H~I~) OF ANTIQUITY 29
Two components unite in the topos of the Sirens, which frequently recurs in
Byzantine letters. In the earlier literature the moral motive seems to pre-
dominate which-sometimes with a reference to the ethical interpretation of
the myth-sees the Sirens as temptations that beset man. 85 Later, the Sirens
with their enticing song are understood as the voice, or rather the letter, of
the correspondent, which has completely betwiched the writer. Examples of
this interpretation are offered already by Synesius (Epist. 146) and Procopius
of Gaza (Epist. 120); in Psellus, Planudes, and Demetrius Cydones it definitely
predominates. Occasionally, the writer is proudly aware of the fact that his
own letters, too, cast a spell which his partner finds difficult to evade. Thus,
for instance, Psellus writes: "Sail past my Sirens !"86 and in another passage
he owns: "I am not able, as was the son of Laertes, to sail past your Sirens. "87
Planudes, writing to a woman, admits that it would betray lack of education
and even be offensive if he were deaf to the Sirens, i.e., her letters.ss Demetrius
Cydones hopes that in reading a letter of his correspondent he will have the
impression of hearing his Sirens.89 Another time, like Psellus, he speaks of these
Siren songs as being reciprocal,90 or, again, emphasizes the importance of those
letters which he received on a long travel, a further tertium comparationis identi-
fying the writer with Odysseus listening to the Sirens. 91 In a few cases we meet
with relations of a different kind, as, for example, when Theophylact Simocattes
juxtaposes the enchanting song of the Sirens with a lament of the Muses
(Epist. 21), or when Cydones understands the Sirens as the personification of
homesickness.92
Finally, quotations from classical authors contribute toward intensifying
the impression of imitation on the reader or listener. Here we would like to
recall that the Byzantines usually quote without giving the name of the author,
or by giving it in a more or less encoded form. That "the poet" \vas Homer,
the son of Olorus, Thucydides, and the man from Paeania, Demosthenes one
learned in elementary school. Titles of works or even hints as to the more
specific context of the quotation are found very seldom. Since one obviously
used to quote from memory, inaccuracies and misunderstandings were inevi-
table. The identification of quotations on the part of the audience seems even
to have been a kind of round game in Byzantium. 93 Furthermore, there were
certain quotations, especially from Homer, that enjoyed a special popularity,
e.g., OVK aya.S6v "ITOAV1<Olpavill' sIs Koipavos EO"TVJ (Il., II.204); oil ve~Ecj\S Tpwas Kai
EVKvi]~15as 'AXOlovs TOlij5' a~<pi YVVClIKi "ITOAVV Xp6vov O:Ayea "ITOO")(EIV (Jl., III.156f.);

85 Cf. Synes., Epp. 32,146 (ed. Hercher, 654, 729); Procop .. Epp. 57, 92, 110 (ed. Garzya and Loenertz,
33, 49, 57); Theoph. Sim., Ep. 82 (ed. Hercher, 785 t.).
8. C. N. Sathas, MEO"<X1Co)V1KT) Bl~i\10.st'\KT), V (Paris, 1876). 296.
8, Psell., Ep. 17 (ed. Kurtz and Drexl, 21, I. 11).
88 Planud., Ep. 6S, 1. Sf. (ed. M. Treu [Breslau, 1890J, 85). Cf. Epp. 21,1. 27ff.; 113,1. 68: here ex-
ceptionally the singular U1Pt'\V, in the same meaning as 'OpcpeVs and ivy~.
s·Dem. Cyd., Ep. 27, 1. 33f. (ed. R- J. Loenertz, I [Citta del Vaticano, 1956J), at the end of the
letter .
.. Idem, Ep. 84, 1. 23ff., also at the end of the letter.
"Idem, Ep. 106, 1. 7ff. Ct. also Epp. 10,1. Sf.; 17, I. 42t.; IS, I. 10; 33, I. 38f.
.. Idem, Ep. 237, 1. 33f. (ed. Loenertz, II [Citta del Vaticano, 1960]).
'3 Ct. Hunger, Reich der Neuen Mitte, 342f.

93
 30 HERBERT HUNGER

ev5E1S, 'A,pEoS viE 5a1'l'POVOS hm05cq.10l0· 011 xpi] iTaVvVxlOV ev5ElV 13ovATJ'l'0pov o:v5po, 4>
Aaoi I' E-rrli'Elp6:q:>aICll Kai loO'O'a ~E~TJAe (ll., 11.23-25 = II.60-62); 011 ~EV TTCuS VVV fO'lIV
c'rn6 opv6s ovo' c'rn6 iThpTjS (ll., XXII.126; Od., XIX.163), etc.
Since Byzantine authors writing in the high language liked to parade their
learning, they quoted abundantly, sometimes using a mass of quotations, which
one might call a kind of mannerism. Here I may also recall the fact that in
such cases the mixture of pagan and Christian quotations was a pattern that
was popular with many writers and can be traced back even to Clement of
Alexandria. 94 As late as the early fourteenth century, Joseph Rhacendytes
the Philosopher expressed his opinion on this "pasting together" of quota-
tions-in connection with the epistolary style-in his Synopsis Rhetorices,
chapter 14.95
Sometimes it is a hardly extraordinary allusion or word combination that
signalizes a quotation. Yet, when Procopius in his A necdola says of the Emperor
Justinian, em lolilECilpOS ap.geiTj Kai aepo13a-roiTj, one will immediately think of
Socrates in his basket, in Aristophanes' ClOttds. 96 When, however, in the same
chapter the verbs 6:yplClivo~CIl, or 6:yploo~al, and O'aipCil are jointly used, only a
dictionary will refer them to a corresponding passage in Aristophanes. 9 ' Re-
ferring to Justinian with KVKWV yap aei Kai ~vv-rap6:O'O'Cilv aVEO'o13El E'l'E~iis crnav-ra,
Procopius alludes to a passage in the Knights.98
We now have to turn to that kind of imitation which concentrated, without
regard to the contents, on classical forms, on style. For the Byzantine author
there was, of course, no conscious division between the different possibilities
of imitation that we have pointed out. On the contrary, they usually went
together. Our division was made only for the sake of a better arrangement of
the material. Formal imitation pertains 1) to all linguistic phenomena, 2) to
meter.
In regard to the language, the strictly and stiffly preserved Attic model
affected phonetics, morphology, and stylistics. Until the end of the Byzantine
Empire educated authors took pains, though with varying consistency, to
write double tau instead of double sigma in 'l'VAO:HCil, KTJpVHCil, yAwHa, etc., as
well as occasionally ~vv instead of O'vv, and yiyvo~al instead of yivo~al.99 The
Attic dual, which had already fallen out of use in the literary koine of the
Ptolemaic period, \vas, from Aristides up to the fall of Constantinople, the pride
of many an author who boasted of his learning 100 Authors writing in the high
language retained the dative to a much greater extent than was required by
"Ibid., 302f. Examples of clusters of quotations, for instance, in Nicetas Choniates: Bonn ed.,
336 and 640; Speech Ko. 9, trans. Grabler, op. cil., 151 f. See furthermore F. Grabler, "Das Zitat als
Stilmittel bei ?\iketas Choniates," in Aklen des XI. Inlernalionalen Byzanlinislmkong"esses, J1Jiinchen
1958 (Munich, 1960), 190-93.
B5 Walz, op. cit., III, 558f.
.. Procop., An., 13.11 (ed. J. Haury, III, pt. 1 [Leipzig, 190611; Arist., .1I'ub., 225.
" Procop., An., 13.3; Arist., Pax, 620.
"Procop., An., 9.50; Arist., Equ., 692 .
•• Statistics of individual authors should be set up only on the basis of a thorough knowledge of
the entire tradition. Even then it w01l1d be difficult to decide what exactly has to be attributed to the
author and what to the copyist.
100 E. Schwyzer, G,.iecli. Gmmmalik, I ("'IUDich, 1939), 127.

94
 ON THE IMITATION (MIMH~I~) OF ANTIQUITY 31
the conditions of the actual development of the language. lot What was to be
regarded as Attic, or un-Attic, or even solecistic was decided by the grammar-
ians, who on the whole followed the doctrine that had been developed in the
time of the Empire.102 Contracted forms of nouns and verbs, Attic declension,
the indefinite pronoun (particularly in its shortened forms TOV, T~), reduplica-
tion instead of the pluperfect with auxiliary verb (E-rETCXxaTo), the middle voice,
the optative, the accumulation of negations, the pleonastic use of particles
(especially Kat), the use of abstracts (TO Vrr~KOOV, TO .9fjAV) instead of concretes,
the perfect with present tense meaning, the figura etymologica, and many
others were preferred because they were held to be Attic. lo3
An important, although negative, element of this linguistic imitation was
the avoidance of all colloquial words and forms, and also of those often techni-
cal terms which-in the course of the development of the koine and later,
during the centuries of the Empire--had penetrated from Latin into Greek.
Though not all Byzantine authors followed this principle with equal con-
sistency, most of them endeavored, in accordance with the imitation of classi-
cal models, to keep their own work free from all elements of the actually
developing language, to keep it, I should say, germ-free, i.e., "sterile." Thus
the names of peoples and tribes that had not been members of the Byzantine
Empire were, in every case, taboo. They had become known, it is true, in the
early and middle Byzantine centuries, but had been unknown to the ancient
historians. Now one tried, usually at random, to identify the contemporary
tribes with those races whose names could be found in Herodotus, Thucydides,
and other ancient historians, the spatial coordination or an imagined kinship
or identity evidently deciding the issue. Considering the methods employed, we
are not astonished to find that three, or four, or even more archaizing names
were attached to every ethnical group by diHerent Byzantine authors.104 In a
similar way the Byzantines rendered foreign titles and offices: Seljuk and
Serbian governors,I05 and even their own generals, were called satraps simply
because this term had been familiar to the ancient historians.10 6
The name that was usually applied to one of the Circus parties, the "Blues,"
j3evETol in colloquial language, is "explained" by Procopius (05" Kvaveov E<IT1),107
who usually emphasizes also the Latin origin of certain terms as, for instance,
OO/-lE<IT1KOV TOu-rOV KaAOVC)'l 'PW/-laiOl,108 or, ,OV KaAOV/-lEVOV KOlai<ITwpos.1°9 This attitude
finds its particularly manneristic expression when, in spite of the enormous
101 In the case of Cinnamus, this was demonstrated by F. Hormann, Beitrage zur Syntax des johml-
nes Kinnamos, Diss. Miinchen (1938), 85-105; in the case of Eustathius of Thessalonika, by P. Wirth,
Untersuchungen zur byzantinischen Rhetorik des 12. jahrhu"derts, Diss. Miinchen (1960). 60-74.
102 Cf. G. Bohlig, Untersuchungen zum rhetorischen Sprachgebrauch der Byzantiner, mit besonderer
Berucksichtigung dey Schriften des Michael Psellos (Berlin, 1956), 1 ff.
103 Ibid., 34ff., 67f., nff., 85, 94ft., 98ft., 162£, 20lf£., 215, 228, 234ff.
10< On this familiar phenomenon, d. Gy. :\loravcsik, Byzantinolurcica, 2nd ed. (Berlin, 1958), II,
13ft. Idem, in Polychronian, 3121.
10' Ibid.
106 H. Hunger, Der byzantinische Katz-i\lause-l\rieg. Die Kaiamyamachia des Theodoras Pradramos
(Vienna, 1968), 113, on I. 293.
107 Procop., De bellis, n.ll (ed. Haury, I, 203, I. 31.).
108 Ibid., rII.11 (ed. Haury, I, 361,1, 161.).
109 Procop., An., 6.13; 9.41. -
95
 32 HERBERT HUNGER
expansion of monachism in the whole East, and also in the capital, during the
fourth and fifth centuries, in the middle of the s~!'th century Procopius still
paraphrases consistently the most common word for monk, lJovaxos, by writing:
ovcrrrep 1<aAEiv lJovaxovs VEVOIJIKOGI, and similar examples.llo Theophylactus Simo-
cattes tries to avoid the offensive colloquial !3EVETOI by paraphrasing it clum-
sily.11l Nicetas Choniates speaks of the "one in charge of the mixing jug," in
order to avoid the title lTIyxEPIlTlS (cup-bearer), which is of Latin descent, and
talks of "internal treasure houses" (TO: £Gc,,) TOlJleio) so that he would not have
to use the technical term (Oi1<EIOKOV !3eGTlaplov) for the privy purse of the em-
peror.l12 In practice, of course, it was hardly possible to stick to this mannerism
consistently. Eustathius, for instance, vivaciously declares, in his report on
the conquest of Thessalonica by the Normans, that he preferred to say 1<01J1'IS
instead of KOVTOS because he hated the barbaric, i.e., the Latin-Italic, expres-
sion.ll3 In the same work, however, Latin terms like 6:SVOVIJIOV or ll'pol3a occur,
with which the author found no fault.u 4
From the time of the Empire, Attic dictionaries served as an important
device for the avoidance of non-Attic words; in Byzantium, too, these lexica
enjoyed great popularity. It has been estimated that about 2,500 Attic glosses
from older dictionaries of this kind were included in the Byzantine lexica. m
How great the interest in this subject must still have been at the time of the
Palaeologi becomes obvious from the new compilations by Manuel Moscho-
pulus and Thomas MagisterY6 It meant the highest possible praise for a
writer if his style was acknowledged to be perfectly Attic; such tribute was
payed to Theodore Metochites by his pupil Nicephorus Gregoras in his epi-
taphll7 and by Nicephorus Chumnus. ll8
The strong trend toward the imitation of classical literature in Byzantine
writing also made itself felt in the development of Byzantine prosody. Since
the classical iambic trimeter had been accepted as the most common meter,
a grotesque situation, though typical of Byzantium, ensued during the early
Byzantine centuries, when gradually the accenting meters gained ground. Al-
though the syllables were no longer measured but counted and the sense of
longs and shorts became more and more lost, which is reflected in the un-
certainty as to the use of the so-called dichrona (the vowels alpha, iota, ypsi-
Ion), most of the poets, impelled by their classicistic ambition, tried to present
to the educated reader a trimeter that was perfect even in the sense of the
classical quantitative meter. In this case the imitation of the ancients resulted
in a discrepancy between poetry as it was read and poetry as it was heard,
no Cf. the index of Haury's edition (III, pt. 2 [Leipzig, 1913J), 387.
III Theoph. Sim., ed. de Boor, 296, I. 24f.
III Nic. Chon., Bonn ed., 384 and 708.
m La espugnazione di Tessalonica, ed. Kyriakides, 110,11. 16-18.
ntIbid., 66, I. 17; 68, I. 13.
110 H. Erbse, Unte~suc'!lmge1l .::u den attizistischen Lexika (Berlin, 1950), 6.
IU Manuel Moschopulus, l:VAAO)'1\ 6vo~trTwv CrrnKC,V, ed. A. Asulanus (Venice, 1524); Thomas
:Ylagister, 'EKAo)'1\6vo~trTwv Kat {>1)~trTWV CrrrlKC,V. ed. F. Ritschel (Haile, 1832).
117 Nic. Greg., X.2 (Bonn ed., I, 477).
118 J. F. Boissonade, Anecdola nOM (Paris, 1844), No. 133, p. 156; cf. also H. Hunger in Byzallt.
Zeitsc,...;!t, 45 (1952), 9.
96
 ON THE IMITATION (MIMH~I~) OF ANTIQUITY 33

and developed a "poetry for the eye" which was no longer related to con-
temporary language and versification. The frequently occurring imitations of
the classical dactylic hexameter and elegiac distich, of trochaic octosyllabics
and anacreontics show the same tendency. Of course, it would have been
better had secular poetry made itself independent of the rules of quantity, as
religious hymn-poetry had done by following the example of Romanus Melodus,
and had it cultivated the genuine Byzantine meter, namely the "political"
fifteen-syllable line. Actually, however, the hexametric attempts of later schol-
ars, like for instance John Chortasmenus (first half of the fifteenth century), are
indications of an utter inability to appreciate the peculiarity of classical meter.1l9
A characteristic feature of Byzantine art and literature, as yet perhaps
hardly noticed, is the balance between a strict adherence to an acknowledged
and accepted tradition-in our case the imitation of antiquity-on the one
hand, and the greatest possible variation of detail on the other; in the best
works of art and literature this is excellently done. The ingenuity of the writer
will express itself in an abundance of stylistic details and phrasings of his own
coinage, which, however, have to be sought; the superficial observer will see
nothing but the repetition of well-worn cliches. I first pointed to this phenome-
non in my book on the prooimion, five 'years ago; in studying the preambles
of the documents, the composition of which was entrusted to men who were
extremely well versed in literary matters, I had noted this combination of
traditionalism and abundant variation. 120 The antinomy of strict imitation in
regard to the whole and broad diversity in detail is of consequence in the
much abused centos, the patchwork poems, to which a short discussion shall
be dedicated here from the viewpoint of the Byzantine literary historian.
Two epigrams by the philosopher and mathematician Leo (fifth century)
that are contained in the ninth book of the Palatine Anthology shall be cited
here as examples of the Greek pagan cento.l2l The first consists of six, the second
of twelve hexameters; each of these lines is a complete verse from the Iliad or the
Odyssey. In the first epigram we can trace only one minor deviation from this
strict rule: in line 4 the nominative yv~v6s had to replace the original accusative,
and line 5 was composed of parts taken from two verses. The first word of the
last verse is used with a meaning different from that in Homer, namely as
an obscene homonym, which is the point of the whole epigram. The second
epigram (Hero and Leander) is similarly composed of complete verses from
Homer. Here the only deviation from the verbally rendered passage from
Homer occurs in line 6 (middle of the epigram), which again consists of parts of
two verses taken from the Odyssey; these two lines were of necessity retouched
for grammatical reasons, but remained otherwise unchanged. In these twelve
lines the poet succeeded in recounting: the story of Hero and Leander briefly,
without apparent effort, and at the same time without grammatical errors
-all this without himself having troubled to compose a single verse or part
110 On this problem, d. Hunger, Der by;;antinisciLe Katz-j\1ause-Krieg, 30ff.
"" Hunger, Prooimion, 17 and 58.
ill A. P., IX.361 and 381 (ed. Beckby, III [Munich, 1958], 222, 236). On pagan centos, d. Stemp-
linger, op. cit., 193-95.
97
 34 HERBERT HUNGER

of a sentence. A praise of such methods will hardly pass the lips of the modem
reader who expects original ideas of a poet's genius. Whatever the case for
originality may be, one will have to admit that such achievements were possible
only on the basis of a most intimate familiarity with the text of Homer, a deep
understanding and knowledge of the language, and an excellent memory. After
all, we cannot believe that the writers of centos went, kalamos in hand, through
the whole text of Homer searching for'suitable lines-or that they even con-
sulted computers! These patchwork poems are, rather, testimonies of an extra-
ordinary performance of the memory and of an active control over the material,
faculties which-though far from indicating genius-belong 'within the range
of a techne, in the classical meaning of the word.
It is true that St. Hieronymus had condemned the cent os as childish non-
sense long before the publication of these epigrams.122 In spite of this, Bishop
Patricius and the Empress Eudocia, the wife of Theodosius II, undertook the
task of molding the Incarnation and the life of Jesus into the shape of Homer-
centos.123 This work undoubtedly required even greater versatility than those
pagan centos composed by Leo the Philosopher which we have already dis-
cussed: here it was a completely disparate subject that had to be rendered in
Homeric lines. Thus, it is not astonishing that things do not end up nearly
as smoothly as they do in Leo's centos. In fact, non-Homeric elements are
repeatedly inserted, and the style as well as the grammar leave much to be
desired. I should suppose, however, that after a closer study of the manuscripts
a better text could be produced. A. Ludwich, the editor, took no interest in
this aspect; in the Latin preface he clearly expresses his contempt for such
"bungling pieces of work. "124 In my opinion, in this enterprise of the Bishop
and the Empress we must see neither narrow-mindedness nor snobism-the
latter could much better be applied to the work of Leo the Philosopher-rather,
we should understand it as the naive and moving attempt to clothe the history
of salvation, which is of fundament'al importance to every Christian, in that
linguistic garb which was the most venerable to every Greek, namely, the verses
of Homer. It is the same spirit to which numerous Byzantine and Russian
icons owe their more or less precious, though artistically often uninteresting,
metal covers.
To a different order belong the two Byzantine dramatic works of the twelfth
century which were intended to be read; of these the one belongs completely,
the other partially, to the category of the centos in a wider sense.
A third of the 2610 lines of Christus Patiens (XP1O'TOS 1TOoxwv)125 is borrowed
from classical models, mostly from Euripides, the Medea and the Bacchae
having the greatest share in the contribution, some others of his plays following
in this order: Hippolytus, Rhesus, and, far less often, Orestes, Hecuba, and
122 Hieron .. Ep. 53.7: Pue,'ilia sunl haec el circulaloyum ludo similia.
123 Edition of chapters 1-13 and 50 (approximately 1950 lines in all) by A. Ludwich (Leipzig. 1897).
124 Ed. Ludwich. p. 87: nam huius gemris libri. qui !laud pauci ad!luc in bibliolhecis latenl. hodie a
mmine digni habentur. qui accumlius explm'entur .... Emendatiuneulae paucae mihi jere invito (!)
e:reide,'wlt inter seribendum: hunc igitur campum. quem videbam nimis sterilem esse. aliis palientioribus
permitto diligentius eolendum.
120 Ed. J. G. Brambs (Leipzig. 1885); A. Tailier (Paris. 1969).

98
 ON THE IMITATION (MI.MH~I~) OF ANTIQUITY 35

The Trojan Women. About three dozen lines are taken from Aeschylus' Aga-
memnon and Prometheus Vinetus, as well as from Alexandra by Lycophron.
In accordance with the old tradition of Christian humanism and the methods
of quoting discussed above, numerous borrowings from the Old and New
Testaments are inserted between the many hundreds of classical quotations.
The author-whose identity remains unknown-was certainly not a genius,
but he was a well-versed writer and an expert on ancient tragedy as well as on
the Holy Scriptures.
The following example should help to illustrate the nature of these adop-
tions:
1127 T[ TCXVT' ci:Mw; TrEIOLEOV Tois aois My0l5
II Hipp., 1182 TI TCXVT' ci:Mw; TrElOLEOV ITCXTPOS MyOlS
1128 EpyOIS S', oa' VrrE8EI~cxs Eis I-IcxpTVpicxv,
// John 10: 25 Ta Epycx 0: Eyw ITOIW EV T0 OVOl-laTI TOU TTCXTPOS I-I0V, TCXUTa
I-IcxpTVpEi TrEpl EI-IOV
1129 WS fOLIV emav aOI SEATlTOV 8vvaTov.
II Ps. 110: 2 I-IEYMa ,0: epycx KVPIOV, E~Tl[ETT\l-IeVa .is 1T():v-ra TO: SEA~l-IaTa aUTov
1130 nOAAWV ,aIJias EOLiv ci:eATT'TWV 8EOS,
1/ }v1ed., 1415 ITOMWV TaIJicxs ZEVS EV 'OMIJTT~
1131 ITOAM: T' ci:EAITTWS ITOAAclKIS Kpaivel 8EOS,
// Jled., 1416 TTOAAO: T' OeATT'TWS Kpcxivoval SEQi
1132 TO: 8' cxu 80KTlSev-r' OVK E'1'evpe wi TEAOS'
/1 Jied., 1417 Kcxi TO: 80KTlSev-r' OVK E-rEAEaSTl
1133 aV 8' ci:80K~'wv CXUTOS evpolS IJOI ITOpov.
II );ied., 1418 TWV 8' 050K"TWV IT<Jpov Tlvpe Seos·
1134 f. Kcxi I-lilv 65' 'IWO"T\TTOS EV CYTTov5ij ITO 50S
CY'TEiXEI, VEOV TI ITpO:yl-l' laws EXWV cppO:aCXl.
II Rhes., 85 f. Kcxi IJilv 68' AivEcxs wi I-IO:ACX aTTov8ij IT050s
OLeiXEI, VEOV TI ITpO:yIJ' EXWV '1'iAOIS '1'paaCXl.
1136 'ATO:P T08' aMo SCXVI-ICX Kcxi TTCXp' EATTi8cx,
/1 Bacch., 248 map T08' aAAo SaVI-ICX TOV ,EPCXCYKOITOV
1137 I-IVCYTT)V vVxlOV T08E cyvv-rpExov-ra ITWS
II John 19: 39 i'jASev 8E Kcxi NIKOOTlI-IOS, 0 EASWV ITpOS
CXv.o v VVK'TOS TO IT PWTO v.
1138 CYKEVT] '1'epov-rcx ITpoa'1'vo: Tij wS08~.
/1 John 19: 39 '1'EPWV l-liYl-la al-lvpvTlS Kcxi ci:i\6T]S WS Ahpas a.:a,ov.
The alternation of pagan-classical and biblical elements is throughout ob-
served. Where lines from tragedies are adopted, the changing of one word often
suffices to establish the necessary meaning. Thus, for instance, in 1127 TTaTp6S
i\6yolS is changed to Tois aois i\6YOIS, in 1130 Zeus has become 8eos, in 1131
Seol has been turned in to 8eos. In 1134 the name is changed (J osepos instead
of Aineas), in 1135 the unsuitable '1'IAOIS is replaced by iaws. The prose text of
the Gospel of St. John had to be transformed into twelve-syllable lines, which
was probably easier work for a somewhat skilled rhymer than finding corres-
99
 36 HERBERT HUNGER
ponding lines in pagan literature. Lines 1136-38 contain such a passage, joined
to a line from the Bacchae.
In the Katomyomachia by Theodore Prodromus the imitation of antiquity
is a parody of ancient Greek tragedy, as I have tried to demonstrate in my
recently published edition of this work. 126 The structure of the play which
consists of elements that are characteristic of classical tragedy, the messenger
scenes, stichomyths, and rheseis in the manner of the classical models, as well
as an abundance of quotations and linguistic reminiscences confirm this
opinion. The comical effect emerges mainly from the circumstance that actions,
speeches, and behavior typical of gods, heroes, and great men in ancient
tragedy are attributed in the play to small, timid animals, namely mice. The
tragic diction befitting the heroism of human beings produces a comical effect
in the mouths of timidly shivering mice. With regard to all of this, I refer to
my edition and quote here only one characteristic verse (218): KoMv TO VIKO:V
(= Eur., Phoen., 1200)---&AA0: OEIAio 1-1' eXEI: "It is wonderful to be victorious-
but I am a coward."
As far as language is concerned, the imitation of classical tragedy in the
Katomyomachia is found in certain forms of words and in whole phrases;
these, too, are listed in my edition. With the accumulation of such quotations
and borrowings in the second half of the Katomyomachia, the text in some
passages (particularly 11. 240ff. and 323ff.) has almost the character of a cento.
When studying these imitations one discovers in many details the variation
mentioned above. One would search in vain for a personage picking up the
dialogue with a KClAWS eCjlTlO"os (122) in tragedy, where it is always KClAWS EAe~ClS.
The quotation from Hecuba (689), CI"ITIO"T' O:rrIO"TO KOIllO: KOIVO: OepKOI-IClI, has the
changed ending 1-101 AEyEIS. The I-ITlvVO"ol KOKO: of the messenger in the Phoenissae
(1218) has been turned into KOKWV I-ITlllVTTis in the Katomyomachia (315). Thoughts
frequently expressed in tragedy are presented by the Byzantine author in his
own words:
Eur., Alc. 1076 0Ui< fO"T1 TOilS .9oVO\ITClS ES CjlO:Os I-Iot.eiv
Eur., Heracles, 297 Koi Tis .sOVO\ITUJV i'jA.sEV e~ "Aloov rrclAllI;
Kat., 269 ovoels .9ovO\ITOS e~eyEipEI TOV TO:Cj>OV.
When the lady mouse in the Katomyomachia does not immediately see the
second messenger because she is so agitated over the death of her son, a
similar thing happens to her as to Euripides' Electra, who, due to excitement,
does not recognize the messenger:
Eur., EI., 767 a< TOI OE{I-IOTOS ovayvUJO"{ov etxov rrpoO"c::,rrov.
Kat., 275f. EK TOV rrO:.sovs Ij"AUJAO Koi o"vllETpi~"v
Koi TCxS KOpOS i11-1~AVVO TCxS Tc;:,V OI-lI-lCrrUJv.
What is more, the expression K6pOI Tc;:,V Ol-il-lCrrUJv occurs three times in the
Orestes of Euripides (469, 1261, 1319).
In the essay quoted above O. Demus has touched upon the interchange-
ableness of the iconographic schemata (e.g., Ascension of Christ-Forty Martyrs
of Sebaste) and of the elements of different pictures. Art historians have
,•• Hunger, Der byzantinisc~e Katz-Mause-Krieg, 51-65.
100
 ON THE IMITATION (MIMH~I~) OF ANTIQUITY 37
noticed transpositions of groups of figures, of individual figures, and even of
individual gestures.127 Mutatis mutandis, this seems to apply also to some pro-
ducts of Byzantine literature in the high language. If we tum from a study of
the Katomyomachia to the contemporary novel we will find that passages
such as the "leader's speech" of Kreillos, delivered at the mobilization of the
mice, the lamentation of the lady mouse over her dead son, or one of the
messengers' reports could easily be taken out of their context and inserted in
the novel of Prodromus (Rhodanthe and Dosicles = RD)12S or in that of Nicetas
Eugenianus (Drosilla and Charicles = DC).129 If one takes a closer look at these
two novels, one finds that both contain elements that indeed fit the respective
context but could just as well stand for themselves or occur in some other
context. In RD this is the case with the speeches of the military commanders,
several letters, and the Helios hymn of Satyrion, in DC it is the case with a con-
siderable number of "lyrical" insertions; among the latter are lengthy laments,
confessions of love, love letters, and songs, as well as all sorts of 8<cppacrlS (de-
scriptive epic). These passages are of particular interest because of the kind of
imitation employed by the author. The letters (in twelve-syllable verses), for
instance, are little masterpieces of Byzantine epistolography. One lament and
two songs are written in dactylic hexameters; occasionally a refrain occurs in
which one soon recognizes Theocritus as the model.
The conscious mannerism of the poet, who varies his use of imitation, be-
comes obvious from three passages imitating epigrams that are contained in
the Anthologia Palatina. A close correspondence exists between A.P., V.2S3
and DC, III.163-72. The train of thought is the same in both cases: the girl
bashfully looks to the ground, yet is fumbling with her belt; Kypris is not in
accord with bashfulness; the girl should nod in assent. Eugenianus used ten
twelve-syllable lines in place of the two distichs of the model and thus his
version turned out to be somewhat longer. The decisive catchwords correspond:
K6:Tw veVovercx ~ VEUEIS K6:TW; lW1Ii)V ... 6:KpOAVlEis ~ lWVT]V '" 6:KpOAVlEis; cx!15ws
voercpl TrEAEI Tfis KVirplBos ~ OUK olBEV cxlBw KUTrpIS; finally, a variation: veVllCXTI TI]v
ncxcpiT]v BEi~ov VTrEPXOIlEVT] ~ fllo! XCXpilOV Kav TO veVila erov IlOVOV. The extensions by
Eugenianus are unessential to the train of thought.
The imitation of A .P., V.259, in DC, III.243-54, is done slightly differently.
Here, not only the main thoughts but also the vocabulary of the two passages
largely correspond. In this case four distichs are transformed into eight twelve-
syllable lines; consequently, no room for extensions was left: OIJIJCXTO eJEV f3apuSoverl
~ f3CXPUVETCXI erov OIJIJCX; TrOSOV TrVEioVTCX ,..... TOV Tr6S0V YEIJOV; wXpos '" WXPOTT]S;
TrcxvwxiDerlv 61JIA~erCXera TrcxAaicrTpOiS '" TrcxAcxicrTPCXIS WlJiAT)erCXS TrCXVvUx0IS; OAf30V '" OAf3IOS;
T~KEI SeplJOS epws '" "Epws cpMyel; eiT)S e!S fIJi: TTlKOIJE1Ii) ,..., EiT)S Trp6S 1)IJO:S 1J00Mov
EKKEKCXVIJEVT]. In spite of this remarkable correspondence, there are also unexpected
variations: EOIKCXS tnrvwv EvBei]s ElvOI, yUVOI has only a distant similarity with
the corresponding passage in the model used; os erE TrEpITrMyBT]V EXE Tr~xeerl is
Demus, op. cit., 95.
1.."7
128 Ed. R. Hercher, in Erotici Scriptores Graeci, II (Leipzig, 1859), 289-434.
129 Ibid., 437-552.

101
 38 HERBERT HUNGER
changed to 6 XElpaS aVTos {lll3aAc:,v aais c:;,;>.fvaIS. Finally, the whole passage is
extended at the end by four lines which are dedicated to the Telephus-motif
(6 Tpc:;,O'as KCXi IOO"ETal; here: vai, 1TaVO'OV, c:;,s e-rpcuaas, T]1TaTO) 1TOVOVS).
The third example of imitation, A.P., V.273 ""'" DC, III.174-96, is limited
to the basic thought (old age takes revenge on coy beauty) and to a few
lexical correspondences: Ocpfi1<E X6:ptv - 6:cpfiKE ... XOplV; llaLOS VrrE1<AlvS" ""'" llacnO)
... \nrE1<A{S,,; 1TECYOV oCPPVEs -- 1TFrrrCU1<EV ocppvs; cpSEyllaTl Y1lPaAE't' ,..., Y1lPaAEOV TO cpSEylla;
1TOA1T]V (last line but one)- 1ToAI1';i (at the beginning); aol3apais (last line),..., aol3apos
(fourili line). The quantity differs (four distichs,...., twenty-three twelve-syllable
verses), the sequence is partly altered: the concluding statement of the epigram
is prefixed as a motto in Eugenianus' passage. A number of lines are in-
tentionally greatly altered. The last third of Eugenianus' passage is skilfully
enlivened by a seeming stichomyth of asyndetic half-lines-rhetorical questions
put to the girl and ironical replies immediately following; this passage has no
counterpart in the model used.
One could extend this survey of the "insertions" in DC, and soon one would
have quite a number of structural elements which could at will be inserted in
other novels. Here we are confronted with a peculiarity of Byzantine imitation
of antiquity which has its striking parallels in the visual arts. As a particularly
instructive object the famous ivory casket of VeroH should be mentioned here,
to which Erika Simon dedicated a thorough and, I believe, pioneering inter-
pretation a few years ago. lao Mrs. Simon succeeded in finding convincing ex-
planations for all the individual scenes on the casket by using the epic poem
of Nonnus as a basis of comparison. Her explanations proved some of the
former interpretations to be completely immaterial. In the group of men
hurling stones, declared by Weitzmann to be a mechanical adoption from the
Roll of Joshua, which turned out instead to signify the assault of Typhon upon
heaven, we have again an example of the interchangeability and ambiguity
of such groups of figures. Similar considerations are relevant in the case of
the "quotations" from Euripides-which are of particular interest to us-
namely the pairs Phaedra-Hippolytus and Bellerophon-Stheneboea. With good
reason E. Simon does not regard these as illustrations of plays but as symboli-
cally condensed pictures which are in some way or other related to the pur-
pose of the object (nuptial casket). The occasionally Christian interpretation
of pagan-mythological scenes (boy extracting thorn and Phaedra entangled
in sin) also fit very well into the frame of our argumentation. As in the case
of the casket of Veroli, where-according to this new interpretation-a number
of parts which might have been used in different ways were composed to form
a new unity (here a nuptial casket), so we may understand several works of
middle Byzantine literature, e.g., the novels, to be similarly constructed.
In conclusion it can be said that, however manifold and varied in respect to
quantity and intensity the imitation of antiquity may have been, it certainly be-
longed to the essential features of Byzantine literary works in the high language.
130 E. Simon, "Nonnos und das Elfenbein-I~astchen aus Yeroli," Jahrbuch d. Deutschen Arclzaologi-
schen Instituts, 79 (1964), 279-336.
102
 THE LANGUAGE OF BYZANTINE LITERATURE

by
Robert Browning
Birkbeck College, University of London

Much of Byzantine literature appears to the superficial observer to be written in a changeless linguistic
foml. There are no striking differences between the Greek of Procopius in the sixth century and that of
Critobulus in the fifteenth. And where a line of dcvt'lopmcnt c~n be traced. it not infrequently seems to
go into reverse. Photius in the late ninth century writes more 'classical' Greek than Theophanes at the
beginning of the century. Psellus in the eleventh century is more 'classical" than Constantine Porphyro-
genitus in the tenth, and Anna Comnena and Nicetas Choniates in the twelfth outdo Psellus in certain
classical fealllres l

This impression is due in part to the fact that very different kinds of Greek were being used for literary
purposes at one and the same time. The point is obvious in. say. the fourteenth century, when the skil-
ful imitation of ancient Greek which enabled the work of Thomas Magister to pass for that of Aelius
AristideSZ coexisted with the use of many of the linguistic features of Modern Greek in the ballads of
chivalry. But it is equally true of other periods. The Chronicle of John Malalas is contemporary with the
History of Procopills in the sixth century. The De Adminislralldo Imperio of Constantine Porphyrogeni-
tus is contemporary with the hagiographic corpus of Symeon Metaphraste's in the tenth.

Another source of confusion is the difficulty of distinguishing between language and style in Byzantine
literature. Eustathius-and others before and after him-classify under the head of 'Attieism'. a stylistic
category. fcatures which ~re linguistic, such as the use of the dual number or the accusative absolute. 3
In fact the distinction is often an artificial one. The kind of variations described are not freely optional,
nor is the content of the message unaffected by the selection made. The concept of 'register' much used
in modern descriptive and socio-linguistics provides a better framework for the analysis of such variations
than those of language or style.4

Byzantine writers were themselves highly conscious of distinctions in linguistic usage, and often refer to
them. and to the supposed reasons for the choice of this or that linguistic form, both in their criticism of
other writers and in their own programmatic statements. The following is only a representative selection
of such references, which could easily be multiplied.

Constantine Porphyrogenitus, addressing his sonRomanos in the proem of the De Administralldo Imperio,
says, "If in setting out my subject I have followed the plain and beaten track of speech and, so to say,
idly running and simple prose, do not wonder at that, my son. For I have not been studious to make a

103
 104 Robert Browning

display of fine writing or of an Atticising style, swollen with the sublime and lofty, but rather have been
eager by means of everyday and conversational narrative to teach you those things of which I think you
should not be ignorant, and which may without difficulty provide that intelligence and prudence which
are the fruit of long expericnce."s The emperor prefaces his De Caerimoniis with a similar apology: "So
that what I write may be clear and easy to distinguish, I have employed a conversational and simple style,
using the same words, and the appellations lon'g attached to each object and uttered in regard to il.,,6
Constantine's father Leo VI had made a similar apology for the style of his Tactica. 7 The principal-un-
acknowledged-source of Leo's work, the seventh-century Strategicon of Maurice, opens with a similar
disclaimer: "As I said, I take no thought for well-chosen words or ornaments of specch, since it is not a
sacred work; I have rather had regard of the facts and brevity. Therefore I frequently use Roman and
other words in common military use so that my rcaders may understand me clearly."8 Theophanes Non-
nus, who compiled a medical encyclopaedia probably at the behest of Constantine Porphyrogenitus in
the tenth century declares: "No one who has the slightest ability to reason will censure me for using in
the present work nouns and verbs taken from the market place and the cross roads. For it is not through
ignorance ... that I have employed from time to time barbarous and corrupt expressions.,,9 It will be
noted that the works thus far discussed are all in some sense technical manuals. Cecaumenus' counsels
to his son are in essence a manual on how to succeed-or at least to survive-in the eleventh-century
Byzantine world. And like the De Admillistrando Imperio they are not for publication. It is hardly sur-
prising that they include a long apology for their language and style: "I have no part in letters; for I
never studied Hellenic culture to get well-turned speech and learn eloquence. I know that some will cen-
sure me and cavil at my ignorance. But I have not composed this as a work of literature for others, but
for you and your brothers, my sons, flesh of my flesh, whom God has given to me. I did not compose it
with ornate words and mere sophisticated fables which have nothing good in them: rather I have set out
what I have done and had done to me, seen and learned, real events which happen every day. My style
may be unprofessional, but if you attend soundly to what I say you will find it most true."tO A new
motive is here introduced, the writer's own lack of education.

Though most religious literature was couched in classicising form, writers of saints' lives often drewatten-
tion to their use of non-classical language and style, motivating their choice either by their own lack of
education, by their desire to be intelligible to the uneducated, or by the lofty nature of their theme,
which needed no meretricious attraction of style: as Synesius observed, "the Holy Spirit despises literary
hair splitting."I! Thus Mark the Deacon in his Life of Porphyry of Gaza writes, "I shall expound his
praise in unadorned language; for it is not fine style which lends adornment to the life of such men, but
the nobility of their deeds which enhances the style."12 And Leontius of Neapolis in his Life of John
the A imsgiver observes that he has written in his pedestrian and unadorned and humble style so that even
the unlearned and the illiterate may profit from what he recounts.13 Similar programmatic statements
are found in many other hagiographical texts up to and including the ninth century. Thereafter they are
rare, and generally frowned upon by the ecclesiastical authorities. Thus a Life of SI. Paraskeve the
Younger, a Thracian village saint of the tenth or eleventh century, was ordered to be destroyed by Patri-
arch Nicolaus Muzalon (1147-1151) on the grounds that it was written l61WTlKW~ 7rapa TWO~ XWpITOV}4
And, as will be seen, a wholesale programme of replacement of saints' lives which were linguistically or
stylistically unsatisfactory was undertaken in the second half of the tenth century by Symeon the Meta-
phrast.

In the twelfth century men of letters were sharply aware of the choice of linguistic register open to them.
One of the speeches of Nicetas Choniates, delivered at Nicaea after the capture of Constantinople by the
Latins has a note appended to the title: €~Eo6eT/ OE oa<!>T/vda~ 7rAilp€~ TO 7rap(JV 7rpoa<!>wvT/paliui nlV TWV
dKpoaTWv da8eVElI1.v. 15 Whether this observation was made by Nicetas himself or by the editor who
compiled a corpus of his speeches and letters shortly after his death is immaterial.

104
 The Language of Byzantine Literature 105

The numerous normative lexica and grammars composed in the Byzantine world are themselves a testi-
mony to the consciousness of writers and readers that there was more than one way of saying things,
though they did not all enjoy equal esteem. The same sense of possible choice emerges from the obser-
vations of men like Eustathius and Michael Choniates on the difference between learned and popular
spcech. 16

It is tempting but misleading to draw a parallel between the linguistic situation in the Byzantine world
and that in Greece since 182 117 Byzantine diglossia is essentially a literary phenomenon, while that of
modern Greece is political and educational. It never occurred to a Byzantine to demand that the public
business of the state should be transacted in popular speech, or that his children should not be taught to
read classical Greek authors. Demotic translations or paraphrases of parts of the Bible are known from
the late Byzantine period, but they did not occasion riots in the streets of Constantinople, like those
caused in the streets of Athens by the publication of Pallis' demotic New Testament in 1902. Within the
limits imposed by the occasion, the literary genre, the expected audience, and his own education a Byzan-
tine author could choose between different linguistic patterns. His readers or hearers had no such choice.
This is not to say that Byzantine diglossia-and that of the post-Byzantine world from 1453 to 1821-did
not contribute to the creation of modern Greek diglossia. It certainly did. But the one was not a mere
foreshadowing of the other. It had its own raison d'etre and its own rules.

The study of the language of Byzantine literature is hindered by the rarity of descriptive grammars or
lexica to particular authors or texts. Some excellent works of this kind have appeared in recent years. IS
But in general all that is available are partial and impressionistic observations on points of grammar and
the very incomplete and unreliable indices verborum of the Bonn Corpus. There is no lexicon of Byzan-
tine Greek of anything like the scale and excellence of Liddell-Scott-Jones or the Lexicon of Patristic
Greek. The student is thrown back on Du Cange-now three centuries old, incomplete, and using a system
of references virtually impossible to check-or Sophokles-admirable in his time, but now nearly a century
old, and specifically excluding the period after 1100. Kriaras' admirable A€~lKO rile; Mwatwl'tKiI,'EAAl1-
I'tKiic; Ar/l.l(..)oou, fpallllaTEUIt excludes texts in all varieties of the learned language, which constitute
the bulk of Byzantine literature, and in three volumes has got only half way through Alpha. Successive
International Congresses of Byzantine Studies have urged the necessity of a new lexicon of Byzantine
Greek. But in the absence of funding the various projects have petered out.19

It might be urged that the .time is not yet ripe for the production of a definitive lexicon, as we lack a
sound critical text of many Byzantine writers, including those as important linguistically asJohn Moschus
and the continuators of Theophanes. Here again there have been notable advances in recent years, but
much more remains to be done.

If I have appeared to labour these rather negative points, it was in the hope of making it clear how dis-
advantaged the student of Byzantine literature is, not only compared with the classical Hellenist, but also
compared with the student of western European medieval literatures. It follows that much of this paper
must be provisional in character and subject to revision in the light of fuller and more reliable information.

II

In order to understand Byzantine attitudes to the use of language in literature we must examine those of
the Church Fathers, whose prestige as models to be imitated was immense throughout the Byzantine
period. And we cannot understand the choices open to the Fathers without glancing back at the use of
Greek as a literary language in the Roman empire?O

105
106 Robert Browning

The Hellenistic world had seen the spread and general use for prose of a common language-the KOWtj
OLdA€KTO~-whose basis was Attic but which diverged in many respects in morphology, in syntax and in
vocabulary from the language used by the Athenian prose writers of the fifth and fourth centuries B.c.
The Koine was merely the literary version of the Greek spoken, at any rate by the urban upper classes,
in the eities from Mesopotamia to Gaul. Like all living languages it was in continuous process of develop-
ment and change, and was not constrained by the pressure to imitate a pre-existing model, though writers
naturally imitated one another. In particular its vocabulary was not 'closed' but was constantly being
augmented by neologisms and loan-words. It had various registers, the use of which depended in part
upon the subject of the discourse, and in part upon the competence of the speaker or writer:21

About the end of the first century B.C. a classicizing tendency set in. Beginning with authoritative writers
on grammar and rhetoric, such as Donysius of Halicarnassus, it spread with astounding rapidity through
the schools and soon coloured the literary education of the upper classes throughout the Roman empire
and beyond its eastern frontier. What the proponents of classicism called for was a return to the manner
of the great Attic writers of the fifth and fourth centuries B.C., who became models to be imitated. The
origins of this Attieist movement are probably to be sought at many different levels in the mentality of
the Greek-speaking world, and it is not part of the present investigation to analyse them in detaiIP

The earliest Atticists were concerned primarily with style, and only secondarily with language. They
wished to replace the often loose and rambling pattern of Hellenistic prose by something more taut. And
they reacted also against the tinselly, over-ornate style of oratory favoured by many distinguished rheto-
ricians of the age, and advised a return to the manner of Lysias and Demosthenes, who made their point
more effectively with less obtrusive use of figures of speech. But features ofstyle slip through the fmgers,
while features of language can be listed and counted. By the second century A.D. what was being taught
in schools was a rigourous linguistic purism, both positive and negative in character. Words and construc-
tions used by the Athenian classical writers were to be esteemed and imitated. Those in current use but
not to be found in the models were to be rejected. To this end lists of linguistic prescriptions were drawn
up, in particular the Atticist lexica of Aelius Dionysius, Pausanias, Moeris and Phrynichus, which listed
both the current words and expressions to be avoided and what were believed rightly or wrongly to be
their Attic equivalents. An extreme adherent of this movement is caricatured by Athenaeus under the
nickname KELToIiKfLT04;; he refuses to eat any dish at a banquet unless its name is attested (KeiTcu) in an
Attic text.

KetTOVKELT04; is a figure of fun not simply because he pushes things to an absurd extreme, but because he
lets his Atticism spill over from literature to life. It is important to remember that the Atticists, unlike
linguistic purists today, were concerned only with certain uses of language-in general tenns with public
speaking and belles lettres. It was no part of their aim to make people speak Attic on the ordinary occa-
sions of everyday life, or to use it to write laws, private letters, technical manuals and the like. Atticism
was a literary movement, which affected in some degree all the Greek literature of the Roman empire.
There were, of course, degrees of purism. Aelius Aristides observes the positive and negative precepts of
the Attieists much more strictly than Plutarch. Lucian satirizes the absurdities of the rigorous Attieists.
But there is a family resemblance between ali these writers.

The impetus of Atticism carried on into late antiquity. Libanius, Themistius, Eunapius, Zosimus, Aeneas
of Gaza, Procopous of Gaza, Choricius, Procopius of Caesarea, are all classicizers, of varying degrees of
rigour. Their work has not been studied as thoroughly as that of their predecessors. libanius seems to
be stricter in following his models than Themistius, but the position of some of the others remains to be
detennined. At first classicism depended on the individual choice. and effort of the writer, his knowledge
and skill. But in the course of time, as it dominated not only education but most public and fonnal use
of language, it acquired by a process of choice and rejection a kind of objective reality of its own, which

106
 The Language of Byzantine Literature 107

detennined the performance of writers. The writer no longer sought out features in his ancient models
to imitate. He confonned rather to a kind of conceptual classicism, which contained features to be found
neither in Koine nor in the usage of Attic writers, such as use of the separative genitive with verbs.

This is the reality underlying Higgins' concept of Standard Late Greek 23

At the same time technical writers and others communicating at a less fonnal level continued to write in
Koine, a Koine which developed and changed slowly, but which was not constrained by the need to imi-
tate any authoritative model, real or imagined, and which retained its links with the various registers of
the spoken language 24 A third variety of language can also be identified in late antiquity, the official
language used for legal documents, governmental business etc. Basically written Koine it was characterised
by an abundance of cliches and flosculi drawn from poetry, from the language of literature, from techni-
cal language of various kinds. In some ways it recalls the language of journalism today.25 It was probably
the only kind of stylised language with which the man in the street, outside of the limited circle of the
rhetorically educated and the members of the learned professions, had regular contact. In the eyes of the
ordinary man it had the prestige of fine language.!t must not be imagined that these different levels of
written language in late antiquity were always kept distinct. Interference between the different levels
was frequent and continuous. Even a self-conscious stylist like Libanius occasionally uses words and con-
structions not favoured by the classicizing tongue. But ideally, and quite largely in practice, they were
distinct.

The Christian community at first made use of the written Koine, often at a level much closer to infonnal
speech than that used by writers like Didorus Siculus. Its members on the whole did not belong to the
educated elite, and the whole structure of social values which were served by classicizing literary Greek
was alien to them. Their own sacred books were composed in Hellenistic Koine, which therefore enjoyed
prestige among them. And their writings had the character of in-group communication quite unlike the
public performance of the rhetorician and the man of letters.

However. when the Christians undertook missionary activity among the upper classes, they had perforce
to lise to the best of their ability the level of language which would win the attention and respect of their
readers and hearers. So apologetic writings tend to approximate in language to the c1assicising tongue of
literature. The point is strikingly brought out by the distinction in language between Clement of Alex-
andria (c. 215) and his pupil Origen (185-253/4). Clement, who addresses himself largely to the educated
pagan world. writes in classicising Atticist Greek 26 Origen, who engages mainly in doctrinal polemic
within the Christian church, writes in technical Koine. comparable with the language used by medical or
mathematical writers. 27

From the middle of the fourth century onwards Christianity made rapid headway among the urban popu-
lation of the Greek world. From being the reli~~on of an inward·turned minority it became that of the
dominant group. The problem of communication within the church changed in scalc and nature. The
new leaders of the church, men like Gregory of Nazianzus, Basil and John Chrysostom were faced with a
choice of the linguistic medium to be lIsed for their pastoral, dogmatic and polemical works and for
their preaching 28 One might have expected that they would decide on Koine in one of its more literary
fomls. It was certainly morc easily understood than Atticist Greek. It was not linked, as was the classi-
cising language, with the defence of paganism and traditional culture. It was the language used for three
centuries by the Christians in communicating with one another: Athanasius (295-373) still wrote his
copious theological and polemical works in Koine. But in fact the late fourth century Fathers chose the
classicising conventional literary language 29 It was appropriate for texts addressed to an audience of the
social elite. It affinned and enhanced the dignity of the occasion-which might well be an address to a
vast congregation in one of the great basilicas which the now rich Christian church was building. Above

107
 108 Robert Browning

all, the new church leaders belonged themselves to the upper class of urban landowners and had received
a long and rigorous tr<lining in grammar and rhetoric. John Chrysostom was a pupil of the great pagan
rhetorician Libanius; Gregory of Nazianzus and Basil had studied in the schools of Athens. Their minds
were formed by a traditional education. It was probably psychologically impossible for such men, had
it ever occurred to them, to make a public utterance on matters of importance in what they had been
taught to regard as an undignified and even contemptible linguistic register. They had to give of their
best, and their best meant the classicizing Hochsprache. A special place, however, had to be made for
those Koine words and phrases which had been sanctified by use in the Septuagint or the New Testament.
They were given full right of citizenship in the language of the Fathers.

The great age of the Fathers at the end of the fourth century and the beginning of the fifth set a pattern
which imposed itself on much of Byzantine literature for the next thousand years. Henceforth the
church spoke in classicizing Atticist Greek. 3D The Fathers themselves became classics, studied, comment-
ed on and imitated by succeeding generations. They took their place aJongside of Lysias and Demosthe-
nes in the handbooks of rhetoric and brought with them the lustre of their own immense prestige.31

Two qualifications need to be made to this general formulation. The first concerns preaching. While
classicizing Greek was the language of communication of the church at the top, at the grass roots it en-
couraged predication in the vernacular tongue of the congregation. John Chrysostom, that model of
classical elegance and force, encouraged his priests to preach in Syriac to the peasants of the Antioch
region and in Gothic to the mercenary troops garrisoning Constantinople. Coptic, Armenian, Georgian
and other languages were also used in preaching. Why then not spoken Greek? The answer is that a
good deal of preaching in Koine, and often in fairly low level Koine almost certainly went on. Not all
clergymen were pupils of Libanius. But it was overshadowed by the use of the classicizing language, and
regarded as something second-rate. And it was not given the permanence and dignity of writing.

The second qualification concerns edifying accounts of the lives of holy men. These were often, but not
always, closely linked with the place where they had lived and died and where their cult was practised.
They were a development and continuation of the accounts of the deaths of its martyrs, which each local
Christian community piously recorded, often embodying the actual words of the official court report,
which would not be composed in classicizing literary Greek. And they were addressed to readers and
hearers for whom the charismatic personality of their own saint was what mattered, and who had little
appreciation of the finesse of the literary Hochsprache. So from the early fifth century we find hagio-
graphical texts surviving written in an approximation to spoken Koine-it could only be an approxima-
tion, since in learning to read and write a man was exposed to the classical tradition. Some of them ex-
pressly reject all attempts at fine language. An example is the Life of Porphyry of Gaza by Mark the
Deacon, which has been already mentioned. Characteristic of this level of language is a simple, para-
tactic sentence structure, infrequent use of the optative, uncertainty in the use of the dative, confusion
of aorist and perfect, frequent Latin loan-words.

So the heritage left to the Byzantine world by the age of the Fathers was a complex one linguistically.
The Hochsprache was firmly established as the proper medium for all important or dignified communica-
tion. But at the same time an undercurrent of rejection of the literary tongue was associated with certain
powerful manifestations of popular piety. And technical writing of aJl kinds, from law to medicine,
continued to make use of a literary stylization of the common language, with no pretentions to the posi-
tive and negative choices which marked the literary tongue.

These preliminary points established, we can now tum to the language of Byzantine literature proper.
It will be convenient, and will correspond to real distinctions, to divide the thousand years from the fifth
to the fifteenth century into the following periods:

108
 The Language of Byzantine Literature 109

Early Byzantine Fifth to seventh century

Dark Age End of seventh to early ninth century
Macedonian Renaissance Late ninth to early eleventh century
Age of the Comneni Eleventh and twelfth centuries
Latc Byzantine Period Thirteenth to fifteenth centuries

The boundaries between periods are not to be thought of as sharply drawn. And the names given to the
periods are purely conventional and embody no value judgements.

III

The death of Cyril of Alexandria in 444 and the Council of Chalcedon in 451 mark the end of an age in
cultural history. The take-over and adaptation of Classical culture by the Christian church was complete,
the amalgam of Byzantine culture formed. The two centuries which follow have as their high point the
long and distinguished reign of Justinian (527-568) and end with the disastrous reign of Heraclius (610-
641), which saw the final loss of Egypt, Syria and Palestine to the Arabs, while the Slavs and Avars re-
moved much of the Balkans from Byzantine control. They were centuries during which the traditional
educational system slowly broke down. The variety and quantity of literary activity, very great in the
days of Justinian, seem to have been strikingly reduced by the end of the period.

Throughout the period the three types of literary language which were in use in the age of the Fathers
remained distinct, but not, of course, without internal change. Poetry raises problems of its own and is
best left to the end of the discussion. In prose classicizing 'Atticist' Greek, literary Koine and popular
Koine each has its own sphere. Historians like Procopius, Agathias and Theophylact Simocatta continue
to write in the traditional Hochsprache, with its strict rules for selection and avoidance.32 Conscious
that he is working in a tradition which goes back to Herodotus and Thucydides and educated at Caesarea
by men whose grandfathers might have known Libanius, Procopius apologizes for the introduction of
words not to be found in his models, distancing himself from the language of his own age by a 'so-called'
or 'as they say'. This means, among other things, that all references to Christian institutions and con-
cepts are made with an air of disinterested detachment, a fact which has led some scholars to suppose
that the writer was pagan in sympathy.33 The argument is invalid, as Procopius treats the technical terms
of late Roman administration in exactly the same way. His vocabulary contains fairly few Hellenistic
words and expressions; his syntax is largely that of the Atticising writers of the second century and
later, even to their 'incorrect' (from the point of view of fifth/fourth century Athenian usage) use of the
optative in phrases like tot n~ t!~ TO dKpcl3f~ 'TTll' j3a.olX€W.l' /iUJ.oKotroiTO, tral/iuiv Twa T~V Kupov apx~l' oliJ-
Of Tal dl'CU. (Aed. 1.1.15). Procopius clearly thought of himself as a second Thucydides. This explains
his regular use of f~ for.el~, and probably his use of-oo-rather than the Attic-and Atticist--TT- (-YAwooa
not -yAwTTa). But his sentence-structure is far from Thucydidean complexity and variety; one or more
participial phrases followed by a main verb, followed in its turn by a final, consecutive or relative clause
is the most frequent pattern of sentence in Procopius. It is essentially the Hellenistic sentence pattern
most clearly exemplified by Zosimus. He uses the connective particles correctly-one of the hallmarks
of the classicizing language-and on the whole handles the traditional Hochsprache easily and flexibly.
He even observes when a word is used in a different sense in classicizing Greek and in the official usage
of his own time. Thus PT/TpUtrOAI~ in Herodotus and Thucydides, and hence in the language of the Attio
cists, means the mother-city of the colony, while in the official language of the late empire it means the
chief city of a province. So Procopius writes ft ou o~ Kal' ei~ PT/TpotrciXEW~ aUw/Ja :qXflel'· ovrw -yap
trUA'" T~l' trPWTT/l' TOV €eVO~ KaAOVOt 'Pwpawt (Aed. 5.4.18).

109
 110 Robert Browning

Agathias, writing a generation later than Procopius, and like Procopius a lawyer with a classical education
in grammar and rhetoric, sets out to imitate his predecessor.34 But the result is a use of language distinct
from that of Procopius. His sentence structure is simple, generally a string of participles preceding a
single main verb or two or more joined by Kal'. His vocabulary includes many poetic and other words
not admissible by the strict Atticists. words whose only common feature is their absence from the living
speech of the age. Examples are li1ropE-rXw1Hippocratic corpus),dvE,),po!-,al(Oppian, Quintus of Smyrna).
€,),Kpip.rrTw (Homer, Herod., tragedy, poetry), oUJ.nATlKTllo!-,al (LXX), civalC:),),PTlOl( (lin. A€')'.: iwalw,),Pfw
is attested in Nonnus). bpoo6vPTI (Ionic form). OVVclPOlO (confusion of aipw and apvv!-'tl. His grasp of
classicizing syntax is also uncertain. Nominative absolute participles, of which he has a number. are not
unknown in the writers of the Second Sophistic. But a dative absolute (32.28 K) is quite another matter.
Agathias not infrequently uses the future optative with liv in potential sentences. he uses the perfect as a
narrative tense, equivalent to the aorist. he uses the future indicative with liv, he confuses €i( + accusative
with EV + dative, he occasionally uses EaVTiw of the first person, and info + dative of the agent. All this
suggests that in spite of his education and his self-conscious artistry he had a less firm grasp of the limits
of classicism than had Procopius. It is also noteworthy that unlike Procopius he favours accentual
c1ausulae in which the last two stressed syllables are separated by an even number of unstressed sylla-
bles. To leave the sphere of language for a moment, Agathias uses far more metaphors than Procopius.
Some of his metaphoric usages amount to scmantic changes.

Theophylact Simocatta wrote the history of his own times about 630. His work was intended as a con-
tinuation of that of Menander Protector (now surviving only in fragments). who himself continued the
History of Agathias 35 His use of language shows a further movement in the direction begun by Agathias.
His vocabulary includes poetic words of every kind, particularly from Homer, and a great many new for-
mations. Almost any word attested in literature or formable according to the rules of derivation, was for
him classical. The only lexical elements he tried-not always successfully-to exclude were those current
in the spoken tongue or the official language of his time. And he goes much further than Agathias in the
use of metaphor. His sentence structure remains fairly simple, very often organized round an antithesis.
but his repetitious piling up of synonyms gjves his work a slow pace and an ornate air. A hasty reading
of a few pages (BK 3 ch. 15-16) revealed the following noteworthy words: EvM!-'vxo( (Tragedy, Nonnus,
KaTaAllTaWW (Hesychius), ilOXclAAW (Homer and later poetry), oTpan'apxo( (Xenophon, as a Laconian
technical term), EVOaWlA€VOllal (Heliodorus, Menander Protector), lirrOOKATJVal (Aristophanes. Menan-
der. Lucian, later poetry), o.rro<l>ol70< , li"arrT1A€VTW~ (') in the typical phrase aKa1fT/AelJTW< aVT<iJ TO TT!<
E~OVOW.< OiaKlI€TO OKG.<I>O~, avu!-'<I>io.oTO( (? ), VOO'<1 oa')'Tlvfvfin'c;. Like Agathias his grasp of the syntax
of the Hochsprache is uncertain. He confuses EV + dat. and fi, + acc., and uses liv + indic. in generic
relative clause. And like Agathias he affects c1ausulae based on stress accent.

What we can see in the three generations from Procopius to Theophylact is a certain loosening of the ri-
gaur of the classicizing literary language, not so much in the direction of accommodation to the spoken
language or the Kaine of technical and official writing, though this does occur sporadically, but rather
in that of a breakdown of the distinction between rhetorical prose and poetry36 Any strange word will
do. The rhetorician takes over the vocabulary and the figures of thought that were traditionally studied
by the grammarian.

Meanwhile writers who did not aspire to produce fine literature continued to make use of literary Kaine.
with its ready acceptance of new technical terms, its syntax which eschewed Attic forms. the dual. most
uses of the optative. and much of the fussiness and coy allusiveness of the classicizing language. and its
tendency to longwindedness and formless composition. Such is the language of Justinian's Greek legal
enactments. of the documents of the Council of Constantinople of 553. of the treatise nEpl1fapa06twv
!-'TlxaVT/J1G.TwV of Anthemius of Tralles. architect of Santa Sophia, of the medical handbook of his
brother Alexander of Tralles, of that of the seventh-century doctor Paul of Aegina. of the military

110
 The Language of Byzantine Literature III

manual of Maurice, of the antiquarian works of the retired civil servant John the Lydian, of the early
seventh century church history of Evagrius (for ecclesiastical history had since Eusebius belonged to the
realm of technical writing rather than belles lett res), and of the anonymous world chronicle of the mid
seventh century known as the Paschal Chronicle. These writers differ in the degree to which they admit
elements of the spoken language, in which radical changes were taking place. John the Lydian and the
learned brothers from Tralles carefully avoid vUlgarisms. They were men of high culture, writing for
their peers. The author of the Paschal Chronicle was probably a monk, writing for his fellow monks,
and neither author nor readers had the squeamishness of the highbrow. So he occasionally-but rarely-
admits forms betraying the breakdown of the old pattern of noun declension. like T17V otilfVoova. TOU,
aAwva" rTj, rOP'l'OVT/, and new verb forms like 7ruiOaL, 1i!J.TJV, €vu<ovv. It is the rarity of such tell-tale forms
that makes us classify the Paschal Chronicle as a text in literary Koine. There is a grey area here in which
firm lines cannot easily be drawn.

The tradition of literature in popular Koine, embodying many of the features of the developing spoken
language, continued to be maintained during the Early Byzantine period. This linguistic form was con-
firmed to works of popular religious edification, probably read mainly by monks. And it was provincial
rather than metropolitan in origin. The content and the audience determined its use, not the linguistic
competence of the writers. These were often men who from their position in life must have been
familiar with and competent in the Atticising Hochsprache or literary Koine. In the prologues of their
works they often affect a different linguistic register from that in which the body of the text is com-
posed. Thus John Moschus composes the prologue of his Spiritual Meadow in a fairly polished literary
Koine. The opening chapters of Leontius of Neapolis' Life of St. John the Almsgiver are linguistically
distinct from the main body of the text. And in any case Leontius also wrote in literary Koine. What
distinguishes these works from those in somewhat careless literary Koine is the frequency with which
features of the spoken language occur and the number of such features occurring. The quantitative dif-
ferences add up to a distinction of quality. To take a single example, the following traits of spoken
Greek-which are all to be found in non-literary papyri of the period-occur in a few pages of the Life
of St. John the Almsgiver:

Nominative plural of first declension nouns in -E~.

Accusative singular of third declension nouns in -avo

Neuter nouns in .w· from ·WI'.

Confusion of weak and strong aorist, e.g. fOK€7raOE"

Omission or misplacement of the augment, e.g. fnofiKvVTo, iiVOL(EV, fKaTEoEXETO.
Confusion of -dw and -EW verbs, e.g. Oa7141'OVVTal, 7rAavovIlE()a.
Replacement of -ow by -"'I'W e.g. ¢oPT"'"w.
Metaplasis of -II' verbs as thematic verbs, e.g. liI'Ilw, OL'IlEI~, OI'llEl etc.
Present indicative functioning as future.
lnrov as relative pronoun.
Forms of the article as relative pronoun, particularly TO.
Confusion of aorist and perfect, e.g. O€OWKCW, 7rErroll8av.
f11l1/11 = fill (fITSt person singular).
Periphrastic tenses formed with E/pl.. and an active participle, e.g. 1!v 7rEIl"'a~.
Dative of agent.
Absolute participles in cases other than the genitive.
The optative is very rare.

111
112 Robert Browning

Other similar texts show in addition such features as dative with verbs of asking, indirect statements
introduced by W~ on, W~ nmTEp, EXW + infinitive as a periphrastic future, aorist subjunctive functioning
as future, genitive in many of the uses of the dative, parataxis with Kai in consecutive, final or con-
cessive sense, (e.g., n'liL'oETf'IiOt Klryw AU/li3dVW O~/lfPOV f~ uvrov T17V €VToMv;). The vocabulary of
these texts is marked by frequency of loan-words, particularly from Latin, and of new derivatives, and
the general absence of the 'Attic' and poetic words which mark the Hochsprache.

It should be noted that none of these features occurs systematically in any of the texts in question.
They are not reproductions of living speech, but macaronic compositions in which features of the new
spoken language co-exist-and are sometimes confused with-those of literary Koine and official language.
The mixture varies from text to text and from section to section within each text. There is no formed
'popular' language of literature, as there is a formed classicizing language and a formed literary or official
Koine.

The following were some of the texts of the period which can be classified as popular, though the ex-
istence of the grey area between careless literary Koine and popular Koine must always be borne in mind:

Mark the Deacon's Life of Porphyry of Gaza (early fifth century)

Callinicus' Life of Hypatius (fifth century)
The Lives of Palestinian Saints by Cyril of Scythopolis (late fifth/early sixth
centuries)
The Historia Lausiaca of Palladius (fifth century)
The Life of St. Symeon the Stylite (fifth century)
John Moschus' Spiritual Meadow (early seventh century)
John Malalas Chronicle (sixth century)
Leontius of Neapolis' Lives of St. John the Almsgiver and St. Symeon the Fool
(early seventh century)
The Doctrina lacobi nuper baptizati (early seventh century)

To the grey area between popular Koine and careless literary Koine belong such works as the Christian
Topography of Cosmas Indicopleustes (sixth century) and the Heavenly Ladder of John Climacus.
Their content and purpose link them rather with the works of popular devotion listed above than with
such texts as the Strategicon of Maurice, which have a metropolitan and official character

During the period in question developments took place in the language of poetry. In classical Greek
tradition each genre of poetry had its own appropriate linguistic form. Epic and elegiac verse were
written in the Homeric Kunstsprache, choral lyric in a kind of vague Doric, personal lyric poetry often
in Lesbian Aeolic, iambic poetry in Ionic, or, in so far as it formed part of tragedy, in a kind of lonici-
zing Attic, comedy in Attic, and so on. By late antiquity many of these genres had become extinct,
including most lyric and dramatic poetry. Epic poetry nourished, and poets like Quintus of Smyrna
and Nonnus of Panapolis (both probably fifth century) developed each in his own way the Homeric
linguistic tradition as modified by the Alexandrians. In particular the linguistic creativity of Nonnus,
who freely coined neologisms on the basis of Homeric material according to the principles of grammar
expounded by Alexandrian scholars, was followed by a whole school of poets writing in epic form,
such as Tryphiodorus, Colluthus, Musacus, and many others surviving now only in fragments or not at
all.3 ? Side by side with these there was a lively production of short occasional poems in elegiac metre
or texamctcrs and in epic poetic language, following the patterns set by the Hellenistic epigrammatists.
The anthology of epigrams by his contemporaries compiled by Agathias in the latter part of the sixth
century bears witness to the strength of this tradition.

112
 The Language of Byzantine Literature 113

At the same time there were signs of a weakening of the link between literary genre and language. Hexa-
meter or elegiac form is used for essentially rhetorical material. encomia, epithalamia and the like. A
number of panegyrics, addresses to local notables etc. by the sixth century Dioscorus of Aphrodito in
Upper Egypt survive in autograph text on papyrus. 38 We hear of similar poems by Panolbius in the late
fifth century and others. Another sign of the same tendency is the paraphrasing in iambic trimeters of
epic poems. About the year 500 Marianus made such paraphrases of Theocritus, Callimachus, Apollonius
Rhodius, Aratus, and Nicander 39 Iambic verse was also used for original poems of all kinds, epic, di-
dactic, panegyric, historical. The epic Kunstsprache required a long and rigorous training for its mani-
pulation. And even to understand it with any clarity was probably beyond most readers and hearers.
The simpler linguistic form of iambic verse-and its simpler metric problems in an age when distinctions
of syllabic and vocalic length were ce,sing to be phonological-made iambic poetry more acceptable
than that in dactylic metre. The iambic trimeter became the verse form par excellence of the middle
Byzantine world. Those who could took account of the traditional quantities of the syllables, those who
could not used it in its Byzantine form, thc rhythm of which depended on the position of the stress
accent before the medial pause and at the end of the line.40 In the second quarter of the seventh century
George of Pisidia, Deacon and chartophylax of the Great Church wrote in iambic trimeters panegyric
epics on the Persian Wars of Heraclius and a long poem of the Creation. His language is essentially
literary Koine, a medium used heretofore almost entirely for prose.41 The classically trained reader has
the impression of reading an endless messenger's speech from tragedy without any of the poetic words
that characterized tragedy, and with far too many prosy abstract nouns. George freely forms neologisms,
as was the habit of literary Koine, and has in fact a remarkably large and expressive vocabulary. But the
special linguistic features which had marked each poetic genre were lost. As a curiosity may be men-
tioned the religious poems in anacreontics by Sophronius, Patriarch of Jerusalem 634-638. They can be
scanned both quantitatively and accentually. Their language is literary Koine. with a certain number of
poetic (Homeric) forms like ¢a€!V6~, A6-yowUJ, il€!-Y€vEO~, aaw\€UJ used for metrical convenience, and an
Ionic veneer created by forms like 1TETPT/, aKOnTt, KOUPT/.

Christians had made sporadic use of various poetic forms inherited from the classical world to praise
their God, to instruct the faithful and to commemorate their saints and martyrs. They had also experi-
mented with simple accentual metres. The sixth century saw the first flowering of a new kind of speci-
fically Christian Greek poetry, the liturgical hymn. The earliest developed form of hymn was the kon-
takion, a kind of sung homily.42 The metre was based on the position of stressed syllables. But it was
very complex, as complex as that of a Pindaric ode. Short phrases of various accentual patterns were
built up into an elaborate strophe, which was repeated throughout the poem. The kontakion appears to
be based on Syriac liturgical practice, and the introduction of the new poetic form is an interesting
example of the openness of early Byzantine culture to influences from the large non-Greek minority
within the empire. But Syriac hymnography does not seem to have known anything as complex and
sophisticated as the kontakion. The acknowledged master of the kontakion was Romanos the Melode
(first half of the sixth century). Romanos writes in a strange language which shows both classicising
Atticist traits, like the use of ,TT- for -00·, and echoes of living speech, like the accusatives xfipav,
I1T/Tfpav, or dxov, "O€AOV + infinitive of unfulfilled potentiality.43 But a quantitative study of his
language makes it clear that he is composing essentially in fairly literary Koine, but with a scatter of
popular features. Mutatis mutandis-and there are many mutanda, for Romanos was a master of his
craft-his language is that of John the Lydian or Evagrius. We find only a single dual (in the cliche €~
ill1¢OLV) few optatives, regularized comparatives like }lflNTfpo~, few examples and these sometimes in-
correct of correlative }lEV, Of, €if; =Ttf;, metaplastic present stems like OI'-yW, 1TVOO}lal, TV1TTttW, clas-
sical future forms coexisting with aorist SUbjunctive or present indicative in future sense and periphrases
with fXW and }l€AAW, a variety of metaplastic forms from athematic verbs, confusion of Elf; + accusative
and EV + dative, confusion between active and middle verbs, and of perfect and aorist, masculine parti-
ciples used with feminine substantives. But the 'spread' of Romanos' language is great. As well as the

113
 114 Robert Browning

vUlgarisms mentioned above he used traditional poetic words like apovpa, a try,) , OEp.a(, Kapa, IJOPO(,
OAfT,)P, T€KO<;, </>0.0<;, {3POTO<;, KfVOP.WV, KAEO(, I'€"fWVflV, {3POTOKTOVO(, OOAtO</>PWV, /JaKap. This strange
mish-mash is partly attributable to the exigencies of a verse form tied to a musical phrase. This seems to
be the consideration which determines the choice between OtiTO<; and aUTo<; as demonstrative pronoun.
But more important was the general blunting of the sense of an appropriate linguistic form for each po-
etic genre.

Romanos established a canon of the language of liturgical poetry which influenced in varying degrees
his successors over the centuries.

The period from the fifth to the seventh century saw the development out of the inherited levels of
literary language of a new 'Byzantine' pattern, marked by a certain breakdown of the boundaries be-
tween the language of prose and poetry, the admission for certain literary purposes of a popularizing
Koine which reflected, however spasmodically, the changing spoken language, and the development of
a 'classicizing' Hochsprache which drew on other models in addition to the canonical Attic prose writers.

IV

The new section, on the period from the, mid seventh to the mid ninth century will be briefer than
the preceding section. The number of surviving works is much less, and there is every reason to believe
that this is not merely the result of an accident of transmission. Less was written. In an age when
Constantinople itself twice underwent a long siege by the Arabs, when Justinian II's march from Con-
stantinople to Thessalonika was regarded as a triumphant campaign. when Constans II thought serious-
ly of transferring the capital to Syracuse, the conditions which permitted and encouraged the high cul-
ture of the reign of Justinian were notably absent.44 Very little in the way of belles Iettres seems to have
been produced. More utilitarian standards prevailed. Pastoral and polemical theology poured forth.
But much of it, including virtually all the theology of the Iconoclasts, is lost. Simple moral and ascetic
handbooks for monks were in demand. Hagiography flourished, though here again the record has been
tampered with by the suppression of lives of iconoclast llOly men. Several world chronicles were com-
posed towards the cnd of the period. Their simple theological view of history excludes all serious con-
side'ration of historical causation or personal motive. We hear of earlier chronicles, since lost or suppres-
sed, but there is very little evidence of any contemporary history in the classical tradition after Theo-
phylact Simocatta, and certianly none has survived. There is a certain amount of official writing, such as
legal codes ami proceedings of Church Councils. This is the total of literature surviving from the Dark
Age.

Furthennore, much of it is still inadl'quatdy edited. There is an almost total absence of specialist
lexica and indices. And with the exception of the Chronography of Theophanes, none of the works of
the period has been the object of serious linguistic study. Thus any general observations must have a
peculiarly im}:>ressionistic and provisional character.

The first impression which this body of literature gives is that of the almost complete abandonment of
the classicizing Hochsprache. Even members of old established official families of the capital, like
Maximus Confessor (580-662), who belongs only marginally to this period, and Theodore Studites
(759-826). no longer write quite like the Fathers of the late fourth and early fifth century45 They use
a vocabulary with fewer' Attic' words and more neologisms, in which occasional poetic words suggest an
insensitivity to distinctions of genre, and they incline-Maxilllus more than Theodore-to let their sen-
tences get out of control. There is much less difference between their language and that of contemporary

114
 The Language of Byzantine Literature 115

leoal codes and canons of church councils than there is between the language of Ps.-Dionysius the Area-
p:gite (? late fifth century) or Procopius and that of the Novels of Justinian or the Acta of the Council
of 553. In other words what survived was in essence the literary Koine rather than the classicizing Hoch-
sprache. But it must be emphasized again that this impression is not based on exhaustive study of even a
representative sample of texts, and that within the corpus of prolific writers like Maximus and Theodore
there are probably significant linguistic differences between writings of different kinds.

Good examples of continuity of official Koine are to be found in the Canons of the Council in Trullo of
692 and the Ecloga, a mid-eighth century law code. Both are written in business-like Hellenistic Greek,
without 'Attic' or poetic vocabulary elements. The Canons introduce classicizing -rr- for -uu- in certain
words e.g. K'r,purTw, ¢vAdrrw; the Ecloga sticks firmly to -uu-. Optatives hardly appear in the Ecloga: in
the Canons they do in formulae as facultative replacements of the indicative, e.g. in three successive canons
we find El /It"(Odl1 . .. KaOatp€t'uOw, ei 1Tapa{3awet . .. KaOatpduOw. el rOA/l.qoOt (! ) ... KaOatp€t'oOw.
The Ecloga is free from optative forms. Neologisms are of the kind permitted in Koine, such as oAoKap1TWOt<;,
frepoOda, o50orarl1<;. Popular forms are very few, e.g. KaT€VaVTt for KaUVaVTtov (which may well be
due to a copyist rather than to the author). Where elevation of language is aimed at, as in the Proems to
the Canons and the Ecloga, it is attained by frequent and striking metaphor rather than by the use of out
of the ordinary vocabulary, morphology or syntax. It should be borne in mind that literary Koine was
by this time far removed from spoken Greek, and had to be learnt by study. The situation was one of di-
glossia. But it resembled the diglossia of nineteenth century Italy, rather than that of Greece today. The
written language was widely understood, was obligatory for certain kinds of 'official' communication.
but probably did not have the elitist overtones of the classicizing Hochsprache, still less the political nu-
ances of modern Greek KaOaoeoovoa.

The tradition of popular Koine was maintained in full vigour for communication of a different kind. Not
only are many short works of asceticism and edification, as well as a number of hagiographical such as
the lives of St. Stephen the Younger and St. Philaretus, or the Miracles of St. Demetrius (a product of
Thessalonika rather than the capital) couched in Koine which shows many features of spoken Greek, but
a similar language is used in the Chronicles of George Syncellus (809), Theophanes (814)46 and Nice-
phorus (814).47 As has been explained in regard to earlier texts of this kind, their language is macaronic,
with elements of literary Koine and elements of the spoken language existing side by side. And it is cho-
sen as a medium by men who are educated and can write a different kind of Greek. Whether the literary
from or the intended audience is the determining factor in their chbice is hard to' say and probably poses
a false alternative.

By way of example, a few pages of the Chronography of Theophanes, probably the most popular of these
texts, show the following noteworthy linguistic features:

airfw + dative
nlv aVrov KE¢aATiv etc. passjrn (for literary Kaine nlv KEri>aAiw atirov)
Confusion of Ek + acc. and EV + dat.
iva + subj. after verbs of ordering
Periphrastic future tense with j.lfAAW + inf., b¢dAW + inf.
Confusion of eKEi and EKEioe.
New words: ihrora-yao17v, 1TtKpo8avuro<;. Kpav-YtKW~, 1TeptAO'y,j.
Loanwords: KOP017, a1TAtKEVW, olv~o/J, TOVAOOV, oov[3Aa.

Words in new senses: 0' vo"'~oOilvat = '(0 be maintained'

115
116 Robert Browning

xp6vo~ = year
aV}JIj;V)(o~ = with all hands
aixpaAWaw. = prisoners
avaaKa1TTw = curse
8opv{3ov {3aAwv = creating a disturbance
Morphological matters: aro}Jw for uro}Jwv, Ct1TOKmv8Et( (already in LXX, NT)

Writers who choose to express themselves in this fashion are making a deliberate altempt to overcomc a
difficulty in communication, real or imagined, with their intcnded audience. The difficulty is perhaps
not so much that more traditional forms of language were incomprehensible to the uneducated-the
Ecloga was presumably not a closed book to the man in the street-as that they were inadequate to estab-
lish the emotional rapport desired by the writers, who were in quite a different position from civil servants
drawing up official documents, or theologians writing for their fellow-clergymen.

With the middle of the ninth century a change began in Byzantine attitudes to literature and language, a
change for which the ground had been prepared by men of the previous generation like Ignatius and Dca-
con.48 It is symbolized for us by the restoration, after perhaps quite a long gap, of official support for
institutions of higher education in the capital (first by Theoctistus the Logothete shortly after 843, then
by the Caesar Bardas in 855). And among its manifestations were the searching out of manuscripts of
authors long unread and the compilation of lexica to facilitate reading them. Men began to reach back to
and try to reconstruct the culture of the age of Justinian and of the Christian Roman Empire of the fourth
and fifth centuries. This is part of a larger movement which comprises the visual arts as well, and which
is best seen as the ideologjcal reflection of the growth of Byzantine power and wealth as the Arab threat
receded and the frontier began to advance eastwards, while the southern half of the Balkan peninsula was
restored to Byzantine control. But I have no intention in the present paper to enter into discussion of
this complex and disputed problem.

To get an idea of the change we can look both at what men say about the language of literature and how
they actually use language themselves. Photius in his Bibliotheca makes a number of judgments of the
language and style of other writers, classical and Byzantine. His critical vocabulary and the concepts
which it embodies are borrowed from the rhetoricians of the Roman Empire. He does not systematically
follow anyone school. But the standpoint which he takes up is that of an age which esteemed imitation
of classical models and rejected the Hellenistic Koine, with its links with the living spoken tongue 49 His
hero is Libanius, whom he calis Kavwv Kat OTti8}J7) A(yyOV 'ArTlKov (cod. 90, p. 67bI4), and he gently
censures Choricius for deviating into poetic style and Himerius for obscurity. What he admires is tradition,
not innovation. Of the historian Agatharchides (second century B.C.) he observes that he uses no new
words, but only olli words in new senses, so that Ka(V07o}JW.V }JflIiOK€ill Elva( KawoTo}Jw.V (Cod. 213,
p. 171 a36). Of Arrian, whom he much admires, he says that h Ka(voT7)~ TWV A€~€WV mixt €ie TO 1rOPPW
V€WT€p('l'€ra(. Sophronius he condemns because €VV€WT€P(lft . . Ka8ti1T€p TI~ 1rWAO~ €1T(raupovll€VO~ TO(C:
aK(pTfI}Jaal (Coli. 231, p. 286bI6). Of the Patriarch Nicephorus' History he says-strangely to our
judgment-

"EaTl liE rflv q,ptiaw Ct1r€P(TT6~ T€ Kat aaq,r,~, Kal\l\(Af~~ Tf Kat OVV8r,KUA(yyOV
OVTf A€ AVllfVU OVT€ au 1TtiAW aV}J1rf1rl€OllfVU 1r€p(fprW~ K€XP7)}J€VO~, aAA'
oi{! UV XPT70a(TO 0 P7)TOPlK~\W~\CtA7)8w~r Kat'/TfA€wdaVT7P' 1'01 T€ /rap: V€W-

116
 The Language of Byzantine Literature 117

7EP01fOLOV EKKAWEt, Kat' TO apxauJrp01fOV Kat' E~1JaK1Jl'ivov ou 1fapaTPfxn.

'En Of Kat' hoov" KEKpaTat aVTOU auv xapm roi( AD-yOt(.

(Cod. 66, p. 33b). He rejects what he calls the vulgar tongue, by which he probably means official Koine
as well as spoken language. The heterodox theologian Basil of Seleucia is rebuked for his XVOad)T1J~ and
his p-ql'aTa €K TPWOOV (Cod. 42, p. 93 I 07, p. 88a). Agapius is OVO€V Ot€V1JVOXW~ TWV €K TPWOOV (Cod.
179, p. I 25a I 9). John the Lydian's style is xal'al1fE7i1~. Kat I'1JOfV t'xovaa TWV fK TPWOOV 1fA€OV
(Cod. 180,p. 125b7).

This is a program, however naively expressed, of restoration of the classicizing Hochsprache in all its At-
ticising purity and rejection of Hellenistic Koine or any other intermediate linguistic register. If we turn
to Photius' own performance we see that though he does not attain his end he makes considerable pro-
gress towards it. His Homilies display a highly organized sentence structure, based on parallelism rather
than subordination; few complex periods, much hyperbaton. New words are strikingly few. Much of the
vocabulary is found also in the classicizing prose writers of the Roman Empire. Poetic words are rare,
and mostly come from the tragedians, whose works had probably begun to be studied again in some
schools. Thc following are some of the linguistic features noted in Photius' fourth Homily on the attack
on Constantinople by the Rhos in 960 (pp. 40-52 Laourdas): €1flO1ftAatw and imEpaKovntw are not else-
where attested-the soundness of the text has been queried: A1JtoV is a poetic word, but it is attested in
Arrian, whom Photius admires as a model of prose; f1flOTa8l'clOl'at is an Aeschylean word; f~Oq,pvow
does not seem to occur in surviving classical prose, but it is cited by Hesychius and in the Etymologicum
Magnum, probably compiled in Photius' circle; f1f1JtaI'E8a shows an aorist formation which is poetic and
unAttic, but of which an example occurs, probably through a scribal error, in Thucydides 2.97.3: tl81JPwirro
Kat €t1J'YPLWTO shows unclassical equivalence of imperfect and pluperfect: ollia on ... ftrLOawtAEVaoLTo is
an incorrect use of the optative after a primary main verb, but for Photius the future optative was proba-
bly in itself an 'Atticism', as it certainly was for later Byzantine writers.

This harking after 'Attic' language and style affects most of the literature of thc later ninth and tenth
centuries in varying degrees. Of course the "Atticism' which they pursued had very little to do with fifth
and fourth century Athens. As Gertrud B6hlig has so well shown, 'Attic' language in the Byzantine world
is characterised negatively by not being acceptable in the spoken language and positively by being used,
or thought to be used by some writer of classical or early Byzantine times.5o The list of authorities for
'Attic' usage grows wider and wider until even Homer becomes a model of 'Allie' language. This process
had alrcady advanced some distance by the ninth and tenth ccnturies. A writer like Arethas of Caesarea
(early tenth century), who collected and annotated manuscripts of classical prose writers from Plato to
Lucian and the Christian apologists, studs his tex t with obsolete words. The Patriarch Nicolaus Mysticus,
even in his diplomatic correspondence with the formidable King Symeon of Bulgaria, uses unusual, book-
ish words, rare middle forms of active verb-the middle voice itself was regarded as peculiarly •Attic'51 -
and other refinements of the kind. The authors of the continuation of Theophanes not only learned from
Plutarch, Isocrates, Xenophon and Polybius how to organize their material, how to depict character, how
to analyze motive.52 They also borrowed wholesale words, phrases and expressions from these authors.
Evan a man like John Cameniates, who had no time for Homer and whose cast of mind was radical and
monastic, did not altogether escape the pervading inOuence of the times. In his account of the capture
of Thessalonika by the Arabs in 904 he writes in general rather undistinguished Koine: periphrastic verb
forms, double augments, confusion of future indicative, aorist subjunctive and aorist optative, confusion
of EKEi, fKEiae and €K€LIJEv, uncertainty in the handling of participial constructions, wide use of the in-
finitive where classical usage would require some other construction, use of optative as an optional variant
for the subjunctive, etc. But in addition there are spasmodic signs of classicism: TTJ/J.€POV appears side by
side with O1il'epov, "(AWTTa with "(Awaaa. middle forms replace active, the optative is freely used in
final sentences introduced by tva and we av.

117
 118 Robert Browning

An example of what was regarded as fin~ writing is provided by the long letter written from exile by Leo
Choerosphactes to the emperor Leo VI about 910.53 This is a pastiche of the manner of the rhctoricians
of the Second Sophistic, complete with quotations from Plato and the tragedians, and allusions to events
in the history of Athens thirteen centuries earlier. The syntax is, with a few trivial exceptions, Attic, and
includes a number of correct uses of th~ optative. The letter is full of unusual words, for which parallels
have most often to be sought in the lilL'rature of the Second Sophistic: G.IlElO",C; (Plutarch, Oppian),
al/JaAroc; (? ), G.vEoproc; (Alciphron), a.e"'P€VTOC; (Xenophon), G.J1'J)rKpTj/lVOC; (Euripides), G.1lq,lTCiAavroc;
(G.Il</JlTaAaVTfVw Nonnus), €7rltrAT1KTI'\Oj1aL (trA1)KTllollaL Homcr), Karaq,Avapfw (Ps.-Lucian, Diogencs
Laertius), trOAVa'YKLOTPoc; (Oppian), G.VOpOAQ'Y€W (Lucian, Alciphron), 01)PWTPOq,€W (Alciphron), G.KpaT-
Ilollal (Aristophanes, Philostratus), trfPLOO{3€W (Aristophanes, Menander, Alciphron), aAOVTw. (Eupolis;
the ordinary word is G.AOVOW.), d~lvTjv orvov Kai EVTPo-rrW.V (fvTpOtrw.e; orvoc;, Ii d~t'V1)e; Suda EKrpOtrw.v Kal
d~lV1)v trlVO/lfV (Alciphron), Utrfpq,Ev(Tragedy), oIKalwnipwv(Plato), KaTatraVVVXIIW (Alciphron),KvpwK-
T6voc; (Josephus) KaTaTVpfVW (rvpfvw Dcmosthenes, Lucian, Philostratus)' G.KfpaOKOIl€W (aKfpaoKoll1'/c;
Homer, Pindar, Nonnus), pwalc; (Libanius). q,IAaKpoallwv (?). The 'Attic' vocabulary is complemented
by such choice pieces of syntax as nominative for vocative (axr,lla 'ArTudJV according to Apollonius Dys-
colus).54

This remarkable pastiche is not a scholarly joke, but a serious appeal by a high official whose career was
in ruins and whose life was in danger. It is ironical to reOect that the overt reason for his downfall was a
charge of 'Hellenism' (=pagan sympathies). Not all classicizers had the skill of Leo Choerosphactes. Mucll
ninth and tenth century prose is awkward. cumbersome and obscure, in spite of being peppered with
'Attic' words and constructions. An example is provided by the works of Arcthas of Caesarea, who never
approached the lucidity and simplicity of the models which he had copied and annotated with so much
loving care.

Technical manuals and other works intended for practical use continued to be written in informal Koine
which was not constrained by the need to imitate models and which often made greater or lesser conces-
sions to the spoken language. Constantine Porphyrogenitus underlines the distinction at the beginning of
the manual of statecraft which he addressed to his son Romanos:

"\ have not been studious to make a display of fine writing or of

an Atticising style, swollen with the sublime and the lofty, but
rather have been eager by means of everyday and conversational
narrative to teach vou those things of which \ think you should
not be ignorant" (D. A. I. 1.1, Greek text in Note 5).

A glance at any passage of this work will illustrate the point. \ take at random Section 27 on Italy, and
find confusion of fie; + acc. and fV~ dat., Kpar€W c.acc.,o' trPOpP1)OEl~, fiv roie; Kalpoi,; EipnVT)<;, fitoouitw.
traKTa = tribute, direct speech introduced by 071 fAtrllW + aorist infinitive of future expectation,
Kar1)VaAWOa, t"xw + infin. as future periphrasis, 01)A01rOI€W, cbaJ,lIAW., KVptEVW c. acc., 1rpl,),Kmarov,
trpalrwpwv, €PT1116Kaarpov. llaarpO/lIAT/<;, KaTftravw, KauTpov, plural verb with neuter plural subject,
KaaTEAAUl. This is a very different kind of Greek from that of Arethas or Leo Choerosphactes, and the
author consciously chooses it in function of the subject matter. Similar texts are the Taclica of Leo VI
and the De Caerimoniis of Constantine Porphyrogenitus; in both of these there is a similar apology for
the use of 'everyday' language. But even at this level there was a feeling that one should not depart too
far from classical linguistic propriety. Constantine Porphyrogenitus mentions that he found a work on
court ceremonies by an otherwise unknown Leon Katakylas unsatisfactory, as the author was I'OVaIKr,<;
'EAA1)VIKr,C; al1€roxoc;,55

Hagiographical texts are still sometimes written in a very informal Kaine. But there is not so much ap-
proximation to the spoken language as at an earlier period. Clearly in an age in which literary circles

118
 The Lallguage of Byzalltille Literature ; 19

valued 'Atticism' something of this classicizing attitude would filter down to almost all those who read
and wrote. Informal, unc1assical Greek would be felt to be unworthy of the nobility of the subject and
out of place. In fact we find that shortly after the middle of the tenth century a wholesale rewriting of
hagiographical texts used for liturgical purposes was undertaken by Symeon the Logothete, known from
his work as Symeon Metaphrastes. It is illuminating to see how Symeon rewrote the old texts.56 First of
all, he reiects or modifies certain recurring formulae. The old lines often began with til' mi~ i}Jl.fpat~ TOV
Nfpwvo~ or (3aOtAEiIOVTO( T oaialJOv or the like. Symeon replaces these jejune formulae with a variety of
phrases based on IiLErrEtlJ apxrw. oWrrtrTpa IifxmO(J4.oKi/1rTpa EXELV, OKT!trTpa IiUO{WEtlJ, aKT!tr7'pa AaJl.(3d.vEtlJ,
ElT! TWV JKT,trTPWV KaOt'omoOaL, and so on. Similarly, in the description of the death of the holy man,
alTolitliovat 7'111' >i/vXT,V is replaced by a number of more colourful phrases like lTPO~ 7'd~ OK1Jvd~ Jl.Em(3awELV
Td~ aCIiWv~, TOV KaAov vlTvwoat I'm 1'01', lTPO~ TOV lTOOOUJl.EVOV XPWTOV fKIi1JJl.Eiv, Ei~ xEipa~ OEOV TT,V >i/vxiw
lTapanOEVat and the like. Next, linguistic forms obsolete in the spoken language are introduced-hands
become dual instead of plural. optatives and pluperfect indicatives are frequent, participial constructjons
are substituted for the paratactic constructions of the original, adjectival participles replace simple adjec-
tives, VEW<: replaces vao<:. crasis is introduced, 'Attic' particles like TE.lTEP as well as the commoner 'YE,
apa,1/ Jl.T,vare scattered over the page. Lastly, rhetorical figures and word-play of every kind replace the
simple n~rrative prose of the earlier lives. In this process of rhetoricisation much of the concrete detail of
the original-names .. places, offices and the like-is thrown overboard. to be replaced by elegantly turned
doctrinal pronouncements of impeccable orthodoxy and long catalogues of the works of the devil.

The Metaphrastic Saints' Lives soon replaced the older tex ts in the Menologies and other liturgical books,
and many of the pre-Mctaphrastic originals are now lost. The new, classicizing texts corresponded to the
taste of the ruling groups in society. Their exclusive use among monks, who must have formed a substan-
tial part of the literate population, helped to disseminate a new sense of literary propriety. The tradi-
tional diglossia of the Greek world took a new turn in the period of the Macedonian Renaissance. Only
the classical, the archaic, the poetic was henceforth to be acceptable to a man of education, and the more
archaic it was the better. Easy-going, informal Koine was reduced to purely technical functions, and the
long tradition of popular religious writing in popular language was brought almost to a close. But once
again it must be emphasized that this was a literary matter. not, so far as we can discern, a socially divisive
issue.

VI

The eleventh and twelfth centuries saw the classicism introduced by the Macedonian Renaissance come
to full fruition. Writers like Michael Psellus. Anna Comnena and Nicetas Choniates cultivated language as
far removed as possible from spoken Greek, with a vocabulary, morphology and syntax imitative of that
believed to be used by classical models, which now ranged from Homer to the Fathers of the Church. Its
mastery required devoted study and long training. From such works as Gregorius Pardus on Syntax, the
schedographical passages attributed to twelfth century teachers. and Eustathius' monumental commen-
taries on Homer we can form an idea of the way in which facility in the literary language was produced.
It can never have been easy to use. Writers often seem to be observing their own performance and draw-
ing attention to their own cleverness. As we shall sec, however, the new flowering of the Hochsprache
and the virtual elimination of informal or formal Koine as a literary language is only one feature of the
linguistic development of the period.

Michael Psellus (1018-1096/7) was by any standard an immensely gifted writer. The canon of his writings
is not yet definitively established. But his speeches and letters and his Chroll()graphy together form a

119
 120 Robert Browning

substantial corpus of writing in genres which called for the most elevated 'Atticist' use of language. It is
typical that the mention of the joint emperors Basil II and Constantine at the outset of the Chronography
elicits a firework display of dual nouns, adjectives and verbs: r,OT71V Of iillq,w r,071 IlfV 7rapfA71AaKOTf Tilv
ii1371v, olaq,6pw Of TO ~OO~, etc. The language of these works has been described in detail by Rcnauld
and analyzed in terms of the precepts of the Atticists by Bbhlig 57 Rather than summarize or repeat
what they have said I would prefer to direct attention to a short work in which Psellus displays his own
critical principles at work, his Encomium on Symeon Metaphrastes (Kurtz-Drcxl, Michaelis Psel/i scripta
minora, I, Milan 1936, 94707). Earlier hagiographers, he observes, fell short of the grandeur of their
theme, since they had neither KaAile'vvolll nor flhrpf7rfl~ A€~€l(. Readers could not bear to read their
texts, or found them laughable because of TO Tij( oVVOi)KT)( aKaAA€(, TO avaKOAOVOOV Tij( €WOUl<:, and
TO Tij( A€~fW( fVrfA€~. Symeon brought to his task the necessary elevation of thought, dignity of com-
position, and a A€~l~ IlfTa TOU oaq,ov~ Ol71PIl€V71 TO Tf 1I'lOaVOv. €XOvoa Kai aAi)Owov Tf Kai e'1lq,VTOV. Many
of oi OOq,WHKWTfPOV riI -yAWTTT/ KfXP71Il€VOL, Or( q,voLKa llaOi)llara 7r€pw7rovoaora, were not pleased
with his work and found fault with its figures, its style and its vocabulary. Such critics want everything
to be written 7rPO( E7rLOfl~W, but Symeon's aim was the communication of facts; he had no need of
-ypallllarLK'17 OfWpW.. He desired to attract his learned (€AACryLIlOL) readers rei> ... PvOllei> Kal'Tei>KdAAEL
Tij( Af~fW~, and to hold the attention of ai iOLWTLOf~ aKoal' by Tc;, oaq,a' Kal' 1I'OTtP4> rij( favrou q,wvij~.
One could write 1I'fPlfP-YOUpov, but one would not be accessible to all one's audience. In descriptive and
other passages Symeon rose to the heights of rhetoric, but in general he aims at a middle style.

Pscllus recognizes differences of level within the classicism which he favours, differences which depend
on the audience aimed at. But at all levels choice of vocabulary is an important method of elevating style
and making acceptable to readers of his time.

The language of Anna Comnena has been briefly studied by Georgina Buckler.58 Anna was proud of her
education and her command of the classicizing tongue. Later generations have often found it difficult
and involved. At the level of vocabulary she uses many Atticist words probably not current in the ordi-
nary language of her time. She also uses a great many obsolete words which would not have been accept-
able by those who established the canons of Atticism. These include many Latin loanwords connected
with the administration of the Roman Empire. Perhaps as an emperor's daughter, born in the purple, she
was particularly conscious of the Roman element in Byzantine tradition. She also uses the occasional
Latin loanword current in the living speech of her time, like Ka(3aXAapLO~, 7ropra, oKdXa. Many classical
Greek words are used either in a vague ... ld imprecise sense, e.g. a7rCryovo~, V€WTa odX71or in an unclassical
sense current in the spoken language of her time, e.g. ~Oavw, OW\w (=to be enough), KaprepEw,oLaq,epw.
Her vocabulary illustrates both the all-embracing nature of Byzantine 'Attieism' and Anna's own inability
to follow its negative prescription of avoiding altogether features of the spoken language. Her syntax is
equally wild. Indicative, subjunctive and optative are used side by side in the same construction (\3.12),
aorist optative and future indicative are treated as equivalent (10.8), genitive absolute participles are used
in apposition to the subject or object of a sentence (9.9, 12.5), demonstrative pronouns are used as re-
flexives (8.1,12.2 [bisj, 13.6), an indirect statement after a verb of promising is introduced by W( lva
(13.12), liv is used, as it was by Psellus, as an element of stylistic variation, sentences change their subject
in mid-<:ourse. Her morphology is on the whole classical, apart from her 'Atticist' use of middle for active,
but a form like fq,eiJpOOav (15.2) occasionally slips out.

Anna writes far more complex and ambitiously classicizing Greek than her husband Nicephorus Bryennius
or her predecessor Michael Attaliates. In spite of her outstanding talent as a writer, the classicizing language
which she uses, and which she handles without the creative imagination necessary, is a hindrance rather
than a help to her. Much of what she has to say lacks clarity, and probably lacked it for the educated
elite for whom she wrote her History.

120
 The Language of Byzantine Literature 121

The Timarion, which may be the work of Nicolaus Callicles5 9 and which certainly belongs to the begin-
ning of the twelfth century. is a good example of the self-conscious imitation of Lucian. But it displays
the Byzantine insensitivity to distinctions of genre in c1assicalliterature.60 There are many poetic words
and forms like rrYOpEOVTO, iiAOO~. aAvw, i!1El'pO!1al, !1ap"{dw, O!1WVEn~, 7rOorI"fO~, pEW.; all is grist that
comes to the writer's mill. Many classical words are used in unclassical senses, e.g. ii,,{,,{EAOC;, a7rapx-q
(:beginning), a7rOAo,,{oV!1al (:answer), {3pa{3€vw (: bestow), oulra~lc; (: edict), OAKrj (: weight), cpOdvw
(: arrive). There are also many Hellenistic words like {3PVX1/0!10~, oulrpavo<;, OllLrV7rW!1a, vEKpa"(w,,{oc;.
)(OlPE!17rOPOC; and a few cl7ra! AE"(OI1EVa, including the compound f07rEPOpOpwv. Perfect and aorist are
used interchangeably. There are many middle forms of verbs used as actives in classical Greek, the future
indicative appears in final clauses, strong and weak aorist are confused, and there are hyperatticisms like
fV 'AOrjvT/ol. The language betrays its artificiality by its lack of homogeneity.

Nicdas Choniates wrote nearly a century later than the author of the Timarion and more than half a
century after Anna Comnena. His History and his speeches and letters provide an extensive body of
texts in the middle Byzantine classicizing literary language. He recognizably belongs to the same stable
as Michael Psellus and Anna. If Anna Comnena is difficult and involved, Nicetas, it has been observed, is
sometimes incomprehensible.' This is probably a reflection on our knowledge of Greek rather than on
Nicetas' handling of it, for he is far more in control of his medium than Anna and uses it in a creative
way. He has an immense vocabulary, with many rare and antique words, and many neologisms. In twenty
pages of his History (665-685B) the following striking words were noted: those marked with an asterisk
are not in Liddell-Scott-Jones: a!1cp,xp1/!1VOC; (Euripides), at,,{lAll/! (Homer), a!1<!il7rEplKAW!1EVOC; *,
aVOa!1lAArjTPIlL *, avaVx1/v (Empedocles), A€W"{€VEWC; (Herodotus), axapl'TWTOC; *, 7rpOaaVa7rA€W *,
7rEpWKal'pw (Oppian), E7rE,,{XPl1rTW *, !1f"(aA07rpE7rEVOIJal *, KaravTV"{w01/C; *, fwd)<; *, a7r6liao!1a *,
fTOt.p.OTP€X.qC; *, KaXEKT1/!1a *, PV1/cp€v-q<; (Dionysius Periegetes, Nonnus), aval/!1/Mcp1/TOV * (? read
al/!1/Adcp1/TOV). In the same way the speeches and letters show many new nouns, verbs and adjectives
formed by composition from classical models, and a number of poetic or dialect forms like {01/jJ.l, 7rOAUTAa,.
Van Dieten marks 216 words in the index to his edition of the letters and speeches as iI.l)1/uaupwTa. Both
in the History and in the speeches and letters he exploits synonyms.61 The pots forming the cargo of a
ship are called TPV{3AUJ., A07rdoEC;, KEpd.!1IlL, oKwdpllL, the tomb of Manuel I is called !1v7l!1a, owpoc;, Tdcpoc;
and Al!:lO~ V€KPOO€"{!1WV (an allusion to Aeschylus, Prometheus 153), the rope which Andronicus Com-
nenus asks for in prison has four different names. Nicetas has ten or twenty ways of saying that a man
was blinded, compared with Psellus' repeated TOU, dcp(JaA!10U, EKK07rTElV, TWV dcp(JaA!1WV b,7roOT€p€iv.
In his exploitation of the resources of 'Atticism', as understood by the Byzantines, for the purposes of
variety and novelty he pushes the resources of the classicizing language about as far as they will go. Only
those who shared his wide literary culture and his classical standards could fully appreciate the ever-
changing nuances and allusions of his language.

At the same time he either cannot or does not care to follow the syntactic patterns of Attic prose. The
optative is for him largely a means of obtaining variety, and in particular he likes to use aorist optative
and future indicative side by side both in principal and in subordinate clauses. He is fond of periphrastic
verb forms like €OEiTal Of~d.IJEVO" 1/ v arwv, "{fVOW fucppaWO!1EVoC;. "Av is largely a stylistic adornment;
he likes to use it with the future indicative in a potential sense. There are few recognizable traits of the
spoken language. But he does use EavT6vof the first and second persons, and occasionally admits a form
like €loTjMooav.

These remarks are of necessity impressionistic. A study of the language and style of Nicetas remains to
be made. But one can say with some assurance that, unlike those writers who try to produce a pastiche
of late Roman or early Byzantine prose, he uses the redundancy, the suggestiveness, the flexibility of the
learned language in an original way 61a It was essentially a use of language possible only in self-conscious-
ly literary milieu, a coterie. And he had the misfortune to live to chronicle the destruction of that milieu.

121
122 Robert Browning

Even in the eleventh and twelfth centuries, when classicism dominated the literary world, when Patriarch
Nicolaus Muzalon could suppress a Li}1' of the Thracian St. Paraskeve the Younger because it was written
7Tapa TLVO~ XWPlTOV iOlWTtJ(W~. 62 when writers of handbooks of rhetoric could insist on the virtue of
obscurity as an element of (i€LV6T1)~,63 not all communication could be carried on at this level. From the
mid-eleventh century onwards there is a growing body of texts written in literary Koine-which by this
time had to be learnt at school-without Atticizing pretentions, and fifteen syllable accentual verse. the
so-<:alled political verse. Most of it is educational in character-brief manuals of grammar, or law or the-
ology, elementary allegorizing commentaries on Homer, chronicles of world history and the like-and
much of it is addressed specifically to women readers, usually ladies of the imperial court, who would
not as a rule have had the education to appreciate all the refinements of the literary language. Other texts
are connected with ceremonies of the court, and are written to be chanted by the demes, those emascu-
lated epigones of the circus parties of early Byzantine times. The earliest of all, the Hymns of Symeon
the New Theologian, are the expression of a mysticism which rejects the world and its values. There are
many obscurities in this strange marriage between non-A tticist literary Greek and a meter for which no
ancient model existed, not all of which have been cleared up by M. J. Jeffreys in his recent important
article.64 It is probably that the meter was already used in popular songs and ballads, but the direct
evidence is very slender. It may also have been the appropriate form for imperial acclamations-the re-
mote descendant of the versus quadratus in which the Roman soldiers had chanted scurrilous and apotro-
paic invectives against their triumphant commanders. On these accounts it seems to have been acceptable
to sections of society which had not-or not yet-enjoyed a full classical education. It was certainly a
useful mnemonic aid for matter which had to be learnt by heart. And it fell completely outside of the
normative systems built up by grammarians over the centuries.

It is interesting to observe the apologetic manner in which writers in this new genre introduce their work.
Philip the Monk, author of a devotional work, the Dioptra, written in 1095-7. says that he is writing for
all, monks and laymen, who are uneducated like himself, and not for rhetoricians and olliaaKaAOt, and
later on he says that he has changed the words and the composition of his sources to make them clear,
precise and easy to understand, and paraphrases them in rustic political verses because he is himself
-ypaIJIJaTWV a7Tflpo~. Finally he begs his readers not to condemn him or deride him but to pay attention
to the meaning of what he writes 65 Psellus in his various introductory works written in political verse
stresses conciseness and ease of understanding, as well as simple, every-day language- EV il7TAOvaTt!pat~
A€!;€OtV Kat KaT1lIJa!;€VIJlvat~. 66 John Tzctzes introduces his Theogony and his Iliad Allegories, both
composed for ladies of the imperial fap.';ly, with apologies for the brevity and lack of polish of his work.
But that was what his patronesses had asked for.67 In his Odyssey Allegories he speaks of writing in a
clear, comprehensible and playful (rrat-YVtw(ifaTEpo~). He adapts his language and style to suit the capa-
bilities of his aUdience.68 Constantine Manasses, whose Chronicle in political verse is dedicated to the
same princess for whom Tzetzes wrote his Theogony, speaks of his work as €tJavvo1r"To~ and oa</117<;. 69
What emerges from all this is that these writers see themselves often somewhat unwillingly, in the role of
popularisers, writing down to an audience unable to appreciate, or perhaps even to understand, the classi-
cizing Hochsprache. Philip the Monk may have been, as he says, an uneducated man. But Psellus, Tzetzes
and Manasses delighted in their skill at manipulating the Atticizing literary language. For them to write
simply was a concession-Tzetzes uses the word oiKovoIJta -which was made easier for them by the use
of a verse form which had no classical antecedents and no rules.

This extensive literature in political verse is not-with two exceptions to be mentioned later-couched in
popular or popularizing language. It is written in Koine, sometimes, especially in the poems addressed to
emperors, in quite elevated Koine. It avoids the farfetched poetic words and the pastiche of antiquity of
the literary language, and retains the freedom to form new words by composition and derivation charac-
teristic of Koine Greek. Constantine Manasses' Chronicle is full of neologisms, like (3apvrretOew, (3apvor-
ovax€w, liWPOA1)7rTOVJl.at, EPWTOA1)7TTOVJl.at, AaIJ7TTTlPoV')(fW. IJap-yapol/Jop€W, Jl.1lxavoppal/JovlJat, dpOopp-
TlIJOVfW. ravro-yvwlloVfW, which a classicizing writer would avoid. The same can be said of Tzetzes'

122
 The Language of Byzantine Literature 123

Chiliads. These writers avoid the antiquarian imprecisions of classicizing syntax. Though they avoid care-
fully and fairly successfully the introduction of words and forms from the spoken language, they probably
owe to it something of the freedom with which they form new compounds. It is noteworthy that many
of the new verbs occurring in Manasses' Chronicle arc recorded from Modem Greek, unfortunately with-
out any references, by Dimitrakos and Konstantinides. 70

A man like Eustathius of Thessalonica, who taught grammar and rhetoric in Constantinople, and used
Attieist language in his speeches and technical Koine in his lectures on Homer, was interested in the
spoken tongue and felt no inhibition in referring to it and quoting words and expressions from it fre-
quently in discussing Homer. The idea of using it as a vehicle for the formal expression of his thoughts
probably never occurred to him. But it did occur to others. Ptochoprodromus-whether he is identical
with Theodore Prodromus or not-addressed several poems to 10hn II Comnenus in which an attempt was
made for the first time to write an approximation to the spoken tongue, with its vocabulary, morphology
and syntax which are essentially Modem Greek? 1 The poems are satirical, on the woes of a henpecked
husband, the seamier side of monastic life, the unhappy lot of the scholar in a world where riches are the
reward of the practical man. Their reflection of the language of everyday intimacies and insults adds to
their piquancy. How accurate that reflection is we cannot easily say. The language appears to be rather
macaronic. But perhaps the speech of educated men in the capital was macaronic. We really cannot de-
termine the degree of interference between different language registers. The poems were meant to be a
tour de force. They are introduced by a brief prologue, significantly in classical Greek, in which the poet
describes his work as TLva( 7rOA!TIKOU( al-dTOOl!<; 7raA!V oTLxav<:, aVllforaAIlEvov<;, 7ra!~avra(, aAA' aUK
ilVawxvvTwVTa<;. This is in no sense a literary break-through of the vernacular. merely a demonstration
that it can be done, if one is clever enough, and that the result is amusing.

Michael Glycas addresses Manuel I in a rather less successful imitation of vernacular speech in a poem
appealing to be released from prison72 He is less successful than Ptochoprodromus in maintaining his
register, and drifts rather helplessly back and forward between literary Koine and spoken Greek. Again
there is no question of a revolution. He wanted at all costs to attract the favourable attention of Manuel,
who may well have an interest in the everyday language of his people. Significantly enough, his appeal
seems to have fallen on deaf ears. Other than these two examples of literary exhibitionism, the twelfth
century shows no sign of the literary use of popularizing, let alone truly popular. language.

No mention has been made of the poem of Digenis A kritas, which, whatever its remoter origins, probably
reached its present form as a conflation of two different stories some time in the period under discus-
sion. 73 The earliest mention of Digenis is in one of the poems of Ptochoprodromus. The reason for my
silence is the uncertainty concerning the linguistic fonn of the earliest version of the poem 74 Surviving
manuscripts show texts essentially in literary Koine on the one hand and in near-vernacular Greek on the
other. The discussion, which has perhaps generated more heat than light, is summed up judiciously by
L. Polites. 75 Though the question must remain open, the likelihood is that the original version, clearly
the work of a man who had some acquain tance with rhetorical literature, was composed in non-Atticizing
classical Greek. It would thus take its place among the other texts in fifteen-syllable verse discussed
above. But it would be unwise to use this tentative conclusion as a premise for further argument about
the language of Byz.antine literature. The same is true, mutatis mutandis, of the Spa/leas, a moralizing
poem probably written by a son of the Emperor 10hn II and surviving in a number of redactions of dif-
ferent linguistic form. 76

123
 124 Robert Browning

VII

The final period to be discussed is that from 1204 to 1453. It will be treated in somewhat less detail
than the preceding periods, not because it is less important, but because the situation is generally clearer
and less in dispute, and because the language of late Byzantine literature has recently been the subject of
several lengthy and penetrating studiesJ7 The most significant development was undoubtedly the appear-
ance of a body of literature in the vernacular in the fourteenth century. It is from this point that the
history of Modem Greek literature really begins. And it is in connection with these early vernacular Greek
texts that comparisons with the rise and development of vernacular literatures in other parts of Europe
can most fruitfully be made.

But before turning to this crucial innovation attention must be paid briefly to the more traditional forms
of literary language. The capture of Constantinople by the Latins in 1204 destroyed the political and
social milieu in which the classicizing literature of the age of the Comneni had flourished. But in the
centers of Greek power which sprang up in areas free from Latin rule, and in particular in the Nicaean
empire, whose monarch Michael VIII in 1261 restored Byzantine rule in Constantinople, the culture of
the twelfth century enjoyed immense prestige, and was imitated and as far as possible recreated. To some
extcnt this cultural restoration was a kind of surrogate for the reestablishment of political power. So the
literature of the Nicaean period and of the early part of the Palaeologan period is uncompromisingly
classicizing. The esteem in which the rhetorical literature of the twelfth century was held is demonstrated
by the number of thirteenth century manuscripts in which its disiecta membra were painstakingly re-
assembled for study and imitation n Writers like Nicephorus Blemmydes (1197-1271), George Acropo-
lites (1212-1282), George Pachymeres (1242-c. 1310), Manuel Holobolus (c. I 245-c. 1300) all write a
language dominated by imitation of ancient models. They vary in the way in which they handle this
language. Blemmydes makes extensive use of poetic words and neologisms in his contorted style. Acro-
polites writes simpler and more straightforward Greek of much greater clarity and precision. Pachymeres
models himself more closely on the writers of the Second Sophistic, foreshadowing the classicism of
Maximus Planudes (1255-1305), who writes in a creditable pastiche of fifth/fourth century Attic prose.
Correctness of language was a preoccupation. It is significant that the most distinguished and successful
scholars wrote handbooks of grammar and lexica of Attic words,?9 There were disputes between cliques
and schools, like that which joined in battle Theodore Metochites and Nicephorus Chumnus in the middle
of the first half of the fourteenth cen:·lry.BO But they were disputes among writers and scholars who
were united by a community of doctrine on matters of language. What separated them was the degree of
compromise which they were ready to make. Metochites would have nothing but antique purism, often
of the most ornate kind, Chumnus, a pupil of Gregory of Cyprus, arrived at a simpler, clearer, more con-
cise style. But both rejected not only the spoken tongue-which in any case was not a serious contender
for literary use, but also the more easy-going literary Koine. Nicephorus Chumnus declares that in writing
one should imitate the ancient masters as painters imitate the works of Lysippus and Apelles'll -a purely
conventional comparison, as no works of ancient painters survived! This imitation is to extend to language:
Kat TiI~ twvi)~ ')'fVOV Kat ')'AWTT7)~ TWV apLUTWV EKfWWV Kat J1f')'aAWV, Kal WV ovlilv J1Eifov, lIaAa.wv
o.vlipwv. 2 In accordance with Byzantine ideas of Atticism, the Fathers of the Church were good models:
au Ii' fvaf{31j~ wv Ei~ iaToplac b.pxala~ avarpfpf TWV TOT' EV 6w,),vwaL\t l7)aaVTWV 6avJ1aarwv Kat IiLKaLWV
b.vopwv.B3 Equally Byzantine is the acceptance of elements of classical poetic language: and J11jv Kat
p.qanc 1I0(7)TWV 'En7)V€~ fllELaa,),oval, J1apTVpOVVTWV aVTWV Toic A6-rOI~' tX€L~ Kat avo B4 Theodore
Metochites, in spite of occasional doubts about the value of so restrictive an intellectual tradition,8S is a
more finicky and thorough-going classicizer than Nicephorus Chumnus. He even wtites long autobio-
graphical poems in Homeric hexameters. These are vitiated by his inability to handle quantitative metre
and by his extensive use of analogy in inventing pseudo-Homeric forms. A few lines will show to what a
depth of absurdity and unintelligibility the pursuit of classicism reduced a man of intelligence who usually
had something to say:

124
 The Language of Byzantine Literature 125

XPOUI'OC; o,oiyor; f/10!-YE 'lrEp!Tp!ipaT' EI' TOUrotU!

'lrOVl'EUI'8' 4J.EPOWU! Ka/1UTOlr; OUW8EV OVTW'
TPl' fTca -yap E1rl' OEKa dUlv Q.,wVov8' wv XPfI,
Kal' 'lrau' irKOVE/1EV Q.vu-YX1/ VEOV avopa 'lrap' aAAwv
U-YE/10I'WV oorpl1}r; TE AO-yWI' 8' f~El1}r; rfXI'1/c:. 86
The next few lines of this extremely interesting description of his own education include such monstrous
forms as aKovKa (from Q.KOVW), KaA€O/1Er;, oa/1Wvp-YEtl', AOV-yWV (=A6-yWI'), EuurporpaAl"(aOOI', rppoiJl'nr;,
/1ovprpa (= /10prp-q), 8€lWpW., rplAOOOVrpW..

The historical reasons for this dominance of mandarin·like classicism in the late Byzantine world, when
the erstwhile oecumenical empire had become a small state of the Levant, manipulated and exploited by
Venetians, Genoese, Serbs and Turks, are easy enough to imagine but difficult to demonstrate.8 ? The de-
fensive reaction of an impoverished class of intellectuals, now too numerous to find employment in the
shrunken state, is one aspect of the matter. But men like Theodore Metochites and John Cantacuzenus
were by no standards impoverished. Wistful dwelling on past glories as a consolation for present humilia-
tions is another aspect. The pioneer work of I. Sevcenko points the way to a sociological investigation of
late Byzantine intellectual groups which may lead to surer results.88

Of one thing at any rate we can be certain, that literature of this kind had a limited appeal to a probably
dwindling group of readers. It was essentially coterie literature.89 Yet communication had to extend
beyond the charmed circle of the c1assicizers. It is interesting, and significant of the tension in late By-
zantine culture, that from the late thirteenth century we find paraphrases of works written in the classi-
cizing Hochsprache, designed to make them more intelligible to circles which were literate but did not
participate in the classicizing movement. These are not demotic translations but simplified texts. Obso-
lete literary words are replaced by common words. Participial constructions with inflected participles are
replaced by subordinate clauses. Complex sentences are broken up and simplified. These paraphrases
must have been made by persons familiar with all the requirements of the classical language for a circle
whose needs they were aware of but to which they did not themselves belong. Unfortunately we do not
know yet where, when and by whom they were made. Such paraphrases survive of Anna Comnena,90
Nicetas Choniates.91 the BaulALKoc; avopuic; of Nicephorus Blemmydes,92 Symeon Seth's collection of
fables entitled Stephanites and Jchneiates,93 Syntipas,94 and the Chronicle of Constantine Manasses,95
The last is in a linguistic form much closer to vernacular Greek than most of the others mentioned, and
may represent a step in the direction of the vernacular translations of texts in the learned language current
in the period of Turkish rule. It is interesting and significant that these thirteenth and fourteenth century
writers are doing the exact opposite of what had been done by Symeon the Metaphrast in the tenth
century.

At any rate these paraphrases are indicative of a broadening circle of interested readers not educated in
the classicizing language and indifferent to its aura of prestige. The existence of such a reading public is
a condition of the development in the fourteenth century of original vernacular literature. This develop-
ment occurred in a genre which was outside the zone of influence of the classicizing culture-in poems in
fifteen-syllable accentual verse recounting fictitious events. This emerging vernacular literature has re-
cently been the subject of an extensive and penetrating study by H. G. Beck,96 so I shall make no attempt
to describe or list works, to discuss individual works in detail, or to establish the date and provenance of
each. J shall confine myself to a few observations on the language of these early vernacular texts.

First of ail there are none of them in a developed and consistent demotic Greek. They are all open to
constant interference from the traditional literary language. This interference takes the form not only of
the use of learned vocabulary words, but of the employment of morphological elements of the classical
language side by side with those of present day modem Greek.97 The language is always mixed. What
varies are the proportions. The romance of Cal/imaclrus and Chrysorrhoe has more classical forms cor-

125
126 Robert Browning

rectly used than, say, the Chronicle of the Morea. The Chronicle, on the other hand, is full of pscudo-
classical forms and classical forms wrongly used-like OrrydTT/P as an accusative. Certain of the romances
of chivalry which are clearly adaptations from occidental originals have a great many Italian and French
loanwords. Callimacilus and Cilrysorrhoe has scarcely any. Belthandros and Chrysantza is more demotic
than Callimachus alld Chrysorrhoe, and imperios and Margarona more demotic than either. Lybistros
and Rhodamne combines a richness in demotic fonns with sporadic and frequent archaism.

Second, it is impossible to detect any regional differences between different texts. And none shows any
particular correspondence with any of the later dialects of Modern Greek. It is as if we had to do with a
kind of common demotic language of poetry, marked by uniformity and by frequent interference from
the classical tongue.

Third, the dating of all these works is far too uncertain for any kind of chronological development of the
language to be observed. We must be on our guard against supposing that more popular means later. This
would be to repeat the methodological mistake which vitiated much of Psycharis' work on the history of
the Greek language.

Fourth, it has sometimes been assumed that this breakthrough of the vernacular took place in areas under
Latin occupation, where the prestige of the literary language and the society whose needs it served was
less, and where the example of the western European vernacular literatures was there to be followed. Now
some works, such as the Chronicle of the Morea, were certainly, and others, like the translation of Benoit
de S. Maur's Roman de Troie, were probably composed in Frankish Greece, by poets who identified
themselves with the conquerors rather than with the conquered in whose language they wrote. But all
that can be said about the Ballads of Chivalry, the Achilleid, and many other works, is that they were
composed in a society in which Frankish influence was strong. This would be true of the areas still under
Byzantine control, and of the capital itself. Callimachus and Chrysorrhoe was probably written by an
imperial prince, a nephew of Michael VIII, who belonged to the highest circles in Constantinopolitan
society.98 Belisarius is certainly of metropolitan origin, and the same is probably true of the poem of
The Four Footed Beasts. Whatever may be the origin of particular works, there is no reason to suppose
that the genre as a whole and the language in which it is couched came into being in peripheral or provin-
cial zones of the Greek world.

Fifth, there is some evidence that at any rate the first steps in the new literary use of the vernacular were
taken by men who were at home in the classicizing literary tradition. Andronicus Palaeologus, the proba-
ble author of Callimachus and Chrysorrhoe, also wrote moral and theological works in the learned
tongue.9 9 Beck has argued forcefully that the first impulse to write in the vernacular came from above,
not from below. His arguments are a counterblast against the romantic simplifications which have often
prevailed in the past. And they are probably a good working hypothesis so long as we bear two things in
mind: first, that there were probably upper-<:lass circles which did not share the ideals of the classicizing
intellectuals, for instance among military leaders or at the imperial court; and second, that the tentative
ovcrture from above had to be met by an answering impulse from below, in the fonn of a public able and
willing to read poems in the vernacular. We really know ex traordinarily little about the early days of this
new reading public, if only because the historians and rhetoricians who chronicled events in the learned
tongue did not deign to notice it. Yet it was an essential element in the fourteenth century attempt to re-
move the con tradiction between language and historical situation, to create literature with the linguistic
resources of the period, and to write for a people rather than for a class.

Sixth, a number of these texts survive in several redactions, differing in language, in length, and in the
dctails of the plot. The situation is similar to that of Digenis Akritas. To this category belong, among

126
 The Language of Byzantine Literature 127

others, Imberios and Margurol/(l, Phlorios and Platziaphlora, the Achilleid, the Chronicle of the Mor",i.
Whether this is due to the casual attitude of copyists, who treated these texts in a more cavalier fashi'"1
than literature in the learned tongue, or is to be attributed to an oral stage in their transmission, is a 111al-
ter of uncerlainty.

To sum up, the fourteenth-century breakthrough of the vernacular was probably metropolitan rather thall
provincial at first, though the language and style of metropolitan texts was later imitated in provinrial
society. II was connected with the desire to communicate with a wider audience than that familiar wilh
the learned longue rather than with the rise to literacy of a sub-literary, oral culture. It did not result ill
a literary language faithfully reflecting the spoken tongue. There were always sporadic features of Ih,'
learned language in vernacu.lar texts. Indeed the presence of such clements may have increased the fl,'~i­
bility and expressiveness of the vernacular and made it more fitted for literary usc, at any rate in the early
stages. The new vernacular literature was confined to poetry and to 'non-serious' subjects. In prose, ;l11d
in the treatment of serious topics, the learned language in its late Byzantine classicizing form was Ill"
firmly entrenched to give way.

It would be interesting, but beyond the competence of the present writer, to compare the developn1l'lIt
of Greek vernacular literuture with that of Italian. IOO In both cases a traditional learned language, SIII'-
ported by a body of doctrine on grammar and rhetoric, enjoyed immense prestige. In both, the beginnill~s
of vernacular writing were confined to poetry-and in the case of Italy were often in local dialects. Yl'I
by the be)!inning of the fourteenth century Dante had created a superbly expressive literary reflection "f
the spoken ton)!ue, in which the most serious of subjects could be treated with dignity and grace. Grl'l'k
vernacular literature became stuck at an earlier stage. There is nearly always something lightweight and
even frivolous about it. Its language too, never attained the unity and the clear distinction from Ih,'
learned tongue which .Italian reached in the late thirteenth and early fourteenth centuries. Like so /11l1dl
else in Byzantine culture, it made a brilliant start but failed to reach full fruition.

In one area alone do we have evidence of the literary use of the spoken tongue with little learned admix-
ture. Leontios Makhairas, an official at the court of the Lusignan kings of Cyprus, wrote in the first h"lf
of the fifteenth century a chronicle of his own times, The Tale of the Sweet Land of Cypms. IOI in "
Greek which shows very many morphological and phonological features of the later Cypriot dialect and
which is full of French loanwords. Its simple, precise, paratactic style owes nothing to classical mod.-ls.
It is probably a faithful rencction of the Greek spoken by the upper classes of the island. But CYPflls
was very much a special case. It had been removed from Byzantine political control since the twelfth
cen tury. Its schools did not transmit the classical linguistic and literary tradition, for which there was 110
need in a community whose upper class was largely bilingual and whose administrative structure was
modelled on that of western feudalism. The Assizes of Cypms embodied feudal law in what was virtually
Cypriot vernacular. Gregory of Cyprus, the future Patriarch of Constantinople, had found it impossihl<-
to obtain a 'Hellenic' education in his native land and had walked to Constantinople as a young man 10
attend schools of grammar and rhetoric. The fact that Makhairas wrote his chronicle in prose is probahly
not unconnected with the influence of the Old French chronicles. He was somewhat younger than Frois-
sart, somewhat older than Commynes. That he was not a mere sport is shown by the fact that he fOllnd
a follower and imitator early in the sixteenth century, Georgios Boustronios.

In Constantinople itself, the great historians of the mid-fourteenth century, John Cantacuzenus and Ni-
cephorus Gregoras, the theologians who debated the pros and cons of Hesychasm, the preachers who
delivered panegyrics upon saints and martyrs, like Isidore Archbishop of Thessalonica at the end of !l1l'
fourteenth century whose panegyrics on SI. Demetrius were recently edited,I02 all write uncompro-
misingly classicizing Greek, complete with dual numbers, Attic tags, obsolete words, optatives and the
whole paraphernalia of Byzantine 'Atticism'. Only in the middle of the fifteenth century do we enCOIIIl-

127
128 Robert Browning

ter a change. Of the four historians who recorded the final years of the Byzantine Empire and the capture
of the city by the Turks in 1453 two. Laonicus Chalcocondyles of Athens and Critobulus of Imbros, are
self-conscious imitators of Thucydides. The imitation is not very successful; but as far as their language
is concerned both employ classicizing Greek in a pastiche of the language and style of Procopius. The
other two, Ducas and George Sphrantzes write a more popularizing Greek. 103 It would be too much to
say, as some have said, that they make a break with tradition. The morphology, syntax and vocabulary
of both is essentially Ancient Greek. not Modern Greek-and by the fifteenth century this was the nub of
the matter. Ducas, who is the more uneven of the two, occasionally uses the dual number or calls places
by their ancient names. Sphrantzes prefers the obsolete iillw to xwp,~ and -rr"Aijll. But both aim at simple
sentence structure. They avoid archaic words and do not hesitate to call modern things by their modern
names. They often use a syntax which is not so much that of living speech as of rather loose, popularizing
Koine. Thus there is frequent confusion between €v+ dat., and Ei~+ acc., and between local adverbs of
motion and rest. Nominative absolute participial phrases-one of the hallmarks of popular Koine since
the New Testament-are common. Sphrantzes occasionally uses on + infinitive and iVa + infinitive.
Both constructions arc attested in a number of early popular Koine texts. Both use the inflected partici-
ple in an imprecise way, e.g. Ducas can begin a sentence with TOTE €M6vra~ Kat Ila(}wv d oovt(40.7). Of
the two Ducas deviates rather more from the nonns of classical Greek. Both were probably animated by
an urge to communicate and an unwillingness to spend time and effort on the niceties of' Atticism' when
their world was falling in ruins about them. But it seems that they turned not so much to the spoken
language for a model, nor to its reflection in the vernacular poetry of their own times, as to the 'popular'
tradition of historiography, that of the chroniclers from Malalas through Theophanes and George the
Monk to Zonaras and Manasses. 104 They were men of the book, and they imitated earlier books rather
than trying to forge a new linguistic medium for Ihemselves. What else could one expect them to do?
Like all medieval writers of classical Greek they occasionally make slips which betray the language in
which they habitually thought. Thus Sphrantzes has 0:/10 and -rrpo with the accusative very occasionally.
Such rare features as these are results of error, not of intention.

VII

It will be seen that the tradition of literary language inherited by the post-Byzantine Greek world was a
very complex one, with varied correlations between linguistic form and subject matter, situation, audience,
literary genre, and so on. A straightforward confrontation between archaisers and demoticists never took
place. To see Byzantine literary usage in that light is to distort history. Rather we should look for
similarities and differences in the development of vernacular literature in the Byzantine world and else-
where, similarities and differences between the Byzantine situation and other examples of diglossia,
medieval and later. It is as a contribution to such a study that the present paper is intended.

If I have to chance my ann by making a premature value jUdgement, I would suggest that classicism was a
positive feature of the Byzantine literary tradition, making for increased expressiveness and flexibility
and strengthening the very real sense of continuity and identity in Byzantine society, until about the tenth
century. Thereafter it became more and more a hindrance to clarity of expression, a barrier to communi-
cation. and a divisive rather than a unifying factor in the consciousness of Byzantine society. If this is so,
the turning point-about the early eleventh century-falls in a period in which many historians have seen
the beginning of the decline in Byzantine political and economic power, and in the creativity of Byzan-
tine culture. Others have described the crucial change in terms of the definitive triumph of feudalism in
the Byzantine Empire. But these are large questions, too great for the present occasion and the present
writer.

128
 The Language of Byzantine Literature 129

NOTES

I. R. M. Dawkins in N. H. Baynes and H. St. L. B. Moss, (edd.), Byzantium, Oxford 1948, ~56; P. Helms, Helikon 11/12
(1971/2) 311.
2. F. W. Lenz, Funf Reden Thomas Magisters, Leiden, 1963, vii-xv.
3. C. Bohlig, Untersuchungen zum rherorischen Sprachgebrauch der Byzantiner, Berlin 1956,90,184-185.
4. Cf. most recently J. Ellis and J. Ure, The Contrastive Analysis of Language Registers, to appear in W. Nemser (ed.),
Trellds in Canstrastive Linguistics (Pre-publication draft, Edinburgh 1974), where a full bibliography will be found.
5. Constantine Porphyrogenitus, De Administranda Imperio 1.1. Ei Of oa¢ei Kal KaTTlIJ~fUlJfV"" :>..&-y"" Kat OWl'. EiKiI
pfovn 7fEItf! KcU G.7fAoU<tf! 7fPO<; nil' rwv 7fPOKEf4.lfJIWV €XPl1ac4<l1v Ii~Awow, 1J11Mv OavlJaoll~, vU. Oti rap f7f~Et~'V
KUAA'rpa¢w.~ ~ ¢paaEw~ i]rrIKtolJfV1/~ Kal TO li<l1PIJ£VOV li'arKoVa~ Kal Vl/ll1AOV 7fOtr,aG.< fa1fOOOaOa, aAM pUAAov
OtO KOwil~ Kal KaOwlJ'Al1lJfV1/~ G.1farrEAw.~ litlia~a, 00, {01ffUOa, ii1fEP oiO{.w,< IiEW aE lJiI G.rvaEw, Kal ii Till' €K lJaKPU~
flJ7fnpw.~ oovw':v n KcU ¢pJIY,law ftilJapw~ aOt ovvara, 1fPO~EVfW. The translation is that of R. J. fl. Jenkins.

6. Constantine Porphyrogenitus, De Caerimaniis 5.2.4. 'n~ iiI' /if aa¢ii Kat EtilitOrvwara dfv ra rErPalJlJEva, Kal KaOW-
lJ'Al1lJfV!l Kat G.7fAOUOr€PIl ¢paof! Kfxp~IJEOa Kal Al~w, rai~ aUra,"i; Kat ovOlJao, TOi~ e¢' fKaaT"" 1fpawan 7faAG.<
trpoaaPlJoaOda, Kat AqOIJ£va,~.
7. Leo VI, Tactica, Migne PC 107, 1093A. Oun KaAAouc; Af~fw~ ¢pOVTl'oavrf~, W~ fV apxil 1fOV Tjpiv «Pl1ra, TaU auvrdr-
lJaT~, oun T-i)~ 1iAAl1~ flJ7JEptlprou avvra~fw~, G.AA· ~ lJovav /if G.7JAa,"i; Kal Kowai~ XP110c4<fVO' ra,"i; Af~fO"', Kat
d7JOw.,~ IJUAAOV Tj orpanwrIKTJ OV~OfUl xp-i)aOa, ErwOfV.

8. Maurice, Strategicon, Proem. 3, p. 44.4-8 Mihaescu. <l>pdoEW~ IJ€V auv iLKptl3aU~ i\ KOIJ7JaV pT/lJaTwv TjIJ;V, w~ etTfOIJEV,
otiOT/IJw. ¢povr,~, oti/il rap fprOV 01' (fPOV, trpawaTwv liE lJuXXOV KcU ovvrolJw.~ A6r~ rfrOVEV' o'tlEv Kal 'PwlJaii<al"i;
7fOAAaKt~ Kal aAAa,~ fV oTpanwTIKV aVV1/0Etlz TETPf4.llJfVG.t~ xp-qIJEOa Af~W' liui Till' aa¢-i) TwvfvrfV~OIJ€VWV KanlAl1 1/1""
9. Nonnos Theophanes, Epitome de curatione morborum, ed. J. S. Bernard, cited by L. Cohn, Bemerkungen zu den Kon-
stantinischen Sammelwerken,Byzantinische Zeitschri[t 9 (1904) 156: fTf!Tf4.l-qon M TjlJwo"lil n" rwv /JIKpd AOr<IW-
Oat bVVUlJfVWV£vrE Ttf! 7Japovr, 7JoVJi/JUT! dvO/JUal' Tf Kat Ihi/JUa, To(~ (n7~ cod.) f~ aropu~ Kal Tpwliov (TptWbf~
cod.) bU'v..T/IJI.lEVO" (-WI' cod.) KfXPl1lJfVO<~ opwv- uti rap arvot{z Af~fwv TWl' KaAA,uTwV Ka' OVV1/0wTciTwv ro(~
"EAAl1a, {3ap{Jdpo" dvOlJIl.o, Kal bU'¢8aPIJEVO'~ lOT'" o"ll Kfxp-qIJEOa - aTOtrov rap KOlJlbfi IJfrPUl TfatliroOlvra~ -, aXX'
IJrrlp TaU 1J11Mv IiUlTfWEW Kat arV01/0iwG.< rtf! (TO cod.) IJTJ TfClJmiTfaal Tfat1i<E>w.~ Tir; 'EAAl1vIKir; IJEraaXDVT!, OVVT/Of-
OTaT)) Af~f! Ka, KaOWIJ'AT/IJ€V!l TOUr"" IilhvwlJEV TfapaOoUVG.< TTJV Elol1ow Ka,' KaTaA171/1w TaU TfPOKEI./.,(vov aKOtroU.
On Theophanes, whose work has been little studied, cf. P. Lemerle, Le premier humanisme byzantin, Paris 1971,296.
10. Cecaumenus 76. P. 272 Litavrin. 'Erw rap iipOtP~ Eip, A&-yOU' 0"
rap 7Jatli£w.~ {AAl1vIKil~ fV O)(oAiI rhova, iVa
arpo#w Mrwv 7JOP"ow/JG.t Ka, EtTyAWTTUw IitliaxOw. Kal ollia, on fTff4.lflJl/lovral· lJoU TLVE~ IipaaaOIJEva, nlv ilpaOtllv
IJOU, itAA' frw otlx W~ TfO''17IKOV Tooro OWf1a~a 7JpO<; aXXOlX; rtvci~, G.AM TfPO<; ae Kat TOVe; b.1iEA~ oov, ToV~ flJobo;
7JaUia~, TOU<; fK TWV f/1WI' arrAdrxvwv, dfx: a &0<; 1J0t bfSWKfV. kvvlra~a lil ravra 0"
KOIJl/lo(~ 7101 Mrol~ Ka, aEao-
¢WIJEVO'~ 1JiJ(}0,~ lJovav KallJl1Mv araOov {xovaw, aAM Tailra f~EOllJl1V Ii n frro,r,aa Kat {"aOov KcU dliov KalllJll.Oov,
Tfparpara o.A1/Oil, a KaO' fKciaTTlv TjlJfPav TfparrovrG.< Ka' rwovrG.<. Kal Ei Taxa U,,",",TIKOt' flaW 0/ Arr,Ol, TfAilv fav
tTy'W~ TfpoaEX1I~ TOr~ AErOIlEVOI~, aA1/OwTaTouc; f"p,jaf".

II. Migne, PC 66.1561B nVEvlJa SEWV VrrEP¢oPP-IJIKpOAarw.v avrwa¢IK,jv.

12. Mark the Deacon, Life of Porphyry 3.1 Off. kunpal/lw /if Till' €KfWOV EV¢l1IJUw aU KOIJTftf! A&-y",,' olm rap i) Ka:>"A':>'"
E~W. rov l1tov TWV TOWVrWV KOalJfW fiwOfV, aAA' Tj UpETii TWV EprWV KcU aVrilv Till' ¢paaw IJEraMvEI.
13. Leontius, Vita Ioannis p. 3 Ceker iVa rei> €VVTfdpXOvrl TjlJiv TfEIei> Ka'
iLKaAAWTfIOIJEv"" Kat XUIJl1:>"ei>XClpaKTijp, S<'1r11_
OWIJEOa fk TO IiwaaOat Kat TOV U,,wT17V Kat arpa..lJaTOt' fK TWi; AqOIJ€VWV W¢E A170ilVG.<.
14. H. C. Beck,Kirche und theologische Literatur im byzantinischen Reich, Munich 1959,640.
15. Nicetas Choniates, Orationes et epistulae, ed. 1. van Dieten, Berlin 1972, 170.
16. Eustathius ad Odyss. 1748.52; Michael Choniates ed. Sp. Lampros, ii, p. 44.25-26.
17. On the distinction cf. A. Mirambel, Diglossie des demiers siecles de Byzance, in Proceedfngs of the XIllth International
Congress of Byzantine Studies, Oxford 1967,309·313.

129
130 Robert Browning

18. Among older works S. Psaltis, Grammalik der byzanlillischen Cizronikell, Gottingen 1913 stands out by the complete.
ness of its information. From the interwar years we have E. Renauld's vast but somewhat disappointing Elude de la
langue el du style de Michel Psellos, Paris 1920, S. von Stepski·Ooliwa's Studien zur Syntax des byzantinischell Histori-
kers Georgios Phrantzes. Munich 1935, and F. Hormann's Beitriige zur Syntax des lohannes Kinnamos, Munich 1938.
Scandinavian scholars-building upon the work of Liifsledl and others in Latin-have done most valuable work on late
Greek, but they have generally dealt with particular problems rather than providing an exhaustive descript:on of the
language or of an aspect of the language of a particular writer. Noteworthy among their contributions are D. Tabacho·
vitz, Sprachliche und lextkritische Studien zur Chronik des Theophanes Confessor, UppsaJa 1926, and t'tudes sur Ie
grec de Ia basse epoque. Uppsala 1943. H. Ljungvik, Beitriige zur Syntax der spiilgriechischen Volkssprache. Uppsala
1932, S. Linner, Synlaktische und lexikalische Sludien zur Historia Lausiilca des Pal/adios, Uppsala 1943, and K. Weier-
holt, Sludien im Spraclrgebrauch des Malalas, Oslo 1963. Among the most recent studies of medievallilerary Greek
are E. Mihevc-Gabrovec, Eludes sur la syntaxe de Ioannes Moschos. Ljubljana 1960, K. Mitsakis, The Language .of
Romanos Ihe Melodisl, Munich 1967, and the linguistic surveys forming part of recent editions of early demotic texts,
such as S. Krawezynski. '0 1l0UAOAD-yO(, Berlin 1960 and V. Tsiouni, llatotoq'>paoTO( Otrj'Y1jOt<; TW" ~';'W" TW" 7fT-
panoow", Munich 1972. Finally mention should be made because of its exhaustiveness ofO. Lampsides, Td prj/lam
f" T4J AE~tAO')'t<';T1i<; Xpo"udl<; kWOl/ifW<; Kw"oTa~wOU TO;' Mwaoa1j, Ilu~a~wd 5 (1973) 187-268.
19. The XIVth International Congress of Byzantine Studies at Bucharest in 1971 appointed a commission for the prepara·
tion of a project for a Lexicon of Byzantine Greek, the chairman of which is Professor Anastasius Bandy.
20. By far the best summary account of the history of Greek literary prose language until the end of antiquity is A. Wif·
strand, Det grekiska prosasproket, Eranos 50 (1952) 149-163.
21. On the Koine cf. A. Thumb. Die griechische Sprache im Zeitaller des Hellenismus, Strassburg 1901; L. Radermacher,
Koine, Sitzungsberichte der oSlerreichischen Akad. d. Wiss .. Phil.·his!. KI., 224, 5 (1947): A. Oebrunner, Geschichte
der gnechischen Sprache II, Gnmd[ragen und Grundzuge des nachk/assischen Griechiseh, Berlin 1954. There is a vast
literalure on the language of the New Testament, the Septuagint and certain other early Christian lex IS, to which refer-
ence will be found in F. Blass and A. Debrunner. Grammatik des neuteslamentlichen Griechisch. 12 te Aunage mit
einem Ergiinzungsheft von D. Tabachowitz. GOllingen 1965.
22. On Atticism cf. W. Schmid, Der Allizismus in seinen Hauptvertrelern, 4 vols., Stuttgart 1887-1897. There are many
more recent studies of particular authors, which vary in scope and value. A very full bibliography is to be found in
Jaakko Fro";n, Prolegomena /0 a Sludy of the Greek Language in the First Centuries A.D., Helsinki, 1974,231-288.
13. M. J. Higgins, The Renaissance of the First Century and the Origin of Standard Late Greek, Traditio 3 (1945) 49-100;
G. Anlauf, Slandard Late Greek oder Attizismus, Diss. Koln 1960.
14. M. Arnim. De Phi/om's Byzanrii dicendi genere, Diss. Greifswald 1912; J. L.lleiberg, Uberden Oialekt des ArchinJedes,
Neue lahrbucher, Suppl. 13 (1884) 542-566; P. Melcher, De sennone Epicteteo quibus rebusab A ttiea regula discedal,
Diss. Halle 1906; G. Ghedini. La lingua greea di Marco Aurelio I: Fonelica e morfologia, Milan 1926.
25. H. Zilliacus, Zur Abundanz der spalgriechischen Gebrauchssprache [Societas scientiarum Fennica. Commentationes
II umanarum Litterarum 41.2 J, Helsinki 1967.
26. H. Steneker, IlEtBOV<; 01j/lIOUP')'w', Observalions sur la fonction du slyle dans Ie Protreplique de Clement d'Alexandrie,
Nijmegen 1967, where references to the earlier literature will be found.
27. J. Borst, Beitrage zur sprachlich-stylistischen und rhelorisehen Wurdigung des Origenes, Diss. Munich 1913.
28. C. Fabricius, Der sprachliche Klassizismus der griechischen Kirchenvater: ein philologisches und geistesgeschichtliches
Problem,Jahrbuch fur Anlike und Christen/urn 10 (1967) 187-199.
29. For a study of the classicizing language of the Fathers of C. Fabricius, Zu den lugendschriften des lohannes Chrysos-
lomos: Untersuchungen zum Klassizismus des vierten lahrhunderts, Lund 1962.
30. A. Wifstrand, Fomkyrkan och den grekiska bi/dningen, Stockholm 1957,65ff.
31. P. J. Bratsiotes,'H ou! /lfOOU TW" aiw"w" fmjl.wOL< TW" TptW" ifpapxW", 'Enfr1jp<c flwEnWT1j/lWu 'A81jvw", July-
August 1931; cf. also the indices to Wall, Rhetores Graeci, Stuttgart 1834.
31. Gy. Moravcsik, Klassizismus in der byzantinischen Geschichtsschreibung, Polychronion. Festschriji F. Dolger, Heidel-
berg 1966,368.
33. Averil and Alan Cameron. Christianity and Tradition in the Historiography of the later Roman Empire, Classical
Quarterly NS 14 (1964) 316·328.

130
 The Language of Byzantine Literature 131

34. On all problems concerning Agathias cf. Averil Cameron,Agathias, Oxford 1970.
35. O. Veh, UllierSlichullgm zu dem byzantinischen Historiker ThcophFiaktos Simokattes, Furth in Bayern 1957.
36. A. Wifstranu, Det ~rekiska prosasprokel, 1o'ral1os 50 (1952) 162-163.
37. The valuable Cambridge doctoral thesis of G. Giangrande on the language of Nonnus, which is a mine of material on
the linguistic usage of the poels of late antiquity, has unfortunately not yet been published. On the popularity of
classicizing poelry in lale Roman Egypl cf. A. Cameron, Wandering Poets: a literary movement in Byzantine Egypt,
/IislOria 14 (1965) 470-509.
38. Texts in E. Heitsch, Die griechisc"el1 Dichter/ragmenle der romischen Kaiserzeit, Gottingen 1961, 127-152.
39. Suda s.v. MapUlvO<;.
40. P. Maas, Der byzantinische Zwblfsilber, Byzan/inisclle Zeilschri/t 12 (1903) 278-323.
41. The London doctoral thesis of J. D. Frendo on the language and style of George of Pisidia has unfortunately not been
published.
42. P. Maas, Das Kontakion, Byzantinisehe Zeilschri/t 19 (1910) 285-306; K. Mitsakis, Bu\avT<VT) ufJ.vO"fpa¢<a I, Thessa-
laniki 1971, 171-353.
43. K. Mitsakis, The Language ofRomanos the Melodisl, Munich 1967, in the index to which will be found references to
most of the features listed below.
44. P. Lemerle, Le premier humanism byzantin, Paris 1971,74-108 provides a balanced and sensitive discussion of the cul-
ture of the Dark Ages, with copious references to the specialist literature.
45. On the language of Theodore Studiies, cf. P. Lemerle, op. cit. 123-128; P. Speck, Theodoros Studiles: lamben au/.'er-
schiedelle Gegenstiinde. Berlin 1968,95-99.
46. Cf. D. T abachowitz, Sprachliche und tex tkritisc"e Sludiell zur Chronik des Theophanes Confessor, Uppsala 1926.
47. P. J. Alexander, The Palriarch Nicephorus o/Conslantinople, Oxford 1958,156-188.
48. R. Browning, Ignace Ie Diacre et la tragedie c1assique a Byzance, Revue des etudes grecques 81 (1968) 401-411; Wanda
Wolska-Conus, De quibusdam Ignaliis, Travaux et memoires 4 (1970) 329-360.
49. G. Hartmann,Pholios' Lilerariisthetik, Diss. Rostock 1929,4-12.
50. Gertrud Il6hlig, Ulilersuchungen zum rhetorischen Sprachgebrauch der Byzantiner mit besollderer Beriicksichtigung
der Schri/ten des Michael Psellos, Berlin 1956, 1-17; id., Das Verhaltnis von Volkssprache und Reinsprache im grie-
chischen MittelaIter,Aus dcr byzantinislischen Arbeit der DDR 1, Berlin 1951,1-13, and in parlicular 3-4.
51. T. Hedberg,EustathiosalsAttizist, Uppsala 1935,123-134,139-143.
52. R. J. H. Jenkins, The Classical Background of the Scriptores post Theophanem, Dumbarton Oaks Papers 8 (1954)
13-30.
53. C. Kolias, Leon Choerosphactes, magistre, proconsul et patrice, Athens 1939,99-107.
54. R. Schneider and G. Uhlig. Apollonii Dyseoli quae supersunt, Leipzig 1878-1910,2.301.
55. De caerimoniis p. 456 Bonn.
56. H. Zilliacus, Zur stilistischen Umarbeitungstechnik des Symeon Metaphr.stes, Byzantinisehe Zeilsehri/t 38 (1938)
333-350, from which most of the following examples are taken.
57. E. Renauld, Erude de la langue el du slyle de Michel Psellos, Paris 1920; id. Lexique choisi de Psellos, Paris 1920;
Gertrud Bohlig, Untersuchungen zum rhetorisehen Sprachgebraueh der Byzantiner mil besonderer Beriieksiehtigung
der Schri/len des Michael Psellos, Berlin 1956.
58. AnnaComnena, Oxford 1919,481-508.
59. R. Romano,Pseudo-Luciano. Timarione, Naples 1974,25-31.
SO. Most of the material here presented is taken from H. Tode, De Timarione dialogo Byzal1lino, Diss. Creifswald 1912,
44-86.
51. A. P. Kashdan, Kniga i pisatei' v Vizantii, Moscow 1973, 112-119.

131
132 Robert Browning

61a Dr. Margaret Alexiull reminds me that the writers of the twelfth-century learned romances also make a creative and
original use of the classicizing language. The example of India shows that a lively literature can Oourish for a long
period in a language which is nobody's mother·tongue. But post-classical Sanskrit literature is addressed to a closed
coterie or caste.
62. H. G. Beck,Kirehe lIlId the%gisehe Literatur im byzantinischen Reich, Munich 1959,640.
63. John Siceliotes, Welz, Rhetares graw' 6.451.22: John Geometres, ibid. 2.226.11. Cf. G. L. Kustas, Studies in Byzan·
tine Rhetoric, ThessaJoniki 1973,95.
64. M. 1. Jeffreys, The Nature and Origins of the Political Yerse,Dumbarton Oaks Papers 28 (1974) 141-195.
65. S. Lauriotes, '0 "A8we 1. (1920) 12-13,224,228-9 [Texts in Jeffreys 162-163J.
66. M. J. Jeffreys,op. cit. 164-165.
67. M. 1. Jeffreys,op. cit. 152-153.
68. M. J. Jeffreys, op. cit. 155.
69. Y.8.
70. O. Lampsides, Td pryl1um EV r0 Af~<AOrLY nje XpovLK~e Lvvol/lfwe Kwvoravnvou roo Mavaoo~, Bu\uvrwa 5 (1973)
187-268.
71. H. G. Beck, Gesehiehte der byzantinisehen Volksliteratur, Munich 1971, 10 I-I 05.
72. E. Th. Tsolakes, MLXa7]A nUKa onxoL aVc f'ypul/lf KaO' d'v KarWXf01/ xpovov, Thessalonike 1959; H. G. Beck, op. cit.
108-109.
73. G. L. Huxley, Antecedents and Context of Digenes Akrites, Greek, Roman and Byzantine Studies 15 (1974) 317-338.
74. The literature on Digenis Akrites is immense. The best bibliography is that in H. G. Beck, Gesehiehte der byzan-
tinisehen Volksliteratllr, Munich 1971,65-68 and in the notes to pp. 69-96. Several important studies have appeared
since the publication of Beck's book, in particular L. Polites, L'epopee byzantine de Digenis Akritas, Problemes de la
tradition du texte et des rapports avec les chansons akritiques, Atti del Convegno Internazionale sui tema: La poesia
epica e ifl sua formazione, Rome 1970, 551-581; A. Pertusi, La poesia epica bizantina e la sua formazione: problemi
sui fondo storico e la struttura letteraria del Digenis Akritas, ibid. 481-544; E. Trapp, Digene. Akrites. Synoptische
A usgabe der liltesten Version en, Vienna 1971; Niki Eideneier, t:.wpOwrLKci oro KEL)JfVO TOO t:.Lrn'li '*
KPVTfTrxP€PP1/e,
'EAA1/VLKU 23 (1970) 299-319; L. Polites, Digenis Akritas. A propos de la nouvelle edition de I'epopee byzantine,
Senptorium 27 (1973) 327-351 ;and G. L. Huxley's study mentioned in the previous note.
75. L. Polites, L'epopee byzantine de Digenis Akritas (reference in previous note) 571-581.
76. H. G. Beck, Geschichte der byzantinisehen Volksliteratur, Munich 1971, 105-108.
77_ S. G. Kapsomenos, Die grieehisehe Spraehe zwischen Kaine und Neugrieehiseh, Berichte zorn XI. Internationalen
Byzantinisten-Kongress, Munich 1958 II i; K. Kriaras, Diglossie des demiers sieeles de Byzance:naissance de la lit-
terature neo-hellenique, Proceedings of the XIIlth International Congress of Byzantine Studies, Oxford 1967,283-299;
J. Irmscher, Soziologische Erwagungen zur Entstehung der neugriechischen Literatur, ibid., 301-308; Andre Mirambel,
Diglossie des derniers sieeles de Byunce ,ibid., 309-313; E. Kriaras, Bilinguismo degli ullimi secoli di Bisanzio; nascita
della letteratura n e o e lle n ic a , Bolleltino del Centro di Studi di Filologia e Linguistica Siciliana 11 (1970) 1-27; H. G.
le tte r a tu r a neoellenica,
Beck, Die griechische volkstiimliche Literatur des 14. Jahrhunderst. Eine Standortsbestimmung,Actesdu XIVe Con-
gres International des Etudes Byzantines I, Bucharest 1974,125-138; M. J. Jeffreys, The Literary Emergence ofYer-
nacular Greek, Mosaic 8/4 [On the Rise of the Vemacuiflr Literatures in the Midd/e Ages] (1975) 171-193.
78. E.g. cod. Oxon. Barocci 131, cod. Scorial. Y-H-lO, cod. Laur. conv. soppr. 2, cod_ Vat. Gr. 305, cod. Yindob. philol.
gf. 321.
79_ E.g. Maximus Planudes' IIEpi rpUJ1/1UrLK~~ and HEpl aWTa~Ewc in L. Bachmann, Aneedota graeea 2, Leipzig 1828,
3-166; Manuel Moschopulus' 'EPWT-q/1UTa, 'ElTLTOI1>J vEa "PUJ1I1UrLK~~, and LVAAor>J dVOl1aTWv'ArrLKwv, Thomas
Magister's 'Ovol1arwv 'ArrLKwv ovva,),wrry (ed. F. Ritschl, Halle 1832).
80. I. Sevcenko, Etudes sur la polemique enlre Theodore Metoehite et Nicephore Choumnos, Brussels 1962 ;J. Yerpeaux,
Nicephore Choumnos, hommc d'etat et humanistc byzantin, Paris 1959,52-62.
81. IiEpi A/rrWV Kpt'afW( Kal ip,),aaws, ed. J. F. Boissonnade,Anecdola graeca 3, Paris 1831,357.
82. Ibid:, 360.

132
 TheLanguage
The Languageof
ofByzantinc
B) 'zan tincLiterature
Literature 133
133

83.
83. Ibid
Ibid....3(,3
363.
84. Ibid.
84. Ibid.,363
363.
85. TheodorosMe[()chircJ.
Beck,Thcouor<Js
85. H.H.G.G. Beck, Metochiles.Dic
DieKrise
Krisedes
desbl'zanrinischcll
byzalllinischeni1'c/{bildcs
We/lbildesimim14. Jahrhunderl.Munich
14.lahrhundcrt, Munich1952.50-75.
1952.50·75.
86. M.
86. M.Treu, Dichtungendes
Treu,DichlunxclI desGroH-i-ogothclcn
(;ross·Logotheten Thcodoros
TheodorosMc(ochi/cs
Melochiles jProgramm
(Programmdes
desVictoria-Gymnasiums
Victoria·GymnasiumszuzuPotsdam I
Potsdam)
(1895),AA354-358.
(1895), 354·358.
87. Cf.
87. Cf.S.S.Runciman, TheLast
Runciman,The LastByzantine
ByzantineRenaissance,
Renaissance,Cambridge
Cambridge1970.
1970.
88. I.I.Sevcenko,
88. Sevcenko,Society
Societyand
and Intellectual
Intellectual Life
Lifeininthe
the XIYth century,XIVe
XIVthcentury, XIVeCOllgres
COllgresIntanational
Internationaldes
desEtudes
ElUdesByzalllines.
Byzalllines.
Rapports I,I, BuchareSI
Rapports Bucharest 1971,7-30;
1971,7·30;cf. cf. also
also E.E. Werner,
Werner,Gesellschaft
Gesellschaftund KullurUn
undKultur XIVenJahrhundert:
imXIYen Jahrhundert: Sozial.oko·
Sozial-<iko·
nomische Fragen,ibid.,
nomischeFragen ,ibid.,31·50.
31·50.
89.
89. II.II.Hunger,
Hunger,Klassizistische
KlassizistischeTendenzen
Tendenzenininder
derbyz.ntinischen
byzantinischenLiteratur
Literaturdcs 14.lahrhunderts,ibid.,
des14.lahrhunderts, ibid.,83·95.
83·95.
90. Gy.Moravcsik,Byzantinoturciea
90. Gy. Moravcsik,Byzantinoturcica22I,I,Berlin
Berlin1958,220.
1958,220.
91. J.J. B.B. van
91. van Dieten,
Dieten, Noch
Noch cinmal
einmal uber
iiber Niketas ChoniatesByzantiniseilc
NiketasChoni.tes ByzantiniscilcZeitschrift
Zeitschrift57
57 (1964)
(1964) 302·328;
302·328;H.H. G.G.Beck,
Beck,
Geschichteder
Gesehiehre derbyzantinischen
byzantinischenVolksliterarur,
Volksliteratur,Munich
Munich1971,6.
1971,6.
n. Migne,P.
n. Migne,P.G.G. 142,611ff.
142,61 Iff.
93. L.L.O.O.Sjoberg,Stephanitesulld
93. Sjoberg,Slcphaniles zmdlehnelates,
lehnelales,Stockholm
Stockholm1962,122-132.
1962, Ic2·132.
94. Geschichteder
Beck,Gesehiehte
94. 11.II.G.G.Beck, derbyzalltinisehen
byzalllinischenVolkslirerarur,
Volksliteratur,Munich
Munich1971,47.
1971,47.
95.
95. K.K. Praechter,
Praeehter,Eine
Einevulgargriechische
vulgiirgriechischeParaphrase
Paraphraseder
derChronik
Chronikdes
desKonstantinos Manasses,Byzantinisehe
KonstantinosManasses, ByzantinischeZeirsehrifr
Zeitscirrift44
(1895)
(1895)272-213,
272-213,
96. Beck,Geschichre
96. H.H.G.G. Beck, Geschichteder
derbyzanrinischen
byzanlinischenVolkslilerarur,
Volks/ileratur,Munich
Munich 1971:
1971:cf.
cf.also
alsoid.,
id.,Die
Diegriechische
griechischeYolksliteralur
Volksliteratur14.14.
Jahrhunderts,XIVe
Jahrhunderts, XIVeCongres
CongresInternational
Inlernationaldes
desEtudes
EtudesByzantines,
Byzantines,Rapports
RapportsI,I,Bucharest
Bucharest1971,67-82,
1971,67-82,and andreviews
reviewsofof
Beck'sbook
Beck', bookbybyE.E.Trapp, Jahrbuchder
Trapp,lahrbueh deroSle"eiehischen
oste"eichischenByzantinisehen
ByzantinischenGesellschaft
Gesellscha/t22 22(1973) 3 I 5·319,J.J.Irmscher,
(1973)315·319, Irmscher,
DeutscheLiteraturzeitung
Deutsche Literaturzeitung9494(1973)
(1973)322·324,
322·324,A.A.P.P.Kazhdan, VizantijskijVremennik
Kazhdan,Vizanlijskij Vremennik34 34(1973)
(1973)278·284,
278·284,G.G.Yeloudis,
Veloudis,
SUdost·Forschungen32
Siidost·Forschungen 32(1973) 506·510,J.J.Koder,
(1973)506·510, Gnomon46
Koder,Gnomon 46(1974)
(1974)508·510.
508·510.
97. Browning,Medievaland
97. R.R.Browning,Medieval andModem
ModemGreek,
Greek,London
London1969,76.91.
1969,76·91.
98.
98. M. Pichard,Leroman
M.Pichard,Le romande
deCallimaque
Callimaqueeletde
deChryso"hoe,
Chryso"hoe,Paris
Paris1956,
1956,xv·xxxi.
xv-xxxi.
99. M.
99. M.Pichard, op.cit.
Pichard,op. cit. xxv·xxvi.
xxv·xxvi.
100. E.E. Kriaras,
100. Kriaras, Die
Die Bcsonderheiten
Besonderheiten der
der letzten
letztenPeriode
Periodeder
dermittelalterlichcn Literatur, Jalzrbuchder
griechischen Literatur,lahrbuch
mittelalterlichengriechischen derOster·
Oster·
reichischenByzantinischen
reichisehen ByzantinischenGesellsehaft
Gesel/schaft88(1959)
(1959)69-85.
69-85.
01. R.R.M.
101. D~wkins,Leolllios
M.D;wkins, LeontiosMakhairas,
Makhairas,Recital
Recitalconcerning
concerningthe
thesweet
sweetLalld
Landof
ofCyprus
Cyprusentitled
entitledChronic/e,
Chronicle,Oxford
Oxford1932.
1932.
02. Y.V. Laourdas,
102. Laourdas,'lotSwpou iJ.pXlETflO.07rOVeEOoa"ovu(11~
'!ol8wpovapXLElTw.OlTOU e.aaaAov""l~Opt""" Ek rd~
O!JlA.a...k rd~ ,oprd,
foprd~rou
rouirrwu
a'l'wvt>T1PT/TPWU,
l>1)!J1)TPWV,Thessaluniki
Thessaloniki1954.
1954.
03. On
103. On the
the language
language cf.
cf. Stephanie
Stephanie von
von Stepski·DoUwa, Studien zur
Stepski·DoUwa,Studien zur Sylllax
Sylllax des
desbyzantillischen
byzantillischenHistorikers
Historikers Georgios
Georgios
Phrantzes, Munich
Phrantzes, Munich 1935;
1935;ononthat
thatofofDucas
Ducascf.cf.M.
M.Galdi,
Galdi,LaLalingua
linguaee1010stile
stiledel
delDueas,
Ducas, Naples
Naples 1910
1910(inaccessible
(inaccessibletoto
me);E.E.Chernousov,
me); Chemousov,Duka,
Duka,odin istorikovKonts.
odinizizistorikov KontsaYizantii, VizalltijskijVremennik
Vizantii,Vizalltijskij Vremennik2121 (1914)
(1914) 171·221;
171·221;Y.V.Grecu,
Grecu,
Pourune
Pour unemeilleure
meilleureconnaissance
connaissancededeI'historien
I'historienDoukas, MemorialLouis
Doukas,flfemorial LouisPetit,
Petit,Bucharest
Bucharest1948,
1948,128·141,
128·141,ininparticular
particular
138·139.
138·139.
04. H.H. G.G. Beck,
104. Beck,Zur
Zurbyzantinischen
byzantinischen Monchschronik, Ideell und
Monchschronik,ldeen undRealiriiten
Realitiiten ininByzanz.
Byzanz. Gesammelte
GesammelteAufsitlze,
Au/sittze, London
London
1972,No.
1972, No.xvi.
xvi.
05. Hel~ne
105. L'ideologiepolitique
Ahrweiler'sL'ideologic
Helene Ahrweiler's po/itiquededei'empire
{'empirebyzantin,
byzantin,Paris
Paris 1975,
1975,draws
drawsattention
attention totomany
manyfeatures
featuresofofthe
the
Byzantineself·image
Byzantine self.imagewhich
whichmay relevanttotochanging
maybeberelevant changingattitudes
attitudestoward
towardlanguage.
language. Cf.
Cf.ininparticular
particularpp.
pp.60.74,
60·74,107.
107.
114,115·118.
114,115·118.

133
133
 Oral Tradition, 1/3 (1986): 504-47

The Oral Background

of Byzantine Popular Poetry
Elizabeth and Michael Jeffreys

The popular poetry of Byzantium first appears in the form of

consistent surviving texts of some size in the middle of the twelfth
century, at the courts of the emperors John Komnenos (1118-1143)
and Manuel Komnenos (1143-1180).1 Little or no such poetry
seems to survive from the thirteenth century, when Byzantine
energies were occupied in the reconquest of Constantinople and
other parts of the empire from the forces of the Fourth Crusade.
The next preserved examples seem to date from early in the
fourteenth century, and the popular poetic tradition then continues
through to the end of Byzantium in 1453 and beyond. Insofar as
evidence permits us to speak about the places in which this
material was composed and written down, it seems that the
western-ruled states surviving from the Crusades on Greek lands
were at least as fertile ground for its production as the area ruled
by Byzantium, under the last dynasty of the Palaiologoi. The
total volume is not large, but it covers a number of genres. We
shall discuss in the conclusion of this paper the difficult question of
the continuity of this tradition in Greece under Turkish rule.
These texts may be regarded for most purposes as the first
preserved material of any length in Modern Greek, a language
which bears much the same relationship to ancient and medieval
Greek as does Italian to ancient and medieval Latin. The
linguistic pressures of Byzantium are not dissimilar from the early
history of many Western European language groups: the steady
development of spoken Greek is hidden from us by the
conservatism of writing, which made efforts to keep up the illusion
that Greek had not "declined" from its great past-the classical
Greek of the fifth and fourth centuries B.C. and the "Koine" Greek
of the New Testament and the Septuagint version of the Old

134
 BYZANTINE POPULAR POETRY 505
Testament. Nevertheless, linguistic pressure of change in popular
speech was building up behind a dam of the linguistic censorship of
Byzantine education? That dam was first seriously breached in the
twelfth century by the popular poetry which is the subject of this
article.
Byzantine popular poetry has not been much studied and
more rarely still within a useful conceptual framework.
Byzantinists may be tempted to regard it with the same
contemptuous eye that Can be observed among contemporary
Byzantine intellectuals on the few occasions when they deign to
notice poetry in popular linguistic and metrical forms. The
information which may be derived from these poems about
Byzantine history and society is quite considerable, but needs
analysis of a rather different kind from that appropriate to legal
documents or learned historians, and so tends to be discounted as
unreliable. This poetic genre is also studied in the first chapters of
histories of Modern Greek literature (e.g., Dimaras 1965, Vitti 1971,
Politis 1973), and in introductory lectures to university courses in
the same subject. In this case the dangers are obvious: it is very
difficult to view these poems within a meaningful context, whether
that of all Greek literature being produced at the time or the
wider context of contemporary European literature.
There is even a problem now of national identification of
Greeks with these first products of Modern Greek literature. There
has always been some reluctance in Greece, dating from before the
Revolution of 1821 which created the Modern Greek state, to
accept as an integral part of the Greek heritage the culture of the
Byzantine period. For the extreme nationalists, there was a
tendency to speak and write in a way which minimized the gap
between 1821 and ancient Greece, as if the death of Alexander the
Great were one of the last significant events in Greek history
before the Revolution. More progressive forces saw in Byzantium
the epitome of all that was wrong in the modern Greek state, and
preferred to view it as a kind of extension of Roman occupation
ra.ther than something essentially Greek (see Fletcher 1977). Mter
all, the Byzantines called themselves Romans and reserved the
word "Greek" for ancient pagans. The nature of the popula.t
poems themselves, as we shall discover, does not help in national
identifica.tion. The epic Digenil Akritas, ha.iled a.t its discovery as
a Greek Roland or Cid, is ambivalent about its allegiances, since
the hero has a.n Arab father and fights more Christia.ns tha.n

135
506 ELIZABETH & MICHAEL JEFFREYS
Arabs. Other poems are translations or adaptations of French, or
in one case Italian, originals. Worst of all, the Chronicle 0/ the
Morea is the foundation epic of a Western state on Greek soil,
speaking at length of the faithlessness of the Byzantine Greeks who
were the main power against which the Moreot principality had to
defend itself.
Byzantine popular poetry thus has fallen outside the normal
frameworks of literary understanding, being regularly ignored in
discussions of European vernacular literature without finding a
secure place within a Greek national context. The writers of this
article, together with several other students of the genre, have
recently attempted a re-evaluation, trying to find ways of
developing a framework for studying it and of understanding the
purposes of those who chose to compose in it.
The most important element in this re-evaluation is the
proposal that we should see in the popular poetry of Byzantium
the written remains of a tradition of oral poetry.3 This statement
must be very carefully qualified. We think it most unlikely that
any of the surviving texts are the verbatim record of creative oral
performances, taken down by the methodology of the "oral dictated
text" (see Lord 1953) -though in one or two cases this possibility
cannot be excluded, as will be discussed later. We believe, on the
other hand, that it is almoSt impossible to explain many features of
the language, meter, and style of this genre of poems without
assuming that they derive in a fairly direct way from a language,
meter, and style developed by oral poets for use in oral poetry.
We would suggest, therefore, that Byzantine popular poetry was
produced by means which approximate to those of conventional
literature, but in a genre most of whose products were orally
composed and disseminated. This genre was the only one available
to poets who wished to write in a way which would be
immediately intelligible to the uneducated majority of their
audiences.
However, before seeking to support this proposal, it is only
fair to point out how little direct evidence there is for it and how
dependent it is on theoretical arguments by analogy with other oral
traditions. Students of medieval Greek can only envy the vast
mass of material available to scholars working in medieval French,
not the least because of the opportunities it provides for defining
the profession of jongleur, particularly from the direct statements
made within the texts of the chansons de geste and by

136
 BYZANTINE POPULAR POETRY 507
extrapolation from the practices observable there (see Duggan
1984). Greek evidence of this kind is extremely limited; perhaps
the best example is the following from the Chronicle 0/ the Morea:

If you desire to hear of the deeds of good soldiers, to

learn and be instructed, perhaps you will make progress.
H you know letters, start reading; if, on the other hand,
you are illiterate, sit down by me and listen. And I
hope, if you are sensible, that you will profit, since
many of those who have come after them have made
great progress because of the stories of those great men
of old (lines 1349-55).4

Other minstrels' comments in this genre are little more than

formalities, like the first words of Belthandros and Chrysantza:

Come, listen for a moment, all you young people. I

want to tell you some very beautiful stories, a strange,
most extraordinary tale (lines 1-3; Kriaras 1955, our
trans.).

The pattern of narration set is that of one storyteller to an

audience whose interest needs to be aroused. We have no way of
checking that this pattern represents the physical reality of
performance rather than a convenient narrative fiction.
Nevertheless, the singing of songs seems to be a significant
feature of the life described in the songs. Digenis Akritas, for
example, sings several songs to his beloved, and takes a musical
instrument rather than a weapon when he sets out for adventure
(Trapp 1971: Ms E 711-12). Five songs are included in the long
text (Ms N) of the Achilleu, and we are told that many more are
sung in celebration of Achilles' first victory./) Libutros and
Rhodamne too is full of short love songs. s In the Romance 0/
A,oUonios 0/ Tyre the ability to improvise saves the virtue of
Apollonios' daughter-though many of the details of the episode
derive from the original text of which this is a translation (Wagner
1870: pp. 63-90; 11. 594-97, 601-4). Unfortunately, no Demodo-
koe appears in any of the poems of this genre.
References within the works themselves are not much
supported by external references to singers or minstrels within
Byzantine society. We may begin with the fact that several

137
508 ELIZABETH & MICHAEL JEFFREYS
troubadours formed part of the Fourth Crusade and remained in
Frankish Greece afterwards. One may mention Raimbaud de
Vaqueiras and Conon de Bethune, and a mysterious "Prince de
l'Amoree" (see Longnon 1939). We have also noted two references
to oral poetry profe89ionals in the Frankish states of Greece-richly
clad "juglars" at a ceremony in the court of Thebes for Guy II de
180 Roche in 1294 and a pair of "menestreux" sent by Thibout de
Cepoys, agent of Charles de Valois, to a wedding among the
Catalan mercenaries whom he was trying to use in his master's
interest. 7
The following list of references concerns singers in a more
purely Greek cultural context. It makes no pretensions to
completene89, but it is unlikely that it could be more than
doubled, say, in length (cp. Beck 1971:50 and Beaton 1980:75-77).
Monks during the iconoclast controversy were mocked by the
associates of the loathed emperor Constantine V Kopronymos
(741-45) to the sound of the kithara, presumably in verse (Vita:
col. 1116). In the time of Michael III (842-67), an improvised
song was used as part of a trick to capture a town (Bekker
1838:72). Next in chronological order comes the most hackneyed of
these references, the complaint ,of Arethas of Caesarea about the
"Paphlagonians" (meaning "windbags," not a geographical
reference), who put together songs about the achievements of
famous men and go round houses singing them for money (Kougeas
1913-14:239-40; Beaton 1980:77). John Tzetzes in the mid-twelfth
century tells the same story of his own day (Leone 1968:ill, n.
218-67). Neophytos, a contemporary Cypriot hermit, heard a
singer singing what sounds very like a modern folk song of exile. s
Niketas Choniates tells us of a song improvised by Andronikos
Komnenos in 1185, as he tried to win over the servants of his
successor who had captured him. He sang in alternation with his
wife and mistre89 (van Dieten 1975:348). Maximos Planudes in his
"Dialogue on Grammar" says that laments in mteen-syllable verse
were sung by "Ionian" women a.t funerals (Bachma.nn 1828:98).
During the civil war between Andronikos II and III, Nikephoros
Gregoras tells us of a journey he made through an area of
terrifying ravines. Some of his company were singing of the "deeds
of men" and the ravines re-echoed antiphonally, in a way he found
appropriate (Schopen 1829:vol. 1, 377). Gregoras later writes of the
sequel of a famous dream of John VI Kantakouzenos (1347-54): a
creator of songs who was present sang a prophecy (ibid., vol. 2,

138
 BYZANTINE POPULAR POETRY 509
705-6; cpo Magoun 1955}. A horoscope from Trebizond (1336)
promises prosperity to singers in their composition to win their
audience's attention, prophesying that December will bring them
eloquence, and probably suggesting that they were improvisers
(Lambros 1916). Further information from an unpublished treatise
of Gregoras had recently been reported by A. F. van Gernert,
together with a published comment by the Patriarch Philotheos
Kokkinos (van Gernert and Bakker 1981-82). The conclusion to be
drawn from the two passages is that at the time of writing (1353)
songs of lament, accompanied by both wind and stringed
instruments, could be heard every day in Constantinople. Michael
Apostolis, in a letter written on the Dalmatian coast in 1466-67, is
reminded of Cretan funeral and festival customs when he hears
antiphonal singing (Noiret 1889:80).
However, even though these lists may be lengthened
somewhat, it is most unlikely that they will ever be fully
convincing about the existence of an oral tradition in Greek lands.
The time-span covered is long, and the geographical area wide.
Although the censorship of Byzantine education can be used to
explain the removal of all non-learned material from Greek writing,
it is an insufficient excuse to prevent a firm negative conclusion.
In our search for direct signs of singers and their songs, we have
found little evidence, scattered over different times, places, and
kinds of song and singer. This serious deficiency demands in
compensation really convincing evidence of other kinds to make the
case for the existence of an oral tradition.
The first part of that case derives from a consideration of the
meter which is employed in nearly all popular Byzantine poetry,
and its connection with the first appearance of poems consistently
written in Modern Greek. It is usually called the ftfteen-syllable
or dekapentasyllabos, less often the political verse (a name derived
from a puzzling label given by some metrical scholars of the
Byzantine period).9 The meter is sometimes called the national
meter of Modern Greece, because it has dominated Greek folk-song
since at least the last century of Byzantium and probably long
before, and has been used by most Modern Greek poets, often for
their solemn poetry on national themes. We have spoken above of
a general lack of national identification among contemporary Greeks
with Byzantine popular poetry. That indifference does not extend
to its meter or, of course, to its language.
It is no accident that language and meter are extremely

139
510 ELIZABETH & MICHAEL JEFFREYS
closely connected in the manuscript evidence available to us. The
overwhelming majority of Modern Greek surviving in complete
texts from the Byzantine period is in fifteen-syllable metrical form.
Prose and other metrical patterns are both equally rare. It is no
exalleration to say that Modern Greek ftl'l!Jt broke the dam of
linguistic censorship in this metrical form: it is logical, therefore, to
examine the social and educational connotations of the form to see
why it was so successful (see further M. Jeffreys 1974).
The first observation to be made is that the metrical form
was used for writing at more learned levels of language well before
the mid-twelfth century, the date of the first Modern Greek
vernacular poetry. The first dated specimens of the verse in which
identification is secure were written in the year 913, a lament for
the death of the emperor Leo VI in respectable Byzantine Greek. lo
It is interesting that the text is not purely fifteen-syllable: there
are some half-lines too, which suggest that the form is not yet
fully stable. Numerous attempts have been made to trace the
earlier history of the verse, but none is fully convincing (summary
in M. Jeffreys 1974:146ff.; see also Politis 1981, Lavagnini 1983, and
Luzzatto 1983). This metrical shape, for example, is certainly
common among the multifarious rhythmical patterns of the hymn
called the kontGkion (Koder 1983), but it is difficult to say whether
any conclusion -even the most tentative-can be based on that
fact, granted the possibility of coincidence. Origins for the
fifteen-syllable pattern have been suggested in a variety of different
periods and in Latin as well as Greek tradition. It is certainly
striking that the most common rhythmical form of Latin medieval
verse is also a fifteen-syllable form with a reversed accent pattern
(trochaic rather than iambic), and the second and third most
common stress meters of medieval Latin and Greek are similariy
mirror-images, with the same syllable numbers-a twelve-syllable
and an eight-syllable pattern (M. Jeffreys 1974:191-94).
However, for those who place emphasis on surviving texts,
there exists a prima faeit case that the fifteen-syllable was
invented at a linguistic and educational level above that of the
vernacular poems, and that it spread downwards and outwards
from the social and educational center of Byzantium: from the
imperial court in the tenth century to folk song in the fourteenth,
when it first becomes possible to make secure predictions about the
form of Greek folk song (see, e.g., Politis 1970:560-63, Koder 1972,
and Baud-Bovy 1973). But to anybody with a knowledge of

140
 BYZANTINE POPULAR POETRY 511
medieval society in general and of Byzantine society in particular,
such a judgment is self-evidently dangerous. Byzantine culture
was centralized to a most disturbing degree around the city of
Constantinople and the imperial court, and the popular culture of
the countryside was systematically excluded from its written records
at all stages, with very few exceptions. We know of no rival
metrical pattern likely to attract the ear of the illiterate or
half-literate Byzantine between, say, the sixth century A.D., when
linguistic changes must have completed the destruction of popular
appreciation of ancient meters based on long and short syllables,
and the period of demonstrable popular dominance of the
fifteen-syllable in the fourteenth century (M. Jeffreys 1981). How
are we to react to this gap? Should we assume that folk song
ceased to exist? It is obviously preferable to test the alternative
theory that the primA /Aeie case which we mentioned above is
merely one of many demonstrations of the nature of Byzantine
culture and society, that evidence is preserved in the center long
before the periphery, perhaps even that the culture of the
countryside had to be accepted and written down at the highest
social and educational levels before there was any chance of it
being preserved in a peripheral, rural form. The distribution of the
evidence may thus tell us nothing about the nature of the meter
and its origins.
If one reads carefully the surviving examples of the verse
from the tenth to the twelfth century In the different varieties of
learned Byzantine Greek, one may find some support for this point
of view. When, for example, the verse Is used by some of the
capital's most pedantic literAti, there is often a disclaimer or an
apology. Let John Tzetzes, the most pedantic of all, speak on
their behalf, in his Theo,onfl, a simple mythological handbook in
fifteen-syllable verse addressed to the sister-In-law of Manuel
Komnenos:

You want to know of the Greek and Trojan generals:

anything more is redundant, full of labour and
effort-both for the listeners, and still more so for the
writers, particularly when they have written in playful
verses. For a mind which Is carrying out a gr@at task
wlll often grow numb, when in matters where It should
win praise it seems rather to be providing faults for its
detractors, who have no regard for the fact of

141
512 ELIZABETH &; MICHAEL JEFFREYS
oikonomia. Indeed, forbearing to write the superfluous
facts of heroic genealogy in the writing of apes (7), I
am putting down here clearly the most important
points. The rest needs time· and hexameters and, more
important still, a language that will bring pleasure
(496-508).11

Tzetzes here connects the verse form of his handbook with

slapdash, simplistic work, and feels he needs to defend himself
against those who will attack his choice of medium. The word
oikonomia, it becomes plain in other paBSages, involves a
compromise between the ambition of the author and the demands
for simplicity imposed by his patron (M. Jeffreys 1974:151-53).
Learned works in the fifteen-syllable form cluster largely into
three groups, those connected with imperial ceremonial, education,
and religion (ibid: 173-76). But the more one reads them the more
importance one gives to a fourth element, the vernacular and
popular, represented at the same period by the first Modern Greek
vernacular poems of the twelfth century. The poems connected
with imperial ceremonial were designed to appeal to a large
popular audience, and also to communicate, in some cases at least,
with members of the imperial family who were not educated to the
highest standards. 12 It was easier to write intelligibly in this verse
than in prose, for prose always had to conform to the rules of
ancient prose stylistics. The fifteen-syllable had no ancient models,
and the writer could improvise with his own choice of style and
language. The poems connected with education seem to involve a
good deal of rote learning: grammatical rules and even dictionaries
were put into the fifteen-syllable. What metrical form could be
more likely to stimulate the memory than the meter of a popular
oral tradition? The religious poems are often mystical or
penitential in character: in the first case the inspired words fall
from the poet's lips regardless, almost, of metrical form and in the
second the emphasis is on simplicity and honesty of utterance. 13 It
would be difficult, for example, to express penitence in polished
antique hexameters. In all these cases we would suggest that the
reason for the choice of the fifteen-syllable is likely to have been
its status as the meter of a contemporary oral tradition of
narrative songs.
Although this metrical discussion has been somewhat
compressed, we hope that its general pattern has been clear. To

142
 BYZANTINE POPULAR POETRY 513
sum up: the fifteen-syllable meter which has dominated Modern
Greek folk song since the fourteenth century at least made its
debut in the Modern Greek vernacular in the mid-twelfth century,
and is the metrical form of almost all early examples of the
vernacular. The origins of the verse seem to go far back into the
history of Byzantium, perhaps we)) beyond its fil~t appearance in
the early tenth century. During its written history between the
tenth and twelfth centuries, it was used by some of Byzantium's
driest intellectuals. However, this choice of meter is usually
imposed on them by a patron who wants the commissioned work
to be comprehensible. The writers regularly complain and expect
to be attacked for choosing a meter with such low educational
prestige. In the majority of cases, of course, the meter is used
without explicit comment about the writers' attitudes. Here the
genres in which the poets write are compatible with the possibility
that the meter was simultaneously used in a popular oral tradition.
The discussion is not a comfortable one for a sober historian,
even the historian of literature. On the one hand, there is the
complete absence of direct evidence for vernacular fifteen-syllable
verse before the twelfth century and for any consistent use of the
verse before the tenth century. On the other hand, we may set
the determined censorship of Byzantine literature against all
material regarded as below an acceptable linguistic level. But
above all, there is the impossibility of a void of many centuries in
verse expression with meters intelligible to the uneducated. Many
Byzantine specialists would accept tha.t the most economical
solution is the assumption that ora.l rllteen -syllable verse existed
earlier, that it inspired imitation in learned language from the
tenth century onwards and more direct imitation in the vernacular
from the twelfth century. Perhaps the strongest single piece of
evidence is a ceremonial song of four verses from the official
Byzantine ceremonial, which cannot be dated earlier than the
mid-tenth century Book 0/ Ceremonies in which it is contained,
but is described there as a customary part of the ceremony. In
spite of its fairly formal language, this song has struck many
commentators as a rural folk song only slightly adapted for its
ceremonial role. What is more, the song belongs to the genre of
calendar songs for spring-the chelidonismata-which is
acknowledged as constituting the strongest traditional similarity
between ancient and modern Greek folklore.

143
514 ELIZABETH &. MICHAEL JEFFREYS
See, the sweet spring is again returning, bringing joy,
health, life and well-being, valor from God to the
emperors of the Romans, and God-given victory over
their enemies. 1'

Fortunately, the hypothesis of a tradition of oral poetry in

Byzantium depends much less on the general considerations about
the history of its meter than on analysis of the poems themselves,
to which we must now turn. OUf first task is to remove from the
argument two very promising groups of poems, the oldest in the
tradition.
The epic, or epic-romance, of Digenis Akritas was the
Byzantine poem which first caught Albert Lord's eye, and he
devoted to it some rather inconclusive pages in The Singer of Tale8
(1960:207 -20; cpo Lord 1954). The material is promising: six
manuscripts, all showing unmistakable elements of the same story
but with very large variations between them; anonymity up to the
two last-written of the manuscripts, which give the names of their
seventeenth-century redactors; clear references in the story to wars
of the ninth and tenth centuries, 'and a world-view which must
predate the Turkish overrunning of central Anatolia in the eleventh
century; a society which is at times convincingly heroic; one or two
non-Christian supernatural interventions; and the survival of the
names Difenis and Akritas (more usually Akritis) in Modern Greek
folk song. 5 Unfortunately, the poem is nearly as disappointing to
students of oral poetry as it has been to Greek nationalists, as we
have seen.
Under close examination it becomes plain that four of the six
manuscripts cannot be oral variants, but are the product of a
purely literary attempt to combine all available material and to tell
the whole story of Digenis (Trapp 1971:28-29, M. Jeffreys
1975:163-201). The remaining two, G(rottaferrata) and E(scorial),
represent the two sources of that compilation text. G has an
appeara.nce of accuracy and organization, which is unfortunately
combined with a flat, anti-poetic quality of language and style that
make it difficult to read. It has learned elements which it is hard
to imagine being successful in an oral performance. E, though it is
not without learned features, has a freshness and directness which
would a.ppear much easier to translate into oral terms. Its textual
transmission, however, is extremely untidy. Until recently, it was
accepted that this untidiness was the result of dictation from oral

144
 BYZANTINE POPULAR POETRY 515
performance, since some of the distortion consisted of syllables
extra to the basic fIfteen of the verse, a phenomenon which seemed
to represent some of the performance features of contemporary
Greek folk song. This idea has been attacked and probably
disproved, in several recent studies, and the textual problems have
been approached in more conventional and systematic ways
(Karayanni 1976, Chatziyakoumis 1977, Alexiou 1979 and 1983,
Beaton 1980 and 1981a).
We are left, then, with the learned and tidy G, dated around
1300, and the less learned and more untidy E, probably written in
the second half of the fifteenth century. Much of the history of
scholarship on Digenis Akritas may be seen as a contest between
partisans of these two manuscripts, each claiming one version as
more authentic, more original, and older than the other. Recent
studies have shown that both preserve authentic early
information. 1s Although the texts are very different, there is
enough common material to indicate that the two versions have at
some stage been in close contact (e.g., M. Jeffreys 1975 and
Alexiou 1982). There is still much work to do in defining the
nature of that contact,17 the degree of learned influence (borrowings
from written texts) in both versions, and the vexed Question of a
hypothetical original for both surviving versions, whether it may
have been in oral or written form, whether (if written) its language
level was closer to the simple but learned level of G or the popular
level of E with its scatter of learned elements. Discussion is at
present impeded by the fact that for some of the participants the
ideal kind of original text and subsequent transmission, with
connotations of authenticity and reliability, should be oral, while
for others it should be written.
Another interesting Question is the relationship of the two
halves of the poem. The first half is centered on the story of
Dlgenis' father, an Arab emir who converts to Christianity. This
part of the poem has a comparatively structured plot and contains
most of the references to the frontier wars of the ninth and tenth
century; its tone is that of comparatively realistic historical epic.
The second half of the poem, which recounts the adventures of
Dlgenis himself, is a sequence of ill-connected heroic episodes with
a romantic, otherworldly atmosphere, including unexpected
supernatural happenings. Digenis himself, whose heroic prowess is
used to keep the peace by defeating bandits who are Christian,
seems to have no relevance to the history of the Euphrates border

145
.';16 ELIZABETH & MICHAEL JEFFREYS
of Byzantium, which was one of uninterrupted hostility and almost
uninterrupted war between Christian and Moslem. He does,
however, read to us like the hero of a much-told poem long held
in oral tradition. A consensus appears to be forming around the
idea that the creator of the whole poem took a pre-extant oral
poem about the emir and appended to it an unrealistic pacifIst
mission, taking as its hero one who was Di-genis, that is, born of
the two races: "At some moment between the tenth and probably
the early twelfth centuries a single gifted individual must have
conceived the idea of a twice-born hero as a symbol for the
rapprochement of two warring empires, and grafted it on to the
traditional frontier stories of Akrites" (Beaton 1981a:21; cpo Beck
1966:137 -46). This sequence of events seems to the writers of this
report unlikely. The pacifIst vision is noble, but apparently
unparalleled during the period. We would prefer to see the story
of Digenis as the original. We make the hypothesis that a hero
called Digenis is more likely to have demanded an emir for a
father than vice versa, and that the connection between the two
halves is likely to have been based on the pre-existence of an oral
poem with Digenis as its hero. We' have therefore suggested a
much earlier situation in which the story could have arisen (M.
Jeffreys 1978).
Whatever the genesis of Digenis Akritas, it seems that its
present texts, G and E, are not susceptible to the most basic
method of oral poetry research, that is, to formulaic analysis.
After Lord's efforts in The Singer of Tales, we may report other
investigations by Beaton (198130:12-16, 1981b), confirmed by some
preliminary unpublished sampling of our own. The results are
certainly not negative, but far less positive than those from later
texts which we discuss below. We are convinced that Digenis
Akritas, in some form, spent centuries in oral tradition, and it
seems likely from evidence which we will present that that oral
tradition included a high level of formulas, as well as some
idiosyncratic linguistic features. It is disappointing that neither G
nor E has preserved these oral features intact.
The most favored milieu in recent publications for the writing
(or writing down) of both versions of Digenis Akritas is the
Comnenian court of the mid-twelfth century (Oikonimides 1979, E.
Jeffreys 1980)-whi~h is also the date of the first independent
reference to the text. This occurs in the Ptochoprodromic poems
(Hesseling and Pernot 1910), a group of satirical writings attached

146
 BYZANTINE POPULAR POETRY 517
to the name of one of the greatest literati of the time, Theodoros
Prodromos, in which he calls himself Ptochoprodromos, "Penniless
Prodromos." In fact, this attribution is probably a contemporary
fiction, since the poems use the fifteen-syllable at a vernacular
language level in a strikingly different metrical way from the
genuine learned fifteen-syllables of Theodoros himself (H. and N.
Eideneier 1978:1-7, Horandner 1982). The tone of the poems
themselves has been compared to that of the Goliardic Archpoet.
The subjects are varied and lively: the poet complains about his
overbearing wife, about poverty, about the poor rewards available
for learning and, in the persona of a young novice, about monastic
exploitation of the monks by their abbots.
We would be surprised if there is anything from popular
tradition in these subjects or their treatment. It is noticeable that
the poet seems in control of the linguistic medium he is using, at
one moment producing lines which are purer reflections of
contemporary Greek than many of the poems which were to follow,
at the next, and particularly when addressing a powerful, usually
imperial, patron, he raises the language level to a respectful
formality. Phrases are repeated in a way which may be formulaic,
but the repetitions are much too few to constitute a system as we
shall find it later. These poems seem to us the work of a court
poet, who is writing to exploit the vernacular tastes of the
Comnenian aristocracy (who were, after all, connected with
Western noble families prominent in patronage ~f other vernaculars;
see M. Jeffreys 1981:110-11; E. Jeffreys 1980:468-72). The poet
may well have been expert in the writing of learned
fIfteen-syllables for ceremonial purposes; here he added personal
themes and vernacular Greek, imitating contemporary oral tradition
from the outside without being seriously touched by any of its
formal constraints. The same judgment may be made of other
twelfth-century experiments in popular language-the Poem from
Prison of Michael Glykas, the Span~as poem of conventional advice
(though there may be more formulas here) and the Eisiterioi for
Agnes of France. is
The thirteenth century seems to be barren of such material,
but the fourteenth is quite rich. Much had changed in the
Byzantine world since the powerful, confident days of the
Comnenians. The last quarter of the twelfth century combined
defeat at the hands of the Anatolian Turks with a series of civil
wars. The thirteenth century opened with the capture of

147
518 ELIZABETH & MICHAEL JEFFREYS
Constantinople by the Fourth Crusade, the most devastating
psychological blow to Byzantium before the end of its history in
1453. The imperial throne was held by the Count of Flanders, the
Marquis of Montferrat became king of Salonika. many islands and
towns became the possessions of the Republic of Venice, a
Burgundian noble became Duke of Athens and Thebes, and the
Morea, or Peloponnese, fell under the control of knights from
Champagne. The three Greek fragments of the Empire which
remained - Epirus, Nicea, and Trebizond -naturally became
obsessed with the past, with the inheritance of the name of Rome
and of the language of Homer and of the classical period. Gone
was the confidence of the mid-twelfth century, when Byzantium,
like France and England, could experiment with breaking the
linguistic and literary forms which linked them with the legitimacy
of the past. The linguistic and cultural censorship was reimposed.
After a century of Frankish control, lands like the Morea or
Crete had become societies with two parallei cultures, the native
Greek and the superimposed Western forms (see Jacoby 1975 and
1979, Topping 1977). Though our so,urces tend to stress the legal
and even cultural distinction of the two races, we may assume
(and even find some evidence for) considerabie interpenetration.
This is, In our opinion, the best framework in which to view the
first fourteenth-century poem which we wish to discuss, the
Chronicle 0/ the Morea. We remarked ealier on its strong
antl- Byzantine prejudices, remarkable in a work written in Greek,
but we have no doubt that Greek was the language in which the
poem was first composed, and have supported our opinion at
considerable length e1sewhere. 19 In a multi-cultural society like the
Morea, it Is sterile to debate the question whether the poet was
Greek, French, or of mixed race, but he must have been working
under the patronage of the French nobles. For them, the whole
hierarchy of Greek learning would be unknown, or meaningless. A
poet whom they patronized would have to be intelligible-to them,
with the knowledge of spoken Greek which many of them must
have acquired, and to the Moreot population, who were to be
enthused with patriotism for their young state by listening to the
story of its history. The taboos of learned litera.ture would only
build barriers. IWe ma.y expect, therefore, an antl-Byza.ntine
document in a style and langua.ge of genuIne Byzantine oral poetry,
a text which reflects the Greek reality in a French state far more
accura.tely tha.n would be possible in a. Byzantine sta.te, with its

148
 BYZANTINE POPULAR POETRY 519
inevitable censorship. This is, in fact, what we find.
The Chronicle 0/ the Morea is a highly formulaic poem. Its
ea.rliest and best manuscript, H (in Copenhagen) has been fully
analyzed for formulas by the use of computer techniques, and was
found to have 31.7% of formulaic half-lines which were identical or
va.ried only in a carefully defined list of insignificant ways (M.
Jeffreys 1973:163-95). Allowing a rather looser definition of the
formula (but no looser than is often used in formulaic studies), the
percentage rises to 38.4%. These figures are almost meaningless on
their own; however, they acquire meaning when compared to the
analysis of another poem in the same fifteen -syllable meter, though
in a rather more formal language, the Alexander Poem (ed.
Reichmann 1963). This work is dated certainly within the same
century as the Chronicle of the Morea, perhaps closer than that.
By the same definitions of the formula, the Alexander gives
statistics of 9.4% and 12.8%, respectively. It should be stressed
that these statistics refer to what is often called "straight formula,"
not to "passages of the same type" (the dotted underlinings of
Milman Parry's tables of repetitions) which we rega.rd as useful in
the analysis of an established oral tradition but of little use in the
confirmation of the existence of that tradition. In case there is
any value in cross-linguistic comparisons, it is worth reporting the
results of similar studies of Old French conducted by Joseph
Duggan (1966), whose methodology played an important role in
fixing the parameters for our own investigations. The percentage
of formulas found in the Chronicle puts it in the middle of the
chansons de geate, while the Alexander is less repetitious than the
romances which Duggan has used as control poems.
The kinds of repetitions found, as in most non-Homeric
traditions, are rather disappointing for those who have read Parry's
exciting list of Homeric formulas. The most common formulas in a
published list of 63 which a.re repeated eight or more times a.re
simple names of Moreot ba.rons, given in a simple form which filJs
the relevant half-line. Next come some toponyms and some simple
phrases and clauses like "great and small," "with the army he had
with him," or "that was his name." Among the less common
Items on the list a more prominent place is taken by verbal
phrases: "and he said to him," "he informed them," and "they
rejoiced greatly" (M. Jeffreys 1973:178-81). Most of the phrases
have little more to recommend them as formulas than the mere
fact of repetition: they are simple ideas expressed in straightforward

149
520 ELIZABETH & MICHAEL JEFFREYS
language. Chief among the small number of exceptions to this rule
are a few frequent formulas, most of them found also in other
fourteenth-century texts, which are expressed in grammatical and
syntactical forms notably more archaic than the main body of the
text. Some of them, curiously, have survived as fossilized archaic
phrases into modern spoken Greek. A certain amount of dissension
has grown up around these. Mohay (1974-75) and H. Eideneier
(1982) regard them as proof of archaic influence on this genre of
texts. We prefer to see in them some sign of the length of the
oral tradition with which the poems are connected, and its
importance in the history of the development of the modern
language.
This strongly positive evidence must be supplemented by
discussion of other tests regularly made of oral and oral-influenced
poetry (M. Jeffreys 1973:195). The Chronicle is anonymous. A
large proportion of its lines are end-stopped, and enjambement
within the phrase or clause is rare. There are some signs of
marked pauses for a performer to take, a rest at fairly regular
intervals. We have already seen that the poet's avowed purpose is
to tell the stories of the heroes of the Fourth Crusade, especially
those prominent in the conquest and organization of the Moreot
principality. Since the date usually set for the composition of the
Chronicle is in the 1320's, none of the events described in detail is
more than 130 years old. However, the Chronicle is full of errors
of fact which may be detected by comparing it with other,
conventional historical sources, while it retains an easy familiarity
with several aspects of Moreot life which other sources
ignore-particularly those connected with the feudal organization of
the principality. It seems very likely that it was based largely on
oral sources (Jacoby 1968:182-83, M. Jeffreys 1975b:325-26).
On the other hand, an almost negative report must be given
about elements of oral organization longer than the formula, that
is, the motifs and themes which loom so large in discussions of
Homeric oral poetry and several medieval traditions. The
parallelism of wording used at the deaths of the Chronicle's main
characters approaches the status of a theme (lines 2441-67,
2752-57, 7213-39, /7753-810). Equally, there is considerable
parallelism in structure between two major scenes between noble
captives and their captors, that of the Moreot prince William II
(lines 4092-191), and that of the Byzantine Megas Domestikos
(lines 5466-575). There are subtle contrasts here redounding to

150
 BYZANTINE POPULAR POETRY 521
the credit of the Moreot leader; they could be regarded a8
straightforward literary parallelism or a8 the sophisticated use of an
oral thematic pattern. But it is clear that these features are
exceptions and not the rule. The poem is not regularly organized
by thematic structures, and as such seems to diverge from the
products of several other medieval oral traditions. This fact must
be given due weight in discussion of the nature of oral influence on
the text.
Equally, the text which survives (we use ms H, which is the
oldest and clearly the best) describes itself regularly as a book and
the poet refers to his own activity as writing. There are in fact
pairs of formulas which can be used with either the specific "write"
or with the ambiguous "tell," which could as well refer to reading
aloud or pure oral narration: "the one I tell you" /"the one I write
for you," for example, or "I am telling you the truth" /,,1 am
writing you the truth." Even cross-references take on formulaic
form: the pair "Earlier in my book" /"Later in my book" are used
a total of six times.20 This confusion is not unique to the
Chronicle: in the Achilleis, which we shall speak of later, it
reaches the extreme form of repeating three times the phrase, "And
what tongue would be able to write in detail ... 7" ,21 a line which
could survive only at a time of transition and could not have stood
for a moment against the ordinary critical judgment appropriate to
purely oral or purely written poetry.
Before pronouncing a conclusion on the status of the
Chronicle, we must deal with another factor, but one where it is
impossible to summarize the results of a published study because
that study has not yet been written. Our preoccupation derives
from the lesser-known half of the work of Milman Parry on
Homer, that which has to do with Homer's language. In work
which culminated in his "Studies in the Epic Technique of Oral
Verse-Making. II. The Homeric Language as the Language of an
Oral Poetry" (1932), Parry showed that his oral theories could
solve with great precision and elegance the problem of the Homeric
mixture of forms, including some usually identified with the
Arcado-Cyprian and Aeolic dialects, together with the Ionic which
was the spoken language of the milieu in which the poems reached
their preserved forms. Scholars before Parry had looked for an
area in which these three dialects may all have been present at the
same time, but were beginning to realize that the hexameter must
have a role in the way in which the dialects were combined.

151
522 ELIZABETH & MICHAEL JEFFREYS
Parry's solution, simplified to its essential mechanisms, was as
follows. A poet working in an oral-formulaic tradition like the
Homeric and with an inflected langage like Homeric Greek needs
not only formulaic patterns of expression for one grammatical case
or one verbal tense, but a flexible system which reflects the
changing demands of the case and tense inflections of the language.
Homeric ships, for example, tend to be "equal" in the genitive
singular and "black" in the dative singular, for purely linguistic
reasons: a black ship in the genitive would break the meter. This
system is complex and subtle, and must have needed long practice
in a young singer. But what would happen when the language
changed, for one of a number of possible demographic reasons, or
perhaps simply because of the passage of time? Where the new
linguistic form is metrically identical, the new would gradually
replace the old. But where the new form is such as to break the
formula, the poets would have a difficult choice: either remake the
formula from the beginning or preserve the archaism. The
evidence of the Homeric language suggests that the latter course
was often followed in that tradition. It ~as a long tradition, and
in some cases the language seems to have changed twice; as a
result one may fmd in Homer three different forms of the same
case of the same noun -one the natural spoken form of the poet
at the moment when the poem reached its final form, and two
archaisms, preserved for metrical reasons to perform two different
roles within the formulaic system and the hexameter line. This
part of Parry's work has never, to our knowledge, been seriously
challenged in principle.
The relevance of this parallel to Byzantine popular poetry
may be established by a quotation from the most authoritative
summary history of the medieval Greek language (which
recommends, it must be said, a different solution to the problem
from that proposed here):

The existence in early vernacular literature of so many

alternative verbal forms poses problems to which at
present we can give no answer. The purist forms ma.y
be eliminated as due to scholarly and literary influence.
But did -oun and -ousi, -don and -oun really coexist
in living speech? They were certainly living forms in
different parts of the Greek-speaking world. This
brings before us the problem of the origin of the

152
 BYZANTINE POPULAR POETRY 523
common spoken language of the nineteenth and
twentieth centuries. Does it go back to a common
spoken language of at any rate the urban population in
late Byzantine times, which is reflected, however
imperfectly, in the language of the early vernacular
poetry? Or do these poets write in an artificial
amalgam of forms belonging to different dialects, which
they have heard on the lips of uneducated speakers? In
other words, is their poetry a kind of incompetent
attempt to imitate living speech by men whose only
familiar mode of expression was the literary language?
To answer a blunt 'yes' or 'no' to any of these
questions would be to over-simplify the matter
(Browning 1983:82).

The author uses the Chronicle of the Morea more than any other
Byzantine text to show the complexity and range of the linguistic
macaronism of the period. To explain the observed practice, he
reminds us that the poet may well be a Hellenized Frank:
"Perhaps they [the variety of forms] were genuine alternatives in
the language spoken by the writer. But it is more likely that he
is using a mixed language, the result rather of a lack of feeling for
the language than of conscious effort to raise his style above that
of everyday speech" (ibid:74). We do not find it easy to believe in
this Hellenized Frenchman with a defective feeling for Greek who
has left a poem which is a kind of museum of all the various
Greek linguistic forms used in the Middle Ages.
The language of these poems has been of interest in the
question of the history of the Greek language, the battle between
supporters of the vernacular dimotiki level of the language and
those of the purist katharetJousa, which has only very recently been
settled in favor of the former. In the years around the end of the
last century Yannis Psycharis, the linguist, novelist, and passionate
supporter of the dimotiki, spent many years charting the
development of the language of these poems, which he thought was
the oral vernacular of Byzantium at the time, the unwritten
history of the dimotiki. He found that there was a real
development from earlier to later, with the percentage of early
forms being high in poems dated early in this genre, and later
forms dominating at the end of the period in a remarkably regular
way.'n His opponent Georgios Chatzidakis, from among the

153
524 ELIZABETH &; MICHAEL JEFFREYS
supporters of katharevousa, showed, however, that whatever
Psycharis was measuring it was not the development of the spoken
language. By assembling all of the other available evidence for the
vernacular of the time, he was able to make a convincing case that
the changes from one form to another in Psycharis' tables gave an
apparent date for linguistic change far too late to reflect accurately
the real development of the spoken language. Several of the forms
which, according to Psycharis, increase in numbers during the
fourteenth century and become dominant only towards its end,
were probably already the primary, even the only, forms used in
vernacular speech in some areas in the twelfth century (Chatzidakis
1892). In this statement of the linguistic problem, one last point
should be made: there is no Byzantine popular poem with the
specific characteristics of any Modern Greek dialect, though it is
likely that some features, at least, of those dialects had developed
by the fourteenth century (Browning 1983:126).
A full proof that the mixed language of these texts is an oral
poetic Kunstsprache, like that of Homer, 'will be very arduous, and
can certainly not be attempted here. It will be necessary to
examine, with detailed statistics, at least a score of the linguistic
variations found in these texts, updating Psycharis' tables and
fitting them into a complex framework of proof. A good deal of
work is also needed to establish the history of each variation from
all other available evidence, following the work of Chatzidakis.
The following sample is offered as a sketch of the general lines that
the argument will take.
We must begin from a description of the fifteen-syllable
verse, which, in contrast to the Homeric hexameter, is a rather
simple meter. It is based on syllable numbers and stress accents
like English metrics, unlike ancient Greek prosody. Each line is
divided by an invariable break after eight syllables. Each of the
resultant half-lines is regulated in accent position towards its end:
in the first, a word-accent must fall on the sixth or eighth syllable
(of the eight), or on both, while in the second half-line, it must
fall on the sixth of the seven syllables, the fourteenth of the whole
line. The remaining/ word accents of the line nearly all fall on
even-numbered syllables, confirming the underlying iambic rhythm.
Most of the exceptions, the word accents on odd-numbered
syllables, fall on the first and ninth syllables of the line, making
the opening two syllables of each half-line metrically unregUlated
(Koder 1969, vol. 1:87-94; Horandner 1974:128-33):

154
 BYZANTINE POPULAR POETRY 525

x x (0) U) (0) / 0 U)
or x x (0) U) (0) U) 0 / x x (0) U) (0) / 0
123 4 5 6 7 8 9 10 11 12 13 14 15

A moment's examination of this table will show that one

syllable, accented or not, added to the seven syllables of the second
half-line will produce an acceptable first half-line, and vice versa
with one syllable removed. Thus the simplest pattern of usefulness
to an oral poet which one could imagine would be two noun- or
verb-forms equivalent in every way except that one would have an
extra syllable at the end. If the accents are in the right place, the
longer form could then be used at the end of the first half-line
and the shorter form at the end of the second. We should like to
propose that much of the diversity of language in poems like the
Chronicle 0/ the Morea may be explained in this way, in fact that
this diversity is a practical working system for composition in the
fifteen-syllable.
Let us take the first of the examples of macaronism quoted
from Browning's description above. Third person plural verb forms
ending in unaccented -oun and -ousi are found in the Chronicle of
the Morea as alternative terminations for the present indicative
active and both forms of the active subjunctive. In general terms,
the -ousi ending is that of ancient Greek, the -oun ending that of
Modern Greek. Examples of -oun can be found in the Egyptian
papyri before they peter out in the seventh century, and Byzantine
grammatical treatises warn that it should be avoided. On the
other hand, the -ousi ending is found in several peripheral modern
dialects, particularly those of Crete and Cyprus (Browning 1983:6;
cpo 1976). It is occasionally found, apparently metri causa, in folk
songs from other areas.
In ms H of the Chronicle of the Morea both these forms are
used systematically. In fact, if one disregards the nature of the
verb-forms concerned (whether, for example, they are indicative or
subjunctive) and merely collects examples of the terminations, it is
possible to find as many as 59 stems from 53 different verbs which
show examples of both alternative endings, as well as many more
which give examples of only one of the forms. In the case of the
59 stems, one may find 280 examples in all using - ousi and 491

155
526 ELIZABETH & MICHAEL JEFFREYS
using -oun. Now the -ousi ending would fit well at the end of
the first half of the line and -oun at the end of the second, and
we do, in fact, find a number of phrases adapted for both halves of
the line by the use of this convenient variation.
The next stage of the argument involves an attempt to show
that one alternative form or the other is used in the poem partly
or completely because of its metrical characteristics, like an Aeolic
form in Homer. In the case that we have chosen as an example,
it is obviously desirable to isolate as many as possible of the forms
ending in -ouai at the end of the first half of the line, though
there are other possible kinds of metrical usefulness too complex to
explain and justify here. It so happens that the -ousi form,
though much more restricted in its use than the -oun form, cannot
be shown to be an archaism to the poet of the Chronicle. The
reasons are probably two: first, that the comparatively simple and
relaxed form of the nrteen-syllable leaves the poet greater scope for
initiative than does the hexameter, and so allows the use of archaic
forms away from the limited situations 'which have forced their
preservation within the poet's linguistic repertoire; second, that the
-ousi form may not have been seen purely as an archaism, since it
may still in the fourteenth century have been in use in wider areas
of the Greek world than now.23
The example chosen is of average value in the proof of the
oral genesis of the linguistic mixture of these texts. Of the various
sets of linguistic alternatives to be studied, it has a higher than
average range of application through the linguistic usage of the
Chronicle-the 59 verb stems we have mentioned. On the other
hand, it is less clear than is usually the case that the alternatives
studied include one which is an artificial form to the
poet-presumably an archaism, preserved because of its specific
metrical usefulness. There are several other verb-forms and a
number of noun-patterns which would give a less ambiguous result,
but would need longer discussion here. Experts in contemporary
Greek folk song may respond to our example and to the whole
linguistic proposal that we are making by saying that it is no new
suggestion that unusual linguistic forms may be used in the
fifteen-syllable for metrical reasons. We would agree, but would
claim that, in the Chronicle of the Morea at least, we are not
dealing with exceptions, with a minority of linguistic distortions,
but rather with a complete linguistic system regularly based on the
needs of the meter. For us, this is one of the firmest pieces of

156
 BYZANTINE POPULAR POETRY 527
evidence that the Chronicle's language and style are those of a
contemporary system of oral poetry.
It is time to state formal conclusions about the role of oral
tradition in the genesis of the Chronicle of the Morea. We do not
believe that the poem as it stands is a text dictated in
performance by an oral singer. Its own insistence on its status as
a book and on writing, and the confirmation of this fact in
formulaic phrases, are decisive. The absence of thematic
organization also seems important, indicating perhaps that the poet
is dealing with material which did not come to him in poetic form.
Much of Moreot history is here told from the point of view of the
lawyer and the diplomat. We have an impression that the oral
style is being extended in length and subject-matter beyond its
usual range, which was probably more restricted to heroic
narrative. The spectacular charge of Geoffrey of Karytaina at the
battle of Pelagonia (lines 4018-72), containing two of the very few
images found in the whole poem, seems likely to reflect a short
oral song of a type more conventional for the tradition.
Yet we regard this poem as a more genuine reflection of
Greek oral style than any other we have examined. The evidence
of the formulaic level (whether the formulas are oral formulas or
specially created for the writing of this text in formulaic style)
combined with the linguistic evidence is enough to convince us that
oral narrative poems with similar characteristics could be heard in
the thirteenth- and fourteenth-century Morea. We are also
confident that, after removing some Franco-Greek la.ngua.ge and a
good deal of anti-Byzantine sentiment, it is possible to see this
poem as a good reflection, perhaps the best we have, of the oral
material which lies behind the whole genre of Byzantine popular
poetry. There exists a later, less authentic text of the same kind
as in the Chronicle, giving the history of the Italian Tocco family
and their conquests in the Ionian Islands and Western Greece (ed.
Schiro 1975; see also Koder 1982 and Zachariadou 1983). We have
already mentioned the paradoxical judgment that Byzantine popular
forms can be seen in undistorted form only in circumstances which
sweep aside the rest of Byzantine culture, for otherwise the
recording of genuine popular material is usually blocked.
Similar judgments can be made about much of the remainder
of the fourteenth-century material in this genre, especially the
romances. There are about a dozen poems which fall within the
category of romance, some of them very long and preserved in

157
528 ELIZABETH & MICHAEL JEFFREYS
numbers of manuscripts. Together they make up a considerable
proportion of the manuscript remains of popular poetry of the
period. They are divided approximately equally between poems
which are translations from sources in Western European literature
and those which seem to be Greek, or at least for which no
western originals have been found. The question of national
identification again causes problems of critical approach. The
romance is to some extent a Greek genre, in spite of its name:
the earliest surviving examples are the novels of Chariton, Achilles
Tatios, Longus, and Heliodoros from the second to the fourth
centuries A.D. (survey in Ha.gg 1983). Then, after a long break,
there appear four more romances in the twelfth century, in the
learned language, those of Theodoros Prodromos, Konstantinos
Manasses (in fifteen-syllable verse), Niketas Eugenianos and
Eustanthios Makrembolites (see H. Hunger 1978, vol. 2:119-42).
This resurgence of the romance must be in some way connected
with the simultaneous flowering of roml),nce in the west (the
direction of influence is not clear; see E. Jeffreys 1980), but the
works themselves are almost pure products of the Greek tradition.
However, the greatest problems of critical approach are caused by
the fact that several of the fourteenth-century translations are of
originally Greek materia.l-the French Roman de Troie, the Latin
Apollonios of Tyre, and Boccaccio's Theseid.
The disconcerting feature is that it is the translations,24
particularly the distorted Homeric material in the War of Troy, the
French Imberi08 and Margarona (Pierre de Provence et la belle
Maguelonne) and the Italian Florios and Platzia-Flora (Fiorio e
Biancifiore) , which produce, in our opinion, the best reflection of
the Greek oral style. The poems which are not translations
include the Alexander poem, which we have already seen used as a
half-learned and non-formulaic contrast to the Chronicle of the
Morea, and Kallimachos and Chrysorrhoe, possibly ascribed to a
relative of the emperor Andronikos II and written without many
formulas in a language rather more correct than that of the oral
tradition. Belthandros andI
Chrysantza and Libistros and Rhodamne
give a similar impression, but are somewhat closer to the oral
pattern, in both formulas and language level. Only with the
Achilleis and the Trojan poem edited as a Byzantine Iliad (ed.
Norgaard and Smith 1975) do the original Greek works reach the
same impression of oral authenticity as the translated romances
mentioned above.

158
 BYZANTINE POPULAR POETRY 529
These romances and the relationships between them have
caused a good deal of scholarly disagreement in recent publications
(survey in E. and M. Jeffreys 1983). There are numerous
similarities in wording between them. At one end of the scale,
these include formulaic phrases like "great and small," which is
found repeated, sometimes many times, in nearly every poem of
this genre. At the other end of the scale there are similarities
which resemble the sudden appearance of a repeated oral theme
between two poems, which is most surprising in a genre where
repeated themes are rare. In one particular case, a striking phrase
from Florios and Platzia-Flora, which appears to be a direct
translation of the equivalent phrase in that poem's original, Horio
e Biancifiore, is taken over into Imberios and Margarona. 25
Basing himself on a solid core of evidence like this last
example, and combining it with long lists of less surprising
repetitions (1976, 1977-78, 1977, 1979; in addition to 1975),
Guiseppe Spadaro has built up an extensive and patient case in
favor of systematic plagiarism among all the poets of this genre:
"E' evidente, infatti, che i poeti della letteratura greca medievale in
lingua volgare si sentivano legati da uno stesso indirizzo poetico,
appartenevano ad una stessa seuola poetica, per cosl dire, e quindi
subivano, ovviamente, il fascino delle opere precedenti, aIle quali
spesso si ispiravano e dalle quali attingevano . . . oltre che motivi,
emistichi, versi interi, insomma tutto quel formulario che
all'occorrenza utilizzavano con molta comodita e grande vantaggio,
sebbene a scapito dell'originalita" (1978:9). This is a school of
poets who read and copy each other's work in a purely literary
way. From Spadaro's tables of similarities it is possible to
construct a complex network of influences among most of the
poems we have mentioned in this article.
We have explained at length elsewhere why we are unable to
accept this account of the way these poets worked (E. and M.
Jeffreys 1983). In the first place, it does not allow enough for the
influence of the copyists of our surviving manuscripts, as we shall
see below. Further, Spadaro's arguments seem designed to show
that simila.rities between the poems are not the result of chance, a.
position which no scholar in this field would wish to challenge.
But his position is exposed to criticism on the opposite flank: he
does little to prove that the phrases he collects are literary
influences rather than oral formulas from a tradition known to all
the poets. Spadaro concentrates attention on phrases repeated from

159
530 ELIZABETH & MICHAEL JEFFREYS
one text to another, often ignoring internal repetition within the
individual text, and so understating the repetitious nature of the
poetry as a whole. He also does not compare each of the poems
of the tradition systematically with all of the others. If he were to
do so, he could find many sure examples of influence at least as
convincing as some of those which he has published. The complex
network of interrelationships which he implies would then become
even denser and .more bewildering. We find the parallel phrases he
cites, in most cases, unlikely to stimulate the memory in a literary
way, and prefer to ascribe their repetition to the mechanical
processes associated with oral-formulaic style. Finally, the
plagiarism which he suggests assumes the existence of considerable
numbers of manuscript copies, so that each poet could read, and
be influenced by, the work of nearly all his predecessors. However,
a recent study by Manolis Chatziyakoumis (1977:247-48) has
suggested that manuscripts were comparatively rare until around
the year 1500, when all the poems dealt with here had long been
composed.
Spadaro's work is a useful counterbalance to those who might
feel that the hypothesis of the influence of an oral tradition
prevents any further investigation of links between the texts under
discussion. It is plain that connections can be established in
several cases among those texts he examines, whether they result
from common authorship, similar circumstances of composition, or
even the conventional literary influence which Spadaro assumes
(espec. 1975:307-9). One of these cases of similarity has recently
been investigated with great care by A. van Gernert (1981), who
ascribes many of the connections to the intervention of two scribes
rather than to links between the original versions of the poems.
As we shall see later when dealing with the textual traditions, it is
vital to examine common lines for the possibility of scribal
intervention. Van Gernert's article is a good example of the large
volume of work which needs to be done.
Chronicles and romances do not exhaust the genres of popular
poetry in the fourteenth and early fifteenth centuries, though they
are far the most impressive in bulk and seem to us more relevant
than any others to the discussion of oral influence. There seems
no point in listing the other, less relevant genres, since Beck's
handbook does this most efficiently. However, before passing from
the texts themselves to their manuscript traditions, there are two
other comments to make about texts which have not yet been

160
 BYZANTINE POPULAR POETRY 5:n
mentioned.
The first is the Belisarios poem. This work has a historical
basis in the story of Justinian's famous general, and is certainly
not without romantic elements. Beck, however, categorizes it as a
didactic poem with a moral purpose, and that seems the best way
to describe the apparent motives of the surviving versions. In
contrast to most of the popular romances, Belisarios is
pre-eminently a poem of the populace of Constantinople. It knows
a good deal of the topography of the city and is less and less
convincing the further away it goes from the capital. It knows
very little of Belisarios' real story, but seems to project onto this
great name episodes from the lives of many popular heroes of the
city, from the fifth century (a hundred years before the historical
Belisarios) to the fourteenth century (when the first surviving
version of the story seems to have been written). It is also one of
a handful of surviving Byzantine texts which show a clear bias for
the common people and against the nobility, whose role it is to
poison the emperor's mind against the great popular hero. The
message is a tract about the power of envy, the envy of the
aristocrats for Belisarios. The whole gives the impression of being
a.n urban folk song, expressing in one composite story the feelings
of a thousand years of the capital's inhabitants for their heroes.
From the point of view of formulas and language the text seems to
have considerable oral characteristics, and it is tightly involved in
Spadaro's tables of influences (see also van Gernert 1975).
A rather peripheral position in this discussion must be taken
by a number of beast fables in fifteen-syllable verse, of which two
have been analyzed in some detail by Hans Eideneier (1982:301-6)
for oral influence, particularly for formulaic patterning. These
poems are structured, as Eideneier says, not only by the pattern of
speech and counterspeech, but also by schematic patterns of abuse
and self-encomium. Perhaps as a result, there are few repetitions
which fill the complete ha.lf-line, and so fulfill the formal
requirements of a formula as defined in our work on the Chronicle
of the Morea. Eideneier's various techniques of analysis do not
depend on full half-line correspondence, and he succeeds in
describing several patterns of linguistic usage which provide some
support for a theory of oral influence, at least that the poems were
conceived for oral presentation.
The twelfth-century examples of fifteen-syllable verse in the
vernacular were court poems written by learned men outside the

161
532 ELIZABETH & MICHAEL JEFFREYS
range of influence of the oral tradition which we think provoked
them. Most of the fourteenth- and fifteenth-century poems were
quite different. The poets, in general, seem to be away from the
influence of Byzantine learning, particularly when we consider the
Frankish propaganda of the Chronicle of the Morea and some of
the translations from western European texts. In the latter case,
for example in the War of Troy, it is very instructive to compare
the Greek text with its French original.26 The Greek text has a
high formulaic content (our samples give 29.3% and 35% compared
with 31.7% and 38.4% for the different levels of repetition defined
above for the Chronicle of the Morea). The French too is
formulaic, but at a lower level of frequency. Sometimes Greek
formulas seem to be made up in translation to reflect the French
formulas; sometimes they are introduced by the Greek translator
and one may assume, especially if they are phrases found in other
poems, that they are Greek oral formulas. There is a large
intermediate group of related phrases abput which one cannot
pronounce (E. and M. Jeffreys 1979:131-36).
All these poets, to our mind, were writing in the only style
available to them which would make for easy communication in
circumstances where the learned languages of Byzantium had lost
their hold on published literature. The meter, the formulaic
phrases, and the lan~uage mixture were all parts of the style which
would be unquestioned by any writer who was accustomed to
listening to Byzantine oral poetry. Whether any of those whose
writings have been preserved for us also had singing skills, we can
only speculate. If it turns out, as we suspect, that the language of
the oral tradition is particularly closely reflected in the Chronicle of
the Morea, then it could be suggested that the poet who wrote
that book also performed parts of it as creative oral poems to the
French and Greek inhabitants of the Moreot castle where he lived,
even before he wrote them down as a text.
It is a cliche for all those who write about oral poetry that
such poems do not have fixed texts, that they exist in some kind
of inchoate form realized from time to time in performances which
usually differ from each/ other, sometimes significantly. This
argument may, of course, work in reverse: if a poem is preserved
in several manuscripts which differ from each other in significant
respects, then it is tempting to conclude that each manuscript is a
separate realization of an oral Gestalt, that each somehow reflects a
separate performance of the oral material. Those Byzantine poems

162
 BYZANTINE POPULAR POETRY 533
which are preserved in more than one manuscript regularly show
changes, great and small, from one manuscript version to another.
Therefore the question has been raised whether each manuscript is
a separate recording of a different realization of the same oral
material. In fact, a history of this period of Greek literature
(Trypanis 1981) has been written from that point of view.
On this issue we should like to sound a note of caution, with
the firmness of those who have been converted to a conservative
viewpoint by long and hard experience. We have been involved
(with Manolis Papathomopoulos) in the edition of the War of
Troy, the longest of the popular Byzantine texts, which also
survives in five manuscripts and two substantial fragments, all of
which show considerable variations from each other. Our initial
approach to this substantial task assumed that we were unlikely to
be able to print a single critical text based on all the manuscripts,
and that our methodology was likely to have far more to do with
Lord than with Lachmann. After 14,000 lines of editing and
interminable discussions, this assumption has been totally
abandoned. It seems to us that in this text the classical
methodology of textual criticism is the correct way to reconstruct
the original translation: both this judgement, and the
reconstruction itself, are much more secure because of the survival
of the French original, which often puts the critic in the position
of being certain as to which of two alternative readings was found
in the original text of the translation?7
Questions of oral poetry, however, are certainly not irrelevant
to the reconstruction of the text within this overall logical
framework. Some of the "errors" which are met represent the
replacement of one formulaic line or half-line by another, and so
seem rather gross to those who are used to textual transmission in
which oral tradition has played no part. Furthermore, it is
difficult to escape the conclusion that the existence of an oral
tradition is also responsible for the number of "errors" found in the
manuscripts. We are here in a totally different world from that of
Lord's Yugoslavia, where the fIXed, printed text tends to impose on
oral poets, when they learn to read, the corrupting idea of the
fixed text: "If one looks at the surviving manuscripts of the War
of Troy . .. it is most unlikely that they could ever have imposed
upon their readers the concept of a stable text. The appearance of
the page naturally varies greatly. There is no standardization of
orthography so that the spelling fluctuates alarmingly in almost

163
534 ELIZABETH & MICHAEL JEFFREYS
every word. Worse still . . . there is no accuracy in the
presentation of the words and grammatical forms of the text. In
our opinion, this is not a case of literacy imposing its rigid
standards upon a fluctuating oral tradition, but the reverse. The
fluidity of the tradition h~ been carried over into itl'l written
expression, probably because the writers of the manuscripts
recognized an oral style which did not demand word-for-word
reproduction" (E. and M. Jeffreys 1979:124). In our opinion there
is a decisive difference between the world of Avdo Medjedovic in
the Montenegro of the 1930's and that of the War of Troy, a
difference that renders invalid the cultural parallelism on which the
extension of the oral-formulaic theories into the medieval period
was first based. That difference is largely centered on the
invention of printing. Printing changed fundamentally the
relationship between one copy of a text and all other copies: only
after that, it seems to us, would a text pe likely to change in a
singer's mind the relationship between one version of his song a.nd
all other versions.
Let us return to the War of Troy a.nd describe its textual
tradition, explaining why we regard it as basically a conventional
literary tradition with only subsidiary influence from oral poetry.
In the first place, the variants in the texts, although very
numerous, seldom extend beyond the individual line. Our line
concordance of the manuscripts very rarely shows that all extant
manuscripts agree over the precise form of anyone line (let alone
the right way of spelling it). Variations of detail are so constant
that one is forced to conclude that word -for-word accuracy was
not one of the copyists' goals. On the other hand, it is equally
rare for the concordance to show inserted lines, expansion or
contraction of episodes, or the replacement of one episode by
another. Our impression is that the copyist would read a passage
of perhaps one to six lines, and would then write it out in a very
similar form, with the same number of lines and similar
vocabulary. In all other respects, however, he can have had no
conscious policy of checking the precise form of his model:
inessential words would vary, articles would be inserted or omitted,
the word order would change, and a different choice would be
made among the linguistic variations permitted by the mixed
language.
The constant change would not be purposeful. Recognizing
the language and form of a fluid oral tradition, the copyist would

164
 BYZANTINE POPULAR POETRY 535
merely relax his standards of accuracy. In fact, in the whole of
Byzantine popular poetry there are only one or two cases of
accurate copies by the· conventional definition. The relaxation
would be assisted by the flexibility of the fifteen-syllable line,
which must have offered the poet-in oral or written
expression-greater scope for small-scale initiatives than, say, the
Homeric hexameter. The whole orally-based system-meter,
formulas, and language-was extremely flexible, and this flexibility
was dominant enough to overcome the demands for precision
usually associated with the act of copying. It is interesting that
several of the manuscript variants are written in hands which
betray the practiced scribe. Several identifications are being
investigated, and it seems to be only a matter of time before we
are able to compare the practice of the same scribe writing both
learned and popular material.
There have been a number of detailed studies attempting to
show the influence of scribes in causing apparent verbal echoes
from one text to another. Several of these have concentrated on a
6O-line interpolation at the end of ms N of the Achilleis
(Michaelidis 1971-72; Spadaro 1977-78:252, 267-78), the connection
of which to more than one other text has been hotly debated.
The researches of H. Schreiner (e.g., 1966) into the relationships
between the varous texts always allowed for the influence of the
copyist, while those of G. Spadaro (1975:313, note 5) do not do so
systematically, leaving some of his results open to objection. This
last point has been recently made in van Gernert's article
(1981-82:95) on the relationship between the Achilleis and
Belisario8. He makes great progress in the analysis of the textual
transmission of both poems and in defining the nature of the
phrases that they have in common, suggesting several cases which
are probably due to the intervention of copyists. His results in the
comparison of these two texts demand caution from all those who
look for similarities between poems in this tradition-whether we
are able to prove that the poems concerned have been copied by
the same scribe, or whether we must merely remember to leave
that hypothesis open as a possibility.
Ultimately the choice of editorial methodology in the
publication of an individual text must depend on the relationships
observed between or among its surviving manuscripts. Van
Gemert's analysis of the Achilleis and Belisario8 is, in conventional
critical terms, attempting to construct a stemma codicum by

165
536 ELIZABETH & MICHAEL JEFFREYS
finding common errors. We too, after some initial reluctance, have
constructed a detailed stemma for the War of Troy and tested it
rigorously in establishing a text of over 14,000 lines. Our testing
will have been more searching than that possible in most textual
traditions, because the results of our stemmatic reconstruction of
the archetypal translation, more often than not, can be confirmed
or denied by reference to the French original. Our stemma works
in the overwhelming majority of cases of serious variants, and the
few apparent contradictions can be explained by a coincidence in
choice of alternative formulas, a phenomenon which is not
surprising in a transmission dominated by an oral tradition.
Unfortunately, the stemm a does not permit the easy reconstruction
of the details of the text: in nearly every line there are tiny
linguistic variants too insignificant, in a fluid tradition like this, to
be solved by stemmatic means, especially in the not infrequent
cases where every preserved manuscript bas a slightly different
variant. But the general conclusion seems to us certain. In spite
of the formulaic density of this text, to which we referred above,
its surviving witnesses lead us back by a written tradition to a
single, written translation from its French original. The oral
tradition intervenes only to relax the definition of accurate copying
aimed at by the scribes.
Before leaving the subject of textual tradition, it is necessary
to mention two obvious cases where differences between versions of
a poem are considerable greater than those described above for the
War of Troy, and there is some greater chance at least of the
redaction being due to a more direct form of oral intervention.
The first example, of course, is Digenis Akritas in its two earlier
versions of G and E, described above. One may add the different
versions of the Achilleis, with their very different lengths.28 Each
of these examples is based on manuscripts containing other popular
material, which also shows some signs. of the same textual
distortion. The Escorial manuscript, '" IV 22, which contains
Digenis, also includes Libistro8 and Rhodamne and the bird-fable,
the Poulologos. The Oxford manuscript, Misc.Gr.282-7, which
contains the Achilleis, also has a version of Imberio8 and
Margaronaj there is also the London manuscript, BL Add. 8241,
which gives the Achilleis and Florio8 and Platzia-Flora. We fear,
however, that a detailed examination of these cases will show not
oral variants but a systematic popularizing of the poems and
purification from their learned linguistic elements, as we shall see

166
 BYZANTINE POPULAR POETRY 537
later in the proposals of Chatziyakoumis (1977:247).
* * *
Thus in some Byzantine century before the tenth, possibly as
early as the sixth, a tradition of oral poetry arose, based on the
stress accent which had replaced long and short syllables as he
basis of Greek metrics. The dominant rhythm came to be the
fIfteen-syllable. Tradition and meter gave only the most indirect
signs of their existence before the tenth century, when the meter
appears in the learned language. From then until the twelfth
century, it was used by learned men, but for specifIc purposes and
often with disclaimers which indicate its status as a meter below
literary contempt. In the twelfth century it appears for the fIrst
time in the vernacular, but the writers are still mainly learned men
and there is little direct reflection of oral style.
In the fourteenth century we are able to follow for the fIrst
time long texts which must have a real resemblance in language
and style to the oral material which had been circulating for at
least four centuries. The style is formulaic, and the language
shows the historical depth of the tradition which had forged it. It
is disappointing, though far from unexpected in Byzantine
conditions, that the chronicles and romances which best reflect the
oral style are non-Byzantine or even anti-Byzantine in ideology.
This is not the only moment in Greek literary history when the
true direction of Hellenism bas been best appreciated, even
exploited, by non-Hellenes, or at least by those outside the range
of a Greek classical education.
It remains to say something about the continuation of the
tradition up to and beyond the end of Byzantium. A key
contribution here is that of Manolis Chatziyakoumis, who has
examined the date, provenance, and present place of preservation of
the manuscripts of Byzantine popular poetry (1977, espec. on
Libistro8, Kallimacho8, and Belthandros). He is struck by the fact
that most surviving manuscripts may be dated around the year
1500, that most are now in western European libraries, and that
some at least seem to have been written in the West. He suggests
that we should test the hypothesis that many may have been
written by educated Greek refugees in the West after the fall of
Constantinople, often in a language more uniformly popular than
the mixed form in which the original texts were written, as we
may see from the comparatively few older surviving manuscripts.
It remains to be seen whether there is any statistical signifIcance in

167
538 ELIZABETH &, MICHAEL JEFFREYS
the number of manuscripts preserved in the West, in view of the
large proportion of Greek manuscripts of all periods which have
found their way to the same libraries. However, if this fact is
enough to make at least a prima facie case for the writing of many
of these manuscripts in the West, then it is interestinlt to seek a.
way to include this new hypothesis in an overall picture of the
preservation of oral material. One could surmise, for example, that
the oral-based material in the West, say in Italy-and even the
texts of very limited oral pedigree to which we are referring-was
written down because the writers felt that they had lost contact
with the tradition. AJJ Lord has documented in Yugoslavia
(1960:155-56), those living in the range of a lively oral tradition
feel no need to write, since the tradition itself seems immortal.
However, one could surmise that a Greek refugee in Italy after
1453 might feel the need to preserve some sign of the oral tradition
he had left behind, and could have been stimulated to ma.ke a
copy of a written text in the style and language of that tradition.
The final point to be made has to do with the apparent
disjuncture between the poems we have been studying and most of
the poems of Modern Greek folk song. So far as we are able to
observe the latter, through scattered texts from the Turkish period
up to the eighteenth century, through earnest nineteenth-
century collecting, and through systematic study of the
twentieth-century remains, we must conclude that Modern Greek
folk songs tend to be short and lyrical, rather than long and
narrative like the poems spawned by the medieval tradition. At
first sight, we can only assume that the one tradition must have
died so that its successor could take over. On a more careful
examination, however, there appears a more conservative branch of
the modern tradition, with narrative songs in Crete and particularly
in Cyprus, which can be compared directly with the medieva.l
poems. On this point we may await the enlightenment of
Roderick Beaton, who has made a special study of the Cypriot
tradition. 29

University of Sydney

Notes

IThe handbook of H.-G. Beck (1971) gives a comprehensive survey and

168
 BYZANTINE POPULAR POETRY 539

full bibliography through 1971 of the texts d1scu8lled here, and should be
COJ18ulted for the older literature on the subject. For work between 1971 and
1978, see E. Jeffreys 1979 and 1981, Beaton 1980 (espec. 7-86), Eideneier
1982-83. See also the general discussion in M. Jeffreys 1975a.
2The moet accessible history of the Greek language is Browning 1983; see
also Browning 1978.
3This proposal stems from Trypanis (1963), who has put the idea into
practice, with wider claims than we would dare to support, in Trypanis 1981.
"Schmitt 1904; there exists a translation by Lurier (1964).
5Ms. N 272-77, 970-84, 1223-29, 1290-94, 1540-46 (Hesseling 1919).
6
E.g., E 1541, 1735, 1906 (Lambert 1935).
7Lanz 1844:par. ccxliv. On Thibaut de Cepoys, see du Cange 1657:vol.
2, 855.
8pariially ed. in Tsiknopouloe 1952; see M. Jeffreys 1974.
9 E.g., Eustathios of Thessalonika, in van der Valk 1971:vol. 1:19; trans.
in M. Jeffreys 1974:147.
l°Sev~enko 1969-70:185-228.
llTrans. from John Tzetzes, Theogonia (Bekker 184O:147~9).
12E.g., the empress Eirene, formerly Bertha von Sulzbach, flJ"lJt wife of
the emperor Manuel Komnenos; and the sebastokratorissa Eirene, widow of
Manuel's older brother. The one was certainly and the other probablY of
Western origiJ18, and both were dedicatees of simple introductions to aspects of
Greek classical culture.
13Mystical: e.g., Symeon the New Theologian (Koder 1969 and Kambylis
1976); Penitential: e.g., Nikephol'08 Ouranos (Papadapoul08-Kerameus 1899).
l"Constantine Porphyrogennetos, Le Livre des drimonies (Vogt 1967:vol.
2, chap. 82, pp. 165, 167). For discussioJ18, see Baud-Bovy 1946.
15The voluminous older literature is conveniently available in Beck
1971:63-97.
16For Ms. E, Alexiou 1979:33-35; for Ms. G, Oikonomides 1979.
17An intereeting argument is advanced in Macallister 1984.
18For editioJl8 of Glykas and the Spaneas poem, see Beck 1971:101-9; the
"Eisitirioi- for Agnes of France are edited in Stryzygowski 1901.
19M. Jeffreys 1975b. Different views can be found in Jacoby
1968:187-88, 1976; Spadaro 1959, 1960, 1961.
20 Chronicle 0/ the Morea 3179, 3469, 4683, 4885, 6249, 7556.
21 Achillei8, Ms. N 69, 96, 778.
22The material is most completely available in Psicharis 1886-89.
28Tables and lists of examples to support these statements are given in
M. Jeffreys 1972.
24For details of editiou, see Beck 1971.
25See the discussion and bibliographical detalls given in Spadaro 1975

169
540 ELIZABETH & MICHAEL JEFFREYS
(espee. 3(9).
26Given the continued delays in the appearance of the edition, see the
examples given in E. and M. Jeffreys 1979.
27On the manuscripts and their relationships, see the preliminary (but
still accurate) statement In E. Jeftreys 1916.
28 Ms. M: 1820 lines; Ms. L: 1363 lines; Ms. 0: 761 lines.

29 A first example is provided in Beaton 1980:162-68, 174-78. See also

his essay in an earlier issue of Oral Tradition (1986).
The two writers of this article have invested more of their research
labors than anyone else they can name in attempts to solve the problems of
Byzantine popular poetry, trying to develop comprehensive methods of analysis
from among the range of interests covered by Oral Tradition. Thus it has
proved impossible to carry through an impersonal and objective survey: it has
seemed more honest to present our views clearly, while explaining others'
objections and alternative proposals, and scrupulously noting all bibliography of
which we are aware. We hope to have avoided narcissism in referring to our
own writings, but such references remain uncomfortably frequent. Only the
reader may judge if the result is of any use.

References

Alexiou 1979
S. Alexiou. Akritikll. Ira.klio: Alexiou.
Alexiou 1982
·Paratirisis ston Akriti.- AriGdni, 1:41-57.
Alexiou 1983
·0 Dlgenis Akrltas tou Escorlal.- Prllktika ti, AkadimiGl
Athinon, 58:68-83.
Bachmann 1828
L. Bachmann. Aneedota GraeeIJ. Leipzig. Rpt. Hildesheim: Georg Olms,
1965.
Baud-Bovy 1946
S. Baud-Bovy. "Sur Ie chelidonisma.- B,zantina-Meta-byzantin",
1:23-32.
Baud-Bovy 1973
-I epikratisi tou dekapentasyUavou 8to eUiniko dimotiko
tragoudi.- Ellinik", 26:301-13.
Beaton 1980
Roderick Beaton. Folk Poetry oj Modem Greece. Cambridge: Cambridge
University Press.
Beaton 1981a
·Was Digenis Akritas an Oral Poem?- B,zantine and Modern
Greek St"die" 7:7-27.
Beaton 1981b
• 'Digenis Akritas' and Modem Greek Folk Song: A
Reaseesement.- B,zantion, 51:22-43.

170
 BYZANTINE POPULAR POETRY 541

Beaton 1986
"The Oral Traditions of Modern Greece: A Survey." Oral
Tradition, 1:110-33.
Beck 1966
H.-G. Beck. " Form problem des Akritas-epos." In his Beitr/Jge zur
SfldosteuropaJorschung. Munich: C. H. Beck.
Beck 1971
Geschichte der byzantinischen Volks/iteratur. Munich: C. H.
Beck.
Bekker 1838
I. Bekker, ed. (Theophanes ContinuatusJ. Bonn: E. Weber.
Bekker 1840
-0-"'----" ed. Theogonia, by John Tzetzes. Abhandlung der Berlin,
phil.-hist. Klasse:147-69.
Browning 1976
R. Browning. "Problems concerning the Genesis of Modern Greek." In
XVe Congres international des ttudes byzantinell. Athens. pp. 3-18.
Browning 1978
"The Language of Byzantine Literature." In The "Past" In
Medieval and Modern Greek Culture. Ed. S. VryoniB. Malibu: Undena.
pp. 103-33.
Browning 1983
Medieval and Modern Greek. 2nd ed. Cambridge: Cambridge
University Press.
Chatzidakis 1892
G. N. Chatzidakis. Einieitung In die neugriechillche Grammatik. Leipzig:
von Breitkopf und HILrtI.
Chatziyakoumis 1977
M. Chatziyakoumis. Ta mesaionika dimodi kimena: Ilymvoli sti melcti ke
stin ekdosi tou". Athens: Chatziyakoumis.
Dimaras 1965
D. Dimaras. Hi8toire de la litUralure neoheUenique, de8 origine8 a n08
jOflr8. Athens: IkIlJ'O!!.
du Cange 1657
C. du Cange. Histoire de I'empire de Constantinople sou" les empereurs
Jrant;ois. Paris.
Duggan 1966
Joseph J. Duggan. "Formulas in the Couronnement de Louis." Romania,
87:315-44.
Duggan 1984
"Oral Performance, Writing, and the Textual Tradition of the
Medieval Epic in the Romance Languages: The Example of the Song of
Roland.» Parergon, n.s. 2:79-96.
H. Eideneier 1982
H. Eideneier. ·Zum StH der byzantini.schen Tierdichtung." Jahrbuch der
()Bferreichi8cheR ByzantiniBfik, 32, iii:30 1-6.
H. Eideneier 1982-83
cLeser- oder HOrenkreis? Zur byzantinischen Dichtung in der

171
542 ELIZABETH & MICHAEL JEFFREYS

Volksspraehe." EUinika, 34:119-50.

H. and N. Eideneier 1978
and N. Eideneier. ·Zum fIlnfzehnsilber der Ptochoprodromika.-
In Aphieroma don kathigiti Lino Politi. Theesalonilti. pp. 1-7.
Fletcher 1977
R. A. Fletcher. ·Cultural and Intellectual Development 1821-1911.- In
Greece in TraMtion: Essays in the Hi&tory of Modem Greece 18IH-19t-4.
Ed. S. Koumoulides. London: Zeno. pp. 163-72.
Hlgg1983
T. Hlgg. The Novel in Antiquity. Oxford: BlackweU.
Heeseling 1919
D. C. Heeseling, ed. L'AchiU~ide byzantine. Verhandelingen der k. Akad.
van WeteD.IIChapen te Amsterdam, Afd. Letterkunde, n.r. 19.3.
Heeseling and Pernot 1910
and H. Pernot, eds. Po~mell prodromiquell en grec wlllaire.
Verhandellngen der k. Akad. van WeteD.IICha~ te Amsterdam, Md.
Letterkunde, n.r. 11.
Horandner 1974
W. HOrandner. Theodoroll Prodromo,: Hwomche Gedichte. Wiener
ByzantinistiBche Studien (CsterreichiBche Akademie der WisseD.IIChaften) , 11.
Horandner 1982
. ·Zur Frage der Metrik fr11her volksspraehlicher Texte." Jllhrbuch
-:de-r-O"-st"'7efTeiehi&chen Byzantini&tik, 32, ili:376-82.
Hunger 1978
H. Hunger. Die hochllprachliche profane Literatur der Byzantiner. 2 vols.
Munich: C. H. Beck.
Jacoby 1968
D. Jacoby. KQuelques consid~rations sur les versons de la 'Chronique de
Moree'." Journal du S(JlIGntll [for 1968):133-89 (=Jacoby 1975:VII).
Jacoby 1975
SocUU et dimogrGphie c\ ByzGnee et en RomGnie IGtine. London:
Variorum Preas.
Jacoby 1976
-Les Etata latins en Romanie: PMnom~nes soeiaus et
konomiques (1204-1350 environ)." In XVe Congra,l international de,
itudes .yzGntinell. Athens. pp. I-M (=Jacoby 1979:1).
Jacoby 1979
~--:-..-:' Recherchu.HIr IG MeditefTan~e orientale d. XlIe a. XVe Ili~elu.
London: Variorum Preas.
E. Jeffreys 1976
Elizabeth M. Jeffreys. ·The ManWlCripts and Sources of the W Gr 0/
Troy." In Adell d. XIVe congras international des ~t.dflS b,zGntines, 1971.
Bucharest: Academiei Republicii Socialiste Romania. pp. 91-94.
E. Jeffreys 1979
-The Popular Byzantine Verse Romances of Chivalry: Work
Done Since 1971." Mandato/oroll, 14:20-34.
E. Jeffreys 1980
-The Comnenian Background to the romarlll d'GntiquiU'

172
 BYZANTINE POPULAR POETRY 543
B,zantion, 60:466-86.
E. Jeffreys 1981
"The Later Greek Verse Romances: A Survey." In Byzantine
Papers. Ed. E. M. and M. J. Jeffreys and A. Moffatt. Canberra:
Humanities Research Centre, Australian National University. pp. 116-47.
E. and M. Jeffreys 1979
=-__ and Michael J. Jeffreys. "The Traditional Style of Early Greek
Demotic Verse." Byzantine and Modern Greek Studies, 6:116-39 (=
1983a:llI) .
E. and M. Jeffreys 1983a
_ _ _. Popular Literature in Late Byzantium. London: Variorum Press.
E. and M. Jeffreys 1983b
"The Style of Byzantine Popular Poetry: Recent Work."
HanJard Ukrainian Studies, 7:309-43.
M. Jeffreys 1972
Michael J. Jeffreys. "Studies in the Language and Style of the Medieval
Greek Chronicle of the Morea." Un pub. Ph.D. diss., London University.
M. Jeffreys 1973
"Formulas in the Chronicle of the Morea." Dumbarton Oak8
Papers, 27:163-96 (= E. and M. Jeffreys 1983a:II).
M. Jeffreys 1974
"The Nature and Origins of the Political Verse." Dumbarton
Oak8 Papers, 28:142-96 (= E. and M. Jeffreys 1983a:IV).
M. Jeffreys 1975a
"The Literary Emergence of Vernacular Greek." Mosaic, 8,
iv:171-93.
M. Jeffreys 1975b
"Digenis Akritas Manuscript Z.'" Dodone, 4:163-201 (= E. and
M. Jeffreys 1983a:V).
M. Jeffreys 1975e
"The Chronicle of the Morea: Priority of the Greek Version."
B,zantinische Zeitschn'ft, 68:304-60.
M. Jeffreys 1978
"Digenis Akritas and Kommagene." SIJenaka Forsming8inatitutet
i Istanbul Meddelanden, 3:5-28 (= E. and M. Jeffreys 1983a:VII).
M. Jeffreys 1981a
....--:-----:. "Byzantine Metrics: Non-literary Strata." Jahrbuch der
Osterreichiachen B,Izantinistilc, 31, i:313-34.
M. Jeffreys 1981b
:---"7. "The Eisitirioi for Agnes of France." In Byzantine Papers. Ed.
E. and M. Jeffreys and A. Moffatt. Canberra: HumanItIes Research
Centre, Australian National University. pp. 101-15.
Kambylis 1976
A. Kambylis, ed. HJfTlnen. Berlin and New York: De Grnyter.
Karayanni 1976
I. Karayanni. 0" Vigenia Akritd lou &corial: SymIJoli .di meleti lou
kimenou. Meletes sti Nea Elliniki Philologia (University of Ioannina), 1.

173
544 ELIZABETH &; MICHAEL JEFFREYS

Koder 1969
J. Koder, ed. Hymnes. Sources Chretiennes, 156. Paris: Editions du
Cerro
Koder 1972
-=-_-.,.:. -Der Fnnfzehnsllber am Kaiserlichen Hot um das Jahr 900."
Byzantino61avica, 33:214-19.
Koder 1982
"Die Chronik der Tocco. Zum Metrum und zum VerhlLltniss
zur Chronik von Morea." Jahrbuch der 08terreichi8chen Byzantini8tik, 32,
iii:383-92.
Koder 1983
. "Konta.kion und politiBcher Verse." Jahrbuch der
'O's-te-rr-e-:ichi8cflen Byzantinistik, 33:45-56.
Kougeas 1913-14
S. B. KoUge88. "E en tis scholiis tou Aretha la.ographike eidisis."
Laographia, 4:239-40.
KriaraB 1955
E. KriaraB, ed. Byzantina [ppotika Mythistorimata. Athens: Aet08.
Lambert 1935
J. A. Lambert, ed. Le Roman de Libistro8 et Rhodamne. Verhandelingen
der k. Aka.d. van Wetenschapen te Amsterdam, Afd. Letterkunde, n.r. 35.
Lanz 1844
K. Lanz, ed. Cflro·nica, by Raymond Muntaner. Stuttgart: Literarischer
Verein.
Lambros 1916
S. Lambros. "Trapezoundia.kon oroskopon tou etous 1336." Neo8
Ellinomnimon, 13:33-50.
Lavagninl 1983
B. Lavagnini. "AIle origini del verso politico." [stituto Siciliano di Studi
Bizantini e Neoellenici, Qua.derni 1l.
Leone 1968
P. A. M. Leone, ed. [oanne8 Tzetzae Historiae. Istituto di filologia
cl888ica, Publicazioni, 1. Naples: Libreria Scientifica Editrice.
Longnon 1939
J. Longnon. ale Prince de Moree chan80nnier." Romania, 65:95-100.
Lord 1953
Albert B. Lord. "Homer's Originality: Oral Dictated Texts." Transactions
01 the American Philological Association, 84:124-34.
Lord 1954
"Notes on Digenis Akritas and Serbo-Croatian Epic." Harvard
Slavic Studies, 2:375-83.
Lord 1960
=-_---'. The Singer 01 Tales. Cambridge, MA: Harvard University
Press, 1960. Rpt. New York: Atheneum, 1968 et seq.
Lurier 1964
H. E. Lurier. Crusader8 as Conqueror8: The Chronicle 01 Morea. New
York: Columbia University Press.

174
 BYZANTINE POPULAR POETRY 545

Luzzatto 1983
M. J. Luzzatto. "La Datazione della Collectio Augustana di Esopo ed il
verso politico delle origini." Jahrbuch der Osterreichischen Byzantinistik,
33:137-78.
Macallister 1984
S. Macallister. "Digenis Akritas: The First Scene with the apelatai.-
Byzantion, 54:551-74.
Magoun 1955
Francis P. Magoun, Jr. "Bede's Story of Credman: The Case History of
an Anglo-Saxon Oral Singer." Speculum, 30:49-63.
Michaelidis 1971-72
D. Michaelidis. "Palamedes Rediens: La fortuna di Palamede nel medievo
ellenico." Ri1lista di studi bizantini e neoellenici. n.s. 8-9 [18-19):261-80.
Mohay 1974-75
A. Mohay. "Schriftlichkeit und Mundlichkeit in der byzantinischen
Literatur." Acta Classica (Debrecen). 11:175-8l.
Noiret 1889
H. Noiret. Lettres inldites de Michel Apostolis. Paris: E. Thorin.
Norgaard and Smith 1975
L. Norgaard and O. L. Smith, eds. A Byzantine Riad. Copenhagen:
Museum Tusculanum.
Oikonomides 1979
N. Oikonomides. "1' 'Epo~' de Digenis et la fronti~re orientale de
Byzance au Xe et Xle si~cles." Travaux t:t mlmoires, 7:375-97.
Papadopoul08-Kerameus 1899
A. Papadopoul08-Kerameus, ed. "Byzantina Analekta." Byzantinische
Zeitschrift, 8:66-70.
Parry 1932
Milman Parry. "Studies in the Epic Technique of Oral Verse-Making. II.
The Homeric Language as the Language of an Oral Poetry." Harvard
Studies in Classical Philology. 43:1-50.
Politis 1970
L. Politis. "Probl~mes de la tradition du texte et des rapports avec les
chansons akritiques." In Atti del convegno internazionale sui tema: La
Poesia epic a e la sua formazione, 1969. Rome: Accademia Nazionale dei
Lincei. pp. 551-81.
Politis 1973
A History of Modern Greek Literature. Oxford: Oxford
University Press, rpt. 1975.
Politis 1981
"Neoteres apopsis ya to gennesi ke ti domi tou
dekapentasyllavou." Praktika tis Akadimias Athinon, 56:209-28.
Psicharis 1886-89
J. Psicharis. Essais de grammaire historique neogrecque. 2 vols. Paris:
Bouillon.
Reichmann 1963
S. Reichmann, ed. Das byzantinische Alexandergedicht. Meisenheim am
Glan: A. Hain.

175
546 ELIZABETH & MICHAEL JEFFREYS

Schiro 1975
G. Schiro, ed. Cronoca dei Tocco di Cefalonia. Rome: Accademia
Nazionale dei Lincei.
Schmitt 1904
J. Schmitt, ed. The Chronicle of the Morea. London: Methuen. Rpt.
Gronlngen: Bouma, 1901.
Schopen 1829
L. Schopen, ed. Nikephoros Gregoras. Bonn: E. Weber.
Schreiner 1966
H. Schreiner. "Die Einleitenden Handschriften zu den von der gleichen
Hand Ilberlieferten Texten in Cod. Neap. Gr. III Aa 9 und Co. Neap. Gr.
III B 27." In Polychordia: Festschrift F. D~lger. Heidelberg: A. Hakkert.
pp. 290-320.
Sevtenko 1969-70
I. Sevtenko. "Poems on the Deaths of Leo VI and Constantine VII in the
Madrid Manuscript of Scylitzes." Dumbarlon Oaks Papers, 23-24:185-228.
Spadaro 1959
G. Spadaro. "Studi introduttivi alia Cronaca di Morea I." Sicu/orum
Gymnasium, 12: 125-52.
Spadaro 1960
"Studi introduttivi. . II.- Sicu/orum Gymnasium, 13:133-76.
Spadaro 1961
"Studi introduttivi. . III." Sicu/orum Gymnasium, 14:1-70.
Spadaro 1975
·Problemi relativi ai romanzi greci dell'eta dei paleologi I:
Rapporti tra Imberi08 e Ma.rgarona e Flori08 e Platzia-Floria." Ellinika,
28:302-37.
Spadaro 1976
"Problemi . . . II: Rapporti tra Diigisis tou Achille08. 130
Diigisis tou Velisariou e Imberi08ke Ma.rgarona." Ellinika, 29:278-310.
Spadaro 1977
_ _ _" ·SuI Teseida. neogreco." Folio Neohellenico, 2:157-60.
Spadaro 1977-78
"Problemi . . . III: Achilleide, GeorgillM, Callimoco, Beltondro,
Libistro, Florio, Imberio, e Diigisis genameni en Tria." Elliniko, 30:223-79.
Spadaro 1978
"L'inedito PoIem08 tis Troad08 e I' Achilleide.'; ByzantiniBche
Zeitschrift. 71:1-9.
Spadaro 1979
_ _ _" ·Spaneas e Glikas: Note filologiche." Diptycho, 1:282-90.
Stryzygowski 1901
J. Stryzygowski. "Das Epithalamium des PalloIogen Andronik08 II.·
ByzantiniBche Zeitschrift, 10:546-67.
Topping 1977
P. Topping. Studies in Latin Greece AD It05-1115. London: Variorum
Press.
Trapp 1971
E. Trapp, ed. D ig c n c s A b ritc a : SynoptiBche A usgobe der /JIteBten

176
 BYZANTINE POPULAR POETRY 547

Versionen. Wiener Byzantinische Studien (Osterreichische Akadernie der

Wissenschaften) , 8.
Trypanis 1963
C. A. Trypanis. "Byzantine Oral Poetry." Byzantinische Zeitschri/t,
56:1-3.
Trypanis 1981
_ _ _" Greek Poetry from Homer to Seferis. London: Faber.
Tsiknopoulos 1952
I. P. Tsiknopoulos. "I piitiki paragogi tou englistou agiou Neophytou."
Kypriake spoude, 16:41-49.
van der Valk 1971
M. van der Valk, ed. Commentarii ad Homeri Riadem pertinentes, by
Eustathios of Thessalonika. Leiden: E. J. Brill.
van Dieten 1975
I. A. van Dieten. Nicetae Choniatae Historia. Berlin and New York: De
Gruyter.
van Gernert 1975
A. F. van Gernert. "The New Manuscript of the History 01 Belisarios."
Folia N eohellenica, 1:45-7l.
van Gernert and Bakker 1981
and W. F. Bakker. "I Achiliida ke istoria tou Velisariou."
Ellinika 33:82-97.
Vita
Vita Stephani luniori8, Patrologia Graeca 100, col. 1116.
Vitti 1971
M. Vitti. Storia della letteratura neogreca. Letteratura e civilita, 16.
Turin: Edizioni Rai Radiotelevisione Italiana.
Vogt 1967
A. Vogt, ed. Le Livre des Ctrtmonie8, by Constantine Porphyrogennetos.
2 vols. Paris: Societe d'EditiODB "Les Belles Lettres."
Wagner 1870
W. Wagner, ed. Medieval Greek Text8. London. Rpt. Hildesheirn:
Gerstenberg, 1971.
Zachariadou 1983
E. Zachariadou. "I chilli stichi stin archi tou Chronikou ton Tocco."
/pirotika Chronika, 25:158-81.

177
This page intentionally left blank
 THE FUNCTION AND EVOLUTION
OF BYZANTINE RHETORIC

by George L. K ustas

In his Prolegomena Lo Rhetoric a fifth-century anonymous author remarks

that the function of rhetoric varies with the type of polity. Among the an-
cient Lacedaemonians rhetoric served the aims of oligarchy; among the Athe-
nians, democracy; "we practice it in faith and orthodoxy under an empire."l
The statement expresses in concise terms the new role that rhetoric had come
to play in Byzantine life. ~ot only had the system of government changed;
it had assumed a religious habit. The devices of rhetoric must be adapte.d
to serve not merely an empire but the Christian Empire of the East, with
its political roots in its Roman past and its cultural heritage the educational
ideals and techniques of late Antiquity. My purpose is to review the bases
of the Byzantine rhetorical tradition as it arose in the early Christian centuries,
to indicate the interplay between it and the forms of Byzantine literature,
and to trace the changes that were rung on this pagan legacy in response
to the evolving patterns and challenges of Byzantine civilization.
Byzantine rhetoric is throughout its history the heir of the Second Sophistic,
that movement in thought and letters which extends from the time of Au-
gustus to the end of the ancient world. Within the movement there are wide
variations, and in its later history it exists side by side with the literature
of Christianity which drew on it. If the Christian Chrysostom is the student
of the pagan Libanius, sophist from Antioch, it is equally notable that the
school of Gaza in the sixth century is peopled by Christian sophists. The
Second Sophistic has in fact no clear terminus. One should rather say that
it shades off into the Byzantine world of letters. In rhetoric as in other fields
there can have been little awareness of the break, so useful to modern analy-
sis, between late antique and Byzantine. Precisely because the tradition
remained alive there never developed in Byzantium a uniquely Christian

1 Prolegomenon Sy/loge, ed. H. Rabe (Leipzig 1925) 41.7-9: 'A(1)vaiot M xai ~ TeOV Qt)-
YOeWV &"ci; £;"oAtTH'aayO /liv ,u; tv ory/lO"eaTiq- Aa"toal/lovtol (ii, tV; tv aglaYo,,(!a-
TEq . ~Wi; o£ vvv tVTVXW; tv {Jaal;.eLq .... IOTW, xai oeOooo;WC;. Cf. also 38.8: ~/lti; auv
Otep /lEueXo/ldJa Ti}v T(!Ll!)V (Le. type of rhetoric). I wish to thank my colleague Professor
Westerink for many helpful suggestions in the writing of this paper.

179
56 GEORGE KUSTAS

rhetoric existing as an entity apart and distinct from its Sophistic for-
bears. The changes within this received framework are slow and sometimes
subtle, but, for all that, there is, as we shall see, a clear pattern of develop-
ment.
Starting from the latter part of the second century the literary practices
and ideals of the Second Sophistic begin to be codified into the systems or
textbooks that are henceforth to form the subject of instruction and the
basis of literary performance and analysis for the succeeding centuries. These
are the treatises that make up the Byzantine rhetorical tradition; they are
copied many times 0\-eT down to the late Middle Ages, commented upon,
excerpted from, and on occasion altered in response to Christian demand.
I mention two authors in particular: first, Hermogenes, the second- and
third-century rhetorician from Tarsus, and second, the fourth- and fifth-cen-
tury figure, Aphthonius of Antioch. 2 Byzantium knew and used other hand-
books. One can trace the influence of Theon, Hermagoras, and Menander,
and record the use of rhetorical masters such as Dionysius of Halicarnassus,
Demetrius, Synesius, and Aristides, particularly after the tenth century; but
they come much less to the fore than the eyer present and eyer used outlines
of Hermogenes and his commentator, Aphthonius. 3
The corpus of Hermogenic writings offers little that is original in the crea-
tive sense of the word. Hermogenes' works, however, have the virtue of
clear exposition and arrangement of the subject matter, from which a large
element of their appeal must derive. At the same time the five treatises in
the corpus 4 were the only attempt to cover the whole of rhetoric. The Second
Sophistic had seen a trend away from the practical application of rhetoric
for the law courts and political ceremonies of the empire. Rhetoric was now
cultivated more and more as an academic subject, meant to supply the cur-
riculum of higher education. Hermogenes' work suited the purpose admirably.
The text is well written and its content, by no means easy-it was meant after
all for the higher schools-has the virtue of a studied simplicity, a quality that

2 }Iuch of the \Yalzian corpus is a selection from some of these commentaries on these
two authors. The wealth of the manuscript tradition may be inferred from the massive
lists in the introduction to the Teubner text: Hermogenis Opera, Rhelores graee; 6, and
Aphthonii Progymnasmala, Rhelores graeei 10, both ed. H. Rabe (Leipzig 1913, 1926).
Rabe's two introductions should be supplemented by the valuable accounts of Raderma-
cher and Brzoska in Pauly· Wissowa, Real-Eneyc/opiidie.
3 In focusing on Hermogenes and Aphthonius I do not wish to underestimate other
contributions. The study of Byzantine rhetoric is in its infancy, and in this paper I do not
propose more than a sketch of its main outlines. Future research must work in part towards
tracing the fortunes of other figures through specialized stUdieS on their rhetorical methods
and vocabulary.
4 fleoyvjJ.vaUjJ.aTa, flee; UTaUeWV, flee; eVefueWr;, flee; lOewv, fleel /JefJ66ov 6£1"0-
nITO;. Rabe considers the fleoyv/J"au/JaTa and the flee' ev(!iuewr; spurious.

180
 FUNCTIOI'< AI'<D EVOLUTION OF BYZANTINE RHETORIC 57

was to strike a yery sympathetic Byzantine nerye and one that the author
actually holds out as an ideal of style in his writings.'
Hermogenes had the good fortune to be received not long after his death
into the bosom of the Neoplatonists. The earliest commentary of which we
hear is by the third-century philosopher, l\letrophanes of Phrygia, and no
less a person than Iamblichus further enhanced Hermogenes' reputation by
declaring in his favor over the claims of rival systems.' Our earliest extant
commentary is by Syrianus, the fifth-century Athenian scholar.'
Before we trace the later fortunes of Hermogenes, let us consider his partner,
Aphthonius. The work by which Byzantium knew him was the Progymnas-
mala. The word refers to those school exercises, praeexercitamenta in Latin,
which had since Hellenistic times formed part of the instruction in the an-
cient tongues. A progymnasma is essentially a set composition, in which
one could presumably exercise the techniques of style laid down in such
textbooks as those of Hermogenes. 8 There is an inner logic in the fact that
Aphthonius, through whom the Christian world was to learn so much of its
rhetoric, was, like Chrysostom, a student of the pagan Libanius. Libanius
also wrote progymnasmata. The special success of Aphthonius lies in the sim-
plicity of his exposition as well as in his inclusion of examples for each of
the types under discussion. His examples are hardly his own creation. As in
the case of most rhetorical works, we can hardly ever hope to trace such
formulas back to their origins, and we can only console ourselves that it
would be little gain to the history of ideas to know the name of the school-
master who first thought of asking his pupils to write a composition repro-
ducing Ajax's thoughts before his suicide or Danae's reaction to Zeus's golden
shower.9
The standard number of progymnasmata is fourteen. They are divided
by the scholiasts on Aphthonius into three categories, symbouJeutic, dicanic,
and panegyrical, the tripartite arrangement that is at least as old as Aris-
totle. 10 Cnder the first are ranged 1) f-Iv()or;, myth; 2) XgEia, ethical thought;

5 Note the heavy emphasis in the nEe! ltluiiv on such concepts as ",aOaeOT7]C;, EVy.eiVEta,
aallnlvEta, E:.UiY.Eta, a<p"AEta, aA~OEta, and how to achieve each.
6 For the relation of these and other figures of the time to Hermogenes see Christ-Schmid-
Stahlin, Geschichle der griechischen Lilleralur 7.2.2, ed. 6 (ylunich 1924) 934-936.
7 Syriani in Hermogenem commenlaria, ed. H. Rabe (Leipzig 1892-1893) 2 vols.
8 Tn tii TOii 'A<pOoviov 7l(!oyv!-,vaa!-,aTO Elaaywy~ TIC; OVTO :'teo; il<Eiva (Le., Tn Toii '£e-
!-,oyivovC; {J1{J).{a): Rhelores graeci 2.566.23.
9 Ajax: Libanii opera, ed. R. Foerster (Leipzig 1925), 8.384ff.; Danae: :-iicephorus Ba-
silaces, Rhetores graeei 1.476.
10 Rhelores graeci 2.567.7. See also Matthew Camariotes, Epilome, Rhetores graeci 1.120.
10; Anonymous on Aphthonius, Rhelores graeei 1.127.16. For other, albeit basically similar
ways of dividing the progymnasmata, see the chart provided by Stegemann, s.v. "Theon"
in Pauly-Wissowa, Real-Eneylopadie, cols. 2043, 2044.

181
58 GEORGE KUSTAS

and 3) YVWI1'IJ, a maxim or saying.!1 Four others are dicanic: 1) aVaay.Ev~,

the refutation of a ginn statement; 2) XaTaay.Ev~, its opposite, or confir-
mation; 3) darpor;/J. l'ofloV, a discussion whether a given law is good or bad;
and 4) XOIVO; Tono;, communis locus, by which is meant the amplification of
a giyen topic. These judicial types were among the first to be affected by
the decline of political oratory under the Roman Empire. At the same time
that they were, on one len!, retained as school exercises, they took a new
lease on life by lending their resources to new purposes. 'Avaaxw~ and
xaTaaxfV~ were early turned into weapons of historical criticism. The Chris-
tian apologists used the techniques to counter the myths of the pagans,
and Byzantine theological literature followed suit.12 Cicero tells us that the
communis locus plays on two emotions in particular: indignalio and miseri-
cordia. I3 Like the Elarpo(jG. V0I101·,14 it had comparatinly little independent
existence in Byzantium, but, insofar as the appeal of many a homily and
epistle rests on the evocation of just these two states of mind, it had a long
and yibrant echo through the whole of the Middle Ages.
The panegyrical types are four: 1) Eyxwflwv, encomium; 2) its opposite,
tpoyo;, which censure, helped supply the long vocabulary of Byzantine in-
yecth'e before which one can only stand back in amazement; 3) avyx(jlal:;,
comparison; and 4) ij()o:wlta, characterization. "Ey.rp(!aa.;, description (of
people, places, and things), and oI1IY'lJl1a, narrative, are common to all three
types, while eiau;, the posing of a question of general interest, partakes
of both the symbouleutic and the panegyrical. Although all the progym-
nasmata were current throughout Byzantine history, Byzantium paid more
attention to the symbouleutic and the panegyrical than to the dicanic.
The clear correlation between these preferences and the main categories of
Byzantine literature is striking proof of their vitality.
Letter writing had not been particularly developed as a genre in ancient
Greece. Its heyday belongs rather to the Roman age and goes hand in hand
with the emphasis on indiyidual portraiture and character expression that

11 Apthonius's definitions, which are not original with him, are as follows: xe£<o EUTi"
anoflvllflOVfVflO avvToflov EvaToxw; in{ Tt neoawnov ava<pieovaa: 3.21; YVWflTl eaTi },oyo,
tv ano<pavawl y.E<pa},alwO'rjr; in{ TI n(!OT(!i:lwv ij ,i;.wT(!Enwv: 7.2. See also Theon, the oldest
of the writers on pl'ogymnasmata available to us (second century of our era), Progymnasmata,
ed. L. Spengel, Rhetores graeci (Leipzig 1854) 2.96.19: x(!da eaTI avvTOflor; ano<paalr; tj
ne!Ulr; flET' €V(J1oxiar; ava<pE(!Ofliv'rj €i; TI w(!laflivov n(!oawnov ij ava).oyovv neoawntp .
na(!Uy.ElTal Oi aVTfi YVWfI'rj uat anoflv'rjflovEvflaTa . nuaa yd(! YVWfITl avvToflOr; Ei; n(!o-
awnov dva<pE(!Ofliv'rj x(!dav nOIEt.
12 H. Lausberg, Handbuch der literarischen Rhetorik (Munich 1960) sect. 1125.
13 De inn. 2.16.51.
14 This particular progymnasma, as a matter of fact, was from the beginning not quite
on a par with the rest. See Hermogenes 26.11: y.at T~V TOV I'0flOV da<po(!dv Tanova{ TIVEr;
iv yt'/lVUaflaa.. Similarly Aphthonius 46.20. One of the commentators, Rhetores graeci
1.59i-648, omits it altogether.

182
FUNCTION A:-ID EVOLUTION OF BYZA:-iTI:-iE RHETORIC 59

is so marked a feature of Roman portrait art. During the Second Sophistic

epistolOgraphy invaded Greek literature more and more: we need only think
of such large collections as the letters of AJciphron and of Libanius. In the
process, rhetoricians undertook to distinguish its various forms and to define
the rules of composition. By the fifth century the pseudo- Libanius tradi-
tion recognized 41 distinct types of letter,IS but even this number was to
prove inadequate to express the variegated refinements of the Byzantine
mind. A late tradition lists 113 different possibilities.I6 There is hardly a
Byzantine author without his collection of letters. Epistolography is one of
the most widely used and most successful of medieval literary forms. From
the point of view of rhetorical theory, it falls under the heading of ~(jo­
nodaP the progymnasma par excellence which gave the freest scope to the
expression of personality traits.
This particular legacy of the Second Sophistic, however, did not reach
Byzantium only through pagan channels. Christianity had introduced itself
to the world in the form of a letter. IS The importance of Saint Paul's Epistles,
as witness the extensive and frequent byzantine commentaries, cannot be
exaggerated. The religion of Jesus had accentuated the sense of the indivi-
dual person or circumstance by emphasizing the private and unique rela-
tionship of the human soul to its Creator. In addition, the deYelopment of
Christian theology had demanded of the Fathers a clear and close exposition
of views by way of refining orthodox belief and combating heretical opinion.
It is not without reason that Gregory of ~azianzus, Gregory of :\yssa, Basil,
and Chrysostom are incorporated into a Christian canon of epistolography
and frequently cited by the rhetoricians of the later centuries together with
their pagan predecessors. The Cappadocians gave to epistolography its emi-
nently practical Christian character," and the lesson was not lost among that
long parade of emperors and priests who chose to express themselves on a
limitless range of subjects through the medium of their letters.
What is more, this emphasis on the personal was not restricted to letter
writing. Many other literary forms, such as general histories, theological
treatises, or even scientific tracts, come to be not merely dedicated, as in
Antiquity, but now addressed to some particular person. 1 he influence
that 1](}onotta exercised upon other forms of literature and the high esteem
in which it was held in Byzantine circles are nowhere better illustrated than
in an anonymous scholium on Aphthonius that ~(jonotta is the perfect kind
of progymnasma and in this capacity contributes to the lntaTo;.tf1a!o~ Xa-

ed. V. Weicheft (Leipzig 1910) 1.J.

15 'EmGTo).tfl Qiot xQeQXTijl}£;,
16 Ibid. 34.
17 See Nicolaus, Progymnasmala, ed. J. Felten (Leipzig 1913) 6i.2.
18 J. Sykutris, Pauly-Wissowa, Real-Encyc/opddie supp. 5 s.v. "Epistolographie" col.
219.24.
19 Ibid. 219.39.

183
 60 GEORGE Kl:STAS

l!aXT~I!'W Not only the epistle but the homily as well had a large stake in it.
Originally, under the Second Sophistic, the person of the orator displaying
his virtuosity had become more prominent as against the virtuosity of the
production. The Christian homilist through his address to his congregation
now achieved the same effect as the pagan sophist.
At the same time, ¥)o:rolta is practiced extensively as a form of literature
in its own right. Here we can distinguish two types of compositions: those
that followed the old pagan models and those that adopted Christian themes.
The origin of the Christian topics is as obscure as that of their pagan equi-
valents. It is clear, however, that the techniques of Christian education were
elaborated in the fourth century and developed and multiplied throughout
the fifth and sixth centuries. That the school of Gaza played an important
role is more than likely.2l The writings of the Gazaeans continued very po-
pular in Byzantium, and in their texts one will find quotations from Homer
side by side with passages from Isaiah. The pious Christian could now
release his imagination to ponder not Ajax, but what Samson said upon
being blinded, what the Virgin might have remarked upon seeing her son
change the water into ~;ne, or-a mixture of pagan and Christian-Hades's
remarks on learning of Lazarus's resurrection. 22 The most that can be said
for these pursuits is that they may have promoted the cause of Christian
piety. Yet behind the 1jF)o:rolta, as well as behind the forms of literature
it affected, lurks the informing presence of Aphthonius, rhetorician from
Antioch.
Just as with the ~()o:rolta, so too in the encomium one can detect three
different levels of transmission from Rome to Byzantium. The first keeps
to the letter of the old tradition. Encomia continue to be formed on the
basis of the pagan prescriptions. In the second, the outer structure is again
retained, but the author heaps his praise on Christian models. The third is
the most fruitful, for it uses the resources of the encomium to guide and adorn
other forms of literature. One thinks of the panegyrics composed by the Cap-
padocians and later Fathers in honor of Christian martyrs; the catalogues of
praise that are a vital ingredient of the saints' lives; the appreciation, often in
the form of a letter, of the virtues of the Fathers and, in the secular sphere, of

20 Rhelores graeei 2.52.1.

21 ~Iuch work still needs to be done to clarify the text tradition and to assess the con-
tribution of the Gazaean authors. See the old but useful account by K. Seitz, Die Schule
von Caza (Heidelberg 1892), as well as Aly's recent discussion of Procopius in Pauly-Wisso-
\Va, Real-Eneyc/opadie; also G, Downey. "The Christian Schools of Palestine," Harvard
Library Bulletin 12 (1958) 297-319,
22 One of the more interesting treatments of such topics is by Nicephorus Basilaces in
the middle of the twelfth century, Rheiores graee; 1. 466ff., where such ,,(Jonottat are pre-
sented side by side with those of pagan themes, whether mythological (Zeus, Ajax) or
historical (Xerxes 487).

184
Ft::"CTIO:" A~D EVOLt:TJON OF BYZA:-;T1NE RHETORIC 61

learned or otherwise distinguished contemporaries; the admiration and rever-

ence with which the Byzantine exegete approached the Gospel text; and,
not least commonly, the praise of the Lord of Heaven and other persons of
the Christian pantheon through homily or hymn. The attitudes and functions
that define this religious literature were nourished early in the career of the
schoolboy and later university student by rhetorical handbooks, of which
the summary of Aphthonius was the most popular. I do not mean to sug-
gest that the Byzantines owed the nobility of their religious vision or their
cultural achievements to what is admittedly a thoroughly pedestrian compen-
dium of rhetorical rules. Rather, the retention of such legacies came about
because they had a certain natural affinity with Byzantium's own interests.
We cannot hope to decide at any given point whether we are dealing with the
Christianization of Hellenism 23 or the Hellenization of Christianity; so in-
timately entwined are the two strands and so inseparable a complement do
they form to one another in the tapestry of medie\'al culture.
The third of the panegyrical forms is the aVyxl]tat;. It cuts across the other
rhetorical types and affects them all. A simile is one of its simplest expres-
sions. A Plutarchian parallel Life is another. LVYXl]tat; is indispensable
to the encomium (how better to stress the virtues of your subject than by
comparing him to a lesser man), and it is common in the bapQaat; (how
much better this mosaic is than that of Zeuxis; this icon puts Pheidias
to sh'1me).24 The habit of the comparison is built into the system of pro-
gymnasmata itself: aVaaXEV1) is followed by xaraaXEVI), encomium by 1pora:;,
and (Hat:; suggests avril3wt:;. Although the comparison appears in many of
the genres of Byzantine literatu~e, it is particularly effective in the homilies
with their exhortations to the good life: the works of the de\'il are contrasted
with the works of God, the grossness of heathenism with the beauty of the
Christian revelation, the way of the sinner with the way of the pious. The
technique of comparison is already biblical: witness the parable of the foolish
virgins, and the same sense of contrast must have impressed itself on the
early Fathers who saw themselves ranged on opposite sides of a dogmatic
issue or against the power of Rome as they sought to develop a scheme of
Christian doctrine. In one of his letters, Gregory of :\'azianzus, instructing a
friend in the principles of effective writing, encourages the use of mythology
as a form of aVyxQwt;.25 Finally, the technique finds its most fruitful and

23 :'\othing better illustrates the impact of religion upon the progymnasmata than the
fortunes of the >poyo;, the opposite of the encomium. In their original secular functions,
the two would be on a par and each had a purely human I eference that derived ultimately
from the equal rights of plaintiff and defendant in the courtroom. In a religious society
whose instruments are devoted to the celebration of the Deity. 'p%; often appears as
part of the encomium. serving to attack heretics and nonbelievers.
24 See ,I. Guignet, SI. Gregoire de .Ya:ian:e el fa rhl!loriqIle (Paris 1911) 189.
25 Letter 46, PG 3i.96A.

185
62 GEORGE KlJSTAS

original development in literary criticism. The many comparisons to be found

among the literary notices in Photius's Bibliolheca are guided by his sense
of historical fitness. He will compare only works of literature which are
legitimately matched and will use a critical vocabulary carefully chosen so
as to be generally contemporary with the authors under review. 26
Although the l%rpljaut; could apply as well to people as to things, some
of our best examples describe the realm of nature and, as a special cate-
gory, works of art. In the general decline of poetry during the Second So-
phistic and in some periods of Byzantine history, it supplied a kind of ly-
ric in prose in which the word-painter could give rein to his talents. The
Exrpljaut; has gh'en us some of the most beautiful passages in Byzantine
literature. The particular attraction that it had for the medieval mind is
not difficult to see. As Guignet points out, the immobility and fixity of form
of a work of art permitted the unhampered pursuit of detail and made pos-
sible, at least theoretically, the attainment of perfection in the genre. 27 At
the same time that it sen'ed the cause of art, the ExrpljaUt; in its larger sense
aided the cause of religion. The descriptions of the heavenly majesty in
the religious poetry of Byzantium, the inspired variety of address to the
Deity, the deeply felt recital of the mercy and providence of God, must have
been initially fostered in the Byzantine classroom. Further, the EXrpeaUt:; is
one of the most Christianized of forms. Bet,,'een the sixth and the eleventh
centuries practically all the examples we have are devoted to Christian works.
Then, with the re\'ival of interest in pagan subjects, we begin to find bupeauct;
of classical art. PselJus's description of two representations of the Circe epi-
sode in Homer and Constantine Manasses's of an antique relief of Odys-
seus and Polyphemus are among the earliest. 28
Aphthonius defines the liv()O; as "a false saying which mirrors the truth,"29
the same definition that appears in the lexica of Hesychius, Photius, and the
Suda. The extensive use of the fable is a continuation from the period of the
Second Sophistic, when the collections of fables in prose and verse that have
come down to us were originally compiled. In Byzantium the liV()o; answered
the ethical demands of the religious mind, and its popularity may be due in
part to the momentary freedom that its fictional appeal gave from the con-
finements of dogma and the strictures of doctrina prescription.

26 This subject has been treated in extenso in my "The literary criticism of Photius,
a Christian definition of style," 'Ei.i.l)Vlxd Ii (1962) 132-169.
27 SI. Gregoire 20i,
28 Psellus: T:etwe al/egoriae lliadis; accedllnt Pselli allegoriae, ed. J, Boissonade (Paris
1851) 363-365; M. Pselli scripta minora 2: Epistuiae, ed. E, Kurtz Cllilan 1941) letter 188,
20i-209. For :l1anasses see C. "'Iango. "Antique statuary and the Byzantine beholder,"
Dumbarton Oaks Papers 17 (Washington 1963) 65,
29 1.6: EaT! (ji J-lvOO; ;.6yo; 'f'Eva1j; Elxovi;wv T1jV d;.~eEIaV,

186
FL~CTION AND EVOLL'TION OF BYZANTINE RHETORIC 63

X(!Ela and yvwfl'YJ also fall in the symbouleutic category,30 the one a specific
quotation from a given author, the other a general maxim. Some of the texts
continue to be taken from the ancient tradition. Xicephorus Basilaces, a
twelfth-century writer, professor of Gospel exegesis and author of a rheto-
rical manual, urges Sophocles as a model for the X!!clu; and an anonymous
scholiast cites a Demosthenic yyWfl'YJ.31 The x!!C£a has its Christian echoes
in the exegetical commentaries of Byzantium, with their investigations of
scriptural meaning, as well as in the quotations from the early Fathers and
in the vast body of catenae literature.
The predilection for the important saying, the mot juste, the epigrammatic,
has its other side as well. The lexicographers are an instance of the fragmenta-
tion of learning in pursuit of the subtle and unique. The extensive collections
of the paroemiographers and the excerpts from classical and Christian works
that begin in the ninth century and continue apace through the scholarly
labors of Constantine Porphyrogenitus are affected by the same spirit. In
judging all these efforts, however, we have to make a distinction between
the interest in the unique and in the uniquely representative. When the
Byzantines pursued the former, they often degenerated into mere purism;
when they sought the latter, we see them at their best, as in the great periods
of their art. Both possibilities derive from the same source and for both
rhetoric had a significant role to play.
The highest achievement of the Christian prose literature is the homily.
Many of the techniques of rhetoric contribute to its formation and lend their
resources to its power and beauty. One of the best illustrations is the recently
edited and translated corpus of nineteen Photian homilies. 32 The selection
of the canon, made probably by the author himself, is such as to give represen-
tation to the major categories we have been discussing: two of the homilies are
in the form of an f,apeU01,; two are encomia; one has extensive citations from
the pagan myths in condemnation of those who prefer them to the Christian
re .... elation; two are narratives of the history of the Arian contrO"l;ersy; and
throughout them all we see a free use of scriptural quotation, folk pro .... erb,
as well as character description. Tile author knew Herrnogenes intimately,33
and his acquaintance with Aphthonius is evident from the notice in the Bi-
bliotheca codex 133. 34 To be sure, the corpus of the homilies was not com-
piled with them in mind; but it is clear that one of the most splendid monu-

30 See n. 11 above.
31 Rhetores graec; 1.445, 605.
32 <J>wTlov o,.I).ial ed. B. Laourdas (Thessalonike 1959), C. ~lango, The Homilies or
Photius, Patriarch of Constantinople English Translation, Introduction, and Commentary;
(Cambridge, Mass. 1958).
33 Photius's use of Hermogenes is analyzed in detail in the work cited above, n. 26.

34 Photius prefers, however, the style of the fourth-centuf)' rhetorician, Palladius of

Methone,

187
64 GEORGE KUSTAS

ments of Byzantine literature owes much of its outer form to the rhetorical
tradition.
The influence of rhetoric is not limited to the literary arts. The Greek
language has, built into it. a surprisingly large number of ways of saying "if."
The system of progymnasmata can be regarded as in some ways a codifica-
tion of this speculative spirit. It compels a consideration of all possibilities
in a given instance. The encomium, for example, calls for ever new variations
on the theme of praise; ~fionolta concerns itself with what so-and-so would
have said if-;the xgcfa asks for the exercise of the imagination on the possible
meanings in a given phrase; the fiiot:; ponders such questions as cl nAEvodov,
cl yaflTjdov, should one take a trip, should one marry; the OVYXgIOt:;, by jux-
taposing parallels, enlarges the range of experience of the possibilities of life.
This rhetorical education, in addition to marking the educated man, was
the key to entering the imperial bureaucracy. If politics is the art of the
possible, one may be justified in seeing a correlation between the educational
system and the vaunted successes of Byzan~ine diplomacy, which by adroitly
balancing one adYersary against another, kept at bay the host of enemies
that through so many centuries ringed the Empire on every side.
Rhetoric for the Byzantine was not simply an educational force but a way
of life. nad)fia means both education and culture. Neither the Hellenistic
nor the Byzantine world knew the set of associations which we give to a
college "commencement." Life is one, under the guidance of Divine Providence.
Such a philosophy could yield brilliant successes, but it had also its pitfalls
in irrelevance and retreat before new challenge. The fortunes of poetry are
a case in point: on the one hand the sublimity of a religious ode; on the other,
poetry as merely an aid to the work of prose. The failure of secular poetry
comes about not because, as is sometimes said, Byzantium could not appre-
ciate it, but because prose, not poetry, was the proper medium of educa-
tion, and education was all in all.
The prescribed style in which the literature we have been examining was
to be written continued to be Attic. The definition of the term, however,
changes as we move down in time. The basic distinction is between a-rTlKW';
(or erJTOet"w,;) and "OtVW~.35 Starting about the middle of the tenth century,
the scribes of the Byzantine renaissance begin to copy not only prose works
but poetry as well. As a result the compass of Attic style expands so as to
include the ancient tragic poets. Bohlig's recent study has shown that much
of what earlier lexica call poetic, Tzetzes by the twelfth century calls Attic. 36
This tendency pushed into the background the difference between prose
and poetry and helped continue the hegemony of prose. The introduction

35 G. Blihlig, Untersuchungen zum rhetorischen Sprachgebrauch der Byzanliner (Berlin

1956) 3ft.
36 Ibid. 14.

188
 FUNCTION .-\:>;D EVOLUTION OF BYZANTI:-.IE RHETORIC 65

of poetry into the definition of Attic meant that more and more the literary
tradition was intended for the discriminating few. Since poetry and prose
together become opposed to what is XOIVOV, and since XOIVOV means "ordi-
nary" as well as "in general use," its opposite is therefore "extraordinary" or
Attic. 37
Concurrent with the introduction of poetry Byzantium witnesses the steady
enlargement of the canon of Attic prose. The scholars of the Renaissance turn
their attention to manuscripts not only of the classical authors but of later
Greek literature as well. The Bibliotheca, for example, reviews not only the old
canon of the ten Athenian orators but records another list, created probably in
the fourth century, which includes Dio, Herodes, Philostratus, and Aristides. 38
Into this continuum there is fitted by the ninth century a third canon of
Christian authors. The models of Christian epistolography come to be Basil,
Gregory of ='iazianzus, and Isidore. 39 Beginning with the ninth century,
a kind of Christian classicism appears, engendered in part by the triumph
of orthodoxy over iconoclasm, which comes to be thought of as the last heresy
to plague the Faith. Hence the need to hold out what is orthodox becomes
apparent not merely in dogma but in literature as well. By the twelfth cen-
tury there is added to this curious mixture of classical and early Christian
figures a Byzantine name. Joseph Rhacendytes's Rhetorical Synopsis lists
as model epistolographers the three Cappadocians, Synesius, Libanius, and
"the most learned Psellus. "40 Another writer gives as models of the panegyrical
form Basil, Aristides, Themistius, Procopius of Gaza, Choricius, and, he says,
particularly Psellus. In epistolography, his exemplars are the three Cappa-
docians, Synesius, Libanius, and Psellus. 41 Indeed, Psellus became the prime
model for all types. Thus, if Homer and Demosthenes and Chrysostom and
Psellus are Attic, the word has come by the twelfth century to refer to any
author worthy of imitation, contemporary or ancient. 42
The way to achieve Attic style was to follow the precepts of Hermoge-
nes. His JIe(!1 lbewv recognizes seven qualities of style: amp/veta (clarity);
a~twp.a A,0Yov (loftiness); "QAAO, (beauty); YO(!YOT7); (conciseness); 1700,;
aA,1jOEta (sincerity); and, the pinnacle of stylistic excellence, bEtVOT1/' (force).

37 Ibid. 16. See also T. Hedberg, Euslhaihios als A,liizisl (Uppsala 1935), and
P. Wirth, Untersuchungen zum byzanlinischen Rhetorik des =wOl{len Jahrhunderls (:\Iunich
1960).
38 See A. :\Jayer, "Psellos' Rede iiber den rhetorischen Charakter des Gregorios von
Nazianz," Byzantinische Zeitsehri{t 20 (1911) 82.
39 See B. Laourdas, "IIa(!aTT)(!~(1El, btl roii xa(!axT11(!O, TWl' i:tt(1TOAWl' Tot; tPwTiov,"
'EnETT)(!i, 'ETal(!ela; BvCal'Tlvwv Enovawl' 21 (1951) i4-109.
40 Rheiores gratci 3.559.12.
41 Rhelores gratei 3.5i2.25ff.
42 Bohlig (n. 35 above) 16.

189
66 GEORGE KUSTAS

Not all of these are equally emphasized at every period, but inasmuch as
their selection defines the oYer-ali pattern of Byzantine literary culture, the
fortunes of Hermogenes are a good guide for tracing the evolution of rhe-
toric.
We can distinguish four periods. After Syrianus our information about
Hermogenes is scanty. The school of Gaza made little use of him. His in-
fluence, it would seem, was particularly felt in Athens and Alexandria; with
the closing of the Athenian school in 529 and with the Arab conquest of An-
tioch and Alexandria his fortunes faded. Constantinople must have initi-
ally been least aware of him, and he does not figure in the works of such
writers as Nicephorus and John of Damascus. The manuscript tradition,
however, may perhaps be interpreted as bearing witness to an undercurrent
of use in the schools. The present corpus of five treatises was first put together
in the late fifth or early sixth century, and many progymnasmata written at
the time became attached to his name. 43 That Photius relies heavily on Hermo-
genic concepts without ever citing the name of the rhetor suggests that Her-
mogenes continued to form a staple of the educational system. The systematic
search for manuscripts following the Arab defeat in 868 and the expansionist
policies of the Byzantine court no doubt helped bring him once again to the
fore.
With the ninth century begins the second period of Byzantine rhetorical his-
tory. The direction it takes and the principles it espouses are the immediate re-
sult of the iconoclastic age. The struggle had ended with the triumph of image
worship. The questions that had for centuries been pressing for solution had
now to be resolved. One of the major contributions of the opponents of images
had been to substitute for religious art a cycle of representations drawn from
the ancient tradition. In so doing they gave strong reinforcement to the
classical elements in Byzantine life. The pagan tradition had, however,
through the use of such textbooks as Hermogenes, rigidified the lay sector
of Byzantium, and in its realm Christianity had done the same. Nothing
could be clearer than the prescription of the Trullan Synod of 692, which
requires homilists to draw from the Fathers rather than compose their own
sermons.44 Thus iconoclasm challenged the Church on its home ground, and
with the victory of the iconophiles the stage was set for defining the rela-
tionship between Christianity and classical culture. Photius offers a syn-
thesis in which Christianity is the ruling element, effecting itself in concert
with a carefully culled classicism.

43 Praefalio to Rabe's edition of Aphthonius, XIV. See n. 2 above.

44 Sacrorum coneiliorum nova el amplissima col/eetio, ed. J. D. Mansi (Graz 1960)
11.952CD: 01 TWV b,,,hW'wv Jt!?OeI1TWU; lliiJ.;.ov lv TOUTO'; (I.e., the Fathers) £v6o","~­
TwGav ij )'dyov; oba:{ov; avv1uTTOVTf;.

190
FUNCTION AND EVOLUTION OF BYZANTINE RHETORIC 67

This interaction of classical and Christian can be detected as well in the

educational reforms of the day. The patriarchal school had existed in Con-
stantinople side by side with the university. It had not dealt exclusively
with theology or the training of the clergy, but obviously did less than the
university with the profane disciplines. In 861 Photius restored instruction
in the profane sciences along with theology. Two years later Caesar Bardas
reorganized the university and, as a result, both institutions now supported
profane studies. Before his elevation, Photius had been a professor at the
university; thus his support of secular learning was natural. 45
The style of writing practiced in both these schools no doubt followed
the Photian principle of adaptation from a Hermogenic base. Selection is
to be made of only those qualities of style which correspond to the ideal of
Christian character. Photius calls for a gracious and noble simplicity-a
Hermogenic concept-of both words and action and, in doing so, deliberately
turns away from the cornerstone of Hermogenes' system, bELVOTTj;, as well
as from some of the means by which it is acquired, such as TeaxvTTj~ (rugged-
ness), arpobeOTTj; (intensity), and beLflvTTj; (pungency), all of which could
suggest the very opposite of Christian modesty and humility. Like Hermo-
genes, he recommends a proper mixture of the ingredients of good style.
This feeling for propriety (TO OiXEiov), which he shares with the Second So-
phistic, is deepened in response to a profound historical sense in the judgment
of life and letters; and, indeed, the mixture is of something far greater than
stylistic effects. It is Christianity and classical culture itself. 46
The tradition of letter writing must have played an important role in the
formulation of the Photian ideal, for some of the terms that he retains are
similar to the requirements laid down by Gregory of Xazianzus for epistolo-
graphy: avvTofl(a, aarp~vE:ta, XaeL;, TO neE::WV and UTTLxwflo;,47 Isidore of
Pelusium in the fifth century continues the tradition,48 as does Photius in the
ninth, listing (JgaxvTTj;, aarp~vfta, XaeL" and a."ti.OTTj; as the qualities by
which to convey the ethos of the letter writer and his subject. 49 So too, a

45 See F. Dvomik, "Photius et la reorganisation de l'academie patriarcale," A.nalecla

bollandiana 68 (1950) 108-125 passim.
46 This paragraph is a summary from the detailed analyses to be found in the work cited
in n. 26 above as well as in my "History and theology in Photius," Greek Orthodox Theolo-
gical Review 10 (1964) 37-74.
47 Letter 51, PG 37.105.
48 Bk. 5.133, PG 78.1404B: 6 E:'IIIJTOAIIWio, xaeaKT~e I'~U :WVTa..-WIJIV dxoIJI''1TO, [aTw
I'~u I'~V d, fiev,!"v XEKOal''1p.ivo; 1; TeV'P~V, TO I'tv yde flJUi.t;, TO bi d:tEl!'oxa).ov . TO
oi WT!,iw, xExoap.ijafial Kai n!,o<; X!,eiav Kai nllO, Kd)')'o, dgXEl.
49 B. Laourdas, "IlaQaTT}(!~(]fl; c:ti TOV XaQQ"'TijeO:; TWV i::r.UJYO;.WV TOU I/JWT/Otl, H '£:7.£-

''1(!/!; 'ETaI(!da; BvCavTlvwv E;rovowv 21 (1951) 81.

191
68 GEORGE KUSTAS

late anonymous writer reminds us that n()onotla should be xa()aga., and

without ruggedness. 50
The third period extends from the eleventh to the end of the twelfth
century. The :!\facedonian renaissance had introduced Christian scholarship
on a solid and equal footing with the classics. The succeeding centuries
saw an expansion of both. If nothing else, the sheer weight of the multi-
plication of knowledge may be enough to account for the dissolution of the
Photian synthesis. The period comes to be marked by a sharp polarity be-
tween the two traditions. The key figures of the age, Psellus, Tzetzes,
Eustathius, have in common a curious ambivalence that makes it possible
for our handbooks to divide their productions into two parts, those on pagan
versus those on Christian themes.
Psellus's works well illustrate this division. He has left us two different
treatments of the style of Gregory of ~azianzus. One cites him as part of the
Christian canon that includes Chrysostom and the other two Cappadocians,51
just as the ninth century had done. Psellus is here at pains to show that Gre-
gory's prose and poetry meet the ancient criteria, that Gregory can stand on
an equal footing with the models of the Second Sophistic such as Aristides,
and in short, that he is the Christian equi .... alent, preferable for the Christian
because he writes about Christian things and writes about them well. This
is a legitimate appreciation of Gregory as a Jiterary figure and as a Church
Father. In the other treatise52 Psellus sees Gregory as the acme of a long,
unbroken tradition of Greek letters, superior to Sappho, Thucydides, and
others with whom Gregory could have nothing in common. Here, rhetorically
speaking, the avyxQwt; principle has gone awry. In an excess of pious zeal,
Psellus has failed to fit his subject into the proper literary and historical
context. Interestingly enough, the treatise that brings to bear the greater
amount of classical knowledge is not the first, which, despite the sounder
e.... aluation, is rather standard and restricted in its analysis, but the second.
Here the learned Psellus draws on wide resources. In addition to Hermogenes,
he makes critical use of the rhetorical .... ocabulary of Dionysius of Halicarnassus
(whose work De composilione verborum he had partially excerpted),53 of

50 Rhelores graeci ;{.595.11: ovdi ""AJa(!u" T~V <pedol" h~(!,)uav <pi.fY/1alvovu; TU :'loUU
xu; Tl!aXVVO/1EVO/ T!i i.i;€l . xat TOt!TO (it tH' hddEL;/v, O/1W; o,)x i/1E/1<P()'WUV TOt; :'loi.ioot;
dta TO 'Pli.OTlfLOV TOV i.drot' %al :tOQIJ.lOl', iaw; ~i ;eal T]flEi;, C1JAwaOIlCJ' T01JTOU; itoni
,<a/eo •.
51 Xal!",'Tiie f ; re'Ii'o(!iot' TOV ()wi.oyot', TOt! /1fya}.ov Bauli.Eioll, TOV XeVUOUT0/10V
xa; rl!'1yoeIO" TOV }""VUU?);. ed. J. Boissonade in M. PselJus, De operatione daemanum;
accedunl inedilo opllscuia I:'\uremberg 1838, repro Amsterdam 1964) 124-131.
52 Text and commentary by A. \layer, "PselJos' Rede iiber den rhrtorischcn charakter
des Gregorios von :'\azianz." By:anlinische Zeilschrifl 20 (1911) 27-100.
53 Rhelores graeci 5.598. :\'ote also Hcrmogenes rendered into political verse. Hhdnres
graeci 3.687, and Psellus's summary of the flEe; 'IDEr;"', Rhelores gracci 5.601.

192
FUNCTIO:-.1 .\ND EVOLt:T10:-.1 OF BYZANTINE RHETORIC 59

Demetrius, and Longinus; and he borrows phrases in abundance from Phi-

lostratus and Synesius. The work displays a heightened interest in pagan
rhetoric as such, side by side with a strong religiosity; but the discussion
fails because the one does not inform or relate to the other.
Psellus's two main interests in the pagan tradition are rhetoric and phi-
losophy. He pursues two objects, "to improve his stylistic eloquence through
rhetoric and to purify his spirit through philosophy."f>4 In an encomium on
his teacher, John :\Iauropus, he remarks that philosophy without rhetoric has
no grace, and rhetoric without philosophy no content.;;; He berates one of
his correspondents with the remark, "Perhaps you know philosophy and
rhetoric, but you do not know how to put them together; there is a philo-
sophizing rhetoric as well as a rhetoricizing philosophy."56 In another let-
ter, he points out, "Just as Plato in the Timaeus combines theology with
physical science, so I write philosophy by means of rhetoric and fit my-
self to both through the use of both. "57 He recognizes another kind of
philosophy superior to the pagan,58 and there are a number of passages
such as the above which call for a synthesis of rhetoric and philosophy,·9 but
the appeal is self-conscious and almost obsessive. The terms in which it is
couched take us back into the atmosphere of late Antiquity and remind us
of the Neoplatonic concern to relate philosophy to rhetoric in a meaningful
way. Psellus's role as head of the philosophical school organized by Constan-
tine IX in 1045 gave him the opportunity of exploring in these areas without
strict reference to the Christian tradition.
At the same time, there is a Christian Psellus. We are not yet in a position
to date the mass of his worKs so as to establish a pattern of intellectual de-
velopment, but it is clear that he was active in both traditions and that

.4 The His/ory 0/ Pse//lls 36, ed. C. Salhas (London 1899) 6.107.15: ~/ITO!!'xoi; fliv i.o-
i'0t.; T~V i'}.WTTUV :-zi.aaUa(JrH :rgo; Ev:r.einttflV XUl rp,,).oaoq;l'..l %uOcigut TOV 1'OVI.'.
55 Biblio/heca graeca medii aeIJi, ed. C. Salhas (Paris 1876) 5. I 48: WTU voii :'lqOf(HWU;
r,;.WTTOV Q(5tuv6'IToll fl.0VGlV----<) TE yd.!! p6vw:; 'PLA.OaolPo; QI.f1Ql; %(tl ,) Triv Tilvlill (5tne"1-
J.livwr; XOI-':tri;wv TOU xoap.ovvTo; earE(!'tITUl (Jx.YjJ.4aTo;-Xf1T(~lJOW(Je (1VVaV)(l:; Ta bU).G''t1j-
vat 6oxoiivTa XUt TOl' vouv y.uOtjfh'vur:; Toi,; u:toTij~ g1JToQuaj; 7.aQ((Tt TrJl' TE yi.(UTT(1V U€p.-
"uva, Toi; rpli.on6rpol; vO·~flanl.
56 Ibid. 4.12: ai, piv ovv taw; q;,,).oaoq:;luv i:tlaTf1f1(J.t %f1i gJ)TOfH; t)lJ. TO I)' i~ flp..fPoil' (J'vv-
linav dyvoti; . lan 1Ji, OJ iJpnTt, xai rpvJ.oaorpiu; t)rITOf!'X~ %ai QIITOf!fiu rp,}.oaorpo:;. The
passage forms part of a discussion of Hermogenes .
• 7 Ibid. 476: ;.wi Wa:'lfq ,; n;.U.TWV Tfi rpVf1!O).oyirf fliyvvf1! T'WltOyqr.HFWV ;<UTa TOV a,-
;.oyf!drpo. oiJTw O~ xdy6J Til QI}TOQlxfi T~V <jJlAOOOrp{uv avvTi(h}fll x"i :.go; d/lrpOTiqu:; Ot' dw
'PoTlfgwv cif!I,6~oWJl·
58 The Hisfory 0/ Psellus 42, ed. C. Sathas 6.109.26: iaTi TI; xa; V:'lE!! TaUTllv iTi!!" 'P,A-
oaorpia.
59 Ibid. 41: Bib/io/heca graeca 5.480; Encomium on Symeon \lelaphrasles. PG 114.188.-\:
,prpwvov iv' oh w; fl:tW To. voii. xa; { •• ov. T~V y).WTTrlV fiQ;'daflTo: Leller 241, ed.
Kurtz (n. 28 above); e/ al.

193
70 GEORGE Kt:STAS

rhetoric is a yital force in both. Hermogenes is still the key figure. 60 Psel-
Ius, like Tzetzes, renders the text into political Yerse. Of particular inte-
rest is his Synopsis of the rhetorician. One way of achieYing aE/-ll'OTl);,
says Hermogenes. is to talk about the gods (TWV efWV). Psellus keeps the
outline but makes the phrase singular, TOV ewv. 61 The frame of the text
is receiwd without question, and only that portion changed which stands
in glaring contrast to Christian concerns. Of roughly the same date must
be a set of anonymous scholia on the n Eel t()EWV. 62 Once again, the treat-
ment of bE/voTl); follows Hermogenes' pattern. Besides the gods, Hermo-
genes suggests 1) a discussion of diyine matters (Tn eEia neaY/-lara), that is,
cosmology (his examples are taken from Plato's Timaeus); 2) things with a
diyine connection that men care about, such as the immortality of the soul
or the cardinal yirtues; and 3) iPlportant but exclusively mortal things, like
the battles of Plataea, ~Iarathon, or Salamis. 63 The Paris manuscript adds
to these same scholia a note to the effect that Christians have many citations
not available to the ancient writers, such as the beginning of the Fourth Gospel,
and that writers on the Hexaemeron achiew aE/-lvoTl)s by their reference
to creation and the works of God.54 Another curious manuscript from the
twelfth or thirteenth century gives an ordinary analysis of Hermogenes'
text, but the scribe into whose hands it fell erased the examples from Demos-
thenes and substituted where he could quotations from Gregory of ]'\azian-
ZUS. 65

The last period in the rhetorical history of Byzantium is the Palaeologan

age. Here we have come full circle. The scholia of il.1aximus Planudes on
Hermogenes do little more than elaborate the text. 56 Pletho's synopsis also
follows the rhetorician, although one detects here a more original mind, going
beyond the traditional scheme by citing other handbooks, calling attention
to Hermogenes' sources, and occasionally being slightly critical of him.67
The literary quarrel between Theodore l\fetochites and Nicephorus Chumnus
in the fourteenth century has recently been the subject of a monograph by
Sevcenko. 68 il.1etochites is attacked by Chumnus because his style is daarp~c;
(unclear), the opposite of the Hermogenic ideal, and also lacks "&)).0; and ~eoc;.

60 See n. 53 above.
61 Rhetores graeci 5.602.2. aE!-'vOTlJ; is a subdivision 01 a;[w!-'u J.6yov in Hermogenes'
scheme. Tzelzes: Rhetores grueci 3.6ilf!.
62 Rhelares graeei i.2.86lf!.
63 Hermogenes, 2191!.
64 Rhelnres graeei i.2.954ff .. esp. 956, 95i.
65 See Y. de Falco. TraUah, relorico bi:anlino (Rhelorica marciana) (Pavia 1930).
66 The passage on Of!-'VDT7I; lor example, restores the plural, neW IJEWV: Rhetores graeci
5.480-481.
67 Rhetores graeci 6.546-598.
68 I. ~e\'cenko, Etudes sur la pviemique entre Theodore I\Utochite el Sic"phore Choumnos
(Brussels 1962).

194
Ft:NCTION AND E\'Olt:TION OF BYZANTINE RHETORIC 71

In his defense ~!etochites says that his aim is OEtVOT1);, also recommended
by Hermogenes, and notes that this recognized excellence is acquired in part
by T(?axvTI]C; and aq;ob(?OTI];, the very qualities which the ninth century had
tended to exclude. Thus ~Ietochites cultivates da6.<pfla deliberately and can
cite in his defense Hermogenes' own condemnation of excessive clarity.69
This peculiar state of affairs, which elevates obscurity into a literary virtue,
explains the direction of much of Byzantine writing. No one who reads By-
zantine literature at all need be told that, whereas the ideal constantly being
held forth is clarity and simplicity, clear and simple is precisely what the li-
terature is often not. The secular reasons for this can be seen in the set of
associations that come to be attached to the word "Attic." One must, however,
look also for a Christian explanation. As early as the latter part of the ninth
century, Arethas had written an answer to those who had accused him of
obscurity.70 Like ~Ietochites, he defends himself by saying that his writing
appears obscure not because he has abandoned the rhetorical rules, but be-
cause his contemporaries do not understand him. Obscurity has the virtue
of keeping out the crowd. The source for such a sentiment is probably to
be found in the admonition by Photius (his teacher):

I answer in a few words your question, as to just why it is that

prophecy is overshadowed by the devices of obscurity. Prophecy is
not history. The virtue of history is assuredly to speak clearly
and not to contrive. It teaches events done in our midst, which we all
alike, the prominent and the ordinary, experienced at the time and
may learn about now. For prophecy, on the other hand, the function
of which is to reveal the hidden to those who are worthy, but to keep
it from the uninitiated, that is most fitting which is obscure and
enigmatic and screened from "iew. So I think your difficulty is
solved: to wit, if we did not need to learn, we should not be obliged
to speak at all; but if we are to learn, then prophecy should not be
unclear and practically the same as not speaking at all. ~Iy remarks
have now established that it is both proper to learn and necessary
not to speak wantonly and commonly. Things should rather be
said in clarity to the initiated but kept inaccessible and unapproa-
chable to the profane. 71

69 Ibid. 57.
70 q,
Il[!ar; TO,)r; fi; oaatpflav ijpJir; blta'''';)lpaVTar;, €V ou
xai Tic; ij 10Ea phliJEv AOyov,
ed. L. Westerink, .4.rethae scripta minora (Leipzig 1968) 186-191. See Sevcenko (n. 68 above)
169 n. 2.
71 PG, lO1.948B, C: :IvOopivq, Oi aOI Ti o1}nou ij :reorpl'}uia TO'r; Tij; oaarpEia; T[!O:nOtr;
avvEI1Xiaaral, 0111 fJgaxiwv apdfJopa!. dlon ovx lanv iaTO[!ia ~ :zqorpl'}uia. iaTO[!iar;
piv yae, ti:ue n r'ii.i.o, a(lfT~ aQ(pwr; u EinE'V xai P-l'}OEV :w;mEvol'}/JEvov €;E[!i'aaaaOol.
To yde iv Tq, /Jia'iJ i'E"/Evl'}/Jiva oloaa><<I, xat a :naVTEr; opoiw; xai a:zovoaiol "at <paVAO I
"aT' haivo xaLgov ,;:rlOTaVTO, xai vi'v £l~£vaL oux a:-tt;teYOVTaL. :UJO'flfTEiq. bi, ri.; leyov
ra
Toi; d.~iOL; d.n;oxai.l~:lTElV aOTJ}.a, Toi~ Of fJt;{3~;'Ot~ d{1aTQ :touiv, TO OVVEaXLa0I-'EVOV

195
72 GEORGE KCSTAS

Photius's remarks proceeds from a reading of the famous text of Saint

Paul on the prophecy of tongues (1 Corinthians 13.4-14.40). They are prompted
more by a religious than a literary feeling. The sentiment, however, could
and did give him and his student biblical authority for cultivating deliberate
obscurity in the name of both Christianity and of the classical tradition as seen
through the medium of Hermogenes. Psellus was later to say that a man
who drs not haw the proper combination of rhetoric, philosophy, and po-
litical action is as tinkling cymbal. 72 The phrase comes from precisely the
same scriptural context. It was to serve Byzantium well. The Pauline text
that inspired the translation of the liturgy into foreign tongues and the con-
version of the heathen and reYealed Byzantium in her most glorious and suc-
cessful role might also sustain the most serious shortcomings of her litera-
ture. 73
Such were the rhetorical patterns that Byzantium inherited from her Greco-
Roman past and the uses to which they were put. The history of accommoda-
tion to rhetoric is the history of Byzantine culture itself. Byzantium contended
throughout her history with the challenge implicit in her Yery birththright.
There are moments when the two claimants to her soul, Christianity and
classical culture, stand apart, but in the very dualism of her nature there
lay hope for a union and harmony such as the ninth century proffered. The
grand tragedy of Byzantium lies in the fact that the expanded conscious-
ness of both the Christian and the classical traditions in the ninth century
facilitated the selection of one or the other on the part of succeeding ge-
nerations.
Rhetoric was one of the few elements of Byzantine life to survive the wreck-
age of 1453. l\lany of the rhetorical ideals of Hermogenes are adapted by the
Italian Renaissance, and in the later sixteenth and seventeenth centuries
exert an even greater influence on English letters. Aphthonius, through even
more numerous editions and translations, passed into the educational habits
of Europe. Indeed, the techniques of composition practiced in our schools
ultimately owe much to the tradition that he represents, while among the
more conservative quarters of modern seminary education, the retention of

xat aivIYl'arci)(5e; xat TO 0«1 :ta(!a:teTaawiTwv, n(!el!WOiaTaTov. 'E~ rov oll'al Aei,vaBai
aOI TO /L'weov, w;, ell'~ lOet /lalieiv, oVO' o).w; ExeijV £inelv. el O· ixe{Ii'r} /lalieiv, OUX daa-
rpwr; EXEIV T~V :t(!orp'r}uia", xai iv iacp axeoo" wv I"rtr5i cieijalial. olnnj/J'r} ya(! EX TWV de'r}-
ptvwv, oTt "at pa8civ :lQoaij%ov ~1' ~ai dvayxaiov p.~ f3E~"AW; xat X'OIVW; tinEt'" d)J.a
Tol, /lEV /lvaTaIr; el(!ijalial aarpij. CiOVTa Oi Tolr; d/lV~TO/; OlaT1)(!1)()ijvUl xal oux £rpIXTd.
72 Bibliolheca graeca 5.148.
73 It would be a useful and rewarding task to trace the history of this difficult and at

some points obscure text in Byzantine thought. The passage i. crucial to the understanding
of some of the most basic manifestations of Byzantine culture.
74 Hermongenes' contribution to English literature forms the subject of a book by Pro-
fessor A. M. Endicott of the l'niversity of Toronto, to appear shortly under the auspices
of the Princeton University Press.

196
FUNCTION AND EVOLUTIO:-i OF BYZA:-iTINE RHETORIC i3

rhetorical forms such as the (Jiat;, one of the most common of the medieval
progymnasmata, gives proof of the continuing vitality of ancient forms.75
Department of Classics
State University of !\ew York at Buffalo
Buffalo, New York 14214, U.S.A.

75 See Brzoska S.!'. "Aphthonios" in Pauly-Wissowa, Real-Ellcylopadie. for relevant

bibliography. I do not wish to depreciate other more important elements, such as Aristotle
and Cicero, but only to call attention to a neglected aspect of the rhetorical tradition.
The edilio prillceps of both Aphthonius and Hermogenes (except for his Progymilasmala,
not printed until 1;90, ed. Heeren, Gottingen) is volume 1 of the Aldine Rhelorum graeco-
rum collectio (Venice 1508). The first Latin translation of Aphthonius appeared as early
as 1513 (Aldine, ed. J. Catanaeus). After that his editions are very frequent and wide-
spread throughout Europe. Some were to become very popular and influential, as for
example, the Latin version by R. Agricola (Paris 1549). In addition to the lists provided
in the introductions in Rabe's editions, consult J. Fabricius, Bibliolheca graeca 6 (Ham-
burg 1797) 69ff., 95ff.; also Brzoska, loco cit. For the history of rhetoric among Greek
speakers after 1453, particularly as regards homiletics. two recent works are indispensable:
K. Kourkoulas, •H fJEw(!ia Toii 1(1)(!vYI'aTo,; l<aTd TOV'; X(!01l0V; Tlj.; TovQl<ol<QaTiu, (Athens
1957), and P. Trempelas, ·Ol',).llnl<~ 1i 'irITo(!ia I<ui 8ew(!ia TOU 1<1)(!';YI'UTO; (Athens 1950).
I wish to thank )Ir. Laourdas for calling my attention to these two works.

197
This page intentionally left blank
 IH O R S E V t ' E N K O , C A M B R I D G E . MAS S .

LEVELS OF STYLE IX BYZAXTIXE PROSE

r. DEFIXTTIOXS

Some modern linguist" understand "tylistic'" as the study of alternati-

ve mode" of expressing the same (or approximately the same) content. and
an author'" stylistic selection as the choice het\\'een items that mean more
or less the same thingl. Such a formulation should be acceptable to
Byzantinists concerned \vith literary styles, for they. too. ('oncentrate on
their author' s s tylis tic selection. The preceding definitions suggest tha t the
best conditions for. the study of levels of style are offered by texts
displaying alternative forms. but the same or related content: consecutive
re\vorkings of the same text by its author: an original that can be
juxtaposed with its paraphrases. clone by another author: or. finally. ,yorks
of the same author dealing with similar topics or falling "'ithin the same
literary genre. \Ye shall make use of Byzantine examples that belong to
each of these three categories.
Byzantine usage pro\'icies a convenient guide in the search for Byzan-
tine conceptions of style. In the wake of antique theur~'. this usage
expresses the notion hy a limited number of terms. such a" ?PX0c:. /,<;:,c:.
PPXX7f,p. Ep~.U;"2~X. -;:),xcr[.LX or -;:),xcrc::,. the meaning of which is for the most
part clear. Only ~3sx is used ambiguouslyl".

I L. DULEZEL and .J. KRAl'S. Prague School of Stylistics. In' B. B. K\(,HI{('.

H. W. F. :-:T.\HLK'~. edd .. Current Trends in Stylistics (1n7~\' 37: A. HOl.'l;H. St.de and
Stylistics (1969). 6~8; ~. E. E:-;KYI:-;T, On Dpfining Style . . in .r. SPE~C'r~R. ed .. Lingui-
stics and Style (EHi.f). .'5.f. There is little systematic treatment of the le\"els of st\'le in
Byzantine prose. Best to date is H. HlCi'CER. Stilstufen in der Gesehichtsschreibung des
12 .•Jahrhunderts: ,-\nna Komnene und }Iichael Glykas. B!I:antin~ Stlldip., ~ Etudes Byzan-
tines .5 (1 97::;) 139~70 (bibliography). The reports on "diglossie" by E. KRLI.RA:-;.
,J.IR)ISCHER and .-\. }IIR,-'.Y1BEL in: Proceedings of the KUrh International Congress of
Byzantine Studies. Oxford 1967. :283~313 (bibliography) and that by P. WIRTH in XVe
Congres international d'Etudes byzantines, Rapports et Co-rapports, II. I. Athens 1976.
3~.5..j, are of interest. Works discussing individual authors. such as those by REXAl'LD,
BOHLlC (except for important introduction), vo'" STEPSKI-DoLIW,-'., HEDBERG or \VlRTH
are only of indirect use to us; so is the good Studies by Kl'STAS (n. 4.f below) and his Photius
study in Hell 17 (1962) 132~169.

199
:!!lO 5 .2 . N t il s t u f m

On the other hand, it is not useful to follow Byzantine rhetorical

theoriml in attempting to define IcvelR of style, for Byzantines displayed a
great deal of confusion in this mattrr. They WPI"(' awan', of thp two alltiq1H~
doctrines of Rtylistie leveb, that of Demetrim; (first cClltury A. D.?) who
distinguished four of them, and that of Dionysios of Haliearnasslls (first
century B. C.) who spoke of three styleR, grand. middle and plain.
Dionysios was known to several highbrow Byzantine writers- Psellns and
Metochites among them-- and rudiments of his doctrine seeped down to
Byzantine commentaries and introductory treatises on rhetorics 2 . How-
('ver. his system was overshadowed hy that of H('rrnog(~rH's. the dominant
inll,wlwl' ill B.vJ',alltilll~ t,lw()[",Y. who did away with ("l~ dlwtrill\~ of (,1m'"
stylI'S and suhstituted for it the system of "ideas", (hat is, various forms or
devices of Rpeech used by great orators. Hermogencs' system was that of
"horizontal" kimlR rather than "vertical" levels of style. AR a reRult. when
B,yzantine theorists spoke of three styles, they often dcrived them from the
"ideaR" and associated them with various great orators, with the three
partR of rhetorics ~ judicial. deliberative, panegyric -_. or with literary
genres:l . A further difficulty results from the fact that Dionysios, under the
influence of peripatetic rhetorical theory, put the middle Rtyle first. and
Byr,antine commentators and some writers often did the Rame. But in their
practice Byzantine writerR of all erlueationallevels admitted the supremacy
of the high style.
Psellos' theoretical pronouncements prm'ide a eOIl\'enient i\lw,trl1tion
of this confusion. He once or twiee divided styles into high, middle and
lowly. and expressed a preference for the middle one (praeticed, according
to him. by .John ChrYHostom), but in comparing Uregory of NazianZlls to
pagan ora torH, he said that the three stylistic levels applied to the "idea",
Ol' form, of speech. and thlls confw;ed the IC\'I.JR with kinds of Rtyle.
EIHewhere Psellos analyzed the style of Gregory of Nazianzus ill terms of
the tripartite diyisioll of rhetoric. Only once did he divide Gregory's works
ae(:ording to leyeb of understandability: into those "aeeeRRible" and thoRe
"of difficult aeecRs" to "the many"". It iR noteworthy that Psellos'
I" Fulh'r liRt in K l'ST,I". lIeli (n. 1 ahove). I:Hi 1:17: "I". also K OHTH. Die I-\tilkritik
<Ips I'hotios (l!l2!l). index.
2 H. RAilE. ed., Prolegomenon f:'ylloge (I !)31 ), 711. 25-- 80. 7: 141. 8- 21 : 21 a, I \l-- 25;

:\7:l. 1 -I:J (Uoxapater): 377. 7--14: 410, II-ln. I'atmiacus :HHi. fol.:n2'· (I'hilotheos of
:-;pl.l"lllhria). Cf also A. ~gn;1\ YNS, IteehereheH ... I I (lInK) :n:l:l: 71 . -7:~.
:1 RAilE (as in Il.2), 405, 22-26 (i)ikeliotes): 213,19-25; 68, 22--28 (Planudes); 79,
2:i-HO.7.
4 P. LEn, ed., Mich. Pselli de Greg. Theologi eharaetere judicium (1!)12). 94. 5ff.;

1-\1'1'('("h on Gregory's Hly!\'. A.MAYICR, ed. BZ 20 (\\)11) :,(j- :,i; On style of I"at-herR, PO
1 :!~. !)02C IIO:IA: IHliiA; nOlic.

200
 I. ~E\'CE:-rKO, Levels of Style in Byz. Literature 291

excerpts from Dionysios of Halicarnassus' De compositione verborum stop

before the part that expounds the doctrine of the three levels of style 5 .
Left stranded by Byzantine theorists, I shall fall back upon a doctrine
both ancient and timely. It goes back to the Rhetorica ad Herennium,
prevailed all over Europe down to the end of the eighteenth century, and
has been revived today by some modern linguists; it is the Theory of Three
Styles, in which the terms "high", "middle" and "low" do apply to stylistic
levelso. I do it out of scholarly scruple, for a working Byzantinist does not
need a precise definition of levels of style. He percei\-es them instinctively,
in terms of his everyday practice. For him, a work in high style is one that
uses periodic structure; its vocabulary is recondite, puristic and contains
hapax legomena made up on a classicistic template; its verbal forms,
especially its pluperfects, are for the most part Attic; its Scriptural
quotations are rare or indirect and its classic-al ones, plentiful. In a work of
middle style, periods are rarely attempted and fill-words and cliches, more
abundant; it requires the use of a patristic lexicon; and its Scriptural
quotations are more frequent than its classical ones. A work in low style
uses largely para tactic structures; its vocabulary contains a fair number of
words unattested in standard dictionaries or coming from languages other
than Greek; its verbal forms are not Attic; its Scriptural quotations, more
frequently than not, come from the ~ ew Testament and Psalter. In
hagiography, the Life of Theophanes Confessor by Patriarch Methodios and
the Life of Patriarch Xicephorus by Ignatios are works in high style; the
middle style is abundantly attested, and represented by the bulk of the
Lives with the Life of Anthony by Athanasius of Alexandria at the head of
the procession; and the low style, by dozens rather than by hundreds of
Lit'es. such as those of C}Til of Scythopolis - unless one wanted to locate
his works in the subclass of lower middle style - the two Lives by Leontius

5A. ACJAC in REB 33 (1975) 257~275, esp. 259. On Psellos' view of style, cf. Ja.
N. LJl'BAR~KIJ, Michail Psell ... (1978) and the intelligent Belgrade dissertation b:v
C. MILOv,,-,ovIe, M. Psel kao knjizevni teoreticar (1979). Only towards the end of Byzan-
tium do we encounter a description of two styles that. to judge by its illustrative examples.
was applied to levels. Cf. Joseph Rhakendytes in Ch. WALZ, Rhetores Graeci. III, 525-26.
The sources of the relevant chapter of Joseph are moot. Cf. V. DE FALCO in Hi8toria 5 (1931)
634.
6 Cf. H. BLAIR, Lectures on Rhetoric and Belles Lettres. I (1783), 370, E. R. ClJ'RTIC's
in Romanische Forschungen 64 (1952) 57~70; V. P. VO;\IPERSKIJ, Stilisticeskoe ucenie M. V.
Lomonosova i teorija trech stilej (1970), 70-98; 189-200; "V. SANDERS, Linguistische
Stiltheorie (1973),95; 103-104; 108 (three or four levels).

201
292 5.2. Stilstufen

of Neapolis, the original Life of Philaretos the Merciful and much of the
puzzling Life of Andrew the Fool in Christ1.
In this report, I shall roam freely, mostly within the field of Byzantine
prose ia . Some obvious, if important topics, such as the relation between
levels of style and genres or subject ma.tter, will be left out. Those that will
be treated either reflect my personal interests or seem appropriate for a
Congress whose motto points to the future.

II. A\YARE~E:-;:-; OF LEYELS OF STYLE

It has always been taken for granted that Byzantine "Titers were
aware of at least two different levels of style. Recently, however. Professor
Eideneier claimed the lack of such awareness for one author. Michael
Glykas. whose poem ¥Titten from prison is one of the prize examples for the
mixture of standard and "popular" lines in politic verse and for the
appearance of popular language in literature destined for an educated
readership. Professor Eideneier's claim ~ sends us. e,'en at the risk of
restating the ob,ious. on the search for proofs of stylistic awareness shown
by ¥Titers at all periods of Byzantine literature 9 . This awareness can be
most easily inferred from an author's own prad.ice. In other cases, proof
comes directly from explicit statements by writers or literary critics who
discuss works of their predecessors or describe their own production. from
re,Tisions. either by the author himself or by an editor, resulting in two
versions of the same text. or from the use of two styles in the same piece of
"Titing as a literary deyice.

1. Inference from an Author's Own Practice

The practice of different lewIs of style by the Rame author i" so eaRily
demonstrated that the demonstration can be limited to one example. and
proyided by means of translated passages. Take the sermons of Theodore
of Studios. In literary prowess he was an equal to Patriarch Xicephorus

7 Also A. GARZYA. 11alia Sru;ra 22 (1973) 1179-1186, adHleates three lii'elli lhz.guisti-
co-stili8lici in hagiography. In his Athens Report. P. WIRTH (n.t above). 6-7. 24 speaks of
a "Trigraphie" in Byzantine Lit.erature of the 11 th-15th ce.
7a To be more precise. I shall deal with that segment of Byzantine prose whose
practitioners. whether expressedly or tacitly. considered high style as norm. This prose
encompasses the bulk of material that ha~ eom(' down to us and ranges from the preeiosity
of a Patriarch Methodios to the artlE'ssness of a John Kananos.
8 BZ 61 (1968) 5--9. The merit of the claim is t.o haw set us thinking. For objections.

cf. Hn;GER, Stilfstufen (as in n. J). 167-9.

9 On Anna Comnena and Gl~·kas. Hl':IWER. StilstufE'n (as in n. I).

202
 1. SEV(;E~K(), Levels of Style in Byz. Literature 293

and may even have surpassed him as a stylist, It therefore stands to reason
that he practiced high style: what endears him to us is that he practiced it
in moderation, avoiding the excesses of his younger contemporaries Metho-
dios and Ignatios, the first representati\~es of super-high style in Byzan-
tium, and that he did not remain within the high style's narrow confines,
Theodore would thus begin his PanegYTic pronounced on the anniver-
sary of St. Arsenius the Anchorite: "Rising today at its commemoration,
which recurs every year, en ever-shining star is sending forth its supremely
bright illumination to the universe by means of a completely incorporeal
light: it is not counted [among the stars] of the Zodiac, but is firmly
embedded in the sky of virtue''lO, This must have been routine high style,
for Theodore's biographer merely listed his hero's Panegyric sermons
without comment 11,
The twenty-first Oreat Catechesis opens \"ith a simpler, if still fairly
complicated, sentence: "There being in existence such a great sea of
instruction, expressed in words by our God-inspired Fathers, we should not
be talking at all. for their [\vords] on the subject [at hand] are sufficient for
those among the listeners who would wish to open to them the ears of their
hearts", Theodore decided to speak nevertheless and used a proverb as a
transition to the main bodv of the Catechesis1~,
The Small Catecheses introduce us into a different world of discourse:
"Brethren and Fathers, Since the day of Annunciation has arrived, I ilhall
attempt to tell you briefly ilomething concerning the Annunciation".
Another sentence is not much more invoh'ed: "Brethren, I pray you
(7:XPXXXAW), let us safely keep our fai th , . , for you know that the Devil is
breathing against us by night and by day" 13,
It is not easy to say who the audience of Theodore's Paneg,\Tic
Sermons may have been, and we can only guess to what extent this
audience understood them, vVe are on firm ground with the Catecheses: they
were delivered before Theodore's monastic community and we may safely
infer that at least the Small Catecheses were easily understood by his
monks, The audience was thus the same for both kinds of Catecheses and the
difference in their stylistic levels was due to the different circumstances in

10 P(; 99, 849B.

II pr; 99, 264B.
12 .J. Cozz..>.. ~ L FZI, ed., in: A. M . u, Novae Patrum Bibiliothecae . , ., IX. 2 (1888), 58.
13 E. .-\lTR,.\.Y, ed., ... Theodori .,. Parva Catechesis (1891), 6;"), 1-2 (p, 227): 22, 71

and 74-75 (p. 83). - Another pertinent set of examples would be the Apoioyeticu8 and the
.d.polOfjeticu8 ;.lfinor on behalf of the Images by Patriarch Nicephorus (PG 100, 533-832,
[high style] and 834-850 [middle style] respectively). Note the abundance of perfect forms
in the former and their scarcity in the latter,

203
294 5.2. Stilstufen

which their author composed them. Theodore's biographer tells us that the
Small Catecheses were improvised (crXEa,c>:cr&Wrwv) during usual encounters
with the brethren; as to the Great Catecheses, they were not improvised (ou
crI.Ea,c>:cr·nxw~) but were a result of reflection and were "Titten in solitude
(xc>:&' EC>:UTOV). The biographer found that both sets of Catecheses gave forth
Di"ine and celestial grace of eloquence, and left it at that. His chief praise
went not to the Catecheses or the Panegyrics but to Theodore's Letters,
,,"hose epistolary style he extolled in terms ultimately borrowed from
Hermogenes l4 .
In moments of discouragement, many a modern reader of Byzantine
Homiletics in high style must have asked himself: What were the sermons
like. that were surely understood by their audiences? The answer to this
question is: they ,,'ere like Theodore of Studios' Small Catecheses. The
majority of sermons gi,'en in churches, even the more important ones,
throughout the Empire's existence must have been of this type. If we no
longer read these sermons, it is because they were not deemed worthy of
being taken down; if it were not for Theodore's social and doctrinal
importance. en'n the Small Catecheses might not have come down to us.
The tFanny of high style, as much as an author's social position, heavily
fa,·ored the survival of texts whose impact at the time of delivery was
limited to a fraction of the elite.

2. An Author's or a Scr£be's Editorial Work

Late in his career, Pa triarch N icephorus began reworking the first
version (prior to 787 1) of his historical work, the Breviarium. The editorial
task - abandoned after a few initial folios - was almost exclusively
st~"listic in nature and consisted in adding fill-words and transitional
phrases or in substituting high style for low style expressions and construc-
tions, such as XC>:TEd;fLC>:'vOV for XC>:TETI:OVOUV, E:7:'TW&tX~wv for k~XAEUtX~wV, 7tpo; TO
~c>:crLi'EWV ~crTU for Ev KwVcrTC>:VTlVOlJ7tO',EL and c>:,JTT. TOV rcrov al6:&mVTO XLVaUVOV for
c>:UTT. 0fLOLWc, I.PY,crc>:cr&CH; or perfects for aorists, such as E:YXEI.ElplcrfL~vm for
E::ZElp,cr&~VTE~ 15. In passages of his Breviarium that are parallelled by those
of Theophanes' Chronicle and go back to a source common to both,
Nicephorus' style is higher than that of Theophanes. Thus the latter's de; T~
TOU ';'EV,XO'J j,OYO&e:crLOlJ 7tp6:YfLC>:TC>: XC>:TEcrT"fjcrEv appears in N icephorus as TWV

14PO 99. 264D.

15Compare L. OROSZ in Magyar·(}orog Tanulmanyok 28 (1949) (on London, British
Mus. Add. 19390). lines 6; 56: 60; 96-97; 9 with C.DE BOOR, Nicephori ... opu8cula
(1880).3,10; 4. 28: 6,1; 7, 24·-25: a, 14. Howeyer, already the first draft has pretensions
and uses duals: tIlE' re\'iKion keeps them. Cf. OROSZ. line 9 and DE BOOR. a, 14.

204
 L SEVCE:\KO, I,evels of Style in Byz. Literature 295

8'Y)f.toaiw'l AOYLeJ~'1, 0'1 TO ih;f.tw8e:r; YE'ILXO'l AOYO&E71JV XOCAOUCJL, XOC7EeJ7Y]eJEV In. On

the strength of what we know about Nicephorus' "elevating" work on the
style of the first version of his Breviarium, we may infer that he raised the
style of his source as well, while Theophanes copied it unchanged.
M i c h a e I G I Yk a s transposed simpler passages from his Annals, a
world chronicle purporting to be composed for his son, into a more ample
and eloquent diction, 'when he reused this work in a letter addressed to a
sinful princess: In the letter, the Annals' CP0eJ(J<X70V became eJTpoc,6v, and the
chronicle's passage "Therefore the Emperor petitioned monks of \~arious
places on account of his sin" was rendered in the letter by "He besought
the holy men of the region by letter and put them forward as his
intercessors before God'·li. Glykas also assumed a more recherche language
in his Controverted Questions on passages of the Holy vVrit, when he
addressed learned members of the aristocracy 18.
Even copyists of a popular romance such as Libistros and Rhodamne
showed an awareness of higher and lower levels of style. Some of them,
such as the scribe of the Parisinus Graecus 2190, considered higher forms as
more desirable, and consciously, if clumsily, raised the text to the more
learned level; or else, they introduced learned glosses into the lines at the
price of violating the verse forms. Other scribes, notably that of the
Scorialensis 'P'-IV-22 - which also contains a version of Digenis Akritas -
consistently demoticized their text, and again in the process created lines
of more or less than fifteen syllables 19.
3. Explicit Statement.s Concerning Levels of Style
In his prologue to the Life of John the Almsgiver, Leontios of X eapolis
asserted that he would write not in the learned high style adopted by his
16 Cf. Theophanes, Chron., 367, :22-23 ed. DE BOOR with Xicephori ... opuscula ... ,
37, 18--19 ed. DE BOOR. - For changes in the direction of more Attic forms and more
recherche vocabulary that ~iketas Choniates made in the versions of the final part of his
History, cf.. e. g., WIRTH. Athens Report (n. 1 above). 43--44.
17 Cf. Annales, 508.12-13 and 17-18 Bonn with the Letter to Princess Theodora.
ed. K. KRDIBACHER, Michael G1ykas. Sitzungsber. d. philoB.-philol . ... CI. d. K. bayer. Ak.
WiS8. 1894. fasc. III, 458, 16 and 25-27. For other examples of improved level of style.
compare Annals, 457,15-16: 22-24: 24-28; 458. :2-3, 27-:29; 459,1-5,16-20 with
Letter, 477.1-2; 2-5; 5-6; 509,1--4; 510, 6-9.
18 Cf. e. g., Aporiai, 33 and 34, to pansebastos sebastos Constantine Palaiologos.

S. EUSTRATIADES, ed., I (1906) 379--461.

19 M.K. CHATZEGIAKOUMES, Tit. fLEO'(xLumxit. OYjfLWO'" xEifLe,,(X ... I (1977). 111-115;
122-138. On the mixture of linguistic levels in vernacular poetry (archaic formulae being
preserved for metrical reasons) and on differences between translated (lower style) and
"original" (higher style) romances, cf. M.JEFFREYS in Mosaic 8 (1975) 171-193 and
E.JEFFREYS in Byzantina Australiensia 1 (1981) 116--127. To what extent did vernacular
poets consider the "standard" politic verse as norm?

205
296 5.2. Stilstufen

predecessors, John Moschos and Sophronios, but in a style which would be

prosaic, unadorned and lowly, so that even readers who are "unlearned and
ignorant men" (Act. Ap. 4: 13) might derive profit from his narrative.
Moschos' and Sophronios' Life of John the Almsgiver is lost; but we may
compare the style of Leontios to that of these two authors, for we can still
read the anonymous writer who excerpted Moschos' and Sophronios' Life
and we know that he followed his model closely. A comparison between the
AnonYIDus and Leontius reveals that the laner was not merely indulging in
a topos of modesty, but did in fact affect a lower style than that used by
Moschos and Sophronias 20.
To cite a celebrated case, Photius' judgements in the Bibliotheca refer
to works whose style. in his "iew. aspired to a better level than that of
others; in other eases a text was said by him to be in a more lowly style
than the one anal~'zed just before; to hold a middle ground between a
precise and negleeted and lowly style; to be more neglected than precise in
style; or ~ such was the style of Cosmas Indicopleustes ~ not to be up to
ayerage standard~l. Photius created a special category of "ecclesiastical
style" of Clement of Rome and Polycarp of SmFna, and described it as
simple and uncomplica ted ~~.
Kicephorus Basilakes differentiated between the styles of his two
speeches to Alexios Aristenos and his two Discourses to members of the
Imperial family respectiyely2:l.
Niketas Choniates. the paragon of high style and a tacit user of
Philostratos, \\Tote in 1210---11 a Speech celebrating Theodore Laskaris for
having slaughtered the Sultan of Ikonion. He remarked in the title of that
work that he had adopted a more relaxed level of style in it. The Speech, the
title said in part, was "full of clarity on account of the [literary] weakness
of the listeners". The author's habits must have stood in the way of
carrying out his intention to \\Tite in a more accessible fashion. for the
Speech has two hapax legomena (?IXUQ"(~(j/\W"";"e:pO:;, fle:y:x/,e:yxe:wr;-:-r,:;) and is
replete with rare words. StilL Choniates made some concessions to middle
style: he did not resist punning on Theodore's name, he freely quoted from
the Holy Writ; and we may be sure that the sentence "But, 0 Emperor,

~o A..J. FEi3TU(;IERE. ed., Yie de Symeon Ie Fou ... (1974), 344, 58-69. The
rt"spectiYe expressions are O,VW7E?W:; and "f?flY)A(;, XG1:PG1:Y.7r,Pl. On the Anonymus, cf. ibid., 315-
319.
21 Cf Bibliotheca, codd. 257; 199 yersus 198; 256; 258; 36.
22 Bibliotheca, cod. 126. The terms are ixY.I''iO"lG1:O"Tty',,:j XG1:pG1:y'7T,PO~ or E:xx).r,mG1:O"-:tXOV -rr,c;
epp.:r;vd:xc; 't"U7t'OV.

A. GARZYA, ed., Boll. del Cornitaio per 10 preparazione ... dei Classici Greci e Latini,
23
K. S., 19 (1971) 56ff., lines 258-2i4.

206
 1. SEVCE.'IKO, Levels of Style in Byz. Literature 297

may you live forever, if possible, and may your Empire be immortal" could
be understood by e\~eryone. Once Choniates made one of his rare bows to
low style: among all his rhetorical works a construction of (;rr,:-,-~ with o~,
occurs only in our Speech ~4,
One final example: In his bibliographical testament addressed to
Nicephorus Gregoras, Theodore Metochites described his own oratorical
oeuvre as being in ornate style and consisting of "showpieces of euphonious
eloquence", while his philosophical works, he said, were unadorned, "sim-
ple and unpretentious and \vithout any beautiful and charming style" 25.

4. Cse of Different Levels of Style as a Literary Device

About 13:24 }Iichael Gabras \\Tote a letter to an Archbishop. He
opened it with a greeting in low style, standard for informal corresponden-
ce between ecclesiastics: "Most honorable Father and Bishop. may you be
healthy in body". Thereupon, and without warning, he turned to high style,
"according to my habit" and, he added, the former habit of the once
rhetorically skilled addressee, By using ;'habit" (au'l+,ih:~:x) to describe high
rather than lo\v style - one of that word's accepted meanings - he
obtained a surprise effect. vvnen Gabras came to the substanti...-e part of his
letter - a request for a written opinion concerning his protege - he
announced that he would now lower the level of his discourse (x:x-
[l:Xl~+,AW<; ... O(IXAS"(Sa&lXl) to comply with the requirements of his subject
matter. Accordingly, the substantive part of the letter was composed in
simple syntax; for "there", Gabras used the word :x,j,,6&(, characteristic of
fourteenth-century partriarchal documents and of informal correspondence.
Having finished with the substantive part, Gabras announced that he
would close his letter in high style. "a language which escapes the many"
- he proved it by quoting Homer - and at the letter's end reminded his
addressee once more that its middle part had been reserved for those who
"are raised in rustici ty" 26.
The Schede of the 11110use (twelfth century) parodies literary forms more
than it satirizes life. It is a mock-school exercise in two parts: the first part

"+ .J.~L. \'.-L'\ DIETEN, ed., Nicetae Choniatae Orationes ... (1972). 170-175; hapax
legomena: 170, i and 171, 26: puns: 174, 14 and 175. 15; simple sentence: 175, 24~25:
rp<x",e" ... ;;'n 173, 7~8. On Choniates and PhilostratoH, ef. A.FATOCROS. JOB 29 (1980)
165~186.
25 Poem 4, 2.f.f~249, Parisinus Graecus 1776, fol. 66r.
26 Ep. 362, A. FATOlJROS, ed., Die Briefe des Michael Gabras ... II (lrBS 10, 2). Wien
1973, 568~570. Cf. Lines 2 and 26 for "my auvr,&ec<x", meaning high, and "the most common
(f\JV+,&EC<X", low style. The low style part is in lines 30-36 (or 30-45). Cf. also lines 29~30;
32; 34.

207
298 5.2. Stilstufen

spoofs secular rhetorical texts in high style; the second is a pastiche of

ecclesiastical middle style prose 27.
The Spanos (fifteenth century?) is a parody whose central part, the
Synaxarion of the Beardless, derides the hagiographic genre. The author,
who was fairly educated, fairly familiar with rhetorics and certainly with
liturgical, legal and hagiographical formulae, obtained much of his effect
from the simultaneous use of two disparate stylistic layers; of the middle
level language and formulae of a standard Saint's Life and of ribaldries
couched in purely demotic garb 2B .

III. THE TYRA!\!\Y OF HIGH STYLE

1. The Case of the Scriptures

By the fourth century, Christian writers practicing the high style were
following the norms of the second sophistics, a movement modeled on the
works of pagan 'Hiters and followed by pagans. The Word of God, on the
other hand, had been preached in works which appeared to these pagans as
stylistic monsters. Early Byzantine litera ti were aware that this discrepan-
cy between the style of the Scriptures and that of a Demosthenes or a
Libanius posed a grave problem. The problem '"as solved, if to the
disadvantage of the Scriptures; it also was solved quickly, although it
continued to trouble the conscience of Byzantine men of letters until at
least the ninth ce~tury.
At an early stage, the low level of the Scriptures' style was readily
admitted, and the triumph of solecism m~er the clever pagan stories was
presented as the triumph of unadorned truth over rhetorics, ,,,hich obscured
that truth's beauty and embellished lies with eloquence. The author of the
spurious correspondence between St. Basil and Libanius put it in a nutshell
when he described the Old Testament as "in substance true, although in
style unlearned" ('JCJ~v f.LEV !XAYj&r" AE:~'V OE !Xf.Lrx&~)29.
John Chrysostom granted that St. Paul's testimony about his own
"rudeness in speech" did apply to matters of eloquence, and that Paul
should not be compared with Isocrates, Demosthenes, Thucydides and

27 Ed. by J.-Th. PAPADEMETRIOr in Illinois Studies in Language and Literaiu.re 58

(1969) 210~222. Cf. 1st part. O:yi,aofloprpa 15; t1.npaYfl".,."u7WC; 17; parison 18-19; ehiasm 19~·
20, with 2nd part, o",x ",i 11 ; xavow1. 69; 01; tra~~oc",o'J 70 = Luc. 18; 12; Lva rpltOCtrW = reach 87.
H. HUNGER sees the whole Schede as a satire against monks, cf. Der byzantinische
Katzmausekrieg ... ' (1968),59--60.
2A Cf. H. ElDE~EIER. ed., Spanos ... (1977), esp. pp.13~14; 105~117; 187~205.
29 Ps.-Basil. Letter 339, e. g., Loeb ed., IV (1970). 298.

208
 I. SE\(~EXKO. Levels of Style in Byz. Literature 299

Plato in style 3tJ . But already in the fifth century Isidore of Pelusium
tempered this admission with praise of pagan high style and proposed a
compromise. The style of Christian wisdom was lowly, but its meaning
reached to the heavens; pagan wisdom's style was resplendent but the
deeds it described, lowly. However, one who managed to combine Christian
meaning with pagan style, would be considered a wise man indeed. For
volubility could be made into a tool of supra-worldly wisdom 3l .
Photius went beyond compromise. He not only stated that good style
did no harm to piety, but in three consecutin Amphilochia praised the
excellence of St. Paul's style in terms of Hermogenes' theory. As for
St. Peter, Photius showed in another chapter of the Amphilochia that the
verb E'YXOfl.~W0a.cr&O:l. "be clothed with", used by the latter, was far from
barbaric: it had been used in antiquity by Epicharmus and the comic writer
Apollodorus of Karystos:J~. Photius' vindication of the Apostles' style was
in fact a further defeat of the Scriptures' claim to become a literary norm,
for his defense consisted in proving. against plausible evidence, that Peter
and Paul followed the rules' of pagan rhetorics and used classical voca-
bulary.
By the fourteenth century it was no longer necessary to fly in the teeth
of evidence. In his Ecloga of Attic nouns and verbs Thomas Magister
quoted words and idioms from the Scriptures a mere ten times. Of those, six
quotations were cautionary, and either pointed to a unique or rare usage, or
reminded the reader. more or less tactfully. that he should not employ a
given Biblical expression in his own '\Titings. Once, the New Testament
usage was even described as barbaric 33 . \Ve are therefore not surprised that
in his textbook on rhetorics .Joseph Rhakendytes mentioned the Old
Testament but once ~ he quoted the Octa teuch and the four books of
Kings ~ along, of course, with the Iliad, Herodotus and Aesop. as texts
from which one might conveniently cull historical examples 34.
In Christian Byzantium the Scripture never became a predominant
model of style at any level, except, and there rarely, for the lowest forms of
hagiography 34a. This created an impasse. The way out of the impasse was
311 De Sacerd. -1, 6. PO 18, 668--69. on II Cor. 11 : 6.
31 Isidore of Pelusium. Ep., 5: 281, PG 78, 1500D. On all this. cf. E. NORDEX. Die
antike Kunstprosa, II (1909),516-534.
32 Amphilochia. 91-9:3: 86 (also appearing as letters). PG 101, 570D-5.,)7AD. On
Photios' use of Hermogenes, cf. B. Wess in <viI/Bellm Helveticnm 12 (195.5) 236-251.
esp.241.
33 Ecloga, 112,9; 165.:3; 192,10; 273, 9; 312 . .5; :3:37,17 ed. F.RITSCHL (1832). Of
course, there is no en try 1:yxOfl~6()',,()'j).,,~.
34 WALZ, Rhetores Graeci, III, .525.
34. The best example: Vita Nicolai Sionitae, BHG 3 , no. 1347.

209
300 5.2. Stilstufen

first found by Byzantine practitioners of high style and later sanctioned by

its theoreticians. It consisted in setting up a canon of style, which would be
Christian, but based on those Christian writers who thoroughly espoused
the precepts of the second sophistics - John Chrysostom and the Cappado-
cian Fathers. Of these, Gregory of Kazianzus was declared the supreme
source of stylistic inspiration. Psellos repeatedly called him the best stylist
of all, superior to Demosthenes, Demades or Isocrates; and authors of
textbooks of rhetorics, beginning with Sikeliotes (eleventh century), quoted
frequently from his works. Sikeliotes even substituted illustrative examples
fI'om Gregory's orations for those customarily drawn from Demosthenes
and found that. compared to Gregor:". Demosthenes was a mere child 35 . By
placing imitation of ancient models under the high patronage of the fourth-
century Fathers who had put a limited seal of approval upon the formal
aspects of pagan literature, the Byzantine practitioners of high style could
consider this imitation as a legitimate enterprise.

2. The Ca8e of Hagiography

Byzantine hagiographers proyide us with specific e,idence for the
prestige of high style. Symeon Magister's Metaphrastic collection of 148
Saints' Lives. most of "'hich had originally been written in simple diction,
is the most telling example of a wholesale stylistic upgrading. This
upgrading was done all at once, as it was produced, upon imperial
instigation, by a team of scribes and correctors superyised by a high
officiaJ35a. Other examples stretch across time. Sometime the pre-Metaphra-
stic L~fe, its intermediate reworking and the subsequent Metaphrastic
version are all preserved - this is the case of the three Ll:ve8 36 of John the
Almsgiver - which suggests that Byzantin,e audiences were receptive t{)
different levels of style and found use for each of them. If we have 1\"'0
versions of a LIfe that neyer entered the Metaphrastic collection, the la ter

".', l'~pllos: BZ 20 (1911) 48~~60 (cf. 57: tD exalt Gregory', paneg:Tics. Psellos puts
Thucydides' Epitaphios down a, being below that historian's usual penormance and not in
the best of styles: to us, a monstrous judgement): J. Fr. BorSSONADE. ed., De oper.
daemonum (1838). 50~... 51; K X. SATHAf;, ed .. Mes. Bibl.. V (1876). 150. Sikeliotes
Prolegom. Syll.. 414. 20ff., ed. R.'\BE. The fullest list of Sikeliotes' quotations from Gregory
i~ in A. B. POY1\T01\. Gregory of X az. and the Greek Rhetorieians. Oxford 1933.
35& Det.ails on the erea tion of the Metaphrastic collection are gi\'en in Psellos'
Eneomium of Symeon, 10.5, 16-27 (for ed .. see n.3fl).
36 BH(i3. nos. 886, 8S7\' , 888; cf. also 887 and E. LAPPA-ZIZICAS in AnRall 80 (1970)

265-278. Another example of three versions is the Life of Xicephorus the Monk: BHcP
no. 1331 (simple); BH(P no. 1334 (high style "metaphrase '. of no. 1331 by the early 9t.h c.
theoretician of rhetDrics, .1ohn of Sardes); BH(i3 no. 1332 (Metaphrastic nrsion, based on
John of Sardes).

210
 r. ::;r;\·('E:\KO. Levels of Style in Byz. Literature 301

version is more elegant. Such i" the case of the Life of Saint Philaretos the
Merciful, originally \Hitten in 812 and revised, with the obliteration of the
autobiographical details. sometime before the twelfth century 37. In the
reworkings just mentioned, the chronological sequence is clear and the
tendency the same as in the Metaphrastic ones: toward more oratory and
higher style at the expense of concrete detail regarded as obsolete. All this
means that in hagiography the prestige of high style was in the ascendant
throughout a large part of Byzantium's existence. There wail a tendency
through the centuries for specimenil of the genre to become progressively
more, not less, high flown and obscure. \Ye must remember that Byzantine
hagiography begins in the middle. rather than high style. ail witnessed by
the Life of Anthony by Athanasius of Alexandria, and is not mails-produced
in high style until the ninth century. This tendency to translate texts into
high style is not true of all forms of Byzantine literature, for in historiogra-
phy and political theory the opposite is attested in late Byzantium. when
several high style texts were pro\ided with middle style paraphrases. There
must have been attempts in hagiography to help those readers who found
the high style heavy going, but I can,recall only two offhand 3s . \Ye have to
wait until the post-Byzantine period for a low-middle style corpus, the
Synaxaristes by ~ikodemos Hagiorites. a counterpart, as it were, to .Jacob
de V oragine' s Legenda A urea.
Most Byzantinists consider Symeon }Ietaphrastes to have been a
practitioner of high style, but Byzantine critics of a generation. or perhaps
two, later than Symeon, would not have agreed. In his Encomium of
Symeon, \VTitten less than a century after his death. Michael Psellos did
hold the unadorned style of pre-Metaphrastic Dices up to ridicule; they
were the laughingstock - he said - among the refined spirits of Symeon's
own time. Psellos himself \iewed Symeon's exercise as an unprecedented
achievement, but he fudged a bit on the subject of Metaphrastes' style. He
himself described it as both clear - a dangerous compliment when coming
from a high style literary critic - and sublime, but he had to report that
the most sophisticated litterateurs of his own time faulted Symeon for not
reaching philosophical depth and - and this is crucial for us - for not
skillfully mixing different kinds of styles. By this they implied that Symeon
did not live up to the precepts of Hermogenes. Finally, critics found fault
with Symeon for not using figures of speech prominently enough and, in
general. for not writing in an epideictic fashion. Psellos defended Meta-

:;; HHU 3 nos. 1511z and 1512.

38 Glosses throughout Monacensis Gr. 366; Vatopedi 114.95"-114", a simplification
of BHG 3 , no. i15 (done by the 9th c.).

211
302 5.2. Stilstufen

phrastes by saying that he did not aim at spurious elegance and wrote
successfully for both the elite and the many. The former could be charmed
by his rhythms; the latter would appreciate the abundant clarity of his
prose. Psellos admitted that Symeon was no Isocrates or Thucydides. He
was aware of works done in better style, but he insisted that almost no one
could approach Symeon's skill in writing for all kinds of audiences 39 .
In the light of Psellos' words we should think twice before assigning
the label "high style" to the Metaphrastic corpus. If we insist on conti-
nning to do so, we shall then have to establish yet another category, called
"super-high style", in order to account for the improved reworkings of
Live8 that go their already high style models one better, such as Nicephorus
Gregoras' version of the Life of the Empress Theophano. Gregoras impro-
ved on its fairly high style model of about the year 900 with both lexical
refinements and rhetorical developments+u.
A rough sociological comment on the producers and consumers of
hagiography in high style would run as follows: It was a branch of
literature produced by members or associates of the upper class, for the
members of the upper class, and, more often than appears to meet the eye,
abo u t members of the upper class. I t was also a literature that made use of
special skills to serve as a distinctive badge either of membership in the
upper class or at least of association with it. \Ve can provide e"idence for
the noble pedigree of some of these authors, for their membership in the
upper class, and for their wealth. \Ye can argue for upper-class origins for
many - perhaps the majority - of saints and their families and we can
quote a hagiographer or two with an elitist outlook on his literary prowess.
But were there enough sophisticates in Byzantine society to account
for all 693 surviving manuscripts of Saints' Lives in Symeon Metaphrastes'
reworked version? \Ve must assume, given the democratic character of
church services, and the existence of communal life in the monasteries, that
sometimes the uneducated, too, listened to productions in high style. We
suspect that they only half understood them and we fear that they paid
little attention. A short text from the Monacensis Graecus 225 bears out
our fears. In it, Manuel Philes, about to recite a hagiographical piece by
Nicephorus Blemmydes, implored members of his audience not to lie on the

:10 Encomium in E. Konz-F. DREXL. Mich. Pselli scripta minora, I (1936), 94-107,
esp. 102-103. The terms for "elite" and "the many" are EAAOYLfJ.OV hpoex.,.T,v and t3LWTL3ex_
hoa.<; respectively.
40 Both Lives ed. by E. KURTZ in JIimoires de l'Acad. Imp. des Sciences de St.-
Petersbou.rg, C[asse hist.-phil., 3, 2 (1898).

212
 r. SEVCENKO. Levels of Style in Byz. Literature 303

church floor and not to sleep during his performance t1 . Byzantine elitist
literateurs were aware of their tenuous contact with the many.

3. Obscurity
Ever since Aristotle. clarity was considered a virtue of stylet~.
Textbooks define it repeatedly as just that, on the strength of its being
listed as one of Hermogenes' Forms-!3. Positive evaluations of obscurity by
Byzantine theoreticians of rhetoric are infrequent obiter dicta that qualify
sentences devoted to clarityH. Nicephorus Basilakes, a difficult ,niter,
praised the clarity of his own style. and hardly a Byzantine author,
however abstruse. made a virtue of his obscurity.!". Not Methodius. who had
to defend himself on this account.!ti. and not Arethas in his counterattack
against those who accused him of obscurity. There. he merely showed off
his knowledge of Hermogenes' terminology, claimed for himself a mixture
of both clear and obscure styles and said that he avoided both too rare and
too vulgar a vocabulary -l7 }Ietochites. writing pro dOIno sua, came the
closest to endorsing obscurity as a desirable characteristic of style, but he
did it indirectly by praising ,\Titers like Synesius and Philo. not to speak of
Thucydides, and by pointing to a writer's need to distance himself from the
vulgar mob. But even he upbraided Aristotle for the intentional obscurity
of his style; defending himself, he fell back on the doctrine of the 3E~',67"r..;,
the mixture of styles, dispensed according to the requirements of the
moment 48 . Nevertheless. a modern reader instinctively subdi,ides Byzanti-

-I, In Hum. the study of style alone dispose~ of the view that Byzantine hagiography as
a whole is a popular genre. Social historians have seen the truth: cf. E.PATLA(;E ..\.X in
Annales 23 (1968) 107; the conventional view still survives in A. LUOl". Charanis Studies
(1980) 84.
42 Rhet., III, 2. 1. Aristophanes' Frogs contains the first Greek polemic on style.
There, Euripides accuses Aeschylus of obscurity. Cf. e. g., 927. :\1. Fn1R~IA...'1N, Obscuri-
tas ... , in: Immanente Aesthetik, II: Poetik und Hermeneutik (1966) Ji-92 found no
antique theorist who would praise obscurity. Only Longinos admitted its uses.
43 Prolegom. Syll., ed. R.'\'BE (1931),126,17; 177, 15; 201, 21.
H In his good pages on Obscurity, G. L. KUSTAS, Studies in Byzantine Rhetoric (1973)

91-93 makes too much of these utterances.

45 Basilakes (as in n.23), line 223; cf. A.GARZYA in JOB 18 (1969) 68. Only for the
Holy Writ or prophecies obscurity was admitted, but endowed with higher purpose or
explained away. Cf., e. g., Photius. Amph. 204, PG 101, 948BC.
46 Cf. my "Hagiography .. ", in: A. BRYER--J. HERRIX, edd .. Iconoclasm (1977) 126
and n. 99.
47 L.G. WESTERINK, ed .. Arethae ... scripta ... , I (1968) 187,4; 189, 14ff.; 188,

22ff.; 190, 11-22.

48 Details in "Querelle sur Ie style", my Etudes sur la polimique (1962), 51-67; cf.
also Metochites, Miscellanea. Ch. G. MDLLER-Th. KIESLING, edd. (1821), Mis. 3; 16-18.

213
304 5.2. Stilstufen

ne high style according to the ascending degrees of its obscurity. His

instincts are right, as were those of the high court dignitary who criticized
Methodius, of the enemies of Arethas and of Nicephorus Chumnos, Metochi-
tes' literary and political foe. Once esoteric style becomes a norm to which
all men of letters aspire. and its proper handling is seen as a sign of
belonging to the literary elite, the more esoteric the style, the higher the
position of its practitioner within that elite. 'Yithin such a system of values
some members of the eleventh-century elite could fault Symeon Metaphra-
stes, whom we blame for his preciosity, for not affecting high enough a style
in his hagiographical reworkings. Given such a system of values. literary
theory had to yield to rules of one-upmanship that served to establish one's
position within literary society. Thus, to gin' an example from epistolary
theory, a widely read textbook quoted from Philostratos and prescribed
that letters should be concise, ~Titten in middle style, and excel by clarity,
that "special guiding principle" of letter \niting4!l. This rule was broken by
too many Byzantine epistolographers. In the fourteenth century, Joseph
Rhakendytes regretfully noted it in his Summary of Rhetorics 5(1 and any
reader of 'Manuel II's eorrespondence realizes it today.

n-. Sl'GGEf'TlONS FOR THE STrTlY OF LEYELS OF STYLE

Owing to the meritorious efforts of our colleagues and predecessors, we

already know in large outline that educated "Titers subscribed to the ideal
of mimesis with various degrees of suc(:ess. while the less educated or
uneducated tried to live up to the same ideal. hut failed to achieye success
within the rules of the same game. '\'1" may nmy proceed to further tasks,
employing both customary and less conventional approaches.

1. Cu~~tomaTy Approaches
a. The Diachronic Study of Styles
Throughout the thousand years of B~-zantium. theoreticians and
practitioners of high style professed their adherence to roughly the same
norm. Thi8 norm was derived from the \nitings of authors who were seen as
stylistically contemporary with each other. eYen though they ranged from
PIa to to Philostratos and thus covered the span of over fiye centuries. They
represented antiquity, hmvever vaguely understood this term may have
been. On the whole, Byzantines were not considered worthy of imitation-

1!1 \', \\"EICHlmT, ed,. Demetrii et Libanii ". t~-)loi, ,(1910) 19. J4ff.
50 \\".\LZ. Rhetores Graeei. Ill. 559,

214
 1. SEV(~E~KO, Levels of Style in Byz. Literature

the Cappadocian Fathers and George of Pisidia being notable exceptions.

Joseph Rhakendytes who mentioned Symeon y[etaphrastes, Psellos, Nico-
las Kallikles and the Ptochoprodromos in his lists of desirable authors was
a welcome, if rare, dissenter'>l. Theodore Prodromos summed up the
prevailing stance in one sentence of dubious taste: "I ha \~e been thoroughly
instructed in oratory, not the kind which the frigid Simocattae fart forth,
but the one which Aristeideses and Platos breathe"·>". Given such an
attitude, the ,evolution of high and even middle literary style is especially
difficult to detect, more so than the stylistic evolution in Byzantine art,
But it is precisely the diachronic study of style that should constitute a
challenge for the immediate future, We should not be discouraged by
Thomas Magister's pseudo-Aristidean orations - whose fourteenth-centu-
ry date escaped us for so long. The Philopatris has been firmly anchored in
the latter part of the tenth century 5:l, and Professor Kazdan's theory that
John Kameniates' work on the Siege of Thessalonica in 904 is some five
hundred years later than its professed date seems as appealing on stylistic
as it is on historical ground~54. The style of the Life of Andrew the Fool in
Christ reminds one more of the seventh than of the tenth century - and
detailed analysis should bring us nearer the solution of this conundrum;;5.
b. The Study of Individual Styles
Given the iron grip of the doctrine of mimesis, distinguishing indivi-
dual styles of Byzantine authors is even more difficult than establishing the
styles of various periods, but this is not an impossible task. Some styles are
unmistakable. Methodius the Patriarch and }Ietochites are cases in point,
and the latter provides us with statistically significant numbers of tell-tale
shibboleths·i .;". Psellos, too, has revealing habits such as putting wCJ';-:sp
behind the noun to w'hich it refers or showing preference for rare technical
terms, such as cr,o~~&~<.0 and its various compounds. These habits enable us
to attribute controverted texts to him - such as the short treatise on
rhetoric published as anonymous but being surely by Psellos 5fi ,
.i1 W.\LZ. Rhetores Graeci, Ill. 526 and 562.
52 On fate, PG 133, 1297B (on his olVn education).
,,3 Cf. nolV ehr. A~GELIDE in Hellenika 30 (1971-78) :34-;jiJ.
54 BZ 71 (1978) 301-314. Cf. 314, on parallels with Anagnostes. }Iy intuiti\-e
judgment is that Kameniates does not \nite in 10th c. style.
55 The work is in middle-low style; I suspect a seventh-eighth century date: Professor
:.\Iango advocates that of seventh century. The conventional dating (e. g. by L. RYDEX) is
10th C.
55. Cf. my Etudes (as in n.48), -!O, n. I. One of the shibboleths is x[J.1)ys;rr" derided by
Lucian in Lexiphanes, 21.
56 L. SPE~GEL, ed., Rhet. Graeci, I, 2 (1894), 208-212. Cf. n7z"7m~C("[J.s'Jov 212, 3.
The manuscript tradition, too, assumes Psellos' authorship.

215
306 5.2. Stilstufen

Even when lin author writes in two styles - Ignatius of Nicaea did it
in his hagiography - he may leave enough fingerprints, as it were, to
reveal his identity. The lofty Life of Nicephorus and the more lowly Life of
Gregory the Decapolite share enough expressions and phrases to make the
attribution of the latter Life to Ignatius a matter of certainty 57.
Ignatius, Psellos, Metochites - these admittedly are the most promi-
sing subjects for the study of individual styles, for all of them were forceful
literary personalities; we understand why Psellos has been repeatedly
studied - even if what has been investigated is more his grammar than his
style. But even where an author displays faceless correctness - Constanti-
ne Acropolites is one such author - an analysis of his style will be of
importance, for it may enable us to establish a norm for a given stylistic
level of a period and thus provide a point of reference for appreciating the
style of that author's other contemporaries.
c. Rhetorical Textbooks and Rhetorical Practice
The study of Byzantine style has been pursued along two fairly
independent lines: the im-estigation of Byzantine rhetorical theory (ex-
pounded either in textbooks or in critical essays on individual authors, such
as the Church Fathers, George of Pisidia, or Euripides) and the analysis of
the style of indi,-iduals. The next step may be to correlate the precepts of
rhetorical textbooks, norms proposed in lexica, and methods of school
instruction (if we ever acquire enough information about them) with the
stylistic practice of individual authors, including those "'Titing in middle
and low style 58.
d. Stylistic Terminology
Now that the antique pedigree of Byzantine stylistic terminology has
been reasonably established, we should examine the degree to which this
terminology was understood and endowed with precise meaning by the
Byzantines themseh-es. Precision in the technical vocabulary of a Photius
or a Psellos is a priori no more likely than is such precision in the usage of a
modern literary critic. Demetrios On Style uses GTWfLVAOC;, in an ambiguous
sense. So does Psellos in his Laudation of Symeon Metaphl'astes, where it
seems to be a positi,-e term 59. But what does Psellos really mean by it? The
ideal result of such examination should be a Byzantine Ernesti, but with a

;,: Cf. my "Hagiography" (as in n. 46), 123 and n. 71.

58 A tag from Demosthenes used as an example in WALZ, Rhet. Graeci, III, 528 may
have been the source of Matthew Gabalas' "On scholars ... ", 2, 17, my Elune8 (as in
n. 48),288.
59 Demetrios: 151, ed. RHYS ROBERTS; PselJos: 104, 21-23, ed. Kl'RTZ-DREXL
(as in n. 39),

216
 1. SEVCEKKO. Levels of Style in Byz. Literature 307

difference: in it, all loosely used or unclear terms should be marked as such,
rather than merely translated by their non-technical equivalents.

2. [' ses of lr1 odern Linguistic Stylistics

The recent pioneer effort of Professor Hunger6() should SpUT us to
undertake similar - and computer-generated - analyses of the sentence
structUTe in works written in various levels of style. We also should explore
the extent to which modern literary and linguistic stylistics can help us in
rearranging OUT materiaL
a. Functional Styles
One concept of modern stylistics which should prove fruitful in dealing
with Byzantine material is that of functional styles. This concept relates
style to the pUTpose for w'hich a message is issued and consequently to the
audience at which it is aimed ti1 . Statements connecting style, pUTpose and
the audience are not unknown in Byzantium. PseUos claimed that he could
vary his epistolary style accorcijng to his addressee and, following antique
terminology, he frequently referred to xcnp6~, "appropriate circumstance"
which should determine the character of a literary work. In one of his
Commentaries on Hermogenes, John Doxapater once stated that he would
use the middle style, abundant in clarity and simplicity, on account of his
treatise's pedagogical pUTpose 61 .
Byzantine practice. however, is contradictory with respect to the
nexus between the style and function of a message. Documents collected in
De Cerirnoniis and De Adminislrando Imperio are in low style, as indeed
they should have been, given their practical purpose. but their Imperial
editor felt obliged. to submit excuses for this deficiencyo:l. Secret letters sent
by Emperor Andronicus II to his nephew and political ally in the late 1320s
show that when the very survival of the old emperor and his party was at
stake, plain talk was considered in order - for these letters were dictated
in sub-literary style, and the imperial patron of the Palaeologan Renais-
60 Stilstufen (as in n. 1).
61 DOLEZEL-KRAUS (as in n. 1) 38-39; S.\~DERS (as in n. 6), 19-20, 108. Caution:
functional styles are often considered as "kinds" rather than levels of style. Another
modern concept applicable to our topic is that of Basil Bernstein's "elaborated" and
"restricted" codes. The latter. corresponding to our middle and low styles, uses smaller
vocabulary, shorter and para tactic sentences, predictable patterns of sentence structure.
neglects the uaT'iatio, and uses cliches. Michael Glykas with his innumerable ,,[ Xp~ X"" AEY","
and "" OE: -:"'~,'" ou""'c; Z;(€' would be a user of the restricted code.
62 For Psellos' x""poc;, cr., e. g., ehron., II, 68, 25 ed. RE:-iAl'LD; Praise of Mauropous,
ed. K. N. SATHAS, Mes. BibL V (1876),150. Doxapater: ProL Sylloge. ed. RABE, 373,1-
13; cf. 161, 12.
63 On all this, cr., e. g., G. MORAVCSIK in SEN 5 (1939) 514-520, esp. 518.

217
308 5.2. Stilstufen

sanee did not shrink from writing XClt O:~ YVWPLo"W in one of them(j4. But what
should we think of Photius' Letter to the newly baptised Bulgarian ruler
Boris-Michael? What did Photius haw in mind in terms of functional style
when addressing this treatise to a man who could not have possibly
fathomed either its doctrinal intricacies or its stylistic refinements?
In their style. the ornate prooirnia to imperial and patriarchal charters
are closer to Photius' lucubration than to the letters which Andronicus II
addressed to his nephew. Owing to the recent studies by professors Hunger
and Mazal 65 we now understand the antique antecedents of these prooirnia.
know the model books from which their repertoire was drawn and can
\isualize the rhetorically educated chancery personnel who penned them.
The purpose of these documents' message, we now understand, was that of
propagating the imperial idea or the patriarchal self-image. The high style
of the prooirnia may haye been appropriate to the loftiness of their
message, but was it appropriate to the function of the charter as a ·whole.
whose dispositi,~e part was what counted? In an analogous case ,Justinian
thought otherwise, for he ordered the prefaces be omitted from the
Constitutions included into his Code. \Yere the prooirnia attentiyely read,
and ifread. understood by the charter' s recipient? It is unclear whether the
style of the message of thp prooimia was functional or merely a reflection of
the inertia of a bureaucracy who acted as custodians of a special craft tha t
contributed to the perpetuation of their position, and who were as much
interested in pleasing their superiors as in reaching the public to whom
those superiors wished to havp their message deli,~eredfl;ia. As for the part
these superiors may have played in the actual \niting of the prooirnia
issued in their names. we should distinguish Qetween those who were able
to \nite w'ell and those - I suspect the majority - who were not. A.rsenios
Autoreianos (deposed 1265) is a cautionary example. As a Patriarch in
office, he must han had at his disposal chancery clerks of high competence.
for the opening parts of his sigillia are couched in highest style 6G ; as a
deposed Patriarch. he had to fall back upon his own \niting or yerbal skills,
and his Te,o,tament is a humanly mo\~ing. but stylistically substandard.
document. Autoreianos construes XP:XO[l:ll with an accusative. uses lV'Y. with
the indicative, puts an aorist instead of a future, writes - or rather
dictates - &.:),W '{vex, and strews dangling participles tlU'oughout his text.

ti< Cantacuzenus, Rist., I. 232. 23-236. 15 Bonn, esp.234, 1.

(;:, H. H l:NGER. Prooimion (lrBS 1). Wien 1964: O.l\L-\zAL, Dip Prooimien der byz.
Patriarchenurkunden (B r 7). Wien 1974.
650 For parallel yipws on professional scribE'S in Byzantine Egypt, cf. R. ZILLlACT" in

Com.mentationes hum.. litt., SocietlUi Sc. Fennica. 41. 2 (1967).18--22.

66 Cf. especially MAZAL (as in n. 65), no. 173 = p. 135.

218
 I. SEVf ' KXKO. L e v e l s o f . St yl e in B v z . L i t e r a t u r e 3(>0

These faults are not redeemed by the occasional occurrenee of a high style
word, such as o[ f.LSiJ.'>·r;,~i:'/ol, of a rhetorical question, of an eceiesiastical
formula. or of an etymological figure inspired by the Septuagint fi '. After all.
Autoreianos was a High Prelate.
b. Metaphrases as X orm
Students of style in a modern language compare their evidence with
some norm of which a ginn style is a specialized or individual nuiant"".
For them, the "com'ersational" or "usual" speech or prose of an average
educated person is such a norm and "high style" is then seen determined a"
a deviation from it. In the \"'eke of our Byzantine predecessors. we
Byzantinists proceed in the opposite way: we assume with them that \\.'orks
in high style provide the Ilorm. e\'en if we can guess that such works had
little to do with the language spoken by their educated authors. Cnlike
modern linguists, we cannot directly confront the subjects of our stud~'.
But we do have at OLU' disposal a large body ofpl'o::;e that can be con::;idered
the Byzantine counterpart of the a\'erage modern educated speech and can
serve as a norm by which to judge Byzantine high style. This prose is
contained in the numerous Byzantine paraphrases or metaphrases 6H of
high-style works, both Byzantine and Antique - some important ones.
such as that of Xicetas Choniates. still remaining unpublished'l). Changes
introduced by the paraphra::;ers are customarily viewed as deviations from
high style: the omission or simplifica tion of difficult passages; the remoyal
of hyperb<tta; the glossing of those \yol'ds that had been taken over without
change; the introduction of prepositional constructions instead of da ti,~es;
of present and past forms for perfects and pluperfects; of periphrastic
cOIlstructioIls with ~:x.Sl'J or '~;:i~Z2W in place of verbs: the elimination of the
classicizing technical terms of the model by functional ones (replacing
(j":-x-:-Y,in by 70")P'/EulO'l and o~oi.o'~~ by u":':XfLS'IX) 71.

,;7Tes{rtmmt in P(; I-!O. 9-!8--957.

68 HUl'OH. Style (as in n.l), 25: +3--±-!; S.\XDERS (as in n. 6), .')3, :57. lil.
69 This term is advocated. with good reasons. by A. PW:';'.-L'ir in Atti della Accademia
Pon/aniana. X. 8 .. U (HJ76) 219-225. esp. 223-,1"
70 J.-L. L\-'i DrETEX, Xic. Choniatae Historia ... (1975), LXXXVI-LXXX\~IIl:
CI\~. and Byz. FOfsch. 6 (1979) 37-77. A choice list of paraphrased authors will gi\'e an idea
of the abundance of the material: Gregory of X azianzus (poems): Synesios: Agapetos:
Syntipas (two): Anna Comnena; .\Ianasses; Choniates; Blemmydes: Pachymeres; not to
speak of prose paraphrases of Homer (anonymous. such as that edited by Bekker - and its
close relative found on Sinai in 197.') - and those by Psellos and :'rloschopulos) and of
Lykophron's Alexandra (two). We should also consider the paraphrasing legends of Scylitz€8
M atritensis, simplifY'ing glosses such as those of Mon. Gr. 366 and Vaticanus Gr. 97i. and
even such trifles as a paraphrase of an epigram by Gregoras (cf. Bessarione 22 [1918] 98).
71 Xic. Choniatae Hist., ed. LI.X DrETEX (as inn. 70),57,

219
310 5.2. Stilstufen

If, however, we ,rjew the work of the authors of the paraphrases as an

attempt to bring the high style down to a norm, we should be able to
reconstruct elements of the standard vocabulary of the Byzantine "usual"
prose by examining what words were consistently used to replace expres-
sions of the high-style models. By comparing the word sequence in the
paraphrases 'with that of their models, we should get an idea of the
"normal" word order prevailing in that "usual" prose.

V. LE\'ELS OF STYLE AND LEYELS OF PRODCCERS AXD CoxsomRS OF

LITER ATt'R E

As a rule of thumb. we connect the use of high style with the educated
few ~Ti ting for the small circle of their equals or dependents; middle style
with the more numerous group of writers producing for a larger audience;
and low st~'le with the less educated few who wrote for their own likes -
again a small number. for we assume that the less educated do not read
much.
As is often the case, the system holds fairly well for the top and the
bottom. \Ye are sure that Metochites. who despised the uneducated -
meaning those less educated than himself - and the lower classes. never
'Hate a line in middle ~la, let alone low style. And we are inclined to believe
John Kananos. who left us the Narratiye on the Siege of Constantinople in
1-1-:2:2. when he confesses that his st~'le if; f;olccistic and barbarous. that he
is inexperienced in letters and that he has 'Hitten not for the learned and
wise, but solely for the "rude folk", being himself one of them. The literate,
he continued. should take his N arra ti,'e,at its face value and read it without
passing critical judgment on it. Faced with Kananos' indiscrimina te use of
T:C:PL or E:x "'ith the accusative or the genitive, with his Modern Greek
futures. and his Italian and Turkish words. we feel that on no occasion
could he have risen to the level of middle st~-le. But we note that he knows
the technical term for "style". and hoped for a learned audienee;~.
Things become eomplicaJed in the middle range. with authors who
'Hite in more than one style and address -~ or pretend to address --

71u EXl'ept in his technical works on .-'l.stroJ1omy or Aristotle.

72 P(i 156, 61AB= E. PIl\TO. ed .. L'assedio di Costantinopoli (1977), lines 6-17.
Kananos tries his best: he coins unattested high,sounding words (lXlfL~.-ro~i)po:; 30, cf.
7.1fLIX701r';,'77,:; in Aristophanes: a,,",xpx;rpovoo"Y,7'O)C 43). repeaLs a high,style cliche twice (i:v fLlqc
"""poi) po:-:r, 32, 395), and uses exclamations and rhetorical questions. For a weleome step in
st~'listic analysis of Kananos, cf. H. H C'l;ER. Die hochsprachliche profane Litera tur ... I
(19i S) 482--484.

220
 f. SEH'E"KO, Levels of Style in Byz. Literattu'e 311

several audiences. \Ve di~believe some, and are puzzled by others, Basila-
kes asserted that his literary efforts were of profit to "the rude" UO~(07(xl)
and Emperors aJike,:J, but nothing in his writing~, Wit even his progymnas-
mata, allows us to as~ume tha,t the rude eouldprofit from his high-flown
prose.
Psellos found Metaphrastes' style appealing both to the elite and to
the many, We ourselves are struck by this style's uniformity, and suspect
that it was way above the heads of the many.
Michael Glykas was a professional man of letters and a member of the
Imperial chancery; he had access to the court, as shown by hi8 addresses to
Manuel I. to that Emperor's relative, and to high dignitaries; and by his
invol\-ement in the antigovernmental plot of 1159/ 4 . He was capable of
applying varying levels wi thin the range of middle style: he even quoted
Pindar/'. And yet, ,ve have not a line from him in high style. Was it a
question of adapting the level of style to the subject matter, or a question
of choice - as his use of the vernacular may have been. if he was indeed
catering to the literary tastes of Manuel r'6? Perhaps Glykas was just a
middle-level functionary of the chancery, unable to ascend the Styli8tic
heights occupied by Imperial orators of his time.
\Vnen we are faced with a paraphrasis that transposes a high style
work into a lower stylistic level, our tacit assumptions are that the author
of the paraphrasis belonged to a milieu considerably lower than that of the
paraphrased author, and that the paraphraser's product was destined for
his own milieu. It is difficult to test these assumptions, for most of the
paraphrases are anonymous. \Ve do, however, know the two authors of one
paraphrasis - that of the Imperial Statue of NicephoI'Us Blemmydes - by
name. It was executed by the Supervisor of Parish ChlITches (cr:xxst.i,iou) of
the Great Church, George Galesiotes, and by George Oinaiotes, "most
learned men and ora tors", "for the sake of more clarity" 77. Galesiotes and
Oinaiotes invalidate at least one of our assumptions. Galesiotes was a
'\Titer in his own right and his position at the Patriarchate was comfortably
high78; and we now have proof that Oinaiotes was capable of writing in a
73 Basilakes (as in n. 23), line 4i.
74 Cf. O.KRESTEN in .JOB 2i (19i8) 69-i2; 93-95.
75 Aporiai, ed. S. EeSTRATIADES, I (1906) 240. For a quotation from the Sibylline

Oracles, cf. ibid., 481, 18-20.

76 H.-G. BECK, Geschichte der byz. Volksliteratur (19il), 109, expressed this view in
passing; M. JEFFREYS, Byzantina AU8traliensia 1 (1981) 105 developed it by adding the
hypothesis of \Vestern influence.
77 A. M.-\.!, Scriptorum vet. nova coll. II (182i) 609~55.
78 On him, cf., e. g., St. 1. KOl'ROl'S~:S in Athena is (19i4-5) :335--3i4 and HCNGER

(as in n. i2), 139-140; 143.

221
312 5.2. Stilstufen

style nearly as high as that of Blemmydes' Impen:al Statue, for he is the

same man as the Anonymus of Florence, the author of an elegant collection
of letters who count~d Theodore Metochites among his addressees and
perhaps belonged to Metochites' milieu .~.
It \I'as certainl~' not for Met.ochit.es and his likes that our two men
composed their paraphrasis. Did t.hey do it for their own milieu? 'IYit.hout
being able to pro\'e it. I suspect that they wrote for a clientele which was
less literarily refined. But for what clientele? For the lower cleries of the
Great Chureh. that is, for people belonging to an established milieu, whose
educational standards were sinking! For students? Or were new. if not too
sophistic-a ted. readers coming to the fore in the first quarter of the
fourteenth c-entury? And how do we explain the faet that the two "most
learned men and orators" misunderstood their original in some places, and
omitted a numher of complicated pas8ages altogether, perhaps because they
eould not understand them at aIls,,! In an~' case their paraphrasis seems t.o
han' Iwen more popular than its modeL for \ye kno\\' three manuscripts of
the Sta/III!'s original and at lea8t six of it:,; reworking.
Relating leyels of st.Yle to social and edueat,jonal leyels of the \lTiters
and of their reading or list.ening publil' is among the most challenging tasks
of OUl' di8l'iplineo l . ~-\.s ~'et, we haye only lJegun to look for solMions: but we
do 8ee the problems.

IAddl:7idum: The book by H. H l-:\(;F:R. ~-\nonyme ?Ietaphrase zu Anna

Komnene Alexiaf; X1- XIII. Ein Beitrai! zur ErschlieBung dpI' h.yzantini-
,.;ehen rmgangsspradlf' (lrB/"; 15). \\'jell 19/'11. eentl'aJ to our topic. reached
ml' after the eomplption of this report.]

~~ fu], proof of idt'lltity. cf. St. 1. KIJl'H()l'''E'';. l\lanouel Cabala~ I (Hli:?) lOlL
E. HE];\'. Die flort'lltiner Briefsammlung ... (1915) 3 :?4- 2i.
'" One I:'xample. they both miwnderstood and ahhre\'ia ted the beginning of chap. 2,
for they did not realizE' that in it RlE'mmydE's \\"[1, di'('us~illg the didionary ddinition of
baRilelis. sl!(·h as it is tram,mitted in the E/YfilO/uf}ic>1i1i JJagnulII and otlwr lexicll.
'IOn the social position of (mustly high style) Byzantine wl'itl'l's illlak pt'rio(l. ef. my
I'(']Jort in Ad" dll X II·' ('(JU!/Ies IlIlerl/u/ill/wl des Etl/des R,IIZIIII/i,tis. J (1!l7+) IiB--\l:? Tlw
timlil1f[ ... '(1'1'1.1' tu (ltl]('r pt'l'iucis as \\'(,11. ('I'. my (ullpuhli ... I)('d) papt'r read at the Colloquium on
Byzantine Rooks and Booklllen ht'ld at DUIIIIJal'ton Oab in 1971. For a stimulating
rec'ollstrudion of the taste's of readers of (mostj~· middle and low sty Ie) ('hronielt's. d.
HI';\,(:EH (as ill n. 7:?). :?57~-:?i8. Cf. aho the pereE'ptin remark>' D.\· E. T'ATL\(:E,-\.,\. alas.
('ouchE'd in the high ~tyle of the Alluale.' (34.2 [1Hi!l] 264--27:,)), on jE'wls of ('ultUI'E' and
literature in 8th~ 11 th CC'.

222
 The ~ow Level' Saint's Life
in the Early Byzantine World *
ROBERT BROWNING

B y 'low level' saints' lives I mean the lives of saints whose activity is largely
among relatively humble people, who have few connections with persons of
power and influence, and who are not - except quite incidentally - members
of the hierarchy of the Church. Such lives are generally composed in an unpretentious
style and in a language little affected by c1assicising tendencies. Like all early
Byzantine hagiographical texts they may be rewritten in more c1assicising and
prestigious style in the tenth century. But both the language and style and the
content of the original versions suggest that these lives were composed with a
popular audience or readership in mind. The common features which they display
- and they are many - may throw some light on how the Byzantine 'man in the
street' saw holy men and what he expected of them.
It must be emphasised that we are not concerned with the saint as he was but
with the saint as he was perceived by the average members of the society within
which he passed his life. The mode of his perception is detennined in large measure
by a whole structure of shared concepts which fonned part of the common man's
view of the world.
There are 'low level' lives of 'dead' saints like St George or Sts Kosmas and
Damianos. But they present problems of their own. We shall concentrate in the
present paper on lives of saints written by persons who knew them in their lives or
who had direct access of those who knew them. The stylites, from Symeon in the
fifth century to Lazaros the Galesiote in the first half of the eleventh are typical
examples. 1 So too are such saints as Theodore of Sykeon in the late sixth and early
seventh century,2 Peter of Atroa in the early ninth century,3 Luke of Stiris4 and
In addition to the sources and secondary studies referred to in connection with pa)1icuIar
passages, I have derived much help from the following recent publications: Evelyne
Patiagean, 'Ancienne hagiographie byzantine ct histoire sociale',Annales ESC (1968),106-
24; P.R.L. Brown, 'The Rise and Function of the Holy Man in Late Antiquity', JRS 61
(1971),80-101; A. Meredith, 'Asceticism - Christian and Greek',JThS 27 (1976), 313-22;
Julia Seiber, Early Byzantine Urban Saints [British Archaeological Reports, Supplementary
Series 37) (London 1977).
1. Cf. H. D eleh aye, Les saints stylites (SubsHag 14 (1923)); P. van den Ven, ed., La Vie
ancienne de S. Symeon Stylite Ie Jeune (Brussels 1962).
2. Cf. A.-J. Festugiere, Vie de ,Theodore de Sykl6n (Brussels 1970).
3. Cf. V. Laurent, Vie de S. Pie"e d'Atroa (Brussels 1956).
4. Cf. Vita Lucae, PG 111. 441-80.

223
 ROBERT BROWNING

Nikon Metanoeite in the tenth. 5 It springs to the eye that none of these saints lived
in or was connected with Constantinople or with any ofthe other great cities of the
empire. The 'popular' saint belongs to a provincial, and largely to a rural, environ-
ment.

Freedom from ordinary needs

What are the common features which the lives of all or most of these saints
display? First and foremost, the holy man is one who withdraws from society, its
bonds, its constraints, and its rewards. These he sees as irrelevant to his overriding
purpose of approaching more closely to God. So he leaves the city of his birth and
takes flight to the desert or the mountains; or he isolates himself upon a pillar or in
a cave. He abandons the city for the desert. Yet the city follows him. His powers
cause a structured society to grow up around him. Often it at first takes the form of
a monastic community of men who share his desire to withdraw from the world.
But soon a large settlement springs up around the holy man, with churches, accom-
modation for pilgrims, permanent habitations, and the rest, sometimes surrounded
by a wall. Examples are Qal'at Sem'an, north-west of Aleppo, and the complex of
structures on the Admirable Mountain west of Antioch. Occasionally the holy man
is incorporated into the hierarchy of the Church, but in general against his will and
often only temporarily. An example is Theodore of Sykeon in Bithynia. For though
the holy man may live in society, he is not of it, and he constantly distances himself
from the manner of life of ordinary people, even those among whom he dwells.
There is always a certain tension between the purposes of the saint and those of his
followers, since in his person two different worlds overlap.
The holy man does not share the needs of ordinary people. Apart from the rejec-
tion of sexual relations, which is common to all ascetics, he has no need of many
other things which are essential to the average man and the object of most of his
striving. First of all food. Symeon Stylites was reported neither to eat nor to drink.
Six centuries later and a thousand miles distant, Lazaros the Galesiote ate only once
a week. Symeon the Younger tried to dispense with food altogether and fmally
succeeded in doing so, save that an angel brought him a kind of rice pudding every
Sunday.6 It is probably futile to speCUlate whether this last item of information is
a rationalisation by Symeon's biographer or a final proof of the saint's holiness. If
the holy man does eat - and most of them, most of the time, do - then he avoids
cooked food, which is the mark of civilisation, of assertion of human control over
a potentially hostile nature. 7 It is almost as if the hagiographers had read Levi-
Strauss. Symeon the Younger from his earliest childhood would eat nothing

5. cr. S. Lambros, 'Ho Bios tou Nikonos tou Metanoeite', Neos Hellenomnemon 3 (1906),
131-228.
6. Vie de S. Symeon Ie jeune 256 (van den Yen, 222).
7: Vie de S. Symeon Ie Jeune 3 (van den Yen, 5).

224
THE 'LOW LEVEL' SAINT'S LIFE IN THE EARLY BYZANTINE WORLD

Anonymous stylite. Stone relief.

Hama Museum, Syria, c. 5th-6th century.
Photo: C. Mango.

225
 ROBERT BROWNING

'prepared'. Peter of Atroa ate only raw vegetables - no meat, cheese, bread, oil or
wine, the staples of ordinary diet. 8
Sleep was another need of the ordinary man with which the saint could dispense.
All of them often pass long periods without sleep, in prayer or in psalmody. The
monks in the monastery which grew up around the holy man would wake in the
night to hear him singing or praying without remission. The stylites did not lie
down, except in extreme old age. 9 If they slept at all, they slept on their feet,
perhaps leaning on the rail which surrounded their lofty platform.
The saint had no need for shelter. He had no house, but lived in the open or in
some natural cavity such as a cave or a hollow tree. Some of the stylites seem to
have had no shelter at all; they could be seen in all weathers standing in the open, a
visible testimony of their special status. Others had a kind of awning or hut con-
structed over their platform, with a window through which they could look and
communicate with their followers or with visitors. to Alypios in the sixth century at
first had a tent on top of his pillar, but later demolished it and spent the rest of his
very long life in the openY None of the stylites used fire, an essential ameniiy in
the life of ordinary men.
Sitnilarly the saints have no need of clothing. Not merely do they usually have
little and poor clothing, usually only a single garment, often a hair shirt. But they
may pass long periods of complete nudity. Symeon the Younger as a child often
gave away his only garment to a poor companion. t2 In later life he once spent eight
months naked on top of his pillar.!3 If the holy man did have clothing, it was not
as that of other men. It was often a trichinon sticharion or a sakkos. And it was im-
bued with the superhuman powers of its wearer. The sweat-rag of the saint or a few
threads from the fringe of his stich arion could be the instrument vf miraculous
cures at a distance.
Lastly, the saint did not, as ordinary men do, require the support of a family.
His path to holiness regularly begins with his running away from his family in child-
hood, and often disappearing for a long period. Luke of Styris is picked up by the
Byzantine police as an absconding tninor the first time he runs away, given a thrash-
ing, and returned to his mother. 14 But later he succeeds in disappearing without
trace, though in fact he had not gone far away. Symeon the Younger, as soon as he
was baptised at the age of two, began to repeat constantly the words, 'I have a
father and I have not a father; I have a mother and I have not a mother'. 1 5 (What
8. Vie de S. Pierre d'Atroa 11 (Laurent, 97). On abstinence from 'prepared' food cf.
Athanasius, Vita Antonii 7.
9. Cf. Delehaye, Les saints stylites, elv.
10. Cf. Delehaye, op.cit., elvi-cbc.
11. Vita S. Alypii 15 (DeJehaye, op.cit., 159).
12. Vie de Symeon Ie Jeune 30 (van den Yen, 30).
13. Vie de Symeon Ie Jeune 77 (van den Yen, 65).
14. Vita S. Lucae lunioris, PG 111. 445n.
15. Vie de S. Symeon Ie Jeune 5 (van den Yen, 7).

226
 THE 'LOW LEVEL' SAINT'S LIFE IN THE EARLY BYZANTINE WORLD

with this and his finickyness about food, he must have been an exasperating child!)
Luke the Sty lite came of a family that was comfortably off. But he took no part in
administering or conserving the family property, and gave away as much of it as he
could to the poor. 16 The holy man does not depend on his family to define his
position in society. He is alone in his confrontation with metaphysical forces.
However there is a curious ambivalence in the attitude of many of the hagio-
graphers, which suggests a certain mistrust of the complete loner. Once they have
established their special status of holiness many saints are reported to have main-
tained friendly and even close links with their female relations. Symeon the Younger
was in constant contact with his mother, who acted as a kind of public relations
officer for him. Finally she attained a somewhat bland and nebulous samthood.
Alypios in Paphlagonia was in close contact with his mother until her death. 17
Theodore of Sykeon, once his holiness was established, maintained touch with his
mother and his grandmother. It was the latter who saw to it that he took the neces-
sarY minimum of food during his exploits of asceticism. 18 It is striking that no saint
has much to do with his father, who is usually rather vaguely outlined and often
dies when the future saint is still a child. Have we here a reflection of Oedipal
jealousy? Or of the tension between the generations which marked a society orga-
nised for the transmission of wealth rather than the creation of wealth, to which
many anecdotes, particularly in the Phi/oge/os, bear witness? 19

Extraordinary powers
The counterpart of the saint's freedom from ordinarY needs is his possession of
extraordinarY powers. The very nature of his life calls for extreme powers of en-
durance and resistance to fatigue and hunger. He nearly always has a precocious
development. Indeed few saints seem to have had any real childhood. Their intellec-
tual and spiritual powers are not those of children. Theodore of Sykeon learnt the
whole Psalter by heart 'in a few days'. 20 Other future saints practise the extremes of
asceticism at an age when most children are preoccupied with play. All holy men
possess the power of prophecy, which they may exercise on an imperial as well as
on a local scale. Theodore of Sykeon foretells the death of the emperor Maurice. 21
Akin to prophecy is clairvoyance, both as the ability to know of distant events and
as that of reading the thoughts and feelings of men. Both types of clairvoyance are
too common in the saints' lives under discussion to need illustration.
Perhaps the commonest manifestation of holiness is the healing power of the
saint. He deals with every physical and mental affliction from constipation to
leprosy, cancer and gangrene. Usually he works by touch, often accompanied by
16. Vita S. Lucae Stylitae 5, 7 (Delehaye, Les saints stylites, 200,202).
17. Vita S. Alypii 15,19 (Delehaye, op.cit., 159'{)0, 162-3).
18. Vie de Theodore de Sykecm 32, 33.
19. Cf. Phi/ogelos, ed. A. Eberhard (Berlin 1869), Nos. 13, 24, 38.
20. Vie de Theodore de Sykeon 13.
21. ibid., 119.

227
 ROBERT BROWNING

prayer. Sometimes prayer alone suffices. Sometimes the holy man works his cure at
a distance by sending to the patient something which has been in contact with his
person. Occasionally he makes a miraculous personal appearance at a distance. In
the accounts of these healings illness is as a rule treated as of demonic origin; but
there are many exceptions to this rule. Often the healing offered by the saint is con-
trasted with that of orthodox doctors. It is instantaneous, while theirs is slow, it is
painless while theirs involves much discomfort or the agony of surgery without an
anaesthetic, and so on. But sometimes there seems to be a tacit demarcation agree-
ment between the holy man and the doctor, who send one another suitable patients.
A case in point is the Life of Theodore of Sykeon. 22 It is striking how many of the
patients treated by holy men appear to be suffering from some kind of hysterical
paralysis. Were such conditions really so common in the early Byzantine world?
And if so, why?
The extreme case of healing power is the resurrection of the dead. Many holy
men do on occasion restore the dead to life, for example Symeon the Younger23
or Peter of Atroa. 24 But those restored to life are always special cases - children,
persons killed by a sudden accident, persons dying in mortal sin, and so forth. The
saint's healing powers are always available to the sick, but only very rarely to the
dead. For death is a normal part ofHfe and not the work of demons, as is disease. It
belongs to the natural order.
The holy man's life is a continual struggle with demons. Ordinary men and
women knew that they were surrounded by demons. But the saint could see and
hear them, and sometimes make them briefly visible and audible to others. Often he
engages in long dialogues with demons. His victory is never in doubt. Presumably if
he failed to overcome a demon he would not be a truly holy man.
The forces of nature are subject to him. He regularly diverts storms, arrests both
floods and droughts, causes swarms of locusts to depart or to die. He has power
over wild beasts too, and engages in ever-victorious confrontations with everything
from snakes to lions.
He regularly has visions or hallucinations, which are quite different from the
dreams of ordinary men. Symeon the Younger has a series of apocalyptic visions
whil~ wide awake. 25 Peter of Atroa sees both Christ and the Virgin Mary.26 Some-
times the visions are of events in far distant places. In addition to the ability to see
things at a distance, the saint sometimes has the power of bilocation. Symeon the
Younger, who never quitted his pillar, was often seen and heard in other places. 27
Peter of Atroa made similar visits at a distance?8 It is sometimes difficult to distin-
22. ibid., 145.
23. Vie de S. Symeon Ie Jeune 129 (van den Yen, 117 -22).
24. Vie de S. Pierre d'Atroa 22 (Laurent, 119).
25. Vie de S. Symeon Ie Jeune 29, 66 (van den Yen, 29, 57-9).
26. Vie de S. Pierre d'Atroa 4 (Laurent, 75).
27. Vie de S. Symeon Ie Jeune 91 (van den Yen, 71).
28. Vie de S. Pierre d'Atroa 41 (Laurent, 155-7).

228
 THE 'LOW LEVEL' SAINT'S LIFE IN THE EARLY BYZANTINE WORLD

guish between the holy man's perception of himself as visiting distant places and
other people's perception of him elsewhere than where he was known to be. The
saint could sometimes become invisible. Peter of Atroa several times evades his
iconoclast enemies by passing unseen through their midst. 29 Sometimes he posseses
indestructibility which goes beyond mere endurance of extremes gf weather and
fatigue. A stylite in Asia Minor entered a ftre which was burning in a fIre-place in
order to refute heretics. 30 The third Symeon the Stylite - whose date is uncertain
- was struck by lightning without suffering any ill effects. 31 And no stylite ever
fell from his pillar, overcome by dizziness or fatigue.
Miraculous provision of food is one of the exploits regularly recurring in these
saints' lives. Symeon the Younger mls the granary of his monastery when it is
empty and the monks are in despair. 32 Luke the Stylite provides a miraculous catch
for the ftshermen of the Bosphorus. 33
The saint is always aware of the powers which distinguish him from other men,
and punishes those who question either their existence or their origin. The punish-
ment is often surprisingly severe. Paralysis or serious illness are the most usual
forms, and the saint grants relief only when the sceptic repents of his doubts. Occa-
sionally he is given no opportunity of repentance. Symeon the Younger punished a
scholastikos (lawyer or merely intellectual?) of Antioch who cast doubt on the
divine origin of his powers by causing him to suffer a fatal seizure in a public place
in the company of two illoustrioi (holders of the highest offices of state).34
Lastly, the holy man enjoys longevity, in spite of his exhausting life style. Daniel
the Stylite lived to 84, Lazaros the Galesiote to 86, Theodoulos the Stylite to 90.
Luke the Stylite was more than 100 years old when he died. Michael, a Stylite in
Mesopotamia, was 105. Theodore of Edessa knew a stylite reI>uted to have spent
ninety-five years on his column. Alypios lived to the age of 120?5
Several different strands can be traced in the life of the 'low-level' saint as per-
ceived by his followers and contemporaries. The most obvious is the imitation of
the Gospel story. The healings, the miraculous provision of food, the restoration to
life of the dead, the dialogues with demons, the apocalyptic visions, perhaps the
close maternal links and slight role of the father, all recall the life of Christ, which
provided a model or framework within which the life of the holy man could be
organised. But there are many recurring motifs and structural features in these
saint's lives which have no Gospel parallels. The most important will be reserved for
separate discussion. In the meantime it is evident that invisibility, invulnerability,
and longevity fall outside the Gospel framework and recall rather the pagan aretalogy.
29. ibid., 66 (Laurent, 195).
30. Cf. F. Nau, ROehr 7 (1902), 605.
31. Joannis Moschi, Pratum spirituale 57.
32. Vie de S. Symeon Ie Jeune 122 (van den Yen, 100·3).
33. Vita S. Lucae Stylitae 16 (Delehaye, Les saints stylites, 212).
34. Vie de S. Symeon Ie Jeune 224 (van den Ven, 194-6).
35. Delehaye, op.cit., passim.

229
 ROBERT BROWNING

The precocious development of the future saint also has pagan precedents - the
infant Herakles strangling the snakes in his cradle and the infant Achilles in Pindar
(Nem. iii. 43ft) are two parallels which spring to mind. Bilocation and clairvoyance
are typical marks of the shaman, who, like the saint, acquires his powers by with-
drawal from society and returns to it in a new status. 36 The cultural amalgam of
late antiquity and the early Byzantine period provided more than enough models
and patterns out of which the humble and oppressed could construct a fantasy of
life rising above the physical and social constraints to which they were subject, a
fantasy which perhaps enabled them the more easily to endure the harshness and
drabness of their everyday life. Perhaps for some of his contemporaries the holy
man played a role not unlike that of Superman, both in the Nietzschean and in the
Hollywood sense of the word, reminding them, if only in an exaggerated and some-
times superficial way, of the range and power which, if God will, human ability can
attain.

The path to sanctity

How did the holy man establish his sainthood and gain his superhuman powers?
Withdrawal from society is always involved. Usually it is physical withdrawal to the
desert or the mountains, or to the aerial regions. Sometimes it is social withdrawal,
as when the saint feigns madness and is so freed from the normal restraints and
tabus. 37 Self-imposed hardships are also both a path to holiness and a demonstration
of it, and are to be seen as a form of withdrawal; the saint seeks out that which
other men avoid. Many saints begin by wlllpping cords tightly round their bodies
until they cut into their flesh. Others voluntarily accept the chains which awaited a
humble member of society if he fell into debt or offended those in authority.
Theodore of Sykeon wore heavy iron chains and passed a long period in an iron
cage. 38 Symeon the Elder loaded his body with chains. 39 So did Luke the Stylite
and Lazaros the Galesiote. 4o These apparently perverse choices of what most men
avoid were unquestioningly accepted as evidence of holiness.
36. Cf. E.R. Dodds, The Greeks and the I"ational (Berkeley 1951), 140-3, 160 (bibliography).
37. Cf. E. Benz, 'Heilige Narrheit', Kyrios 3 (1938). 1-55; J. Saward, 'The Fool for Christ's
Sake in Monasticism, East an:l West', Theology and Prayer, ed. A.M. Allchin fStudies-
Supplementary to Sobornost 3) (London 1975); L. Ryden, Das Leben des hi. Na"en
Symeon von Leantios von Neapolis (Uppsala 1963); S. Murray, A Study of the Life of
Andreas, the Fool for the Sake of Christ (Barna-Leipzig 1910).
38. Vie de Theodore de Syketm 28.
39. Theodoret, Historia religiosa 26.10.
40. Vita S. Lucae Stylitae 5 (Delehaye, Les saints stylites, 200); Delehaye, op.cit., ex.

OPPOSITE
Irons worn by an ascetic (siderophoros), displayed at
Xenophontos, Athos
Photo: Robert Byron (by courtesy of the Courtauld Institute
of Art)

230
231
 ROBERT BROWNING

It is striking how often the acquisition of sanctity involves a kind of substitute

death. Often the young man passes months or years in an enclosed or subterranean
space resembling a grave. Symeon the Stylite lived in an abandoned water-tank,
Alypios in an old tomb,41 St John of Rila in tenth-century Bulgaria lived for many
years in a hollow oak-tree which recalled a coffin.42 Theodore of Sykeon excavated
a spelaion skoteinon under the altar of a church and lived there for some months.
Later he hollowed out another spelaion under a rock in a remote place and remained
there, unknown to all but one disciple, for two years. 43 When his grandmother and
mother found him, his head was covered in dried blood (ichor) and full of worms,
his bones were laid bare, and none could stand near him because of the stench. In
the words of his biographer, 'he looked like a corpse'. The point could not be made
more clearly. Theodore was as it were a man risen from the dead. His holiness
needed no further proof, and the local bishop at once ordained him reader, sub-
deacon, deacon, and priest. The evil smell resulting from the holy man's self-morti-
fication is mentioned again and again in the lives as evidence that he has passed
beyond the normal human state and gained superhuman powers. It clearly symbolises
the smell of death. But the stench of holiness soon turns into its apparent opposite,
the fragrance of holiness. Once his sainthood is established the saint's body and all
that is in contact with it distil an exquisite and unearthly perfume, like that which
flows from the relics of dead saints and martyrs.
So the saint is perceived not merely as one who enjoys all the powers which the
humble do not possess - a kind of Superman. He is also seen as one who has passed
beyond the human condition although he continues to sojourn among his fellow
men. He is beyond sin and beyond the power of evil, and has regained the primeval
innocence of man before the fall. He is a living reminder of the victory of life over
death and of good over evil, and one that is local, visible, and tangible. He has died
and been born again. But his rebirth is the beginning and not the end of his earthly
career. In this respect the pattern of the saint's life is very different from that of the
life of Christ.
Not all 'low-level' saints display clearly all the characteristics that have been enu-
merated. But all show some of them, and many show all of them.

Social and political status

Human weakness and vanity play their part in the perception of the saint.
Naturally he is a man without possessions. But his poverty is usually chosen, not
inherited. His parents are often respectable, sometimes substantial citizens. The
parents of Luke the Stylite owned a considerable estate - perhaps including a
stratiotikon ktema, since his father was a soldier. As a young man Luke distributed

41. Theodoret, Historiil reiigiosa 26.6.

42. J. Ivanov, 'Zhitiia na sv. Ivana Rilskii: GodSof 31.13 (1936), 28.
43. Vie de TheodoredeSykeon 16, 19,20·1.

232
 THE 'LOW LEVEL' SAINT'S LIFE IN THE EARLY BYZANTINE WORLD

his property to the poor, then worked for two years as a sWineherd. 44 Theodore of
Sykeon, though the son of a barmaid and prostitute, was somehow related to lead-
ing members of his local community - protiktores, as they are called4s - and his
unknown father was reputed to be a person of more than local importance. 46 This
emphasis on the economic and social position which the saint abandons in his
search for holiness is partly ethical; poverty is not a moral category unless it is
voluntary. But it probably also contains an element of social snobbery. The man
who embodies so many of the dreams and aspirations of the group must not be in
origin a nobody, a landless, homeless, rootless outsider.
Though essentially a local figure, linked closely with a particular region and
often with a particular spot, the saint sometimes displays his powers at an imperial
level. He prophesies the death or accession of emperors, warns them, is consulted
by them. This is not all fantasy. Daniel the Stylite, who is perhaps not quite a low-
level saint, was really consulted by several emperors - presumably through an inter-
preter, since the saint never troubled to learn a word of Greek. Theodore ofSykeon,
living and working in a region through which a main military road passed, may well
have actually visited Constantinople and become acquainted with persons of power
and influence including three emperors; after all, he was a bishop, however
reluctant. But in most cases the imperial and metropolitan connections of the holy
man must have been a reflection in fantasy of the common man's eagerness to have
influence with the highest authority, an eagerness which cohabited uneasily with an
anxiety to keep out of the way of authority in general.

The counter-hero of the dispossessed

The popular saint appears as the direct antithesis of the ideal citizen of classical
antiquity. He is a rural, not an urban figure. He lacks military prowess, he has no
public career, he is equally devoid of physical strength and beauty and of intellec-
tual distinction. Yet it may be unwise to take him as evidence for a complete
abandonment of traditional ideals. Perhaps he displays a kind of reflection of the
ideal man of classical urban culture, but a reflection in a distorting mirror. His
prowess is in the unceasing war against demons and the powers of evil. He follows a
career which takes him out of this world and into the next by a regular progression
of triumphs over human constraints. He possesses moral strength and spiritual
beauty, often manifested in a kind of glowing aura surrounding his person. And
though he may not read books, he has the power to read men's minds and hearts -
1ioratikotes. 47 He is perhaps to be seen as the counter-hero of the dispossessed and
)f those to whom the high urban culture of Late Antiquity had nothing to offer.

44. Vita S. Lucae Stylitae 5 (Delehaye, 206).

45. Vie de Theodore de SykefJfi 25.
46. ibid., 3; he was a forrmer circus acrobat who had become an imperial apocrisiarius under
Justinian. Was he perhaps a protege of Theodora?
47. Cf. A.-J. Festugiere, Vie de Theodore de Sykeon, ii. 218.

233
This page intentionally left blank
 Imperial Panegyric:
Rhetoric and Reality

George T. Dennis

Above all else the people we call Byzantines prized orthodoxy, correct doctrine, correct
thinking, which they-and we-generally think of in a theological context. But there
was another kind of orthodoxy, political orthodoxy, as some have tem1ed it. I This in-
volves correct thinking about the civil and instirutional life of the empire, the whole
imperial ideology. These orthodoxies, inseparable and sacred as they were, found expres-
sion in ways that were also sacred for the Byzantines. Theological orthodoxy found its
expression largely in the divine lirurgy and was there made known to the faithful. Politi-
cal orthodoxy, in rum, was articulated by a literary elite and communicated to the citi-
zens of the empire through rhetoric.
The use of rhetoric as an instrument of politics has ancient roots. In the fourth cenrury
B.C. Isokrates endeavored to influence Athenian foreign policy in favor of Philip of Mac-
edonia, whom he compared to Herakles, the benefactor of mankind, just as Byzantine
orators, a thousand years later, would compare their rulers to Christ, the lover of man-
kind, philatlthropos. 2 In fact, many of the rhetorical cliches formulated by Isokrates would
be heard over and over again throughout the Byzantine period. The imperial govern-
ment, from Diocletian on, couched its decrees in rhetorical style to render them more
solemn and memorable. 3 To its fmal days the Byzantine administration continued this
practice, the most notable example being the preambles to imperial chrysobulls. 4 In its
most prosperous days, as well as in its most penurious, in victory and in defeat, the
Byzantine upper classes never gave up the srudy and practice of classical rhetoric. The
basic handbooks by Hermogenes and Aphthonios were copied and commented on
throughout the Byzantine cenruries. s The medieval dictionary, the Suda, in its entry on
Hermogenes, notes that "everyone has a copy" of his rhetorical manual. 6 As late as the
fifteenth century, John Chortasmenos prefaced an encomium of Manuel II with the
declaration that he was explicitly following Herrnogenes, and that his oration should be

, See H.-G. Beck. Das byz"ntinischeJahrtausend (Muruch. 1978),87-108.

, H. Hunger, Die hochsprachhche profane Literatur der Byzantiner, 2 vols. (Muruch, 1978), vol. I, 72.
J Hunger. u"teratuT, I, 71.

-' H. Hunger, Prooimion. Elemente deT byzantinischen Kaisen"dee in den Arengen deT Urkunden (Vienna, 1964).
; Hunger, Literalur, I, 76.
'Suidae Lexicon, ed. A. Adler, 5 vols. (Leipzig, 1928-38; repro 1971), E 3046.

235
 [132] Court Intellectuals and Rhetoric

judged, not on its content, but on how closely it observed the rules of rhetoric.' While
fonnal epideictic orations go back several centuries, it was Menander the Rhetorician,
in about A.D. 300, who was credited with setting down the rules governing them. s
The fourth and fifth centuries saw a great flowering of Greek rhetoric in which both
pagans and Christians participated. Names that immediately come to mind are Libanios,
Themistios, Eusebios of Caesarea, Gregory of Nazianzos, Prokopios of Gaza. It was
Eusebios, in his orations on Constantine the Great, who christianized the imperial ideol-
ogy and articulated the political orthodoxy that would prevail until the death of the last
Constantine. The speeches and letters of Libanios served as models of fme writing for
rhetoricians, such as Psellos, and for emperors, such as Manuel II. Gregory of Nazianzos
and others added biblical and Christian motifS and topo; to those of classical antiquity.
Even during the dark ages that followed, it is obvious that men continued the study
and, presumably, the practice of rhetoric, for when, in the ninth century, the sources
become more abundant, we fmd them engaged in their literary activities with renewed
vigor. From then until the end of the empire, the teaching and practice of rhetoric
underwent very few changes. Teachers of rhetoric attracted students; they taught them
to imitate classical models, and, in due time, some, such as Psellos, Italikos, and Basilakes,
were themselves held up as models! Byzantium was never without rhetoricians.
Who were these rhetoricians, and why was rhetoric so very important to them? First
of all, they were not very numerous. They fonned a special class, a literary elite, which
included emperors and empresses, ordinary laymen, secular clergy, bishops, and monks.
The bond that held them together was rhetoric, the "communion of letters" as they
tenned it.' o For one thing, it distinguished them from the ignorant tribes outside their
borders. Manuel II felt obliged to continue his literary activity to set an example for his
subjects, "so that as they mingle so much with barbarians they might not become com-
pletely barbarized."" And inside, it separated them, as Nikephoros Gregoras claimed,
from ditch diggers and tavern keepers. '2 By maintaining classical rhetoric, with all its
peculiarities, this class experienced continuiry with the greamess of Rome and with
Greek culture. In a world of change it gave security, stability, and meaning to their lives.
In addition, proficiency in the art could lead to advancement and material rewards.

7Johannt5 Chorwmenos (Ul. 1370-<a. 1436/37). Briefe, Gedichte und kleine Schrijttn, ed. H. Hunger (Vienna,
1969),225.
• Mmander Rhetor, ed. and traru. D. A. Russell and N. G. Wuson (Oxford, 1981); cf. Hunger, Uteralur, I, 88.
, C( P. Wendland, Commenlaria in Aristocelern Gra=, 1lI, 1 (Berlin, 1901), appelldix; J. Duffy, "Some Observa-
tiOIlS on a Byzantine Author and Title List:' Sixtanth BSCAbstr (Baltimore, 1990), 81.
011 Byzantine rhetoric one must begin with Hunger, Uteralu" vol. I, 63-196, with detailed bibliography,
189-96. Also see G. Kusus,'Studies in Byzantine RktDriJ: (Thessaioruca, 1973); S.Averintsev, "Vizanrijskaja ritor-
ilea," ill Problemy literatumoj teorii v Vizantii i LAtinskom srednevekovje, ed. M. Gasparov (Moscow, 1986), 19-90; G.
Kennedy, Cr<ek Rhetoric under Christian Emperors (Princeton, 1983), although this is less than satisuctory for the
Byzantioe period; Averil Cameron, Christianity and the Rktorie oj Emp;"': Tht Development oj Christian Discourse
(Berkeley, 1991).
I. E.g., Dimetrius Cydones Correspondan«, ed. R. J. Loenenz, 2 voh., ST 186, 208 (Varian City, 1956, 1960),
vol. II, ep. 270, p. 188,1. 47.
11 The Letters oJManuel II PaUuologus, ed. G. T. Dennis (Washington, D.C., 1977), ep. 52, p. 150.

12 S. Be2delci, "Nicephori Gregorae epistulae XC:' Ephcnnis Daco,omana 2 (1924), 239-377; <p. 7, p. 340;

ep. 50, p. 250.

236
 Imperial Panegyric [133 1

Most rhetorical composlOons were intended to be read aloud in a circle of one's

friends or before the emperor and his court. Such readings were likened to a perfor-
mance in the theater and were critically evaluated. Orations, which by their nature were
meant to be delivered aloud, might be composed on the armiversary of a battle or some
other significant event, for example, the reconciliation of Leo VI with his father. 13 Some
pleaded certain causes; others were occasioned by the death of a prominent person; still
others were meant to praise an individual, especially the emperor. It is with this last that
this paper is concerned, the imperial encomium or panegyric, the basilikos logos. I limit
the discussion here to compositions in prose, from the ninth to the fifteenth centuries,
simply noting that much panegyric was in verse or in ceremonial acclamations.
Speeches in praise of a ruler go far back in human history, but the classical encomium
took shape about the fifth century B.C. and by the fourth century A.D. had attained its
definitive form, as codified by Menander and others. It shared the characteristics of all
Byzantine rhetoric, that is, it tended to omit concrete details, especially names of persons
and places. Psellos, for example, states that he does not need to give the name of a certain
rebel, since he is addressing people who know what has been happening. 14 Reference is
made to the Persians, Scythians, and others, rather than to the Turks and the Pechenegs.
Some panegyrics were simply stylistic exercises and never pronounced in public.
The basic format for an official encomium of the emperor, as laid out in the rhetorical
man~als, was fairly simple. The orator was to recall the emperor's place of origin, ~s
birth, his parents, his education and physical appearance, his deeds in peace and war; he
was to pomay him as a shining example of the virtues, especially wisdom, courage,
justice, and moderation. He should stress his philanthropy and piety. Within this frame-
work, of course, a great many variations were possible. One or more of the standard
topics could be omitted. In one oration, for example, Psellos does not mention Mono-
machos' £amily and fatherland, "even though the rules of the art call for it." 15 Other
topics could be dealt with in greater detail than expected, while others could be more
nuanced. The orator could include praise for the empress, other members of the imperial
family, the patriarch, or others. While the Byzantines expected a panegyric to have the
classical characteristics of the genre, they also delighted in little literary surprises, a play
on words, a new twist on an old proverb, a subtle allusion to scripture or classical litera-
rure. Although Byzantine rhetoricians placed great importance on observing the stan-
dard format, their orations are more varied than one might expect.
What other elements characterized Byzantine imperial panegyrics? The modern
reader, perhaps, is most struck by the extreme, almost sickening, flattery in these ora-

\J This was cornmemonted on 20 July 901 or 902: Arethae archiepiscopi Cae",riensis Smpta minora, ed. L. G.

Westerink. vol. II (Leipzig, 1972), or. 65; also R. Jenkiru. B. LlOurdas, and C. Mango, "Nine Oraool15 of Are-
thas from Cod. Marc. gr. 524," HZ 47 (1954), 1-40, or. 9.
" Michael Psellus Orationes Panegyricae, ed. G. T. Dennis (Sruttgan-Leipzig, 1994), 4, 293.
15 Or. Panegyn'", 2,451-52. In addition to works on chetoric, see L. Previale, "Teoria e prassi del panegirico bi-

zanrino," Emerita 17 (1949), 72-·105, 340-66; C. Chamberlain, "The Theory and fucace of Impenal Panegync
in Michael Psetlos," Hyzantion 56 (1986), 16-27. Although focusing on the reign of only one emperor, the best
discmsion ofimpenal panegync is by P. Magdalino, The Empire of Manuel I Komnenos, 1143-1180 (Cambridge,
1993),413-70.

237
 [ 134 1 Court Intellectuals and Rhetoric

tions, which reminds one of the personality cult accorded to certain dictators in this
century. The ever recurring image of the emperor as the sun who brings warmth and
light to the whole world is a prime example. Psellos addressed Constantine Mono-
machos: "0 Sun Emperor. Could anyone fault me for invoking you by this title, which
fits you so perfectly? By your circle of virtues, your quickness of intellect, your natural
magnificence, and your radiant beauty, do you not shine upon the whole earth?" 16 The
empress, of course, is compared to the moon. According to Timarion, the souls in
Hades called Psellos the sun-king because of his excessive use of that metaphor. l7 Psellos
goes on: the emperor's speech is likened to that of Demosthenes, Isokrates, and Lysias.
He charms his audience as much as do the odes of Pin dar and the lyre ofSappho. And
for his deeds, it would take another Homer to recount them. The emperor is lavishly
praised for his benefactions, his piety, his martial valor, and a vast array of other virtues.
The orator marvels at his handsome features, impressive stature, and generally superhu-
man qualities.
One wonders how the person so honored could sit and listen to such unabashed
flattery without feeling some embarrassment. But, as far as we know, only one emperor,
Manuel II, reprimanded his panegyrist, Demetrios Chrysoloras, and ordered him to cur-
tail his extravagant praise. 18 His ancestor, Michael VIII, however, apparently found a
basilikos logos by George Akropolites so annoying that he simply walked out while the
orator was still speaking and went off to dinner.19
One could also question the sincerity of these orators. One could, for example, con-
trast Psellos' effusive praise of Constantine Monomachos in his panegyrics with his de-
cidedly less complimentary comments in his Chronography. Manuel Holobolos was muti-
lated, imprisoned, and exiled by order of Michael VIII, whose ecclesiastical policy he
openly detested, yet, in a Christmas oration, he praised him as an ocean of graces, a river
of gold, and as the sun rising in the East. 20 And in an Epiphany oration, probably in
1273, he lauded Michael, who had attained the throne by treachery and murder, as the
supreme model of virtue. 21 Perhaps, though, we should not be too harsh in judging
these speakers. They were, in one sense, 'Just doing their job." While so much of it
may seem to us shameless adulation at the expense of truth, we must keep in mind that
the speakers and their audience firmly believed that, whatever they might think about
the individual, the position of the emperor was sacred and worthy of all praise. In Mi-
chael's case, moreover, he was the New Constantine who had liberated the imperial city,
and this overshadowed his faults. Finally, today I think we can mute our criticism of
Byzantine oratory when we listen to; say, nominating speeches at a political convention
or those accompanying the conferral of an honorary degree.

" Or. Panegyril4 I, 3-7.

17Ed. R. Romano. TimanoM (Naples, 1974). 45; cf trans. with notes by B. Baldwin, Timanon (Dettoit,
1984),74.136-37.
18 I.erurs ofManUl!I PalaLo!ogus, pp. 137-39: epp. 46, L 10; 47, 11. 10-11; 48, 11. 11-15.
" Gtorgii Aaopolil4t Opera, ed. A. Heisenberg, 2 vols. (Leipzig, 1903); repro P. Wirth (Stuttgart, 1978), 89, vol.
I, p. 188.
20 M. Treu, Manu,lis Holoboli Oraliones. Programma VICtOria Gymnasium (potsdam, 1906), 30-50; (1907), 51-98.

esp.51-52.
21 L. Previale, "Un panegirico inedito per Michele VIII Paleologo;' BZ 42 (1943-49), 1-40, esp. 30.

238
 Imperial Panegyric [135 1

One cannot help but note that imperial panegyric, at least as it has come down to
us-and we have only a fraction of the orations actually composed-seems to flourish
as the empire declines. Only one oration is extant from the reign of Basil II, and the
largest number, by far, date from the reign of Manuel I; then there are quite a few right
up to the disaster of 1204. Again, there was a final flowering in the last days of the
empire, to Manuel II and his sons, John VIII and Constantine XI; these orations con-
tinue to heap extravagant praise upon the rulers and the imperial city with no hint of
trouble or impending catastrophe. 22 It was, one suspects, one way of closing one's eyes
to reality and living an illusion.
Another thing that impresses the reader is the extraordinary amount of liquid. Rivers
of blood are shed by the empire's enemies; John Komnenos returns from campaigning
in the East "swimming in a veritable ocean of barbarian blood." 23 The emperors shed
vast quantities of tears in pleading with God on behalf of their subjects; among these,
Leo VI seems to have been particularly productive. 24 But most abundant of all are the
immense buckets of sweat poured fonh by the emperors as they toil for their subjects in
both peace and war. We read of the sweat of vinue, bloody sweat, and rivers of sweat.2S
There are very few imperial panegyrics that do not dwell on imperial perspiration.
Another characteristic comes from Christianity. The emperor is eulogized as the ideal
Christian ruler, as God's representative on eanh. He is compared with biblical paradigms,
David and Solomon, and with the idealized Christian emperor Constantine. His piety
averts God's anger, gains victories over the foe, and, in the case of Leo VI, brings rain
to relieve a drought. 26 The emperor is commended for his zeal for onhodoxy, his philan-
thropy, his reverence for the patriarch, and his love of monks.
The panegyrics of the eleventh century that have come down to us generally praise the
peaceful accomplishments of the emperor and less so his martial exploits. Constantine
Monomachos, for example, was lauded for his suppon oflearning, his generosity toward
monasteries and philanthropic institutions, and his construction of aqueducts, fountains,
and gardens. When rebels are defeated or foreign foes subdued, it is usually not the
emperor himself who engages in combat. Toward the end of the century, though, as
emperors such as Romanos Diogenes and Isaac Komnenos personally took the field,
one notices a change in the rhetoric. Especially with Alexios Komnenos and his succes-
sors, the orators focus more on personal bravery and military strategy.27
Formal panegyrics were delivered in the presence of the emperor and his coun by
prominent rhetoricians, although it is not always clear how they were selected. The
encomiast of Leo VI, about 901-902, was Arethas, a deacon at the time, who seems to

" See Isidore on Manuel II and John VIII, ed. S. Lampros, Palaiologeia kai Peloponnesiaka, vol. III (Athens,
1926), 132-221; John Dokeianos on Constmtine XI, IbId., vol. I (Athens, 1912),221-35.
2J Nic<phori Basil"",e Orationes et <pistolae, ed. A. Garzya (Leipzig, 1984), or. 3, p. 49, I. 20.

" Arethae Scripta minora, or. 58, p. 7, II. 11-14; or. 59, p. 13,1. 17; or. 63, p. 38, 1. 15.
25 In the orations to Manuel It allusion is sometimes made to the "sweat of virtue," as in Hesiod, Opera et

Dies, 289, by Eustathios of Thessaloruca: Fontes rerum byzantinarum. Rhetorum saeculi .X1I orationes poiiticae, fases.
1-2, ed. W. Regel (St. Petersburg, 1892-1917; repro Leipzig, 1982 = FRB), or. I, p. 2, I. 30; or. 6, p. 124,
II. 24-26; bloody sweat, by Gregory of Antioch, FRB, or. 11, p. 201, 1. 7; rivers of sweat, by Eustathios, or. 6,
p. 124, I. 24; by M;chael Rhetor, FRB, or. 10, p. 169, I. 21.
26 Arethae Scripta minora, or. 61; Jenkins et aI., "Nine Orations;' or. 5.

27 Cf. Magdalino, Manuel I, 418-25.

239
 [ 136] Court Intellectuals and Rhetoric

have been the official palace orator, although no specific title is given. Psellos bore the
title of consul of the philosophers. But a contemporary of his, whose name we do not
know, had held the title of master of the rhetoricians; he "had sat upon the throne of
rhetoric and revealed the secrets of the art" to his students. 28 The same title was held by
a disciple ofPsellos, Theophylaktos, who lauded Alexios I in formal orations.29 One also
encounters the title "teacher of the rhetoricians." Praise of Michael Palaiologos was
offered by Manuel Holobolos, "rhetorician of the rhetoricians and teacher of the
teachers." 30
Orations to honor the emperor were given on a number of special occasions and
anniversaries. Fairly soon, however, Epiphany, the Feast of Lights, became the preferred
date for an annual solemn oration in praise of the emperor. The first such that we know
of was delivered by Arethas in honor of Leo VI, most probably on 6 January 901 and in
connection with a banquet." It then became a regular part of the court ceremonial for
Epiphany.32 It also provided an opportunity for the chosen rhetorician to present his
students to the emperor, as Michael Psellos did on an Epiphany berween 1045 and 1050.
"0 most sacred emperor, you see this holy and philosophical gathering; they are all the
fruits of my planting and have all drunk from my springs, and they have whetted their
tongues for rhetoric." 33 Some years later, his own student, Theophylaktos, concluded
his Epiphany oration in 1088 before Alexios I by inviting his pupils to continue the
praise of the emperor.J4 Euthyrnios Tornikes called upon his students to offer their praise
of Alexios III on Epiphany 1201.'5
What did these orators hope to gain from their lavish praise of the emperor? For one
thing, they wanted to be recognized as an important segment of Byzantine society to
which the emperor should listen. Panegyric orations provided an opportunity to lobby
the emperor in furtherance of their collective and individual interests. To this end they
sought public funding for the support of rhetoric and for themselves. "Be generous, 0
emperor," pleaded Theophylaktos to Alexios I, "as were the emperors of old, to the
sophistic, and with public funds support the sophistic tongue." 36 Some orators, such" as
Psellos, were quite blunt about seeking personal reward. 37 Others went about it more

28 P. Gautier, "Quelques lettres de Psellos inediles ou deja edilees," REB 44 (1986), 111-97, esp. 162, 166.
29 P. Gautier, "Le discours de Theophylact de Bulgarie :i l'aulocralor Alexis ler Comnene (6 janvier 1088);'
REB 20 (1962), 93-103.
'" Treu, Manuelis Holoboli, 78.
" Arethae Scripta minora, or. 63, pp. 35-38; Jenkins el aI., "Nine Orations;' pp. 34-36.
)2 This is described in Comt.lntin Porphyrogenete, 11 Ii"", des dremonies, ed. A. Vog<, 2 vok (paris, 1935-39),

34-35; voL I, pp. 130-36. After the processions and the liturgy, the emperor and the parriarch, with others, are
se.led .t table in the Hall of the Nineteen Couches and have. drink. A banquet follows, concluded by the em-
peror and the p.rriarch having another drink. After everyone leaves, they have still another drink: before de-
parting. The Epiphany oration by Areth.., just mentioned, apparendy took pIace at a banquet in the presence of
the emperor and patriarch. The ceremony of the prokypsis (manifestation of the emperor) developed later:
Pseudo-Kodinos, Traite des ()ffias, ed.1- Verpeaux (paris, 1966), 195-98.
'-' Or. Panegyrita 6,261-92.
,. Gautier, "Le discours de Theophylacte," 120.
lS 1- Darrouzes, "Les discours d'Euthyme Tomikes (12()()-1205)," REB 26 (1968), 49-121, esp. 56-57.

,. Theophylaktos 10 Alexis I: Gautier, "Le discours de Theophylacte," 119; Michael Rhetar 10 Manuel I,
FRB, or. 10, p. 167,11. 18-21.
" Or. Panegyrita 2, 798-825; Or. Panegyriul 12, 45-53; Manuel Straboromanos to Alexis I: P. Gautier, "Le dos-
sier d'un haut fonctionnaire d'Alexis ler Comnene, Manuel Straboromanos," REB 23 (1965), 168-204, esp. 191.

240
 Imperial Pallegyric [137 1
subtly by praising the literary and rhetorical abilities of the emperor and his interest in
those endeavors. Some hoped to advance their careers in state or church. Leo VI re-
warded Arethas by naming him archbishop of Caesarea. Theophylaktos gave a speech
in honor of Alexios I, on 6 January 1088, which marked the high point of his career,
after which he was appointed archbishop of Ohrid. 38 Michael ltalikos gave an oration
for Manuel I, in 1143, which clearly demonstrated his right to the throne and the en-
dorsement of him by the church; not long after he was named bishop ofPhilippopolis. 39
Finally, of course, the orators wanted applause. They performed before an audience
composed not only of the emperor and his courtiers, but also of their colleagues, their
intellectual and literary peers. Impressing them with their verbal wizardry was most
important for their reputation and, perhaps, for the recruiting of students.
It is very difficult to evaluate these panegyrics as rhetoric and as literature. These were
speeches, and, unless we hear them delivered orally and in the language in which they
were composed, accompanied by the appropriate gestures, we miss their full impact.
Were they any better or worse than the baroque orations of, say, the sixteenth and
seventeenth centuries given at royal and pontifical courts? We dislike much about their
speeches, but they clearly seem to have liked them. Certainly the emperors liked them; at
any rate, they kept inviting the speakers back to give yet another speech, as Constantine
Monomachos did at least seven times with Psellos.'" All we can do is judge them, not
by our standards, but by what we can understand of theirs. Did they follow the rhetorical
manuals, and did they deviate from them just enough to please their colleagues? As
mentioned above, Psellos, Italikos, Basilakes, and others were pointed out as models to
subsequent orators, and thus must have met the standards of their contemporaries.
Do these orations have any value as historical sources? Are they merely empty bom-
bast, or do they provide information about real people and events? Obviously, the ora-
tors did not intend to compose historical documents, as we know them, and we cannot
stretch that point too far. Still, it is legitimate to ask whether we can use these speeches
to learn more about the history of the times. Unlike a narrative source, which may have
been written some thirty or forty years after the events recorded therein, these orations
were generally delivered within a year of two of the events they described, and many
can be accurately dated. People in the audience had participated in those events, and the
orator, vihile permitting himself some embellishment, could not present a total fabrica-
tion. True, unpleasant events such as military defeats may be glossed over, but, in general,
these panegyrics are more reliable as historical sources than has been recognized. Some
authors, such as Choniates, even used their orations in compiling their histories. We can
ask what the orator adds to the infornlation given in the narrative sources, more facts,
or perhaps certain details or nuances we might otherwise miss.
In his book on Manuel Komnenos, P. Magdalino has made exemplary historical use
of the orations addressed to Manuel I, and there is no need to repeat that here. A few
other examples might also be illustrative. Some of the orations ofPsellos on Constantine
Monomachos provide historical infonnation not found in other sources. For example,

" G.uuer, "Le cliscours de Theophyldcte," 93.

" .Himel ilaliko5, Lettre5 el di5(ours, ed. P G.utier (Paris. 1972). 26 .
., Or. Panegyri'" 1-7.

241
 [ 138 I Court Intellectuals and Rhetoric

from a panegyric we learn that his father's home was near a large city to which he moved
and in which the future emperor was born' 1 Psellos, of course, does not name the city,
but simply refers to it as "the city of God," Theou polis, which his audience knew-and
we know-is a common epithet of Antioch in Syria. In his second panegyric, Psellos
infonru us that Constantine performed military service under Basil II, and then fell into
disfavor, therefore before 1025, when Basil died, thus setting a date for Constantine's
birth sometime before the year 1000. While the Chronography of Psellos tells us that
Constantine's father, Theodosios, had been imprisoned for conspiring against Basil, the
panegyric implies that he was executed or at least died in prison.
Within a year of his accession to the throne, Constantine had to deal with a major
rebellion in southern Italy. This is recounted in three narrative sources and in one ora-
tion.42 Among the forces making up the rebel army, Psellos notes in his speech, were
infantry and cavalry from Old Rome, as well as the large army of the East and, in particu-
lar, a Russian detachment, a detail not found e1sewhere. 43
Other sources mention Constantine's building activity, but provide few details.
Psellos, though, in his orations, informs us that this emperor had constructed, perhaps
also repaired, canals, aqueducts, and fountains in and around the capital. "He has di-
verted entire rivers, made them flow up and then flow down again."" "You have drawn
rivers together, elevated waters, and linked seas, so that you might assuage our drought
in the summer and give drink to your thirsty people .... The water is forced into a
constricted space, shoots up to a great height; it cools the air and seems to be a new sort
of rain coming down from the c1ouds."45 Allowing for some exaggeration, he is clearly
describing fairly elaborate hydraulic engineering projects.
The historian Attaleiates writes that Constantine liked to entertain his subjects by
displaying exotic animals. Psellos, in an oration delivered in the spring of 1043, speaks
of such animals as gifts from foreign potentates. An elephant, with its driver seated on
top, was paraded around the hippodrome, as was also a giraffe!' An elephant was trained
to kneel before the emperor's throne and touch its forehead to the ground' 7
A century later, in an encomium on John Komnenos, Nikephoros Basilakes provides
a detailed description of th,: emperor's itinerary and campaign in Asia Minor, which
corroborates and adds to the information in the narrative sources. The orator also de-
scribes the emperor's innovative way of protecting his stone-throwing machines at the
siege of Anazarba. 48 In tact, the orator has a number of interesting things to say about

41 Or. PanegyriliJ 6, 45-54.

42 Chronographia, 6,76-88; ed. E. Renauld, Michel Pstlws Chronographie, 2 vols. (paris, 1926-28), vol. 11, 1-18;
ed. S. lmpellizzeri, Imperalori di Bisanzio (Cronografia), 2 vols. (Milan, 1984), vol. 11, 8-22; Ioannis &ylilzae Synap·
sis Hisloriarum, ed. 1. Thurn (Berlin, 1973),427-28; MidJaelis Au"liotae Hisfaria, ed. 1. Bekker (Bonn, 1853),
18-20; Or. Panegyri<a 2, 715-49.
" Or. PanegyriliJ 2, 719 .
... Or. PanegyriliJ 1, 241-43 .
.. Or. PanegyriliJ 4, 406-14 .
.. Or. PanegyriliJ 1, 267-87.
" Or. PanegyriliJ 4, 156-65 .
.. Niaphori Basilacae Oraliones, 61; cf. loannis Cinnami 'Pilome rerum ab loanne 01 Alexio Comnenis gfStarum, ed. A.
Meineke (Bonn, 1836), hk. 1, par. 7, pp. 17-18; Niwae Choniatae HUfaria, ed. J. L. van Dieten, 2 pIS., CHID 11
(Berlin, 1975),1. 26; Mi£helitaliko5, no. 43, p. 254, 11. 8-12.

242
 Imperial Panegyric [139]

the stone-throwing machines.'" One hurled a very large number of stones in rapid suc-
cession, while another shot a stone the size of a wagon. Is this rhetorical exaggeration?
Not necessarily. Twenty years later it is almost certain that the Byzantines were making
use of the counterweight trebuchet, the heavy artillery of the middle ages; perhaps a
slightly earlier date could be postulated for its invention, quite possibly by the Byzan-
tines.'o A traction trebuchet transported by the Byzantine army in 1071, which required
a pulling crew of twelve hundred men, is reported to have fired a stone weighing 96.21
kg (212.10 lbs.).sl Using smaller stones, a traction trebuchet could get off more than
four shots a minuteY A counterweight trebuchet employed by the Crusaders at Damiet-
ta hurled stones weighing 185 kg (407.84Ibs.), and Arab machines are reported to have
shot stones of 259 kg (570.98 Ibs.).s3 Closer examination of accounts of siege warfare
and other military matters in these orations should prove fruitful.
These panegyrics, then, often provide unexpected tactual information for the histo-
rian. Study of them will also tell us a great deal about how the Byzantines looked upon
themselves and upon other peoples. In his second panegyric, Psellos gives a lengthy
account of the emperors from Basil II to Monomachos, most of which is also in his
Chronography, but here with different emphases, showing how he expected his fellow
citizens to view their past. Several orations also give expression to the Byzantine dislike,
if not contempt, of foreigners. S4
TRese panegyric orations, which were so important to the Byzantines, served several
purposes. They provided a forum for the literary elite to present their concerns directly
to the emperor, as well as the opportunity to influence, in varying degrees, the direction
of policy, although this had to be done with extreme subtlety and caution. But they
were primarily expected to talk, not so much about themselves and their agenda-
although they did much of that-but about the emperor and his agenda. Praise of the
emperor was praise of the system, and from the time of Eusebios of Caesarea the tran-
scendent theory of empire held sway. The emperor was God's vicar on earth, and all had
to obey him. While the Christian people endured temporary setbacks, the emperor,
with God's favor, would eventually subject the entire world to his dominion. A sure
promise that this will indeed come about is provided by the submission of foreign rulers,
manifested even by gifts of elephants and giraffes, and by the victories of the emperor
over the barbarians of the day. Whether the emperor is praised for building aqueducts
or for slaughtering barbarians, the intent was the same: to portray the emperor toiling

" Nicephori Basilacae Orationes, 57-59.

50 The first recorded use of the counterweight trebuchet was at the siege of Zeugminon in 1165: Niatae

Choniatae Historia, I, 134. C. Foss argues th.t the counterweight trebuchet reached Byzantium by the middle of
the 12th century: Survey of Medieval Castles of Anarolia, 1: Kutahya (Oxford, 1985), 77-84, but see review of Foss
by R. W. Edwards in Speculum 62 (1987). 678-79.
" A1-Fath ibn 'Ali a1-Bundari in M. Houtsma, Recueil de textes relatift" I'histoire des Seljoucides, vol. " (Leiden.
1889).42.
"At the Crusader siege of Lisbon in 1147: De expugnatione Lyxbonensi, ed. and trans. C. W. David (New
York. 1936). 142-43.
" See Ibn a1-Muqaffa, History of the Parriarrhs of the Egyptian Church, ed. and trans. Y. 'Abd a1-Masih and
O. Bunnester, 4 vols. (Cairo, 1942-74). vol. III, pt. 2, p. 218; A. Melkonian, DieJahre 128711291 in der Chronik
al Yuninis (Freiburg, 1975), 86. A few years ago. in England, medieval enthusiasts constructed a counterweight
trebuchet rut hurled an Austin Mini automobile a considerable distance: Mechanical Engineering (January 1994).
" See, e.g., the speech of Eusuthios to Manuel I, FRB, or. 6, pp. 94-95.

243
[ 140] Court Intellectuals and Rhetoric

for the well-being of his subjects and, in return, demanding their unquestioning alle-
giance. Some orations were designed as propaganda for specific imperial policies or for
generating desired attitudes among the populace, regarding a rebel leader, say, or a n;rili-
tary defeat. It would seem, although there is not much concrete evidence, that these
panegyrics, or at least their message, based largely on the reputation of the speakers,
were widely disseminated throughout the empire, certainly in other major cities, such
as Thessalonica. Although the imperial portrait may not have been carried about from
city to city, as in Roman times, the imperial image and message, unchanging but ever
variable, was conveyed by means of these panegyrics throughout the empire, and even-
tually down the ages to us.

Catholic University of America

244
 Byzantine Hagiography and Sex in the
Fifth to Twelfth Centuries
ALEXANDER KAZHDAN
H agiographv and sexuality-can two notions
be more contrasted, more incompatible?
-Iagiographical works present the entire life or an
'pisode in the life of a holy man or woman or a
~roup of men and 1I'0men, or posthumous mira-
les at their tombs or shrines, in order to provide
he reader with a moral paragon and instruction
'fl holl' to devote one's whole life to God. Hagio-
Taphical works present, usually in a sequence of"
pisodes, the system of Christian values, among
rhich chastity naturally holds a place of honor.
'he gist of the hagiographical message is that the
ody and its "impure" desires should be sup-
ressed and the sexual drive eliminated. The hero
as to forget, in his or her claim to holiness, what
~x he or she was given. A hermit in the desert is
eprived, for a casual observer, of any marks of his
~x, and a woman in disguise enters a male mon-
itery and bravely exercises her piety among the
~presentatives of another sex. I Angels had no
'x; in visionary dreams they resemble eunuchs.
nd the monastic community, an ideal of hagiog-
Iphy, was an angelic, that is, epicene society.'
But in the paradoxical, ambivalent world of By-
.ntium the most edifying genre of literature was,
the same time, the most entertaining one. Meant
indoctrinate and assuming a substantial part in
lurch reading, hagiographical writings were the
In order to avoid excessively long footnotes 1 will refer to
IC, its Auctanum of 1969 and Novum Auctanum of 1984. When
IG indicates several editions, I will refer (in an abbreviated
"m) to the one that I used; only if I used a more recent edition
t included in BHG and its a~claria, will I give the complete
"erence.
E. Padagean, Structure socia/e, /amille, chretiente a Bywnce
mdon, 1981), XI, 597-623.
As C. S. Galatariotou ("Holy Women and Witches," BMeS 9
184-85],85) puts it: "Saintly'women require not only a denial
sexualitv ... but a denial of their very sex."
245
mass media (7hvialliterattlr) of the time.' Thev
were crammed with sujets that attracted the aver-
age listener: travels and shipwrecks, natural disas-
ters, fantastic beasts, incredible healings, murders
and thefts, and the topic of sex was certainlv not
lacking. The Byzantines, admirers of epicene
angels and saints, were far from being epicene
themselves; the average listener of hagiographical
works wanted sexy stories, and he got them in
quite significant numbers.
THE PLACE OF WOMEN IN
HAGIOGRAPHICAL SOCIETY
In her article referred [0 above (note 2). C. Gal-
atariotou emphasized the "patriarch\""' and "mis-
ogvny" of Byzantine society. even though she her-
self runs into a difficulty-the image of the Virgin
Mar\". She based her conclusions on Neophytos the
Enkleistes. On the contrary, hagiography beneyo-
lentlv accepted the role of women. God created
both Adam and Eve, and later appeared in the
world via the female, says the hagiographer of St.
Maninianos.' The biographer of S1. Andrew the
Fool agreed with him: it was not the Devil who cre-
ated woman but God himself; he created woman
in order to augment the world, and every man who
wants to have a wife can have her with God's per-
mission.' There is no distinction between the male
and female in the eyes of God, savs another ha-
giographer" and the female sex is to be praised
beginning with the Mother of God.' Hagiogra-
phers assume that the female gender is weaker and
'Cf. H. Kech, Hagiographie als christliche Unterhaltung5literatur
(Goppingen. 1977).
'BHe 1177, ed. P. Rabbow, p. 291.23-24.
'V,'a of Andrew the Fool, BHe 117, col. 765AB.
6Translation of Theodora of Thessalonike, BHG 1i39. p.
42.3-4.
'Vila of Thomais of Lesbos, BHe 2454, p. 2348.
132 ALEXANDER KAZHDAN
gentler,' but women are able to reject their weak-
ness and act as men 9 To be like a man is singular
praise for a woman: 'O when the Arab fleet ap-
proached Attaleia, the governor of the city com-
manded the entire population to mount the city
walls, and not only were there men holding shields
but young women disguised as men l l Women
could demonstrate more heroism than men: in the
words of deacon Ignatios, the author of Patriarch
Tarasios' biography, during Iconoclasm feminine
weakness turned out to be more steadfast than
masculine strength."
From this equality in origin, sanctity, and sah'a-
tion one has to distinguish the inequality of eyery-
day life: hagiographers knew that seclusion was
the normal status of women, and only an extra-
ordinary emergency would cause the prudent
woman to forget her modesty (ut6wc;) and rush
into the street l3 1t was sometimes dangerous for a
woman to appear in public, as is demonstrated by
the story of a girl whom a local commander
"'anted to drag to slave1'\' "only because of the
beauty of her body" and who ran for protection to
a chu~ch." The father of St. Theophano neyer al-
10"'ed his daughter to leave the house without
chaperons. and she was even sent to the bath-
house either in the evening or in the morning
when the streets were empty, and was accompa-
nied by sen'ants and maids."
MARRtAGE Ar>D ITS PROBLEMS
On the scale of Byzantine ethical nlues, virgin-
ity occupied the topmost place, much higher than
legitimate marriage; it is, however, to be noted that
\irginitv "'as considered a type of marriage. the
yirgin (or nun) being proclaimed Christ's bride. It
goes "'ithout saying that hagiographers praised
celibacy; more interesting is their positive attitude
to"'ard marriage. A legitimate marriage, con-
cluded with the apprm'al of parents. was desir-
able. ,n Empress Theodora is said to have respected
yirginitv, but she held marriage in honor-"O
q "ila of Theodora of Thessalonike, BHG I 'i37. p. 20.1-2.
~ \"ila of George uf Amastris. BHG 668, p. 36.9-10.
I" Palladios. Dialogue, BHC 870. ed. P R. Coleman·Nonon. p.
98.30-31.
,\'ila of Antony the Younger. BHG 142. p. 199.1-4.
"BHG 1698. p. 415.28-29.
'rita of Tarasios, BHG 1698. p. 421.34-~5; Forty-Two Marhn
o(Amurioll. version B.BHG 1212. p. 13.18.
''; \ 'i/a of Peter of Argos. BHG 1504. ed. Ch. Papaoikonomoll.
p. 6~.22-24.
"SHG 1794. p. 3.25-30.
~\·ila ofAndrc\\"lh(' Fool, BHG 117, col. 765C.
246
praiseful marriage," exclaims her biographer'"
Marriage had a twofold purpose: procreation (see
above, note 5), and restriction of the sexual drive
and avoidance of promiscuous fornication (see be-
low, notes 23, 59).
The conflict between chastity and connubial life
is a frequent topic of hagiography. Its simplest so-
lution is an escape from the world to a monastery
or nunnery, an action that could lead to a clash
with the will of one's parents, as is described. for
instance, in the Vita of Nikon Metanoeite. '8
In some cases pious girls managed to escape
marriage. The hagiographer of three brothers-
David, 5ymeon, and George of Lesbos-relates a
noyelette about a rich woman of senatorial rank
who had two virtuous daughters, one of whom
studied poetry and grammar. read the works of
the church fathers, and longed for the monastic
life, whereas the mother wanted her to enter a le-
gitimate marriage. The daughter would probably
have had to yield, but a sudden vision helped her:
she saw a man in glimmering garment and lost her
voice in fear. When 51. Symeon of Lesbos healed
her. the whole familv, together with their maids,
took the habit and founded a nunnery."
Other children obeyed the will of their parents,
as did Martha. the mother of St. Symeon the Styl-
ite.2() The ancient yersion of Symeon's Fila plainly
states this fact. whereas the later short Vila dwells
longer on it: Martha. says the compiler. did not
want. in the beginning, to discard virginity. the
treasure favored by God, even though she was
often told how honorable marriage and the unde-
filed marriage bed is.2I
Hagiography presented a way of compromise
between filial obedience and pious chastity-the
non-consummated marriage. It is well known that
young Theophanes the Confessor persuaded his
bride to leave the world immediately after the wed-
ding. and the same effect was achieved by SI. Ko-
non of Isauria. 2 ' When St. Demetrianos turned
fifteen. his parents decided to marry him off; their
purpose was to secure the sah'ation of his soul,
since, comments the hagiographer, legitimate mar-
I: BHG 1731, ed. A. Markopoulos, Symmeikla 5 (1983), 2,1,
par. 99-101.
18 00 this problem, see A. Kazhdan. "Hagiographical !'Joles,"
B\'1.illltioll 54 (1984).188-92.
. "BHG 494. pp. 234-36.
?('BHG 1689, ed. P. ,'an den Ven, p. 3.21
"BHG 1691. ed. J. Bompaire. p. 79.16-18.
2~l'ila(' of Theophanes: BHG liB7z. ed. V. Latyse\, pp. 9f;
BHG I i89. ed. C. de Boor. pp. 6 f: flHG 1792b. p. 20.:10-37.
Vita of Konon. BHG 2077, p. 301.1-7.
 BYZAI'iTINE HAGIOGR.-\PHY Al'\D SEX IN THE FIFTH TO TWELFTH CENTL'RIES
riage is in this respect second only to virginity."
Even though Demetrianos did not follow the ex-
ample of Theophanes and Kanan. his marriage
was not consummated. and when, in three months,
his very young wife died, she was still a virgin.
A conflict between marriage and chastity could
evolve in the process of connubial life, as hap-
pened with Melania the Younger. After the death
of her two sons she felt an aversion to marriage
(IlLOOe; taD YUllou) and told her husband that she
would stav with him "as her lord and master" onlv
if he agr~ed to lead a life of chastity; if not, sh~
would gi"e him all her belongings and "liberate
her bodv."" In the later Vila of "Ielania, the sharp-
ness of this anti-marital tendencv was reduced. the
"aversion" disappeared. and onll' the call for chas-
tity (a,YELa) remained."
The revision of two vitae, that of Symeon the
Stvlite and that of Melania, shows a change of at-
titude: the praise of marriage as "honorable" was
introduced and, in addition, the aversion to mar-
riage was muted. We can make a more general ob-'
servation: the concept was developed that sanctitv
could be achieved not only in the desert or in the
monaster;' but in family life. !vlaria the Younger
and Thomais of Lesbos (see below) were married
women who deserved the reward of holiness:
Nicholas Kataskepenos. in the Vila of Cvril Phi-
leotes. conjured up the image of a saint who. after
the birth of his child. limited his sexual intercourse
with his wife but did not accept consistent celibacv,
and Kataskepenos' younger contemporary. Eusta-
thios of Thessalonike. in an extraordinary I'ita of
Philotheos of Opsikion not only praised the mar-
ried holv man but proclaimed the saint living in
the world more honorable than lonely anchorites
dwelling far from mundane temptations. 2•
Normally marriage was a union contracted
within family circles. parents plaving a role more
substantial than the spouses-to-be. The state. how-
ever. could interfere with this extremely pri"ate re-
lationship of its subjects. The Vita of Athanasia of
Aegina refers to an imperial pTostagma that com-
manded all single women and widows to marrv the
elhnikoi. ,;
~'BHG 495. ed. H. Gregoire. p. 222.163-67. corrections by
E. Kurtz. AB 27 (1908). 31.
~~HiJloria Lausiaca, II, ed. C. Butler (Hildesheim, 1967). p.
155.8-16.
,; BHG 1241, ed. H. Delehave. AB 22 (1903), p. 11.13-15.
~bOn this evolution, see A. Kazhdan. "Hermitic, Cenobitic,
and Secular Ideals in Byzantine Hagiography of the ~imh [to
Twelfthl Cemuries," GOTR 30 (1985), 484-87.
"BHG 180. ed. F. Halkin. SIX inidlls d'hagiologte byzantine
(Brussels, 1987), p. 181.7-9.
247
133
The wedding feast is frequently recalled in hag-
iographical writings. St. Demetrianos' parents cel-
ebrated his ,all~ALOt; EOQTli-" Morem"er, the wed-
ding appears in a metaphorical context as a symbol
of joy. Thus John of Damascus, after his ampu-
tated arm was miraculously restored, was so happy
that one could think, says his hagiographer. he
were celebrating gamelia, filling the chamber with
the noise of marriage.'9 When the Iconoclast em-
peror Theophilos summoned Michael Synkellos,
the saint came forth fearlessly not as a man enter-
ing the lists but as if he were called to his own mar-
riage festivities.'o A monk is said to ha,'e had a vi-
sion: a crowd of men in white garments on white
horses and a woman in purple among them; their
procession was accompanied with music and clap-
ping, as it is, adds the writer. at marriage ceremo-
nies. 31
Unlike the Byzantine romance, hagiography
does not stop at the wedding; the authors of saints'
vitae knew that marital problems began to arise
after the festivities. These problems could be med-
ical (primarily the barrenness of a marital couple)
or moral (the incompatibilitv of spouses). Hagiog-
raphers gladly deal with barrenness ..-\ frequent
element of the saint's story is the protracted bar-
renness of the hero's parents that must be over-
come by constant prayer or divine intervention.
When Theophano's parents had no child for a
long time, they started visiting a church da:' and
night and besought the Virgin to give them an heir
and participant in their life." The Virgin was an
appropriate protector of childbirth. but a male
saint could also be helpful: the tomb of Patriarch
Ignatios healed the barren spouse of a nobleman
so that she brought forth several children."
When the hagiographer of St. Blasios broaches
the topic of sterility, he uses a very risky, almost
impudent formulation: a high-ranking Roman was
said to have a legitimate wife but no children: he
addressed Blasios, begging the saint to give him a
child "to continue the familv line and to inherit the
property." 34 The Greek CtVum:aOLt; om'Qlla'Wt;,
which I translate periphrastically as "continuation
of the family line," has the connotation "raising up
"BHG 495, ed. H. Gregoire, p. 222.171.
"BHG 884: PG 94. 460AB. Cf. hta of Kosmas of Maiouma.
BHG 395. p. 326.9-11.
"BHG 1296, p. 242.4-5.
" Vita of Elias Speleotes, BHG S81. col. 880E.
"BHG 1794, p. 2.16-18.
"BHG 817; PG 105. 56lD.
"BHG 278, col. 665B.
 134 ALEXANDER KAZHDAN
the sperm," physiologically well connected with the
situation of our Roman aristocrat.
A barren couple could expect some help from
medical doctors, but this help was expensive. St.
Antony the Younger had an adventurous life and
once pretended to be a physician; a rich land-
owner (oLxo/)€on61:1'],) complained to Antony that
after twenty-five years of marriage he had no child
and promised the "doctor" a third of his property
if he helped, but Antony demanded ten war horses
as his honorarium."
Byzantine society could look with incredulity
and suspicion at cases of miraculous assistance that
healed protracted sterility. Thus John Moschos re-
lates the stor\' of the hermit Daniel who blessed a
barren wom~n and she became pregnant; there-
after the rumor spread that the father was sterile
and the bab\'. was sired b,'. Daniel. Then the hermit
worked the second miracle: when the woman gave
birth to the baby, Daniel asked the father to invite
guests to a banquet, and during the festivities Dan-
iel took the baby (then twenty-two days old) and
asked him who his father was. "This man," an-
swered the baby, pointing his finger in the direc-
tion of the husband'6
Barrenness was a physical problem to be solved
by medical or supra-medical intervention. Another
serious problem of marital life was the incompati-
bilit), of spouses. The story of Mary the Younger is
well known:" a faithful wife, she was accused by
her cruel husband of ha"ing a love affair; in a fit
of rage he beat her, and soon thereafter she died.
Another case of conjugal tragedy is the Vila of
Thomais of Lesbos. As did man\' hagiographi-
cal heroines, Thomais preferred virginity and
shunned ph"sical pleasures. Howe\,er, she obeyed
her parents. "bent her head to accept the wreath,"
and joined her legitimate spouse. But her hus-
band, Stephen, turned out to be not her ally but
an adversary; a man of the world, he behaved like
a beast and beat her because she cared for the
needy" Another unhappy marriage is described
in the Fila of Peter of .-\.rgos: the Devil excited in
the wife such a hatred toward her husband that as
soon ·as she heard him entering the house she
would fall down on the floor and quiver;'" only St.
Peter's intervention cured the poor woman.
Whatever the hardships of marriage, it was a
"BHe 142. pp. 195 f.
";John Moschos, PG 87. 2977D-29S0A.
"BHe 1164, cols. 695 f.
"BHe 2454. cols. 235f-236f.
"'BHe 1504. ed. Ch. Papaoikonomotl, p. 69.11-15.
248
holy institution, and spouses were obliged to pre-
serve their bonds. Theodore of Stoudios was an
eager supporter of the concept of the purity of
marriage. One of his biographers conveys that
Theodore insisted that a man could be the master
of one woman only, as well as a woman the com-
panion of only one man, and this union had to be
legitimate and not illegal or licentious' o Theodore
worried that Constantine V I's example (separation
from his first wife) could make divorce a rule." In
their anti-Muslim polemics, hagiographers criti-
cized Islam for allowing four legitimate wives and
thousands of concubines (naAAaxaC), which indi-
cated to them that Muslims worshiped God as pro-
tector of fornicators and even of sodomites" The
author of the Vila of St. Andrew the Fool con-
demns a certain Rafael (the more as he was a dea-
con) who walked out on his wife and began living
with a maidservant."
More complicated was the case of an illegitimate
union, especially that of two slaves, which in the
tenth century was not yet considered a marriage.
The \ 'ila of St. Basil the Younger described a "fam-
ilv" of house sla\'es H Following the order of her
master, Theodora joined her O1)V1:QO<l>o, ("com-
panion," not husband) and bore him two children;
after his death she lived in chastity rearing her boy
and girl; her master gave her a tiny little room in
the vestibule of his house." She had not been
chaste in her youth. and when she died demons
accused her of fornication. The angels, however,
retorted that the slave maid was not blessed by a
priest and had not received a legitimate benedic-
tion, nor was she gi"en to a husband in church and
adorned with a wreath-her pseudo-marriage was
based onl\' on the word and gesture of her master,
and therefore she cannot be blamed for fornica-
tion. 46
BEAt'TY AND NAKEDNESS
P, Brown emphasized that Christian writers re-
jected the ancient reverence toward "the fiery
spirit unleashed in the sexual act."" In the grada-
tion of sins. the illegal sexual act, rroQv£(a, was
4°BHC 1'75..J; PC 99, 2SiC. The passage is Jacking in BHG
1755: PG 99. 144 C.
41 EHG 1755, col. 13iSC.
42The.'\((s o!Sixty-Three Martyrs. BHG 1218. p. 147.15-2'2.
43 BHG J Ii. co1s. (g7e, SOOA.
HSee Chr. Angelide. ~O(jAOl O'tytv KwvatavtlVOU1[OA'l tOY 10°
at., S)'mmeikla 6 (1985). 40 f.
"BHe 264b. pp. 300.39-301.6.
"BHe 263, ed. A. \'ese1o\"kij, I. p. :l2.H-IS.
4, P. Brown, The Bod,Y and Society O",ew York, 1988), 432.
 BYZANTINE HAGIOGR.\PHY AND SEX IN THE FIFTH TO TWELFTH CENTuRIES
given a particular place: all other sins are those of
the soul, but pomeia is bodily wrongdoing. "Every
vice terribly darkens the soul, but pomeia makes
the bodv filthy and disgraceful"'" The body was
considered sexually dangerous, especially if it was
beautiful and naked.
Human beauty was ambivalent: it could reflect
the beauty of the soul and, on the contrary, it could
deceptively cover the spiritual ugliness of a man.
Hagiographers frequently stress the beauty of
their heroes and heroines. The pious girl Hypatia-
Fevronia, in the Vita of the three brothers from
Lesbos, shone in her spiritual and physical splen-
dor." The daughters and granddaughters of St.
Philaretos the Merciful excelled all other women
in their beauty.'" Thomais of Lesbos also was
exceedingly beautiful. so that her inner virtues
flashed through her external appearance." But
beautv in hagiographv is a relati\'e value and can
appear in a suspicious context. St. Nilus of Ros-
sano was handsome and had a beautiful voice, so
that many single women (of course. excited by the
Devil) chased him; one of these Immen surpassed
all the others in both looseness and beauty, and she
managed to seduce him and gi\'e birth to a girl."
The union was probably illegitimate, since imme-
diately after this notice, the hagiographer exclaims
that Providence did not leave Nilus to wallow in
this mire of a life.
"Good" hagiographical beauty is abstract and re-
moved from the body as far as possible; since the
body is shameful. it should remain covered. Helio-
dore, the Faust-like anti-hero of the Vita of Leo of
Catania, made several women accompany him;
they followed him across "deep waters" and natu-
rallv drew up their shirts (XL1WV(OXOl), so that not
only their knees were bared but their thighs as
well. The poor women were humiliated and
laughed at.;3 Even more shameful was the behav-
ior of an indecent woman who remained naked in
the water while her companions, mortified at
seeing St. Dometios, tried to cover themselves.;4
A story told by Moschos shows the shocking fear
that a Christian could experience before a female
body: a Persian girl asked the priest Konon to bap-
tize her; she was so beautiful that the priest "was
"BHG 117, col. 828B. D.
"BHG 494, p. 234.13-14.
"BHG 1511z, p. 141.2-5; BHG 1522, p. 76.17-18.
"BHG 2454, co1s. 234B, 235E.
"BHG 1370, ed. Acta55 Sept. VII, col. 263B.
"BHG 981 b, p. 17.30-33.
"BHG 560, p. 308.33-34.
249
135
unable to anoint her with the holy oil." She stayed
two days in the Lavra of St. Sabas waiting in vain,
and then Archbishop Peter (524-552) suggested
sending a deaconess to celebrate the rite, but it was
found improper. Confounded by his weakness,
Konon fled to the hills. There John the Baptist ap-
peared to him and enjoined him to return, "Hav-
ing removed Konon's clothes. John sealed thrice
with the sign of the cross the part of Konon's body
down from the navel" and thus quelled his sexual
passion. so that the next day Konon quietly bap-
tized the girl "not even noticing that she was fe-
male by nature." 55
Sanctity helped a man overcome temptation
caused by the naked body. When Luke the
Younger sent the monk Pankratios to a sick woman
he ordered Pankratios to take a small vessel of oil
and rub it with his own hands on the patient'S body.
Both the woman and her husband were stunned
bv this inappropriate behavior but eventually
obeyed, and Pankratios rubbed her entire body
from head to toe, following the instructions he had
received from Luke.'6
ILLICIT I~TERCOURSE
In Byzantine society, "'hich venerated marriage
and was frightened by the sight of the naked body,
adultery and lewdness occupied a surprisingly im-
portant place. ,,",owada)'s, complains the hagiogra-
pher of Andrew the Fool. the people plav like
blind and insensible beasts: many fornicate with
rejoicing and commit adultery by giving gifts and
presents. 57 The secular administration took severe
measures to fight this nil. If we may believe the
Vita of Antonv the Younger. a governor issued a
decree ordering every fornicator and every pros-
titute to be arrested, their property confiscated,
and their hair cut off to make them a laughing
stock. 58
Libertines were found in all walks of life. We
hear about a poor young man who could not af-
ford legitimate marriage and therefore was defiled
by improper and illicit intercourse." A married
man had two eunuchs who provided him with di-
verse women-single and married alike, as well as
whores; his habit was to get up before the roosters
and to head to church. but on his way to church he
would stop "to accomplish the deed of the Devil."
"PG 87, 2853D-2856B.
"BHG 994, ed. E. Martini, p. 107.4-13.
"BHG 117, cols. 765D, 800A.
"BHG 142. p. 194.16-19.
"MIracles of 51. Menas, BHG 1254m, p. 148.27-31.
136 ALEXANDER KAZHDAN
And thus his piety was praised by everybody but in
fact he was "a clandestine devil." 60 Another man
"of the new generation" was extremely licentious
(the hagiogl'apher uses the word nOQvoxdnl1Ao"
which is not found in either Lampe or Liddell-
Scott), inclined to incest and lust, even to sodomy,
never visiting a church, wasting his time with
strumpets, drunkards, and musicians 61 Women
could be lewd as well: St. Thomais healed two of
this type, and grateful for her medical help, one
promised to enter legitimate marriage, whereas
the other's vow was more restricted-she swore to
stop disorderly intercourse with men during the
divine and great feasts."
Even some monks were not free from the sin of
fornication. Moschos tells a story of a monk of the
monastery of St. Theodosios who left for Jerusa-
lem and went first of all to a brothel; he then
worked as a craftsman and led a profligate life,
squandering his salan and that of his compan-
ion. 63 Another monk, while visiting his village. suc-
cumbed to erotic desire and at night went to bed
with a local woman; his act did not remain un-
known-miraculously Stephen the Sabaites found
out when and where the man had sinned." In
Capua, due to the omniscience of "!ilus of Ros-
sano, the mother superior of a convent was caught
in bed with a priest." Even decent and saintly
people could be falsely denounced as fornicators.
A former disciple of Stephen the Younger accused
the holy man of insulting Constantine V and of
sexual relations with Stephen's spiritual daughter,
the nun Anna. Anna was arrested but naturally
did not acknowledge her adulten' (the hagiogra-
pher knows that there 'QS no crime perpetrated),
but nevertheless the cruel Constantine ordered
her to be flogged 66
In this atmosphere of pervasiw lewdness, the
harlot's was a necessan and popular profession,
and accordingly brothels frequently appear on the
pages of hagiographical works. The hagiographer
of Andrew the Fool describes a merry com pam in
Constantinople that, after a carousal, headed to
"rU[.IdQLU of crude women," defiling there the
beauty of their souls until the small hours; as they
6(Tila of Andrew the Fool. BHG II i, col. S52B,
"' BHe 117. col. 833Be.
"BHe 2454. col. 238DF.
"PC 87. 2956BD.
"BHe 1670. col. 53Se.
"BHe !:l7D. ed. AclaSS Sep!. \'11. col. 3117BC. Gregor,. 'he
hagiographer of Lazarus Galesiotes (BHG 979. col. 555AC). also
tells (! stor~' about a priest's wife who tried (0 seduce a monk
"BHe 1666: PC 100, I J25BD, 113211C.
250
left they were arrested and whipped 6 ' The word
mimarion is explained below as "brothel." We also
meet in this Vita the term "the inn of actresses
([.Il[.ld6wv)" having the same meaning6' Moschos
relates a story of a hermit who could not resist the
sexual drive, went to Jericho, visited a brothel, and
was infected with leprosy.5O
Prostitutes were impudent in their dress and es-
pecially in their gestures. Gregory of Dekapolis
saw in Syracuse a tower built near the harbor; a
licentiolls woman lived in this tower, and the sailors
who arrived safe from the perilous sea would be
allured by her fake beauty and impudent ges-
tures.'O Gregory succeeded in changing her habits,
and transformed the brothel into a pious dwelling.
The hagiographer of Andrew the Fool also de-
scribes a mimas who noticed two young men silting
in a public place and right away began to make im-
pure gestures trying to arouse desire in them.'1
Although prostitution was a sinful and disrepu-
table profession, some saints nevertheless had har-
lots as their mothers. Thus Maria. Theodore of
Svkeon's mother. earned her living as a prostitute
in a country inn." and a later legend made St.
Helena, Constantine the Great's mother. a harlot
in an inn. although some historians of the ninth
and tenth centuries, shocked by this legend, re-
jected the story of her improper pasl."
Repentant harlots, however, could achieve re-
spected sanctity. 51. Maria of Egypt and St. Pelagia
are famous examples of transformed prostitutes.
Tenth-century morality, which would not put up
with the prostitution of St. Helena, tried to refor-
mulate the past of St. Maria-at any rate, in the
tenth-century \fila of Theoktiste of Lesbos. which
was modeled on the basis of the Egyptian saint's
life. the heroine was not a strumpet but a pious
nun." Other hagiographical prostitutes are less
well known than Pelagia and Maria, but they were
1'7Vita of Andrew the Fool. BHG 11 i. col. 649AB.
f>~ RHG 117, col. 7i6D.
'''PC 87. 2861C.
inBHG ill, ed. F. D\"ornik. pp. 56.23-57.2.
"BHG 117. col. 764D. The \Ilia of Andre,..' the Fool is prob-
ably exceptionally severe IOh'ard "fornicators"; see J. Grosdidier
de Ma[om. "Les themes d'edification dans la Vie d'Andre Sa-
los:' TAl 4 (1970), 324.
"BHe li48, ed. A FeSlugiere, I, p. :1.8-14.
73A. Kazhdan ... 'Constantine imaginaire,'" R":al1tinn 57
(1987).2 J 2-15. For some examples of harlots in hagiographicaJ
works of the late Roman period, see H. J. Magoulias. "Bath·
house, Inn. Tavern, Prostitution, and (he Stage as Seen in the
Lives of the Saints of [he Sixth and Seventh Centuries:' 'En.
'E!. Bt'~.r". ~8 (1971). 240-46.
i~ A. Kazhdan. "Hagiographical Notes,'· BZ 78 (l 9BS). 49 f.
 BYlAt\TI"E HAGIOC;RAPHY At\O SEX IN THE FIFTH TO TWELFTH CEt\TURIES 13i

also destined for sal\'ation, Thus, in a fragment her self-sacrifice, Cornelius gave her 230 nomis-
from the Pmtllm spirituale, we read of a nun who mata and other coins and things and, without ask-
left her convent in Thessalonike, devoted herself ing her name, sent her away.
to fornication, but eventually repenting returned The anchorite Kaioumas had to tackle an anal-
to the entrance of her convent to die 7 ' ogous case. He was summoned as arbiter bv a
A single good action by a fornicator or prostitute council in Cyprus that discussed the fate of the late
could be sufficient to secure his or her salvation, Philentolos, the son of Olympios. Philentolos was a
Moschos tells a tale about an orphan girl who gave generous rich man who helped the poor and even
away her whole inheritance to save a man overbur- founded a hospital but had "a passion for fornica-
dened with debts; then, having no means to exist, tion." Kaioumas explained that Philentolos was to
she became a harlot. but soon "came to her senses" be saved from Hell because of his charitable deeds,
and abandoned this profession, When she decided even though-and Kaioumas is less tolerant to the
to convert to Christianity, the pious members of licentious sinner than the author of the Vita of
the parish, in their indignation, refused to accept Theodoulos-he was not admitted to Paradise: his
a whore, but their earthly morality was overruled soul had to remain with those of unbaptized chil-
by hea\'en, and angels brought her to a church and dren. 1s
ordered her to be ba ptized 7'; Both Cornelius and Philentolos, fornicators as
A similar storv is found in the Fita of St. Theo- thev were. were rewarded for their good deeds,
doulos of Edes;a." The saint heard a prophetic primarily generosity; Pelagia and Maria repented
voice announcing that a certain Cornelius would and became zealous anchorites. The Vita of Theo-
inherit the kingdom of heaven. Theodoulos was dore of Edessa. however, reveals that a prostitute
astonished since Cornelius was a flutist in Damas: could acquire holiness automatically, without any
cus, a representative of a despicable profession, effort on her part. A woman, relates the hagiogra-
according to traditional Christian standards. pher, had a son who was gravely ill. so she asked
l\'evertheless, Theodoulos started the search for e\'eryone to pray for his health; all was in vain until
Cornelius and found him in a hippodrome hold- she met a whore, hurled the boy into the harlot's
ing his musical instrument in one hand and with lap and, genuflecting, besought her to pray. The
the other caressing a bare-headed harlot who wore sinner was ashamed by this unexpected demand,
inappropriate and impious decorations. Cornelius but seeing the boy at the verge of death turned to
confessed to Theodoulos that he spent each day the East and, beating her breast with her hands,
with whores and actors-but he was granted the praved in tears An extraordinarily brilliant light
heaven Iv kingdom because once he had acted with descended from heaven upon the boy and the
sympathy and generosity. prostitute, the prayer was accepted, and the bov
As a matter of fact. Cornelius' story as related to recovered.;9 The licentious sinner could be pure in
Theodoulos is the centerpiece of the whole Fita. the eyes of God.
Cornelius, returning home from a nightly church
service. noticed an incredibly beautiful woman; he UNREQt:ITED LOVE
addressed her with flatteri~g words (treating her
A particular suJet of the hagiographical erotic
as a prostitute) and tried to hug her, but she began
story is the conflict in which one person is inclined
crying and was unwilling to accept his caresses.
to or even longing for intercourse and another re-
Surprised, Cornelius asked her to explain her
sists it. The Acts of the apostle Andrew deals "'ith
strange behavior, and she told him her stOry. She
such a conflict developing within a family, long
was a daughter of honest parents, an orphan from
after the wedding. Maximilla, wife of Aegeatus,
the age of twelve; when she married, she brought
anthypato5 of Achaea, walked "on Christ's road"
her husband a significant dowry. He squandered
and decided to cease sleeping with her husband;
everything and, moreover, went into debt and now
she used her illness as an excuse. Aegeatus in-
was afraid of being arrested and thrown into
sisted, threatening otherwise to execute Andrew,
prison. So the woman decided to become a street-
walker to earn her husband's ransom. Touched by 78 F. Halkin, "La vision de Kaioumos et Ie son eternel de Phi-

lentolos Olvmpiou." ,~B 63 (1945), 62-64. See C. P. Kvrris. "The

nEd. Th. Nissen, "Unbekannte £rzahlungen aus clem Pratum Admission of the Souls of Immoral bur Humane People inw
spirituale," BZ 38 (1938), p. 357,t 1-18, the 'limbus puerorum; according to the Cypriot Abbot Kaiou-
"PC 87, 3097e-3100B, mos." RESEE 9 (1971), 461-77,
"BHC 1785. cols. 752F-754F. "BHC 1744, pp,58 f.

251
 138 ALEXANDER KAZHDAN
who encouraged Maximilla to remain chaste even
though she might die in the name of Christ. No
threat, however, could break Maximilla's steadfast-
ness. The hagiographer describes Aegeatus' de-
spair; the man could not understand his wife's be-
havior and kept calling her back: "Why do you
disregard your parents who entrusted you to me
in marriage, and why do you avoid the [sexual]
union' Your parents appreciated in me not my
wealth, glory, or origin but my honesty and gentle-
ness." 80 This appeal to parents is typical of the Byz-
antine concept of marriage: it was negotiated by
parents, but it required an individual will to dis-
rupt the union and accept chastity.
l\-lore often hagiographers present the conflict
of "unrequited love" in an extramarital situation or
in the case when only one part is bound by marital
ties. The object of desire can be either male or fe-
male: if the object is male, the situation is fre-
quentlv compared with that of the biblical legend
of Joseph and POliphar"s wife, and the man is
called "a ne,\" Joseph." Thus the "edifving story" of
Philotheos evokes the biblical episode. opposing
"rhe just Joseph" and "the impious Eg,-ptian lad,."
adulterous wife of the pa/rikios Constantine, who
tried to seduce the young slave and after her fail-
ure accused him of encroaching upon her honor."
The Vita of Theodore of Edessa contains a
lengthy tale about a handsome young monk. Mi-
chael from Mar-Saba, who would come to Jerusa-
lem to sell baskets. A eunuch of the "queen Seida."
the spouse of Abd el-Malik, liked his baskets, and
once he brought Michael to the abode of "the new
Egvptian lady." Seida let Michael in; she started
"'ith generous promises of freedom and wealth.
but Michael responded that he had already re-
cei"ed evervthing from Christ. Seida took the next
step and asked Michael to become her friend; by
q;(i,o; she meant "lover." l\-lichael retorted that he
loved Christ and that his friends are heavenlv
powers and saints; describing his attitude toward
them. Michael employed vocabulary appropriate
to erotic relations-sweemess, desire, passion.
beaut\·. comeliness. Then Seida changed her tactics
and threatened Michael, but he was not afraid. In
despair she exclaimed: '''You, poor man, deprive
yourself of great benefits-am I not beautiful, de-
sirable, worthy of passionate longing'" Michael
did not accept her boastful assertions; for him she
"'BHG lOa, pp. 344.11-345.9, BHG 99. p. 301.11-14
':BHG 23i2-i3. ed. F. Halkin. "Histoire edifiame de Phi-
lothec injustcment accuse par une femme et miraculeusement
sJme."jOB 3i (l98i). p. 34.82-91.
252
was not beautiful and desirable but ugly, filthy, and
hatefuL He could not find a better way to offend
the beautiful queen, and Seida ordered him to be
flogged; Michael, however, endured the punish-
ment for hours on end. Then she sent him, in
shackles, to the "emperor," and in a note accused
"the impudent monk" of an offensive assault, for
which he was beheaded."
Another version of the "new Joseph" story pre-
sents the woman not as a mighty queen or the pa-
trikios' spouse but as an ordinary seductress. Thus
the Samaritans sent a woman to tempt Sl. James
and paid her twenty gold coins. James let her in,
kindled the hearth, and tried to warm her with his
hands. The woman, enticing the saint, asked him
to stroke her breast (lit., heart), but to avoid temp-
tation James put his left hand into the fire until his
fingers burned down. Shaken by his steadfastness,
the woman converted." Another whore acted on
her own. She heard about the blessed Martinianos
and proclaimed that he would fall like the leaf
from the tree if she wanted; if my beauty, said she,
does not shatter him. he is really a marvel not only
among men but even among the angels. In the rag-
ged disguise of a poor woman. she knocked at his
door and was let in. but the saint retreated to an
inner chamber and locked the door behind. When
in the morning Martinianos came out to see her
off, he did not recognize his guest: the harlot had
taken off her rags. adorned herself. and prepared
for a siege on his chastity. She told him that she
had heard about him and vearned to sate herself
wi'th his handsomeness; she asked him wh,' he was
burying "such a marmoreal youthfulness,''"and, re-
ferring to the Bible, called him to marriage. Mar-
tinianos retorted that he could not marry her since
he was poor and unable to support a family. Do not
worn', she replied, I have everVlhing-a house,
gold, silver. properties. and slaves. The saint. con-
tinues the hagiographer. began to be inflamed with
desire. and he even looked out at the road. which
was deserted at that hour. But God did not allow
him to perish: Martinianos collected some brush-
wood, made a fire, and entered the flame strug-
gling against himself: this fire, comments hagiog-
rapher. typified [he archetypal eternal flame. The
whore was stricken; she realized her wrongdoing,
threw her ornaments into the fire. put on her rag-
ged dress. and fell to the feet of the saint asking to
be forgiven."
"' BHG 1744. pp. 17-28.
'" BHG 770. pp. 27~-81.
~iBHG 1171, ed. P. Rabbow. pp. 278-82.
 BYZANTINE HAGIOGRAPHY AND SEX 11\ THE FIFTH TO TWELFTH CENTL'RIES
St. James and St. Martinianos found themselves
on the verge of a fall and needed the purifying
force of fire to overcome temptation. Later saints
seem to have mortified their flesh to such an extent
that they simply ignored attempts to seduce them.
Elias Spileotes, with a number of monks, had to
flee to Patras, where one of the local residents in-
vited him for dinner. Seeing his clean, bright face,
the wife of the host, "the new Egyptian lady," "with
her heart pierced by the sting of licentiousness,"
not only ogled Elias with her intemperate eyes
but even stretched out her foot and shamelessly
rubbed his leg. Her courtship remained in vain';
"the new Joseph" was not shaken in his hean. nor
did he cherish sweet desire but, as he later told his
disciples. his flesh did not even feel her touch" In
more detail a scene of seduction is depicted in the
Vita of Elias the Younger. His master's spouse. en-
flamed with desire, besieged Elias with a myriad of
tricks: she colored her face brightly as if anointing
it with milk, dyed her cheeks red. painted her eyes
so that they sparkled, let her hair fall and her shirt
trail along; she placed her hands on his bosom,
dragged off his clothes, stripped his breast and
arms, and tried to caress his face with hers-but
"the new Joseph" remained immoyable like an im-
pregnable wall or a stony rock. 86 A similar story is
in Prodromos' Fita of Meletios of Myopolis. A
noblewoman from Thebes. proud of her wealth
and beauty. licentious and insolent. came to ylele-
tios and attempted to incite him by improper
movements and obscene words. The saint paid no
attention and recommended that she communi-
cate with those who wore long golden hair."
Nicholas the Soldier, when he stayed in an inn,
was tempted by the owner's daughter. She came to
his room at night and called him to "an impure
intercourse"; she was mad from eros, but Nicholas
quietly explained to her that she was a victim of
demons who were forcing her to lose her virginity
and become a shameful laughing stockS' The
chaste man should be polite and careful not to of-
fend his seductress, as we can learn from an erotic
story told by Prodromos. Once the monks of St.
Meletios of Myopolis went to a nearby village to
buy wine; the merchant(?) was a good man, but his
wife was a real harlot. She wanted to seduce one of
the monks, so she behaved like a lewd woman. She
"'BHe 581, col. 857AB.
"BHe 580, ed. C. R. Taibbi, Vi'a di ,anfElia i/ Giovane (Pa-
lermo, 1962), p. 16.192-217.
"BHe 1248, ed. V. Va,il'ev,kij, p. 45.14-28.
.. BHe 2311, ed. L. Clugnet, ROC 7 (1902), pp. 323 f.
253
139
took off his shoes and offered him slippers, washed
his feet and dried them. looked gentle, let warm
tears run, and addressed him with the following
words: "Why do you need the desen, why do ,ou
need poverty? These boorish sandals do not fit
you. this rough dress does not suit your flesh. You
could adorn a city and be the first among consuls."
She waylaid him, she waged a hand-to-hand fight
against his invincible soul; she accompanied the
monk to the vineyard and there attacked him in
her "naked passion," but he drove her away using
crude words (and not the cleansing fire, as St.
James and St. Martinianos). When he came back to
the monastery Meletios already knew about his af-
fair. but he did not praise his temperance. Without
mentioning the monk's courageous resistance in
the house and the vineyard, Meletios berated him
for his swearing and quoted Matt. 23:24: \\'h,
being meticulous in straining a gnat. did he swal-
Iowa big camel?89 Did Meletios mean that swearing
was a vice worse than fornication?
Saints could be helpful in preventing young
monks from intercourse. A woman wanted to
arouse a stvlite's pupil to the sexual act; the lad "'as
slow in making a decision, but Gregory of Deka-
polis appeared on the sp'ot and with his staff
chased the evil woman away.90
The ultimate temptation scene occurs in the rita
of .\ndrew the Fool: a whore dragged him into a
brothel where prostitutes surrounded him and in-
"ited him to fornication. The description is natu-
ralistic: they fondled his secret parts. kissed him.
and urged him to sate his soul and his desire. But
the\" were unable to excite Andrew and cried out
hopelessly: "He is dead or [a piece of] insensible
wood or immovable stone."91
The situation could be reversed, with the woman
becoming the object of courting. As in the "ne\l
Joseph" situation, two types of attack occur: en-
treating and forcing. In a short tale, Moschos de-
scribes a monk's attempts to seduce a peasant's
daughter; she managed to persuade him genth'
that the fornication would ruin his soul and she
would have no recourse but to commit suicide after
having slept with him.92 Another story by Moschos
is about a young virgin who lived in her own
house. A man inflamed by love for her (it was cer-
tainly a satanic passion incited by the Devil) con-
stantly loitred in front of her house, so that she
"BHG 1248. pp. 57-59.
"'BHG 711. p. 67.1-9.
"BHG 117. col,. 652D-653A.
"PC 87, 2892AC .
 140 ALEXA!\DER KAZHDAN

could not even go out to pray in a chapel. The vir- tale of the daughter of Haplorrabdes included in
gin sent her maid to invite him in, received him the epic of Digenis Akritas. A parallel stoT\' is to be
sitting at her loom, and asked him why he pursued found in the Vita of Lazarus Galesiotes who met a
her. I love you, he answered. When I see you I am girl moaning and lamenting because she had been
all aflame. What do you like in me that makes you swindled and led away from her native land, and
adore me so passionately, she inquired. Your eyes, her deceiver robbed her and disappeared. I have
he answered sincerely, they have deceived me. She analyzed the similarity between these two tales!·
immediat, :y took the weaver's shuttle and gouged and see no reason to return to this topic: unlike
out both eyes!' thus surpassing the ordeal of St. Digenis, Lazarus did not succumb to the tempta-
James and St. Martinianos. The poor lover, in com- tion of the situation and remained honest in his
punction and regret, took the monastic habit. role of caretaker of a forsaken girl.
Women are often presented as victims of the
sexual drive of the powerful. Many works dealing MAGIC A1\D DEMONS IN THE FIGHT
with the period of the anti-Christian persecutions AGAINST CHASTITY
portra" the sufferings of beautiful Christian "ir-
Libertines suffering from unrequited Im'e and
gins who attracted the attention of the womanizer-
unable to resort to force or deceit looked to magic
judge. One example is found in the Histaria Lau-
for help. The HlslO1'ia Lausiaca contains a story
slara by Palladios. A typical Roman judge tried to
about an Egyptian infatuated with a free married
persuade a t"pical Christian girl to worship emper-
woman who rejected his courting. The man went
ors and idols. She refused. and up to this point the
to a sorcerer who made the woman <j>oQ~ci<;. The
story follows the typical pattern. Then Palladios in-
word means "mare" but also has the connotation
troduced, instead of the traditional motive for tor-
"prostitute." The hagiographer depicts how the
ture. a sexuallv tinged episode: the judge placed
man puts a halter on the woman of his heart and
the girl in a brothel, and instantlv the patrons of
leads her "like a horse" to the desert'"' When a girl
the establishment began to haunt this "workshop
from a noble Cappadocian family settled in the
of destruction." but the girl managed to gain some
convent of Irene of Chrysobalanton, her former
dela" hiding in a secret, foul-smelling room. Then
suitor found a magician, a servant of Satan, who
a young "magistrianus" came to her aid; he en-
promised to satisfy the infatuated man's desire.
tered the room and gave her his clothes so that she
Soon Ihe girl was attacked by a frantic lust for her
could flee. The next day the trickery became pub-
former suitor, so that she threatened to commit
lic. and the "magistrianus" was thrown to the
suicide unless she was allowed to see him. :\one of
beasts." The situation described seems perverse to
the pra"ers of Irene and the other nuns could help
our taste: when love appears, it is a kind of perse-
her, and the Mother of God had to arrange a
cution and vengeance; when we expect true love,
magic operation to quell the magic forces crealed
it is replaced by Christian compassion and self-
by the sorcerer. She appeared in a vision to Irene
sacrifice completely devoid of sexual anraction.
and then sent the martyr Anastasia and Basil the
.-\nother version of the same theme replaces the
Greal who came flying through the air and let
pagan judge with a barbarian suitor. Thus the bla
down a package weighing about three pounds; it
of Peter of .-\rgos relates how the Cretan Arabs
contained a variety of magic devices, including two
took many captives and were ready to give them
lead puppets-one resembling the suitor. the
up for ransom. except for a woman who was young
other the lovesick girl-embracing each other.
and beautiful; they wanted to deliver this woman
These instruments of sorcery (yor]"rElJllata) were
to their phvlarch. The agreement was not reached.
committed to the flames, and immediately the
the Cretans interrupted negotiations, and sailed
woman was liberated from her invisible ties and
a\\'a,,95 The intervention of St. Peter destroyed
restored to soundness of mind."R
their plans.
Gregory, the author of the Vita of Basil the
Instead of force, a seducer could employ a sim-
Younger. recalls his encounter with a witch. She
pler. more down-to-earth tool-deceit. The classic
example of a forsaken and deceived girl is in the !If, A. Kazhdan. "Hagiographical NOLes," ByuwtirHl :~).t (1984).

"PG 87. 29120--2913B.
~~Ed. C. Butler (see above, note 24), pp. 160-62.
"SHC 1504. ed. Ch. I'apaoikonomou. p. 67.23-32.
182-84.
97 Ed. Buder (see above, note 24), pp. 44.2H-45.13.
~~ Fila of Irene of Chr\'sobalanton. RHG 952. ed. J. O. Rosen-
quist. Til, Ll{e of St. lre7le, Abbess of Chr:r:iObalanton {l!ppsala,
1986). pp. 52-64.

254
 BYZANTINE HAGIOGR.-\PHY AND SEX IN THE FIFTH TO TWELFTH CEN1TRIES
was united "by legitimate marriage" with the mis-
thios Alexander but was so licentious that she slept
with all the men from the neighboring proasteia.
Daughter of a witch. she used magic instruments
(~aYlxd) to achieve her impure aims. When she
wanted to entrap Gregory, she began to apply, as
Gregorv sees it, both traditional feminine means
and sorcery: by day she strutted before him. and
at night she sent him seductive visions. The hag-
iographer asserts, however, that he managed to
chase her away. 99
Sorcerers and witches could induce sexual de-
sire in chaste persons. Even more dangerous for
monks, nuns, and clergymen was the Fiend him-
self or a special demon of fornication. Moschos.
for instance, narrates a story of a recluse on the
Mount of Olives against whom the demon of for-
nication waged war: finally. the monk gave up and
called the demon, who immediately appeared and
demanded that the monk cease venerating the im-
age of the Virgin.,on The cult of the Virgin and
chastity are bound together in this story, as they
are in the Vita of Irene of Chrvsobalanton.
The Historia Lausiaca also mentions the demon
of fornication, and tells the story of the deacon
Evagrios, who was entrapped by the "idol of desire
for women." '0' Theodora of Alexandria was an
honest woman living in legitimate marriage. but
the Evil One selected a rich man. incited longing
for Theodora in him. and with the aid of a witch
helped him achieve his goaL Ashamed and upset.
Theodora confessed her sin in a convent and after-
ward. having put on male dress, entered a monas-
tery outside Alexandria."" In the Vita of Lazarus
Galesiotes, there is a tale about a girl in Rome, a
daughter of rich parents, who was possessed by the
demon: she demanded to be brought to the monk
Paphnoutios. Three times she tried to enter his cell
and retreated, but finally the parents left her alone
with the monk. Then the demon stopped tortur-
ing the girl and began to confuse Paphnoutios'
mind, persuading him to have intercourse with the
girL Paphnoutios yielded, perpetrated the sinful
act, and later, in despair, ran to Asia and settled in
a cave. This erotic affair is omitted in the later re-
vision of the Vita produced by Gregory of Cyprus,
who only mentions the name of Paphnoutios 'o,
The spirit of fornication battled with St. Ioanni-
'"
» BBHe pp. 320 f.
H G 264b, pp.
""'PG 87. 2900BC.
,0> Ed. Butler (see above. note 24). pp. 121.3, 117.8-16.
'" BHe 1727-30.
'''BHe 979, col. 521AG: BHe 980. col. 597E.
255
loll
kios. The holy man cured a girl who, while walking
with her mother, was suddenly assaulted by the
rrvfij~a 1:li~ rrogvf(a~, and, unable to bear "the
flame of the demon," began shouting impudent
words. The hagiographer tells another storv about
the spirit of fornication who possessed a woman so
that she lost her reason, tore up her dress, let her
hair fall loose, and lived like a beast, even feeding
on her own flesh. '0' Peter, the hagiographer of St.
loannikios, walks on a tightrope here. since naked-
ness and loose hair could appear in hagiographical
writings in a different context, as a svmbol of the
pious hermit living for decades in the desert. As
frequently happens in Byzantine literature, the
image has an ambiguous meaning. Even though
we do not hear about any licentious behavior of the
possessed woman, Peter em phasized several times
that Ioannikios rescued her "from the hefty chain
of fornication," ministered to her "spiritual shrine"
troubled by the spirit of fornication, imitated
Christ who had saved the suffering harlot. When
Symeon Metaphrastes revised this story, he omit-
ted the sexual element, leaving only demoniac pos-
session.los
A story about the Devil's malice is included in the
Vita of Lazarus Galesiotes. A man from the theme
of Anatolikon was taken captive and led away to a
barbarian land. He entreated God to liberate him
from captivity, vowing that he would never return
to his house but would put on the monastic habit
and go to the Holy Land. His supplication was
heard, and the man was released from captivin·.
But. neglecting his vow, he headed home, and in a
small village a poor woman accommodated him in
her hut. The Devil persuaded the man to have in-
tercourse with his hostess, but after he had sated
his desire, the former captive learned that he had
slept with his own daughter'06
Sexual visions and dreams were powerful instru-
ments in the Devil's hands.,o7 The demon of for-
nication attacked Epiphanios, the chaste friend of
Andrew the Fool; the young man suffered from
these assaults when in his dreams he saw lewd
women and sinfully slept with them.'os St. Nilus of
Rossano, while staying in Rome, saw a German
woman tall and bulky; the demons who produced
'" BHe 936, cols. 41 OG, 398G-399A.
w; BHe 936, col. 399BC: BHG 937, col. 60BC.
'06 BHe 979, col. 529BO.
For brief notes on erotic dreams in Byzantine literature,
IQi
see G. Dagron, "Rever de Dieu et parler de soi," I sogni nel me-
d",,",o. ed. T. Gregory (Rome, 1985),45 f.
'''BHe 117, col. 792BC: see also col. 841B.
 142 ALEXANDER KAZHDAN
this image insistently displayed it to the saint dur-
ing his mental exercises. '09 St. Antony the Younger
was also haunted by "demoniac fantasies." Once as
he was sitting naked near his cell he saw a beautiful
woman, with loose hair (AuoC8gL1; appears in
Liddell-Scott with a reference only to the Byzan-
tine Geoponika) , who shamefully headed toward
him; the saint retreated to his cell. Before he be-
came a hermit, Antony was much tortured by sex-
ual yearning-the trail of the flesh, as his hagiog-
rapher says; so he decided to enter legitimate
marital union lest he be caught by "the nets of for-
nication" and become the object of ridicule." o
1\loschos makes abba Elias confess to erotic visions
incited by the Devil; unable to resist he ran out of
his cell, paying no attention to the stone-scorching
heat. to satisfy his yearning. He was burning with
desire, but suddenly had a vision of another kind:
he saw a deep precipice and corpses reeling out of
it. In panic he fell on the ground.' II
Saints could fight against such visions. Once in
his ,,·anderings. St. Nilus of Rossano met a young
nun who thre,,' herself down in a narrow passage
so that the hoh' man could not avoid touching her.
An innocuous incident' :>;0, Nilus recognized Sa-
tan's trick, hit her with his staff, and quickly passed
by. Despite his success, l\ilus was frightened by this
affair and decided not to allow his brethren to
journey alone. nor would he walk by himself.'"
Laymen, howeyer, could easily be deceived by de-
moniac visions. A story in the Vita of Irene of
Chn'sobalanton illustrates this idea. l\icholas was
a misthios in Irene's com'ent. He wanted to sleep
with a nun of the convent, and finally, at night, the
Dev'il made him believe that he had achieved his
goal: he dreamt that he had passed through the
gate"'ay and entered the cell of the coveted
woman. But instead of this affair he had a stroke,
from which el'entually Irene cured him.'"
The Vita of Andrew the Fool teems with sexual
v·isions. A woman saw in her dream an old Ethio-
pian (a symbol of the Deyil), who started to em-
brace and kiss her as if he were joking, and then
he asked her to sleep with him. She saw a dog (an-
other devilish sV'mbol); he was big and black, and
kissed her mouth-to-mouth, as a man. She also saw
herself standing in the Hippodrome and embrac-
ing the statues, driven to them by a desire for for-
""BHe 1370. ed . .1(10 5S Sept. \'11, col. 2740.
""BHe 142. pp, 207.24-32, 200.17-25.
'" PG 87, 2865(0.
'''BHe 1370. ed. Arlo 5S Sept. "II. cols. 299(-:1000.
JL1BHG 952. ed. Rosenquist (see above, nQ(e 98), pp.66-74.
256
nication; finally came a vision of herself and a
dog gobbling a frog, a snake, and even worse
things '''-objects that were probably considered
sexual stimulants.
The Byzantines seem to have been surrounded
by sexual temptation, and even marriage was not a
guarantee against the seductions that attacked
men and women in reality and in deyil-inspired
dreams. Priests, monks, and nuns suffered from it
no less than secular society. Some saints hoped to
escape the sin by fleeing. Thus St. Martinianos ex-
pressed a clear desire to live in a place to which no
woman had access."s A young nun in Jerusalem
was loved by a youth (Moschos naturally describes
his feeling as "satanic eros"). The nun did not wish
to become the cause of the young man's moral
ruin, so she took a small basket of soaked pulse
and fled to the desert; a miracle confirms that she
behaved properly: the amount of the pulse in her
basket did not diminish and she was able to survive
on this food.'16 Some pious men were able to de-
lelOp such a level of sanctity that they grew insen-
sible to sexual yearning and, moreover, shared this
capacity with their disciples. A certain Kiketas of
Macedonia suffered from "satanic eros," but St.
Meletios cured him so effectively that not only was
he abov'e any temptation but his genitalia became
completely cool and he remained impotent.'"
When St. Luke the Younger and his pupil Pankra-
tios ran away from a hostile invasion, they took
shelter in a cave; two women joined them and were
admitted by Luke, since it was wintertime. Because
of the cold he put the women between himself and
Pankratios, but he treated them as a mother treats
her child. He remained, says the hagiographer,
like a stone or a wooden log, and not even a single
satanic thought flashed through his mind.' IS
COJ\'CLUSION
We have gone through dozens of erotic tales re-
lated by hagiographers, and now we must try to
draw conclusions from these dispersed and evi-
demly non-comprehensive episodes. The first and
simplest statement is that the Byzantines acknowl-
edged the existence and the strength of sexual de-
sire: the body and its yearning was a presumption
from which hagiographers would start. Frequently
II~BHG 117,col. 7BOAC.
'''BHe 1177, ed. p, Rabbow, p. 2~4.34-35.
"" PG H7, 3049AO.
117F1Ia of Meletios of Myopolis. SHG 1248, ed. V. \'asiJ'evskij,
p. 52.19-26.
'" BHe 994; PG III, 468BC.
 BYZAf\:TIf\:E HAGIOGRA.PHY AND SEX IN THE FIFTH TO TWELFTH CENTURIES
sexual desire was interpreted as a result of devilish
intervention, of satanic eros, of demoniac frenzy;
sometimes such an explicit interpretation is lack-
ing and we may consider it the depravity of human
nature. The difference is not so great, since the de-
pravity of human nature could itself be explained
by the role of Satan. Touching on this point, how-
ever, we face a major theological problem, that of
free will. and this is not the place to discuss it.
Sexual desire is the sin of the body, and not of
the soul, and probablv because of this some forni-
cators are granted dispensation: first of all, the
sexual liberty of a slave maid was not sinful: then
a sincere repentance would sa,'e even a profes-
sional prostitute; a single good act opened the
kingdom of heaven for a libertine; finally, God's in-
comprehensible will could transform a harlot into
an instrument of miracle-working.
Of course, what hagiographers recommended
was not indulgence in fornication but avoidance of
it. Legitimate marriage was the natural choice:
l'vlarriage was a blessed union, and we probablv can
observe gradually growing respect toward it. The
earlier concept that the ideal marriage is the one
without consummation was contrasted with the
ideal of the "middle way" (in the style of Cyril
PhileO[es) or even of a normal family propagated
by Eustathios of Thessalonike.
257
143
Monasticism was by definition an escape from
sexual activity, though not from sexual desire: not
every monk or hermit was above his flesh, and
while some of them succumbed to temptation, oth-
ers would turn to physical ordeals (especially self-
torture by fire) not only to quench their passion
but to confound their seducers-a theme that
found its classical expression in Leo Tolstoj's "Otec
Sergej." Some perfect saints, however, reached
such sexual aloofness that they became insensible
to the charms of the other sex-like logs or rocks,
to use hagiographical similes.
Thus sexuality besmears the men and women
affected; in hagiography it is never coupled with
love. Love is placed on a different level-as Chris-
tian devotion or Christian philanthropy. Marriage
is concluded not by interested parties but by their
parents. It is seen as necessary for procreation, for
the very existence of humankind; it prescribes
marital fidelity and requires self-sacrifice when the
material well-being of the family is threatened. BlIt
never was hagiographical marriage based on the
jovs of the flesh, on sexual love. The topic of hu-
man love was reintroduced in Greek literature in
the eleventh century, in Psellos, and burgeoned in
twelfth-centun erotic romances.
Dumbarton Oaks
 TRUTH AND CONVENTION
IN BYZANTINE
DESCRIPTIONS OF WORKS OF ART

HENRY MAGUIRE

258
 INTRODUCTION*

S INCE the Renaissance, commentators have seen abstraction as a major

characteristic of Byzantine art. It was on account of its lack of natural-
ism that Vasari, in the sixteenth century, condemned the Byzantine
style. In his view this "style of lines and profiles" represented a low point
in the history of painting; it was only the genius of such artists as Cimabue
and Giotto that restored painting on its correct course toward the imitation
of nature. 1 In our own century, on the other hand, the unnaturalistic qualities
of Byzantine art have been regarded with favor. The critic Clive Bell, for
example, wrote in 1914 that "Post-Impressionism .... shakes hands across
the ages with the Byzantine primitives."2 He even went so far as to say that
"Giotto's art is definitely inferior to the very finest Byzantine of the eleventh
and twelfth centuries."3 The writing of a pioneer of abstract painting, Kan-
dinsky, testifies that Byzantine art has exerted an influence on the development
of modern art.4 But if we have now come to respect Byzantine art because
of its abstract tendency, it is puzzling that this aspect of it seems to have
been neither valued nor even acknowledged by Byzantine commentators
themselves. Byzantine descriptions of their own works of art, at whatever
period they were written, usually stress the realism of the art, and reveal no
awareness of any lack of naturalism. It has been suggested that the reason for
this discrepancy between the post-Renaissance view of Byzantine art and that
of Byzantine writers is that they were following a literary tradition which
had been formulated in the classical period.'
Ekphrasis, or description, was one of the standard exercises of late antique
rhetoric to which a Byzantine, even as late as the fifteenth century, was
subjected in the course of his primary education. 6 Ekphraseis were written
in verse and in prose. Sometimes they stood on their own, but often they
were incorporated into longer compositions, such as letters, sermons, and
histories. The form of ekphrasis had been defined in the second century A.D.
by the sophist Hermogenes of Tarsus. He says that the description can be
• This article and its planned sequel formed part of a Ph.D. thesis which was submitted to Harvard
University in the spring of 1973. My work on the thesis was generously supported by the Leverhulme
Trust and Jesus College, Cambridge, and by Durnbarton Oaks. I would like to thank all those who
have helped me in the preparation of the thesis and the article (without, of course, implicating them
in my errors), In particular I am indebted to Professors Kitzinger and Sevcenko for supervising my
work, to Professor c.'iango for his valuable comments, to Mr. and Mrs. Roueche for advice on some
translations, and to my wife for encouragement, criticism, and patience.
1 G. 'lasari, Le vite, ed. R. Bettarini (Florence, 1968), II, 37 and 97.
2 Art (London, 1914), 44.
3 Ibid., 123.
• W. Kandinsky, Concerning the Spiritual in Art (New York, 1947), 76.
5 This point has been emphasized by C. Mango, in "Antique Statuary and the Byzantine Beholder,"
DOP, 17 (1963), 65f., and again in The Art at the Byzantine Empire, 312-1453, Sources and Documents
in the History of Art (Englewood Cliffs, 1972). xivf.
• See R. J. H. Jenkins, "The Hellenistic Origins of Byzantine Literature," DOP, 17 (1963). 39ff.,
esp. 43, and M. Baxandall, Giallo and the Orators (Oxford, 1971), 85.

259
114 HENRY MAGUIRE

of " ... persons, deeds, times, places, seasons, and many other things,"7 and
that it should be vivid, and full of detail. He gives, as an example, the outline
of an ekphrasis of a war, saying that it should include "... the levying of
troops, the expenditures, the fears, then the engagements, the slaughters,
the deaths, then the victory, then the paeans of the victors, and the tears
of the defeated and their bondage."B
When classical authors wrote ekphraseis of paintings and sculptures they
evaluated the art with respect to its truth to nature. The more realistic the
work, the better it was, and the more vivid was the literary description.
This standard of judgment was taken over by Byzantine writers and applied
to mediaeval works. One aspect of realism to which both classical and Byzan-
tine authors gave particular attention was the depiction of emotions, such as
fear, joy, and sorrow. Hermogenes of Tarsus had included descriptions of
fear and grief as necessary components of an ekphrasis on a war. Classical
and Byzantine authors also used such descriptions to engage the reader's
sympathy when they wrote about works of art. But many modern observers,
from Vasari's day to the present, have felt that Byzantine art is neither as
realistic, nor as adept in the portrayal of the emotions, as classical, and
especially Hellenistic art. For this reason doubt has been thrown on the
sincerity of the Byzantine writers. The inflexibility of the literary tradition, in
face of the great artistic changes which we know to have taken place between
antiquity and the Middle Ages, puts in question the validity of the ekphraseis.
The Byzantine ekphraseis copied ancient models not only in their general
standard of judgment, but also in their specific language. There was a constant
repetition of cliches, paraphrases, and quotations, some of which were of
considerable length. These topoi cast further doubt on the accuracy of the
ekphraseis. It is reasonable to ask how often Byzantine writers looked at
the works which they described, and how far their descriptions were purely
literary exercises, based on written models.
This article, then, is an enquiry into the relevance of the ekphraseis to the
history of Byzantine art. It falls into two sections. The first attempts to
discover instances of spontaneous observation by Byzantine writers, and of
the appreciation of qualities peculiar to specific phases of Byzantine art.
The second section discusses the extent to which the employment of literary
conventions in the ekphraseis affected their accuracy as descriptions and
their value as expressions of Byzantine attitudes to art.

1. SPONTANEOUS OBERVATIONS IN THE EKPHRASEIS

Modern literature on the Byzantine ekphraseis has concentrated on their

archaeological interest. The ekphraseis have been seen as valuable keys to
7 n VOVTCX\ 512 EK'PpaCTElS 1TpoawTTwv TE Kat 7Tpay~6:-rWV Kat KOtpWV Kat T6,rc:vV Kat Xp6vwv Kat lTOAAWV
ITlpoov. Progymnasmata, 10, ed. H. Rabe (Leipzig, 1913), 22, line 9f.
BnpWTOV !JEV TO: -rrpo TOV lTOAEjJOV tPOVJ,1EV, TaS OTpCXTOAoyioS, TO: aVOAWIJCXTO:, TOV) cp6~ovs, eha Tas
O'\JIl~OA6:S"t TCxS acpayas. 70US SOVCITOVS. EtTa: TO TP01TOlOV, EtTa TOUS: rralo:vas TWV VEVIK1lkOTWV, TWV se. TO:
OclKpva, TIjv 50vAEiav, Progy",nasmata, 10, ed. Rabe, 23, line 2ft.

260
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 115
the reconstruction of lost monuments. 9 Little has been written on the ekphraseis
as expressions of Byzantine attitudes toward art and architecture. The most
important work on this subject is a paper published in 1930 by Oskar Wulff
which draws attention to Byzantine descriptions of architectural spaces.lO
These often seem strikingly apt, when they are assessed in relation to surviving
buildings. However, on the question of the validity of Byzantine descriptions
of painting and sculpture, the opinions of scholars have been divided. Some
authors have stressed the dominance of literary tradition and the discrepancy
between the bias toward naturalism shown in the literature and the apparent
abstraction of much surviving Byzantine arPl Others, taking a less skeptical
point of view, have pleaded for a limited acceptance of the ekphraseis.12
But although the problem of the relevance of the ekphraseis is obviously
one of great importance for our understanding of Byzantine art, it has not
been given the detailed examination which it deserves.
In a number of the ekphraseis the authors are careful to point out in their
introductions that they have themselves looked at the works of art which
they are about to describe. Lucian, for example, introducing his description
of a painting of the Marriage of Alexander and Roxana, declares: "The painting
is in Italy; I myself have seen it, so I may describe it to yoU."13 In the fourth
century Gregory of Nyssa introduces a description devoted to paintings of
the Sacrifice of Isaac with these words: "I have often seen this event in
painting, and I could not pass by the sight without tears, as art brought
the story vividly under my eyes."14 St. Gregory's statements are echoed in
a twelfth-century homily on the Massacre of the Innocents. The author14a
says: "I saw this event depicted in colors on a panel, and I was moved to
pity and tears."15 Another twelfth-century writer, Constantine Manasses,
seems to make a distinction between information on works of art which

• This approach is exemplified by the work of A. Heisenberg, Grabeskirehe und ApostelkiTehe.

I, Die GrabeskiTehe in Jerusalem, and II, Die A postelkiTChe in Konstantinopel (Leipzig, 1908). Heisenberg
attempted a detailed reconstruction of these buildings primarily on the basis of literary deSCriptions.
,. "Das Raumerlebnis des Naos im Spiegel der Ekphrasis," BZ, 30 (1929-30), 531 If.
11 See particularly Mango, "Antique Statuary," 65f. John Beckwith has also stressed the element
of literary convention in the ekphraseis, in Early Christian and Byzantine Art (Harmondsworth, 1970),
161.
12 E.g., A. Munoz, "Le ekphraseis nella letteratura bizantina e i loro rapporti con l'arte figurata,"
Recueil Kondakov (Prague, 1926), 139 If. ; E. Kitzinger, "The Hellenistic Heritage in Byzantine Art,"
DOP, 17 (1963), 9SIf., esp. 109f.
" la-rlv.1'} EIKo,V ev 'ITcxAlCj<, KcIyo, elSov Wo-rE Ked act av Elmiv 'XOI"I. Herodotus sive Action, 5.
a EISov ;roMCxKIS ';rl ypacpiis EIKova Toii ;rcl.Sovs, Ka\ OUK aSQKPvrl TT}v .9Eav ;rapiih.90v, .vapyws Tfis TEXV11S
v;r' OIjiIV cIyou<1T]S TT}V lo-ropiav. De deitate Filii et Spiritus sancti, PG, 46, col. 572C .
... Identified as the South Italian preacher Philagathus by A. Ehrhard, ObeTlieterung und Bestand
der hagiographischen und homiletischen Literatur der griechischen Kirche, pt. I, vol. III, fasc.7 (Leip-
zig, 1943), 631f1.
15 EISov eyo, TOVTO TO ;rcl.Sos xpcb"CXTI yeypa ..... vov !v ;rlvQKI, Ka\ ;rpos olKTOV EKIVt\.9T1v, Kal SOKpva.
"Theophanes Cerameus," Homilia LII, In sanctos Innocentes, PG, 132, col. 924B (= Homilia XXIV.9,
ed. G. Rossi Taibbi, in Filagato da Cerami, Omelie per i vangeli domenicali e Ie teste di tutto I'anno, I
[Palermo, 1969J, 159). Gregory of Nyssa was also quoted in an anonymous description of a scene
from the life of St. Nicholas of Myra in an eleventh-century manuscript in Milan (Biblioteca Ambro-
siana, gr. D. 92, fol. 59): EISov TOVTO ;roMOKIS KEXapay"'vov ev ;rlva~1 ••. Ka\ ~6v'i' T<;I T\i;r'i' Tiis .geas
iAIYY1<'xaas OUK aSQKPVT\ ;rapE;\t\;\v.9a. G. Anrich, Hagios Nikolaos, II (Leipzig, 1917), 166. lowe this
reference to the kindness of Dr. Nancy Sevcenko.

261
116 HENRY MAGUIRE

he has received from books, and his own observations. He prefaces his descrip-
tion of a floor mosaic in the Great Palace at Constantinople with references
to the bronze heifer by the classical artist Myron and the seated Hercules by
Lysippus. He repeats the old story of the heifer attracting a live bull, and
says that the Hercules was shown lamenting his fate. Then he remarks:
"These things have been written in books and are inscribed in the histories,
but I have seen the work of a painter's hand, and my eyes have been bewitched
at the sight .... "16 Constantine Manasses then explains that his description
will make the work accessible to those who have not seen it.
It we are to believe these writers, they have based their ekphraseis on their
own observation of works of art. But one wonders whether this claim, which,
after all, is a convention, is true or false. It is very easy to demonstrate instances
in which the writers of the ekphraseis obviously took no account of the works
of art at all. There are several instances of entire descriptions being plagiarized
from earlier authors. For example, John Phocas, a monk at the monastery
of St. John on the island of Patmos, wrote an account of a journey in Palestine
which he made in 1177. Amongst the sights which he described was a mosaic
of the Nativity with the Annunciation to the Shepherds, which decorated
the Grotto underneath the church of the Nativity in Bethlehem (the mosaic
is now 10st)Y John Phocas discussed this work of art at some length, but
not from his own observation. The passage was for the most part copied
verbatim from a sixth-century description by the orator Choricius of the
mosaics in the church of St. Sergius in Gaza.1 8 The same source was used as
a model by John Phocas for his description of the representation of the
Annunciation at the site of Joseph's house 19
Another Byzantine author who wrote descriptions of works of art which
were entirely derived from earlier ekphraseis was the fourteenth-century poet
Manuel Philes. One of his poems paraphrases the second-century description
by Lucian of a painting of the Marriage of Alexander and Roxana. The deriva-
tion is admitted in the title to the poem. 20 Another poem of the Palaeologan
period, ascribed by its title to Manuel Melissenus, paraphrases the twelfth-
century ekphrasis of a floor mosaic which was composed by Constantine
Manasses. 21 Constantine's ekphrasis was devoted to a mosaic in a bed chamber

16 Tatn-a ~lv o(;v KO:V Tais ~(~A01S yEypO:'I'aTal Kai Tais lo-ropiats &vaypa7TTa 'I'~pETal, !yw 61 1.wYPO:'I'0v
XE1p6S 1pyov 16wv Koi ,.O:s olfiElS KaTay01]TEV!lEiS Til> !lEO:~aTl .... Ed. L. Sternbach, "Beitrage zur Kunstge-
schichte," OJIl, 5 (1902), Beiblatt, col. 75, lines 33ft.
l? Descriptio Terme Sanclae, PG, 133, cols. 957D, 960A-B; K. Krumbacher 2 , Geschichte der byzan-
tiniscken Literatur (Munich, 1897),420.
18 Laudatio Marciani, 1.51 fl., ed. R. Foerster and E. Richsteig, Teubner (1929). The authors of
The Church ot the Nativity at Bethlehem, ed. R. W. Schultz (London, 1910), do not seem to have been
aware of the source of John's description. His ekphrasis is quoted in connection with the mosaic
at Bethlehem, and he is described as a "shrewd and receptive spectator" (page 65). Dalton even found
some of the details described by John Phocas characteristic of the art of the eleventh and twelfth
centuries, and not of earlier times, in spite of the fact that Choricius had described these very details
over 600 years before (page 50).
"Descriptio Terrae Sanciae. PG, 133, cols. 936B-C; Laudatio !vIarc., 1.48f.
20 Manuelis Philae carmina, ed. E. "liller, II (Paris, 1857), 336f.; Lucian, Herodotus sive Ailtion, 5.
21 Manuelis Philae co,.,,, ina , ed. Miller, II, 267f.; ed. Sternbach, "Beitrage zur I{unstgeschichte,"
cols. 74ff., 79ff.

262
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 117

of the Great Palace, which was evidently of the asaratos oikos type and
portrayed an assortment of objects and creatures arranged as if they had
been strewn on a floor. The lemma of the later poem does not acknowledge
the source of the paraphrase. Instead, it changes the representation from a
mosaic into a picture by the celebrated painter Apelles.
It should be added that the ekphrasis of Constantine Manasses itself had
certain points in common with a sixth-century description by the orator
] ohn of Gaza of a tabula mundi, which was painted in a bath house of his
city. Both authors, for example, claim to have seen in the works of art various
sea creatures, which they describe at length. They also both found the figure
of Earth, personified as a woman. 22 The type of mosaic which Constantine
Manasses describes is antique or Early Byzantine in date. Therefore the
possibility that he may have used an earlier literary model cannot be ex-
cluded. 23
The freedom with which written descriptions could be transferred from one
work of art to another is particularly obvious in the case of inscriptions.
At Tokah Kilise, in Cappadocia, for example, the New Church contains a
twenty-line verse which lists the subjects depicted in the fresco decoration.
The catalogue is incorrect; it mentions'scenes, such as the Feeding of the
Multitude, which were not represented, and omits other important episodes,
such as those of the Passion, which were shown. Presumably the inscription
was borrowed from the decoration of some other church. 24 In manuscripts,
inscriptions were copied from one book into another along with the illumina-
tions. Sometimes the paintings were slightly altered in the process, while
the inscriptions remained the same. 25
There is, then, ample evidence that many Byzantine writers on art had
little or no regard for accuracy, even when, as in the case of inscriptions,
inaccuracy would be immediately apparent. But, in spite of this conclusion,
it is possible for us to prove that some Byzantine authors did make original
observations on works of art. The ekphraseis were not entirely the product
of a dead literary tradition; they can also be shown to reflect contemporary
changes and developments in the visual arts.
22 John of Gaza's poem is found in P. Friedlander, Johannes von Gaza und Paulus Silenliarius
(Leipzig, 1912), 135-64. For the description of Earth, see Part II, v. 7ff., and for the fishes, Part II,
v. 79ff. The description of Earth in the ekphrasis of Constantine Manasses is lost, but it is recorded
in the title and in the paraphrase by Manuel Melissenus, ed. Sternbach, "Beitrage zur Kunstge-
schichte," cols. 74, 80. For the sea creatures, see ibid., cols. 77, line 145ff., and 78, line 196ff.
23 Another ekphrasis which may have been based in part on an earlier source is the description
of the mosaics in the Holy Apostles by Constantine the Rhodian; A. Salac, "Quelques epigrammes
de.I'Anthologie Palatine et l'iconographie byzantine," Byzantinoslavica, 12 (1951), Iff., esp.12f.
But see also G. Downey, "Constantine the Rhodian: His Life and Writings," Late Classical and
Mediaeval Studies in Honor of A. lvI. Friend, Jr., ed. K. Weitzmann (Princeton, 1955), 212ff .
.. G. de Jerphanion, Une nouvelle province de i'art byzantin. Les figlises rupestres de CaPPadoce,
I,2 (Paris, 1932), 305 I. Another Cappadocian example of a reused inscription is a description of
Christ Calming the Storm in the church of the Holy Apostles at Sinassos, which had been written in
the fourth century by Gregory Nazianzus, and which also appeared in the church of St. Basil in
Caesarea; R. Cormack, "Byzantine Cappadocia: The Archaic Group of 'NaIl-Paintings," JBAA, 30
(1967), 24.
2S See 1. ~evcenko, "The Anti-Iconoclastic Poem in the Pantocrator Psalter," CahArch, 15-16
(1965-66), 41.

263
 118 HENRY MAGUIRE
The ekphraseis often give precise descriptions of the fonnal layout of
scenes, and of the poses adopted by the figures. There are several such passages
in the sixth-century ekphrasis written by Choricius on the mosaics of St.
Sergius in Gaza. In his account of the Nativity scene, Choricius describes the
Virgin as "... a maiden lying back on her bed with her left hand placed
under her right elbow and resting her cheek on her right hand."26 The Virgin
is shown reclining in this. posture in the Nativity scene represented on a
phial in Monza (fig. 1). The phial probably dates to the sixth or early seventh
century, and was brought to Italy with oil from Palestine. 2? In date and place
of origin, therefore, it is close to the oration of Choricius. The Virgin is also
shown sitting, with her hands in the positions described by Choricius, in the
Nativity miniature of the Rabula Gospels (fig. 2).28 This manuscript was
written in 586 at the monastery of St. John of Zagba, in Syria.29
The Monza phials also provide parallels to the description given by Choricius
of the Annunciation to the Shepherds. He writes: "Some of the shepherds
seem to have no use for their staffs, but the staff of one shepherd, while it is
not employed on his flock, serves as a support for one of his hands. He has
raised his right hand, I should think in wonder at the cry. "30 On two of the
phials at Monza one finds representations of shepherds resting their left
hands on their staffs, and raising their right hands up toward the angels. 31
The description given by Choricius of the mosaic in the main apse of St.
Sergius makes an interesting comparison with surviving apse mosaics of the
sixth century. According to Choricius, the apse contained in the center an
image of the Virgin and Child, which was flanked by two groups of figures.
On the far right stood the donor of the church: "He is asking the patron
of the church, who is close by him, to receive the gift favorably; the other
accedes, and looks on the man with a gentle gaze, placing his right hand
upon one of his shoulders, and is clearly about to introduce him to the Virgin
and her Child the Savior."32 The general layout of the apse mosaic described
by Choricius is similar to that of the cathedral of Paninzo, a work which was
executed around the middle of the sixth century.33 The Parenzo mosaic
shows in the center the Virgin and Child, flanked by angels and saints. The
donor, the Bishop Euphrasius, stands to the extreme right of the Virgin,
accompanied by his young son. In the Parenzo mosaic the donor offers a
model of the church directly to the Virgin and Child, and not, as in the apse
., ... K6p111rpOs Mrl). &.crrrhrroucra Til. lIE. Aala. VrrOSEtcra T~ Tiis htpas ~c\).l, Tij 6E~IGi 6. Til. 1rap.la.
EmtV.lvoucra. Laudatio Mayc., 1.51.
" A. Grabar, Ampoules de Terre Sainle (Paris, 1958), pI. VII (Monza 2) .
.. Florence, Laurenziana, Plut. 1.56, fol. 4v; G. Millet, Reche1'ches sur l'iconogt'aphie de l'Evangile
(Paris, 1916). 101 note 4 .
•, C. Cecchelli, G. Furlani. and M. Salmi, The Rabbula Gospels (Lausanne, 1959). 9f .
.. Tots lIE. al KaAaVpoms &j(peTOI <pal.oVTal, T~ 6, 1rpOs IlEv Til. 1ro11'1IT)v &pyel, Scrrip'i' 6. aul'l'CX)(Ei ToT.
XEpoiv' Til. yap 6E~la. &.aC7)(wv, Il'ol60Kc\), TESaVI'aKE Tilv ~ot\ •. Laudalio Marc., 1.53.
31 Grabar, Ampoules, pIs. II and VIII (Monza 1 and 3) .
•• oVros oOv TOO VEe:, TO. 1rPoo-rCrnj. 6vra 1rA'lCTlo. elJl'E./;;s 61~a0-6al TO 6c\)po. alTEi, 0 6. mlSeTal Kal
yaA'l.~ SECxl'aTl TO. av6pa 1rpOCT~ArneI Tc\). "'1''''. atiToO SaTep't' Til. 6E~lav mltV.I.",v Kal 6ijMs laTI.
aliTlKa auCTTt\CT"'. aVTO. Tij TE l1apSe.'t' Kal T~ 1ra161 T~ I"'TijPI. Laudatio Marc, 1.29f .
.. B. MolajoJi, La Basilica eu/rasiana di Parenzo (Padua, 1943), 26f.; Mango, The Arl of liz.
Byzantine Empire, Jt2-t4S3. 62 note 37.

264
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 119

described by Choricius, to a saint, who acts as an intermediary. For this detail

one has to turn to sixth-century mosaics preserved in Rome. On the triumphal
arch of the church of S. Lorenzo fuori Ie mura, which opened into the sixth-
century apse, there is a mosaic of Christ enthroned and flanked by two groups
of saints (fig. 3).34 To the far right of Christ is the donor, Pope Pelagius II
(578-90). He is introduced to Christ by the patron saint of the church, St.
Lawrence.
A similar composition is to be found in the apse mosaic of SS. Cosma e
Damiano, a church founded by Pope Felix IV between 526 and 530 (fig. 4).35
Here the two martyrs are presented to Christ by SS. Peter and Paul, each
of whom places a hand on the shoulder of his protege, in the gesture described
by Choricus.
There is, then, a considerable coincidence between the layout of the scenes
as described by Choricius and as they are represented in contemporary works
of art. Since we no longer have the mosaics of St. Sergius in Gaza before us,
we cannot state categorically that Choricius gave an entirely accurate account
of this particular monument. But his precise descriptions cannot have been
inspired only by his knowledge of ancient authors. They must to some extent
also have been due to his own observation of the mosaics.
The description of St. Sergius by Choricius is only one of a number of
surviving ekphraseis which were produced by orators of the school at Gaza.
Both John and Procopius of Gaza wrote descriptions of paintings. John
gives an account of a tabula mundi, which, as we saw above, was a precursor
of a twelfth-century ekphrasis by Constantine Manasses. Procopius, who was
the instructor of Choricius, describes a painting of Phaedra and Hippolytus. 36
It is clear from both these ekphraseis that the authors had observed actual
works of art, and did not follow only literary sources. John of Gaza, for
example, seems to have been in doubt about the identification of some of the
figures he saw in the tabula mundiY Moreover, it has been shown that the
allegorical figures which John describes have close parallels in floor mosaics of
the fourth to sixth centuries which have been discovered in Syria and Pales-
tine. 38 In the ekphrasis of Procopius there is a description of four mythological
scenes, which are said to have been represented in a strip on the architrave
above a painted colonnade. The scenes are described in the wrong chronological
order, as if Procopius started his description at the wrong end of the sequence.
This seems to indicate that the ekphrasis follows the actual layout of a picture
and does not depend on literary models. 39 Certain passages of the ekphrasis
.. G. Matthiae, Mosaici medioevali delle chiese di Roma (Rome, 1967), 149£., pI. 89.
3. Ibid., 135f., pI. 78.
3. P. Friedlander. Spatantiker Gemaldezyklus in Gaza. Des Prokopios von Gaza :EK'I'paat, EIKovo"
ST, 89 (Vatican City, 1939) .
., Friedlander, Johannes von Gaza (supra, note 22), 214 .
• 8 G. Downey, "John of Gaza and the Mosaic of Ge and Karpoi," Antioch-on-the-Orontes, II
(Princeton, 1938), 205ff.; G. M. A. Hanfmann, "The Seasons in John of Gaza's Tabula Mundi,"
Latomus, 3 (1939), 111 ff.
3' The scenes are described in this order: the lion hunt of Hippolytus; Theseus and the Minotaur;
Ariadne gives Theseus the thread; Ariadne sees Theseus among the Athenian captives. Friedlander,
Spalantik., Gemaldezyklus in Gaza, 102f.; Procopius, Descriptio imaginis, 4-7, ed. Friedlander, 6f.

265
 120 HENRY MAGUIRE

seem to betray the artistic taste of the period in which Procopius writes.
The description of the jewelry worn by Phaedra is a good example: "The
mistress wears as much jewelry as is fitting for life in the house, necklaces
and armbands, collars around the neck and rings on the ears. A golden head-
band binds her head; it is surrounded by a row of Indian stones."40 This
bejeweled portrait of Phaedra calls to mind contemporary representations
of Byzantine empresses, such as the ivories of an empress now in the Vienna
Kunsthistorisches Museum (fig. 5) and in the Museo Nazionale in Florence,
or the celebrated mosaic portrait of Theodora in S. Vitale, Ravenna. 41 Mosaics
which survive from the late antique period sometimes show mythological
figures clothed in contemporary imperial dress. Weitzmann has noted a
mosaic of Apollo and Daphne at Antioch in which the ardent god is arrayed
in a purple chlamys and a pearl diadem. 42
There is good evidence, then, that the ekphraseis of the school of Gaza
were based as much on visual observations as on literary models. \Vhen
we turn,to the ekphraseis of the post-iconoclastic period, we can also prove
that some authors, at least, looked at works of art for themselves, for they
were aware of contemporary changes in iconography. For example, a brief
epigram on a painting of the death of the Forty Martyrs, which was written
in the first half of the eleventh century by the Byzantine poet Christopher
of Mytilene, specifically describes a version of this scene which in all probability
was first created in the Middle Byzantine period, that is, after the ninth
century. According to tradition, the forty saints met their death by being
exposed to freeze beside a lake in Lesser Armenia. Their sufferings are graphi-
cally illustrated in two famous ivories, now at Leningrad and Berlin. 43 On
each ivory, the naked martyrs are carved in various contorted poses, no two
of which seem to be alike (fig. 6). These ivories are usually assigned to the
tenth century, although later dates have been proposed. 44 A less accomplished
rendering of the scene is to be found amongst the earlier frescoes in the church
of Asinou, in Cyprus, which date to 1106. 45 In the Asinou version, too, the
martyrs adopt a variety of different poses.
These images are exceptional amongst Byzantine scenes of martyrdom for
the violent attitudes shown by the victims. There was, indeed, another, less
tortured version of the Forty Martyrs' death, in which each saint was rep-
resented alike, in a rigid orant posture. This version of the scene is the only

'0 TOO'Ov-rOVyap ti KEl<TTl!lfVT)- TIEplKElTot KOO'J.,lOV, arrav Tij KOT'oTKOV ap~6crE1 olaiTTJ. cqlqnShas TE Ka1 lTEpt-
~pCX)(16.la·"PIlOI TE TIEpl Tij OEP,) Kat TO'S werlv 'AIKTfjPES Kal xpvcrij TalVla TIiv KEq>aAt1v TIEplerq>lyovera 'I.OIK"'.
Alew. olaooxij TIEplMEIETal. Descriptio imaginis, 22, ed. Friedlander, 12.
U Friedlander, Spatantiker Gemaldezyklus in Gaza, 52ft; F. W. Deichmann, Friihchristliche Bauten
und 1I1osaiken von Ravenna (Baden-Baden, 1958), pI. 360; Beckwith, Early Ch"istian and Byzantine
A,·" 36 and pI. 63 .
.. K. Weitzmann, "The Survival of Mythological Representations in Early Christian and Byzantine
Art," DOP, 14 (1960), 52, fig. 12.
"A. Goldschmidt and K. Weitzmann, Die byzantinischen Elfenbeinskulpturen, II (Berlin, 1934),
pI. 3, nos. 9, 10.
U J. Beckwith, The Art of ConstantinoPle (London, 1961). 136f.

"D. C. Winfield and E. J. "v. Hawkins, "The Church of Our Lady at Asinou, Cyprus," DOP,
21 (1967), 261 fl., fig. 9.

266
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 121

one known to have existed before the iconoclastic controversy, and is best
represented by two frescoes in the church of Santa Maria Antiqua in Rome
(fig. 7).46 It is probable that the version of the martyrdom represented by the
Leningrad and Berlin ivories was not created until the Macedonian Renaissance
of the tenth century. The contorted figures seem to have been borrowed from
several other scenes. Some figures may have migrated from portrayals of the
Deposition, of Souls in Torment, and of John Baptizing the People, subjects
which had been rare or unknown in the pre-iconoclastic period. 47 Other poses
came from classical scenes, which we know to have survived into the tenth
century, as they appear in contemporary manuscript illumination. 46
Christopher of Mytilene's epigram stresses the variety of the martyrs'
poses: "See the victorious Forty of God here, standing by the lake, each one
in another posture. But if they are not alike in their postures, in the trial by
frost they have one mind."49 Thus the poem refers to what was probably a
comparatively recent development in Byzantine iconography.
The most important ekphrasis which survives from the post-iconoclastic
period is that devoted by Mesarites to the church of the Holy Apostles in
Constantinople. Nikolaos Mesarites was born in 1163 or 1164 in Constantinople.
It has been deduced from internal evidence that his description of the church
of the Holy Apostles was written between 1198 and 1203. In this period,
in 1201, he is recorded as sacristan of the churches in the Great Palace of the
Emperors in Constantinople, and he had a judicial appointment at the cathedral
of St. Sophia. 50 His ekphrasis on the Holy Apostles is of particular interest
for two reasons. First, it is one of the longest and most elaborate Byzantine
examples of this genre which survives. Second, it describes an important
building and a cycle of mosaics which have completely disappeared, and which
are accessible only, if at all, through written documents and through a few
manuscript illuminations.
The first church of the Apostles seems to have been built in the fourth
century by Constantius. It was rebuilt by Justinian before A.D. 550 as a
cruciform structure with five domes-one dome over each arm and one over
the crossing. Justinian's church was destroyed after the Turkish conquest
to make way for the Fatih mosque of Mehmet the Conqueror, which now
46 It is also known from a fresco at Syracuse. See O. Demus, "Two Palaeologan ::\Iosaic Icons in
the Dumbarton Oaks Collection," DOP, 14 (1960), 87ff., esp. 101f.
" Ibid., 106f. The motif of an older martyr holding up the slumped body of a younger companion
recalls Joseph holding up the dead body of Christ. Compare the Leningrad and Berlin ivories of the
Forty Martyrs especially with the miniature in Paris, Bibliotheque Nationale, gr. 74, fol. 100
(H. Omont, Evangiles avec peintures byzantines du XI'siecie [Paris, n.d.J, pI. 88). The martyrs who
hug their chests with their arms and droop their heads are similar to the souls in torment depicted
in Paris. gr. 74, fol. Sl v (ibid., pI. 41) and in Vat. gr. 394, fol. 12v (J. R. Martin, The Illustration of the
Heavenly Ladder of John Climacus [Princeton, 1954J, fig. 73) .
•, Weitzmann, "The Survival of Mythological Representations," 64f.
•• r ASPEl &.sAoJcpopOVS eEOV EvSaoE TEC'C'apO:t<ovra axill-lClTl aMov !v aAAO;> ecpEa1'aOTas KaT" AII-IVTJV' [Ei
5'/ipJa Kat EV ax,,~ao-lV 015E 1-111 elEV 6~oiol, !v !'lao-avo;> 1TayETOlO 6~6cppova Sv~6v EXOVcnV. E. Kurtz
ed., Die Gedichte des Christophoros Mitylenaios (Leipzig, 1903), 90, no. 133. The title reads: [Eis TOUS
&ylovs] TECTCTapO:t<ovra !v olaM6:Trovo-l axil~aC'l ["'ypa<pT]Sevras .
•• G. Downey, "Nikolaos Mesarites: Description of the Church of the Holy Apostles at Constantin-
ople," TAPS, N, S., 47, 6 (1957), 859,

267
 122 HENRY MAGUIRE
stands on its site. o1 Like the cathedral of Saint Sophia, the Holy Apostles
had to undergo several repairs and alterations in the course of its existence.
One chronicler, Theophanes, says that Justinian's successor, the Emperor
Justin II, "added adornment" to the church. 02 A major repair is recorded
in the ninth century, when the Emperor Basil I, according to his grandson
Constantine VII, buttressed the structure and, as Constantine put it, " ... strip-
ed off its old age, and took off its wrinkles .... "53 The phrase "stripped off
its old age" may suggest a renewal of the decoration on the walls in the ninth
century, but perhaps we should not put too much weight on it, as it is a stock
quotation from the Iliad. 04 The church was restored again by Andronicus II
at around 1300, but was apparently already in a ruinous condition before
Constantinople fell to the Turks. 55
In his ekphrasis of around the year 1200 Mesarites tells us that he is the
first to describe the church. 56 This was far from the truth, as the J ustinianic
building had already been described twice; first, by the historian Procopius,
before 560, and then again by the poet Constantine the Rhodian, between
931 and 944. Procopius gives a relatively short description of the building,
and makes no mention of mosaics. 57 Constantine the Rhodian's tenth-century
account is longer, and it portrays the mosaics in some detail. 58 However, the
selections of scenes made by Constantine and Mesarites do not always coincide.
The major puzzle which has been posed by the mosaics of the Holy Apostles
is their date. Heisenberg, who published the first edition of the text of Mesarites
in 1908, believed that the mosaics were sixth century. 59 Since his book appeared,
several writers have questioned this early dating, and it is now apparent that
some of the mosaics, at least, must have dated to the twelfth century. Of
course, there is no reason to suppose that all the mosaics seen by Mesarites
dated to the same period-just as the mosaics at Saint Sophia do not all
date to one time. The church of the Holy Apostles was large, and it was more
than once repaired. Its mosaics may well have been added or restored piece-
meal. The evidence for the post-iconoclastic date for at least some of them
comes both from texts and from extant works of art. The textual evidence
primarily concerns the artist Eulalios. A marginal note on the thirteenth-
century manuscript of Mesarites tells us that the artist of the mosaic of the
Marys at the Tomb was called Eulalios, and according to Mesarites the artist

51 R. J anin. La giographie ecclisiastique de I' Empire byzantin. Part I. Le siAge de Constantinople

et Ie Patriarcal I1lcuminique: vol. III, Les iglises el les monas/Ares (Paris. 195.3). 47f.
52 hm:6crll1lcn TCrS lKKA1lcrlaS TCrs KT1cr3e(OOS \nrO ')ovCTTIvlavoii, T1\v Te llE)'aA1lV lKKA1lcr(av Kat TOUs ay£ovs
emoCTT61.ovs ••.. Theophanes, Chronographia, ed. C. de Boor, I (Leipzig, 1883), 242.
" .•• Kat a".o~lcras TO em" Xp6vov yijpas Kat TCrs pvtiSas mplE1.c:,v .••. Theophanes Continuatus,
PG, 109. col. 337D .
.. Iliad, IX, v. 445f .
• 5 janin. op. cil., 48 .
.. XII.12. ed. Downey (supra, note 50).
" De aedificiis, 1.4.9-24 .
• 8 Ed. E. Legrand, in "Description des ceuvres d'art et de l'~glise des Saints-ApOtres de Constantin-
ople: poeme en vers iambiques par Constantin Ie Rhodien," REG. 9 (1896). 36-65.
50 Heisenberg, Grabeskirche und Apo.,telkit'che, II (supra, note 2); idem, "Die Zeit des byzantinischen
Maiers Eulalios," PhW, 41 (t921), t 024 ff.

268
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 123
portrayed himself in the mosaic. 60 In addition, a late thirteenth- to early
fourteenth-century epigram by Nicephorus Callistus Xanthopoulus ascribes
the Christ Pantocrator mosaic in the central dome of the Holy Apostles
to this same artist Eulalios. 61 A group of twelfth-century poems by Theodorus
Prodromus strongly suggest, if they do not conclusively prove, that a Eulalios
was a contemporary of the poet. 62
The visual evidence for the late date of the mosaics comes from the parallels
between the descriptions of Mesarites and tenth-, eleventh-, and twelfth-
century works of art.63 Scholars disagreeing with Heisenberg have pointed
out that on several occasions Mesarites describes features in the mosaics which
are not known in Byzantine art before the tenth century. These parallels
have a double significance. They prove the late date of the mosaics, and also
the spontaneity of Mesarites' observations. Unlike John Phocas, he was not
copying a sixth-century literary model.
We may look first at the description in Mesarites of the Transfiguration
mosaic. Here he depicts in detail the poses of the three astounded Apostles
who witness the event. "Peter, the most vehement, springing up from the
ground, since he could ... seemed to speak words .... James, partly rising
with difficulty on his knee, and supporting his still heavy head with his left
arm, still has the greater part of his body nailed to the ground, while his
right hand he holds closely to his eyes .... John however does not wish
to look up at all, but ... seems to lie there in deep sleep .... "64 These distinc-
tions between the poses of Peter, James, and John are characteristic of Middle
Byzantine representations of the Transfiguration. One finds the distinctions
already tentatively expressed in a miniature of a Paris Gregory manuscript
(BibliotMque Nationale, gr. 510, fol. 75), which was produced in Constantino-
ple probably between 880 and 883 (fig. 8).65 Here we find Peter, on the left,
boldly standing to address Christ, while James rises on one knee, and lifts
his hand as if to shield his face. John adopts a stooping posture and looks
down toward the ground. In later Byzantine representations of the scene
the attitudes of the Apostles correspond in more detail with the description of
•• XXVIII.23, ed. Downey .
., A. Papadopoulos-Kerameus, NIKTlq>6pOS KclAAlOTOS zaVS6"TTOVAOS, BZ, 11 (1902), 46, no. 14.
02 N. A. Bees, "Kunstgeschichtliche Untersuchungen tiber die Eulaliosfrage und den Mosaik-
schmuck der Apostelkirche," RepKunstw, 39 (1916), 97ff. and 231ff.; and 40 (1917), 59ff., provides the
documentary evidence to date the artist Eulalios in the twelfth century; A. Salac, "Quelques epi-
grammes de I'Anthologie Palatine et l'iconographie byzantine," Byzantinos/avica, 12 (1951), 1 ff ..
esp. 12f., draws attention to discrepancies between the descriptions of Constantine the Rhodian and
Mesarites, which may indicate that the mosaics were altered between 931 and 1203. The relevant
texts have now been collected and translated by Mango, Art of the Byzantine Empire, 229f1.
.. See especially N. Malickij, "Remarques sur la date des mosaiques de l'eglise des Saints-Ape-tres
it Constantinople decrites par Mesarit:es," Byzantion, 3 (1926), 123ff .
.. 6 IolAv avvrovWTaTOS mTpOS Tiis YiiS Ws elXEV ~~avaOTas •.• ~S6KEl 'l'SEyye<rSaI p"~aTa .... 'ICrKo>flos
~OAts rnl yovv 5tavacrras Kal TTlV KE'l'a;>.t)v flEflapTlIolAVTlV m TVYXclvovaav T41 EVO>W~'i' SlaflaOTclaas i3pexxiOVI
T41lolAv "TTAe(OT'i' IolAPEI TOV aW~aTOS "TTpo01\AO>Tal "TTclAlV Tfj yij, Tfj SE~lC~ SE XElpl TOUs oq>Sa;>.~oUs "TTVKva KaTa-
'lie;; .. ,. 'Ic.:>clVVTlS Se TO "TTapc!rnav ovS' clvaveVaat flEfloVATlTal, clAAcl . .• flaSE"'S <5:yav V"TTVWTTEIV lv TOVT'i'
SOKEi , •.• XVI.3f., ed. Downey. I am indebted to Professor Downey for permission to quote from his
translation.
" On the date, see S. Der Nersessian, "The Illustrations of the Homilies of Gregory of Nazianzus,"
nop, 16 (1962), 197.

269
 124 HENRY MAGUIRE

Mesarites. A mosaic in the cathedral of Monreale, for example, which probably

dates to around 1180, shows St. John crouching on the ground, ,vith his head
lowered, as if in illustration of the words of Mesarites, "John however does
not wish to look up at all, but ... seems to lie there in deep sleep .... "
(fig. 9),66 and a twelfth-century illustration in a lectionary preserved on Mount
Athos (I viron, MS 1, fol. 305 v ) shows St. James, in the phrase of Mesarites,
"supporting his head with his left arm," as he rises from the ground (fig. 10).67
These Middle Byzantine Transfiguration scenes may be contrasted with the
outstanding pre-iconoclastic example, the Justinianic sixth-century mosaic in
the apse of St. Catherine's church on Mt. Sinai (fig. 11). The Apostles here are
labeled, and we find that it is St. Peter who is depicted lying on the ground,
below the figure of Christ. James and John are shown kneeling on each side,
in entirely symmetrical poses. There is not yet the differentiation between
the attitudes of each of the three Apostles that was to appear in post-icono-
clastic versions of the Transfiguration, nor does this distinction between each
of the Apostles appear in any other surviving Transfiguration scene made
prior to the ninth century.
Here, then, Mesarites says something that could not have been said in the
sixth century. One might even go further, and say that he could not have
composed his account before the twelfth century, for it is only in this century
that we begin to find the Apostles in the precise poses which Mesarites
describes. '
The description of the mosaic of the Miraculous Draught of Fish provides
a different kind of eyidence that Mesarites based some of his remarks on his
own obseryation of the mosaics, and not merely on literary sources. For here
he has been ruled by the iconography of the mosaic, to the extent that he
alters the sequence of the Gospel story. Millet demonstrated this by comparing
the account in Mesarites with the fresco in the Miroz monastery at Pskov,
which dates to shortly before 1156 (fig. 12).68 In the Pskov fresco we see on
the left the boat with the disciples casting their net, in accordance with the
instructions of Christ (John 21: 6); then we see Peter in the water, swimming
toward the shore (verse 7). The figures on the right illustrate the next two
episodes of the story. First, Peter is represented going back to the boat,
and pulling the net full of fish to land (verse 11), then on the right, the Lord
feeds the fish and bread to the assembled disciples (verse 13). It is clear from
the description in Mesarites that the mosaic which he saw in the Holy Apostles
resembled this fresco quite closely. First, he describes the Savior commanding
the disciples to cast their net over the side of the boat, then Peter swimming
toward Christ. After this, Mesarites describes the feeding of the disciples
•• Artists from Constantinople probably worked at Monreale; E. Eitzinger, The Mosaics of Mon-
reale (Palermo, 1960), 75f.
.7 The comparison was made by K. \Veitzmann, "The Narrative and Liturgical Gospel Illustra-
tions," New Testament Alal1uscript Studies, ed. M. M. Parvis and A. P. 'Wikgren (Chicago, 1950),
164f.; reprinted in Studies ill Classical and Byzantine ]\~anuscript Illumination, ed. H. L. Kessler
(Chicago, 19'71), 261, fig. 249; for the date of the manuscript, see V. Lazarev, Storia della pittura
bizantina (Turin, 1967),252 note 51.
r.s Millet, Recherches sur I'iconographie de l'Evangile (supra, note 28), 574.

270
 BYZANTINE DESCRIPTIONS OF WORKS OF .-\RT 12;"i

(verse 13). Finally, he turns to the figure of Peter with the net (verse 11),
and says: "But Peter alone, since, I suppose, he received his food from the
hands of the Lord before the others, . " turns again, like a man eager for
activity and mindful of his work, and draws up the net from the sea, bracing
himself with his feet and grasping with his hands the fishing net, which is
completely full of great fish .. " "69 Mesarites goes on to give us his reason for
considering the disciples' eating and Peter's hauling of the net as contem-
poraneous actions, when a reading of the Bible would suggest that Peter
first finished hauling in the net, and only then received food from the Lord:
"And Peter turns his head toward his companions and fellow-workers, calling
to them, I suppose, to lay hold along with him and drag the net out to dry
ground, since he alone has not the strength to draw it out easily himself .... "70
It was, therefore, Peter turning his head toward the next scene, the Feeding
of the Disciples, that made Mesarites alter the biblical sequence of events.
In the Pskov fresco Peter turns toward his companions in just the manner
described by l\Iesarites. Here again, we must conclude that the description
of Mesarites is based on visual rather than literary data.
To this point I have considered only cases of simple description; I have not
yet touched upon the problem of the attitudes toward art displayed in Mesa-
rites's ekphrasis. It remains to decide whether all these attitudes are the
stereotyped responses of antique rhetoric, or whether there is something new,
and only appropriate to post-iconoclastic Byzantine art. In one passage
Mesarites shows awareness of one of the most important distinctions between
the aesthetics of post-iconoclastic art and that of the Christian art of preceding
centuries. In his description of the pose of Christ in the mosaic portraying the
Incredulity of Thomas, Mesarites writes that, as the Apostle stretches out
his hand to touch Christ's side, "the Savior ... depicts in His posture the
wounded man, and bends over and seems almost, so to speak, to fear the
touching of the scar. The hand of Thomas enters in at the side of the Savior
like some spear stretched out far and pressed against an unresisting body ....
The side seems to shrink from Thomas's continued handling of it, and wishes
to pour forth blood and water again .... "71
This description of Christ's attitude corresponds with a number of Late
Byzantine depictions of the Incredulity of Thomas. In the thirteenth-century
fresco at Sopocani in Yugoslavia, for example, Christ inclines his head sharply
toward Thomas, so that he appears to be bending his body away from the
Apostle's touch (fig. 13). An icon of the fourteenth century, which is now in the
" niTpos OE ~ovos, ells ol~al TIPO TIClVTWV Aa~ellv EK -rwv OECTTTO-rIKWV XElPWV TIlV -rpoq>i]v ... aywvlCITIKOS
w<TlTfpEi 'T1S Koi E~'ilpOVTlS epyov KaL rraAlV EX€Tal Ked Tiis .$aAO:-rrllS clVEAKEl TO 6hnvov, ToiS ~EV TIocriv CcVTl[3o:ivwv
TIPOS Eavr6v, -rais
ed. Downey.
0, XEpai 'TiiS IxSvaypas ETIElA'l~~EVOS, 'lTA'lPECITCrr'lS ova'lS IXSvwv ~EyaAwv .... XXXVI.4,

7. a-rpEq>EI OE niTpo, TIlV KEq>OAi]v TIpO, 'TOU, eavroO C'V~~va-ro, Kai 6~0'TEXVOVS, 'ITpOC'KOAOU~VO) ol~Ol
TOV-rOVS Eq>6:ljIoaSoi 01 'TiiS 'TOV OIK-rVOV TIPOS -rDV ~'lp6:v E~OAKiis -re;; ~i] oEovviiaSOI ~6vos 'TOUTO pq:OTEPOV
E~"'KiiaOI •... XXXVI.5, ed. Downey.
71 6 oe O'wniP EO-Xl1~crTIO'I.lEVOS TOV Tpavl-laTiav Kat rrpas EO\J"TOV crvvILavwv KCXt TIEpi T1iv Tfis WiEIATlS ava01<O:-
AEvalV olov EhrEiv OEOl'T'T6~vos. ElaovvEl 'ITEpi -rDV -rOV aW'Tiipos 'ITAEVpaV " XEip 'TOV ew~i3: KoSompEi 'TIS AOYX'l
~aKp6SEV !KTETO~EV'l Koi 'ITpO, aw~" impEtaSEiao EVEVSO'TOV .... 'H 0' 'ITAEVpa olov a'ITOSAl~O~EV'l EK Tij, -rov
ew~a av;(viis ETIoq>iis KO\ nOAlv oi~o Koi uowp pEvcrm ~E~OUA'l-rOI .... XXXIV.5-7, ed. Downey.

271
126 HENRY MAGUIRE

monastery of the Transfiguration in the Meteora in Greece, is a still clearer

illustration of the text of the ekphrasis (fig. 14). Here the Lord bends his
whole body sharply, drawing his side away from Thomas' hand, in a manner
which appears to fit closely the description by Mesarites. 72
Among the surviving Byzantine portrayals of this scene which antedate
the ekphrasis of Mesarites, there are no instances in which Christ assumes
just this posture. Generally Christ stands straight upright, as in the eleventh-
century mosaic from Hosios Loukas in Greece. 73 It is very likely, however,
that the iconographical type of the Meteora icon and the Sopocani fresco
existed in Byzantine art before 1200, even if no examples of it have survived
to this day. First, we have the evidence of a miniature in a late twelfth-century
Coptic Gospel book, now in the Bibliotheque Nationale in Paris (copte 13,
fol. 278"; fig. 15). It shows Christ supporting himself by laying his right arm
over Thomas' shoulders; at the same time he seems to be trying to draw
his side away from the Apostle's touch, and thus stands at what appears to
be a some\\"hat awkward angle. 74 It is probable that this Coptic miniature is
a copy of a more elegant Byzantine model. There is also the evidence of a
tenth-century ivory, which is at Dumbarton OakS in Washington (fig. 16).
Here Christ is turning his side away from Thomas' touch, and at the same
time he inclines his head toward the Apostle. Thus, we have the same elements
in the tenth-century ivory which appeared, vastly exaggerated, in the four-
teenth-century icon. These examples allow us to conclude that the mosaic
in the Holy Apostles which Mesarites described showed Christ with the upper
part of: his body bending down, and his side curving away from Thomas'
hand.
In his description Mesarites says that Christ "depicts in his posture the
wounded man," and that the "hand of Thomas enters" his side "like a spear,"
and that the side of Christ "~ishes to pour forth blood and water." Clearly
Mesarites is making a visual comparison between the pose of Christ showing
his wound, and that of Christ receiving his wound on the cross. This compari-
son is appropriate only to post-iconoclastic Crucifixion scenes, because only
in this period did artists depict the crucified Christ as a wounded man, with
his body bent and slumping on the cross. On some occasions pre-iconoclastic
artists showed Christ inclining his head slightly, as in the miniature from
the Syrian Rabula gospels of the sixth century (fig. 17). But here, as in all
other pre-iconoclastic examples of the Crucifixion, Christ's body is erect. It
is an image which emphasizes Christ's triumph over death, rather than his
suffering humanity. The earliest surviving Byzantine work of art from the
milieu of the capital which portrayed Christ on the cross as a dead man seems
to belong to the ninth century. In a monastic psalter in the Pantocrator
monastery on Mount Athos (codex 61, fol. 98) there is a miniature which

"K. Kreidl-Papadopoulos, "Die Ikonen im Kunsthistorischen Museum in \vien," jbKsWien, 66

(1970), 55, fig. 37.
73 E. Diez and O. Demus, Byzantine Mosaics in Creece (Cambridge, Mass., 1931), fig. 10.
,. Millet, Recherches SUI' I'iconographie de I'Evangile, 578, fig. 631.

272
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 127
depicts Christ on the cross, with his eyes closed. 75 Here, as well as in all other
surviving ninth-century Crucifixion scenes, Christ's body is still upright. But
by the second half of the tenth century, Byzantine ivory carvers were depicting
Christ with his whole body sagging.76 In the eleventh- and twelfth-century
Crucifixion scenes, such as the famous mosaic at Hosios Loukas, the curve
of the body becomes more pronounced (fig. 18). Only an image of this type,
with Christ's body sinking down and his side curved, would fit the description
which Mesarites gives us of Christ's posture in the mosaic of the Incredulity
of Thomas. When Mesarites, in describing the Incredulity of Thomas, visually
recalled Christ's posture in the Crucifixion, he must have had in mind Cruci-
fixion scenes of the post-iconoclastic type, dating to the tenth century or
later. This passage of his ekphrasis, therefore, shows an appreciation of one
of the most important innovations of Middle Byzantine art, the portrayal
of Christ suffering on the cross as a man.
The passages which I have quoted demonstrate that the ekphraseis were
often accurate and in touch with contemporary developments in art. There
was more in them than a dead literary tradition. They reflected not only
changes in iconography, but also the new attitudes which lay behind the
changes. On the other hand, it cannot be denied that the literary element was
strong. Before we can fully assess the ekphraseis as evidence for art history,
we will have to look deeper into their literary aspects and identify common
quotations and conceits in order to find out how the use of topoi affected the
validity of Byzantine writing on art.

II. THE USE OF TOPOl AND THEIR EFFECT ON ACCURACY

From antiquity to the fifteenth century the ekphraseis devoted to works

of art formed a continuous tradition in Greek literature. 77 Nearly all of them,
to some extent, copied each other; there is no need to look further than the
introductions of the ekphraseis to find examples of repetition. The prefaces
often explain the author's purpose in writing the description, and contain
general observations on the visual arts. The innocent reader might hope that
they would give a picture of changing attitudes toward art throughout the
centuries. Unfortunately, they are completely stereotyped. Thus, in'the
twelfth century, Constantine Manasses prefaces his ekphrasis on a mosaic in
the Great Palace at Constantinople ,vith the statement that painting is superior
,. ]. R. Martin, "The Dead Christ on the Cross in Byzantine Art," Late Classical and Mediaeval
Studies in Honor of Albert Mathias Friend, Jr., 189f1., pI. 23, 4. The earlier examples from Sinai have
been discussed by H. Belting and C. Belting-Ihm, "Das Kreuzbild im 'Hodegos' des Anastasios
Sinaites," Tortulae. Studien zu altchristlichen und byzantinischen Monumenten, ed. W. N. Schumacher
(Rome, 1966), 30ft.
"See, for example, Goldschmidt and Weitzmann, Die byzantinischen Elfenbeinskulpturen, pI. 39,
no. 10l.
"On the ekphrasis as a literary form, see L. Meridier, L'influence de la seconde soPhistique sur
l'auvre de Gregoire de Nysse (Paris, 1906), 41ft.; G. Downey, "Ekphrasis," RAC, IV (1959). 921ft.;
G. Pfohl, "Monument und Epigramm," 75 Jahre Neues Gymnasium Nurnberg, Festschrift (Wiirzburg,
1964), 1ft.; A. Hohlweg, "Ekphrasis," RBK, II (Stuttgart:1971), col. 33ft.; Baxandall, Giallo and
the Orators (supra, note 6), 78f.

273
 128 HENRY MAGUIRE

to sculpture. Unlike the sculptor, the painter can reproduce color and shade,
and is thus able to portray " ... the roughness of skin and every kind of
complexion. a blush, blond hair, a face that is dark, faint, and gloomy, and
again one that is sweet, comely, and radiant with beauty .... "78 This passage
is inspired by the ekphrasis of the third-century sophist Philostratus the
Elder. In the introduction to his Imagines there is a similar passage on the
superiority of painting to sculpture, in which the painter's ability to reproduce
shade and color is extolled. The painter, says Philostratus, can imitate the
look of a man who is mad, suffering, or joyful. He can differentiate between
eyes of different colors, and likewise with hair and clothing. 79 The opinions of
Philostratus the Elder on this subject were repeated in the introduction of his
grandson's ekphrasis. 80
The untrustworthiness of the statements which preface the ekphraseis is
demonstrated by a remark which I have noted in the introduction to Mesarites'
description of the interior of the church of the Holy Apostles in Constantinople.
Mesarites, writing around the year 1200, declares that he is the first to describe
the church, even though both Procopius and Constantine the Rhodian had
already written on the building, the former in the sixth and the latter in the
tenth century. Furthermore, Constantine the Rhodian had made the same
claim, saying that he was the first to speak on the church. 81
When the reader of an ekphrasis passes from the introduction to the descrip-
tion itself, he will usually find a large number of topoi which are variations
on one theme, the realism of the work of art. The author may, for example,
declare that the art is so realistic that he has forgotten it is art and not real
action. The third-century sophist Philostratus the Elder, describing a painting
of a boar hunt, claims, "I was carried away by the painting, thinking that the
figures were not painted, but existed and moved .... "82
The same remark is made by Procopius of Gaza, in the late fifth or early
sixth century, in the course of his' description of a painting of Phaedra. 83
And the Byzantine writer Mesarites, describing the mosaics in the church of
the Holy Apostles in Constantinople, twice reminds us that "the events that
we now observe are not real life, but a picture. "84
This sentiment reflects a tendency in the ekphraseis for the authors to go
beyond the works of art themselves, to a fuller narration of the stories which
inspired them. The first example of this technique in Greek literature is

78 ••• TpaxlrrrjTa Oip~CTOS Kal XP6av '!Ta\I'To5a'!T~ IpvSl'J~a Te Kai K6~l'Jv ~avS"v Kai '!Tpo""mov KC'TTVl'JPOV
Kai WPaK'OV Kai envyvov Kai a6S,s /jov Kai xaplEv Kal cnl;'flov 'l'c;'> Ka;';'El ..•. Ed. Sternbach, "Beitrage
zm Kunstgeschichte" (supra, note 16). col. 75, line 16ff.
,. Imagines, 1.2.
80 Imagines, Prooemium 3. This tapas also appears in the writings of Leonardo da Vinci; ed.
H. Ludwig, Das Buck von del' Malerei, I (Vienna, 1882). part I, 38.
81 XI1.12. ed. Downey, and ibid., p. 860; ed. Legrand "Description" (supra, note 58), v. 412; Pro-
copius, De aedificiis, 1.4.9f.
"E~T]XST)V V'!TO T~S ypa,!,~s ~" y'yp6cpSat OOKWV aliTovs, erVat OE Kai K,veicrSat .... Philostratus,
Imagines, 1.28.2.
83 Descriptio itnaginis, 17, line 163f., ed. Friedlander (supra, note 36).
84 00 yap Iv TTp6y~a",v aM',v ypa~~a", Ta vvv TIpOS ,,~wv KaSopw~eva, XXX.2 and XXXIV.8, ed.
Downey.

274
to
•-J

275
ΟΙ

2. F lorence. Bibl. L iuiienz.. P lu t. 1.56, D etail

'"
....
;:..

1. Monza C athedral. Phial

T h e N a tiv ity
276
tv
-.....J
-.....J

4. Home, SS. Coslna e Damiano, Apse 5. Vienna, Kunsthistorisches Museum,

I vory of an Empress
6. Berlin, Ehemals Staatliche Museen. Ivory of the Forty Martyrs

7. Rome, S. Maria Antiqua, Chapel of the Forty Martyrs, Apse

278
 t< 11f (j b 't 'iH Ί ’<' rV /, t //«.* 1
iiY t/r r r t't( U \ W v 4 /
iH t K ift t H ft i- t n i ϊ/n V ' т Я Й К 5 Г І 6 Ш Ш ! з т о м ш ;
!/ti/r<X\vivunt/* . .

279
9. M onreale. Mosaic

8. P aris, Bibl. N at., gr. 510

T he IV ansfiguration
 Й т < А т ш S ιί ·;Κ i U К Ш

10. Mt.
Mt.Athos,
A thos,Iviron,
Iviron,?lIS
MS1 1

w U V ti
І |№ й >
І Ш Ш Щ Ш Ш т

11. Mt.
M t. Sinai,
Sinai, Monastery
M onaster}1ofof St.
St. Catherine.
Catherine. Apse
Apse Mosaic
Mosaic

The
T Transfiguration
he T ransfiguration
280
12. Pskov, Miroz Monastery. Fresco, the Miraculous Draught of Fishes

13. Sopocani. Fresco, the Incredulity of Thomas, Detail

281
-'"c..'"
0
U

"cil
z
:E
~
<n'
'C
ct!
p,.,
vi
.,...

282
17. Florence, Bib!. Laurenz., Pluto I.56

18. Hosios Loukas. Mosaic

The Crucifixion
283
19. Ravenna. S. Apollinare Nuovo. Mosaic. the Betrayal

20. Boston. Museum of Fine Arts. Tapestry

284
21. Paris, Bib!. Nat., gr. 550, St. Gregory of Nazianzus

285
286
22. D ap h ni. D om e M osaic, th e P a n to crato r
24. Paris,
Paris,Bib!.
Bibl.Nat.,
N a t.,gr.gr.74,
74,fo!. 61v,Detail,
fol.61v, D etail,the
th eApostles
Apostlesononthe
th eWay
W aytotoGalilee
Galilee

25. Asinou.
A sinou.Fresco,
F resco,the
th eVirgin
Virginand
andChild
Child

287
26. Monreale. Mosaic, Christ Walking on the Water

27. Chios, Nea Moni. Mosaic, the Anastasis

288
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 129
Homer's description of the shield of Achilles. 85 According to Homer, Hephaestus
managed to represent on this shield a prodigious wealth of detail. Homer
gives detailed descriptions of scenes of weddings, of a trial, and of war, as well
as scenes of ploughing and of harvesting, of vineyards and of pastures. The
shield also showed a dance, and depicted the sea and the heavens. In later
ekphraseis one often finds, likewise, that the author elaborated on the inven-
tions of the artist. Often he heightened the realism of the drama by providing
speech for the mute creations of the sculptor or painter. The tenth-century
writer Constantine the Rhodian, for example, concluded his ekphrasis on the
mosaics in the church of the Holy Apostles with the lament of the Virgin,
delivered at the scene of the Crucifixion. 86 The mother mourns the loss of her
only son, and expresses her faith in his triumph. The form of the speech was in-
spired by the apocryphal Acts of Pilate, in which there is a similar threnas. 87
References to works of art which were so close to nature that they might
be capable of speech are very common in the ekphraseis; we find the conceit
in a sixth-century description by Choricius of an Annunciation mosaic in the
church of St. Sergius at Gaza: "A winged being has just come down, by the
painter's art, from heaven, and approaching the Virgin Mother. .. he greets
her with the good news .... He is posed' as if he were talking to her, but even
if the painter had given him a voice, it would not be easy to hear what was
said, for the intervening distance is large."88 The Byzantine Emperor Leo VI,
in a sermon on the decoration of a church founded at the end of the ninth
century by the official Stylianus, gives a description of an Annunciation
mosaic which echoes that of Choricius: "Here is a winged being just descended
from heaven, who converses with a young Virgin. You would say that the
representations are not deprived of rational discourse. The artist has diffused
so natural a color and character on the faces of the actors that he gives the
spectator a sensation similar to that produced by the sound of voice."89 A
similar statement is found in a Greek ekphrasis in a homily by the twelfth-
century South Italian preacher Philagathus. Here the subjects of the descrip-
tion are the bereaved mothers in a painting of the Massacre of the Innocents:
8. Iliad, XVIII, v. 475f£.
.. Ed. Legrand, "Description," v. 946ff., p. 64f.
81 Acta Pilati, part I, B, chap. X.4, in C. Tischendorf, Evangelia apocrypha (Leipzig, 1853). There
was a tradition of such funerary laments in Byzantine homiletic literature: Gregory of Nyssa, De
hominis oPificio, PG, 44, col. 217D (lament of the the Widow of Nain); Simeon Metaphrastes, S. Mariae
planctus, PG, 114, col. 212; "Theophanes Cerameus," De filio viduae, PG, 132, col. 224Df. (= Hom.
VI, 8-12, ed. Rossi Taibbi [supra, note 15], 40-42). For the lament as a rhetorical form, consult
Meridier, L'infiuence de la seconde soPhistique, 226ff., 270ff.
88 • APTI yap Tt, tl'rro'1TT£po, E~ oupavoO KaTaj3a, Til> l.",ypa<pcp Kat <pol'n'icra, 'ITapa T1'jv avev crvvO!KOV
IlT]TEpa ... TOi, eVayyeAiol, a<1'ITaleTal .... Kat O'J(T]llaTi1.ETal IJfv ola TI, TIPO, TaVTT]v Sla7lryollEvo;, 0:Ar.a
yap Ei Kat 'P"'VI]V EviST]KEV 0 (",ypa<po; aVTil>, ou PG<SIOV Ta 7IryOllEva - 'ITOA\! yap TO ",0'0 V- &KOVeIV. Laudatio
Marc. (supra, note 18), 1.48 .
.. • ApTI IJfV yap KOPlJ 'ITapSEvcp WO'ITTEPOS Tt, crrr' oupavoO KaTtWV SlaAEyETal' ei'ITol, av Kat AOYlKij, 1'1]
al'olpeiv Ta EiKoviO'llaTa SlaAi~e",,· OUT"'S E'ITt TWV 'ITpoO'om",v <pVO'IKOV aVToiS " TE)(viTT]S xpwlla Kat fiSo,
avE.9T]KEV, C\ miSel TOV SeaT1]v tl'ITOAallj3avelV WS apa Tt, crvvaicrST]O'IS Myov OUT"', aUTa XP"'l'aTi(el. Logos 34,
AeoVTo, TOO LO<pOO 'ITavvY'lpIKot [sic] MyOl, ed. Akakios (Athens, 1868), 277; trans. A. Frolow, "Deux
eglises byzantines," EtByz, 3 (1945), 43 £. Four poems ascribed to the twelfth-century poet Theo-
dorus Prodromus which describe another speaking image of the Annunciation are quoted by Bees,
"Kunstgeschichtliche Untersuchungen tiber die Eulaliosfrage" (supra, note 62), 103£.

289
130 HENRY MAGUIRE

"Although the artist was not able to imbue his colors with a voice, he signified
the laments in his drawing. "90
There were many other topoi concerned with realism. The persons or animals
depicted in works of art were often said to be breathing, or to be about to
move. There is an epigram, for example, in the Greek Anthology which
instructs strangers not to touch a stone statue of Ariadne sleeping, lest she
wake and spring Up.9! Another epigram, by the eleventh-century poet Chris-
topher of Mytilene, warns the spectator not to approach a certain bronze horse
in the Hippodrome, lest he be trampled under the animal's upraised hoof. 92
Some images were reputed to be so realistic in their beauty that they
inspired love in the beholder. Both Christodorus of Thebes, writing before the
destruction of the Zeuxippus Gymnasium in 532, and Nicetas Choniates,
writing after the sack of Constantinople in 1204, make this claim for statues
of Helen. 93 A pair of Hellenistic epigrams in the Greek Anthology record that
the bronze heifer made by Myron attracted the attention of a bull. 94 As we
have seen, this information was repeated in the mid-twelfth century by the
Byzantine ecclesiastic Constantine Manasses. 95
If the image depicted a divinity, the writer might declare that the artist
himself must have seen the god in order to produce so faithful a likeness.
This sentiment was applied both in the pagan and in the Christian context. 96
This survey has shown that references to realism abound no less in descrip-
tions of Byzantine art than in those of classical art. Yet, as I have noted in
the introduction, modern admirers of Byzantine art are keenly aware of the
strong element of abstraction which separates Byzantine works from the
masterpieces of Hellenistic and Roman art.97 It is no wonder that it has been
asked whether the ekphraseis are truly relevant to an understanding of
Byzantine art, or whether they are entirely the products of a dead literary
tradition, out of contact with the innovations made by artists. One answer
to this question is to suppose that Byzantine writers really did consider
contemporary works to be "realistic," because their expectations in this
respect were lower than ours are today. The modern critic can compare
Byzantine painting with the greater illusionism of Hellenistic and of Renais-
sance art. But the range of reference of the Byzantine viewer was more
limited. 98
•• Kal bm51'] >11'] ElXEV " TExviTlls cpwvf]V tvSEivat ToiS XPw>lacTtv, tCfTl>laVE TOUS SPTlVOVS Tois yp6:>I~aCftv.
"Theophanes Cerameus" (d. note 14a supra), Homilia LII, In sanelos InnocenlBS, PG, 132, cols.
924B-C (= HomiZia XXIV.I0, ed. Rossi Taibbi [supra, note 15), 159) .
.. Anthologia Palalina, XVI, no. 146 .
•• Ed. Kurtz (supra, note 49), 30, no. 50 .
•• Christodorus of Thebes, Anlh. Pal., II, v. 168f.; Nicetas Choniates, Narratio de staluis, PG,
139, col. 1052D .
•• A nth. Pal., IX, nos. 730, 734.
o. Ed. Sternbach, "Beitrage zur Kunstgeschichte," col. 75, line 24. The source of this tapas is
presumably the legend of the cow made for Pasiphae as a lure for the bull of Minos .
• 0 Mango, "Antique Statuary," 66f.; An/h. Pal., XVI, nos. 81, 162 (statues of Zeus and Aphrodite);
A. Papadopoulos-Kerameus, NtKT)cp6poS K6:AAtCfTOS zavS6;rovAos (supra, note 61), no. 14 (mosaic of Christ).
" See notes 2-5 supra.
o. A discussion of the "horizon of expectation" in art is found in E. Gombrich, Art and Illusion
(London, 1968), 53.

290
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 131
It should also be remembered that many of the illusionistic devices invented
in antiquity survived in Byzantine painting, even if they were used less
consistently. An example is the trick of showing one form overlapping another,
so that there appears to be a recession in space. This arrangement is commended
in the third-century ekphrasis of Philostratus the Elder. In describing a
painting which shows a siege of Thebes and an army surrounding the walls
he says that the artist represents" ... some men complete, some with the
legs obscured, some half hidden, and just the busts of some, and heads only,
and helmets only, finally just the spear points."99 By this means, says Philo-
stratus, the painter obtains an effect of recession. This description recalls the
depictions of annies on Roman triumphal monuments, in which the back-
ground figures are indicated by glimpses of shoulders, heads, and spears which
appear over the tops of the soldiers in the foreground.1° o The technique of
overlap survived in Early Christian and Byzantine art. In the sixth century
Choricius comments upon it in his ekphrasis on the mosaics of St. Sergi us.
He tells us that in the scene of the Raising of the Widow's Son the mourning
women, "being grouped together, block each other partially from view. But
you might suppose that if you set them apart from each other, each one
had been painted completely .... "101 Again, we can find parallels for this
description in contemporary works of art. In the sixth-century mosaics of
Sant'Apollinare Nuovo and of San Vitale in Ravenna, for example, groups of
figures which overlap each other are frequently represented (see fig. 19). The
device was also described by another orator of Gaza, Choricius' instructor
Procopius. He is talking of an attendant in a painting of the palace of Theseus:
"Screening himself behind one of the columns he divides himself, hiding the
lower part of his body, so that he is only half visible. And winning an easy
flight from his master's attention, he thrusts out his head as he looks out,
and his hand as he moves."102 A late antique textile from Egypt in the Museum
of Fine Arts at Boston shows a man drawing back a curtain which hangs
between two columns, so that, like the palace attendant described by
Procopius, he is partially screened from view (fig. 20) .103
These correspondences between descriptions and works of art demonstrate
that passages in the ekphraseis which had literary antecedents may still
have been relevant to contemporary art. A tapas was not necessarily untrue.l° 4

..... "OUs Ilkv ap,.{ovs lTapExEt 6pcrv, ,.olls Be aaa<pEiS,.a O1<EA'1, ,.olls Be l'J><IClEas Kat aT'pva !v'wv Kat KE<paAaS
Ilovas Kat Kopv.$as ~ovas, Eha alx~&s. Imagines, 1.4.2.
, •• See, for example, a battle relief from the "Great Trajanic Frieze" on the Arch of Constantine;
D. E. Strong, Roman Imperial Sculpture (London, 1961), pI. 75.
,., Kal avv'1~llkvat Ilkv!s ,.a"'ov &IJ\';AatS y(voVTat KWAVIl" ,.ou ~l'J TEMws o<p.$fivat· IITTOAa~OtS B' O:v MaS,
EI BtaaTtiClEts &IJ\';AWV, c3AI1V EKa"'l1V yeypa<p.$at .•.. Laudatio Marc., 1.63.
102 ITpoKaA\Jl.Illa Be Ttva ,.wv Ktevwv lTOtWv Ka\ lTPOS EK<lTEP" IlEP'a"s MOV, ,.a Kerrw ..ev &1TOKp.:nn-.t ,.OU
aw~CXTOS, l'J~tTEAl'JS B, TtS <patvollEvOS Kat p"Blav £CXVTC;; Tl'Jv <pvyl'JV!K "fis ,.ou 8Ea1Te,.ov lTpOaAexyxavwv ala.$,,-
ClEWS, lTpoi3aAAet Tl'Jv Ilkv KE<paAl'Jv lTPOs .$eav, Tl'Jv Be xe1pa 1rpOs Klvl1atv. Descriptio imaginis, 14, line 131 ff.,
ed. Friedla.nder.
103 L. Salmon, "An Eastern Mediterranean Puzzle," Boston Museum Bulletin, 67, no. 350 (1969),
136ff., figs. 1, 1t.
,., The same point has been made with regard to conventions of rhetoric by 1. Sevcenko, Etudes
sur la polimique entre TMod01'e Metochite et Niceph01'e Choumnos (BrusselS, 1962), 171 note 2.

291
132 HENRY MAGUIRE
If the ekphraseis were strongly bound by tradition, Byzantine art was no
less so, and many of the illusionistic practices of antiquity survived into the
Middle Ages.
The same point can be demonstrated in the case of another well-known
tapas, the statement that the artist has managed to combine two or more
contradictory emotions in one figure. This tapas has had a long history, from
antiquity to the present day. lOS According to Pliny, the standard of accom-
plishment which later artists had to emulate was set in the fourth century B.C.
by the painter Parrhasius. Pliny, writing in the first century A.D., tells us that
this artist painted a personification of the Athenian Demos, showing it simul-
taneously as " ... variable, irascible, unjust, inconstant, but also placable,
clement, and merciful; it was boastful, ... lofty and humble, fierce and
timid, and everything at the same time."I06 It was more usual, however,
for writers to consider the conveyance of only two conflicting emotions in
the same character a triumph of realism. Thus Philo stratus the Elder praises
a painter for showing Apollo's anger at the thefts of the infant Hermes giving
way to amusemenP07 The second-century novelist Achilles Tatius wrote an
ekphrasis on a painting of the rape of Europa, in which the aspect of Europa's
companions was said to be made up of both joy and fear.10 8 There are also many
epigrams in the Greek Anthology which speak of images showing opposite
emotions. Damocharis says that a painting of Sappho was both gay and
intellectual in its facial expression.109 There is a sequence of poems, by different
authors, which describe images of Medea on the point of slaying her children.
Several say that her expression combines the jealousy and rage she harbors
on account of her husband with the tenderness she feels for her children. llo
This stereotyped description of Medea was borrowed by a Christian bishop,
Asterius of Amasia, who lived in the fourth and fifth centuries, in an ekphrasis
devoted to a painting of the martyrdom of St. Euphemia. Asterius says that
he had previously admired paintings which portrayed the combined rage and
pity of Medea. Now he turns his admiration to the representation of St.
Euphemia, in which are combined the modesty and the courage of the virgin
martyr.lll
It is small wonder that this often repeated cliche received the scorn of
Sir Joshua Reynolds. He remarks that Pliny " ... observes that in a statue
of Paris, by Euphranor, you might discover at the same time three different
characters; the dignity of a judge of the goddesses, the lover of Helen, and the
conqueror of Achilles. A statue in which you endeavour to unite stately dignity,

10. Gombrich, Art and Illusion, 293, gives a recent example.

106 • •• varium iracundum iniustum inconstantem, eundem exo1'abilem clementem miseric01'dem;
gloriosum ... , "fcelsum Izumilem, ferocem fugacemque et omnia paritcr. Naturalis his/aria,
XXXV.69.
107 Imagines, 1.26.5.
108 Leucippe and Clitophon, 1.1, 6ft. See also the account by Callistratus of a statue of Memnon
which expressed both grief and joy: Descriptiones, 9.
10. Anth. Pal., XVI, no. 310.
110 Anth. Pal., XVI, nos. 135, 136, 138, 139, 140, 143. See also Callistratus, Descriptiones, 13.2.
111 In Laude", S. Euphemiae, PG, 40, col. 337A; Mango, "Antique Statuary," 65.

292
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 133
youthful elegance, and stern valour, must surely possess none of these to any
eminent degree."1l2
But in spite of the strictures of Reynolds, there are, in fact, certain com-
monly employed artistic devices which correspond to this tapas. One is the
practice of giving a different shape to the features of each side of the face,
particularly to the eyes, even when the pose is frontal. In a few instances
the descriptions in the ekphraseis appear to refer directly to this device,
although it is obviously not a convincing basis for the more elaborate passages
on mixed passions. Manolis Chatzidakis has drawn attention to a word portrait
of St. Gregory of Nazianzus, written by Elpios in the ninth or the tenth
century, which describes the look of the Saint as gentle and kindly, but of
his right eye as sullen, as it was contracted by a scarY3 In a portrait of the
Saint which adorns a twelfth-century manuscript of his Sermons, the disparity
in the sizes of the eyes is very striking, the right one being the smaller (fig.
21).114 Here, there does appear to be a close correspondence between a literary
tapas and artistic practice.
The ultimate literary source of the description of St. Gregory of Nazianzus
by Elpios is probably the account writt.en by Chrysippus in the third century
B.C. of the figure of Justice, as she was usually depicted by painters and
orators. She was said to inspire fear in the unjust and courage in the just;
to the latter her face was friendly, to the former it was hostile.!15 This charac-
terization was applied by Byzantine historians and eulogists to imperial
persons. U6 It was also used by Mesarites, around the year 1200, to describe the
mosaic of Christ Pantocrator in the Holy Apostles. We are told that "His
eyes, to those who have achieved a clean understanding, are gentle and
friendly and instil the joy of contrition in the souls of the pure in heart ....
To those, however, who are condemned by their own judgment they are
scornful and hostile and boding of ill .... "117 In both the pre-iconoclastic
and in the post-iconoclastic period we can find parallels in works of art which
seem to justify the use of this tapos by Mesarites. In the original late eleventh-
century mosaic of Christ Pantocrator at Daphni there was a striking asymmetry
in the treatment of the eyes and cheeks, and it was partly from this asymmetry
that the image derived its power (fig. 22).118 A pronounced disparity between
the eyes can also be found in certain pre-iconoclastic bust portraits of Christ,
such as an encaustic icon at Mount Sinai, perhaps of the sixth century, and
112 Discourse V (London, 1772).
118 ';lJl'pov J3AE-rrWV Ka\ 1TpOcrT]VE;, S6:Tepov TWV 6q>SaA~wv, Os i'jv S'~I6s, o-rvyv6npo<;, OV Ka\ OVAI') Kena
TOV KavSOv avvfjye' M. Chatzidakis, "An Encaustic Icon of Christ at Sinai," ArtB, 49 (1967),
200 note 11.
116 Paris, Bib!. Nat., gr. 550, fol. 4.
110 Quoted by Aulus Gellius, Noctes Atticae, XIV.4.1f.
111 For example, by Anna Comnena in descriptions of Alexis (Alexiad, III.3.2), and of Anna Dalas-
sena (ibid., 8.3); so also "Theophanes Cerameus" (Philagathus) in an address to the King of Sicily,
Homilia LV, PG, 132, col. 952B (= Homilia XXVII.1, ed. Rossi Taibbi, 174).
111 0\ 6q>SaAIl0\ Tol, &Kcnayvwcrrov KEKTTJ~VOl; TO avVElSO; IAapo\ Ka\ e\rn'p6alTOl Ka\ YAVKaallOV Kena-
vV~W; !vcrraArnOVTE; Tal; ljN)(al; TWV KaSapwv Tij KapSic;x .... 01; S' aVToKen&KplTov T6 oiKElov KPIT11PIOV,
6PyiAOI SVa1Tp6aITO( TtV,; Kal SVaCx1lT1lTOI .... XIV.3ft., ed. Downey.
118 The asymmetry is less evident in the restoration. which, fortunately, took place after the mosaic
had been photographed. G.lllillet, Le monastere de Daphni (Paris, 1899), 105, fig. 48.

293
 134 HENRY MAGUIRE
the images on a group of solidi of Justinian II (type A) which date between
692 and 695.119 In this case, also, artistic and literary tradition appear to
have run a parallel course.
On some occasions, however, the conventional tapas of the combination
of contrary emotions could lead a writer into a false or inaccurate description
of a work of art. A case in point is a homily delivered in 867 by the Patriarch
Photius, in the church of St. Sophia in Constantinople, which vividly describes
a portrayal of Mary carrying her Child: "A virgin mother, with a virgin's
and a mother's gaze, dividing in indivisible form her temperament between
both capacities, yet belittling neither by its incompleteness. With such exacti-
tude has the art of painting, which is a reflection of inspiration from above,
set up a lifelike imitation. For, as it were, she fondly turns her eyes on her
begotten Child in the affection of her heart, yet assumes the expression of a
detached and imperturbable mood at the passionless and wondrous nature of
her offspring, and composes her gaze accordingly. "120 It is almost certain
that the homily of Photius refers to the mosaic now to be seen in the apse of
St. Sophia (fig. 23).l21 This gives the ekphrasis a particular importance, as it is
one of the few which can still be checked against their SUbjects. In the mosaic
the Virgin appears frontally, holding the Christ Child on her lap. She stares
out into space, an attitude which would correspond with the expression of
detachment described by Photius. But there is no indication in the mosaic
that the Virgin is looking down toward her offspring, as the words of Photius
also suggest. It appears that Photius exaggerated the element of maternal
affection in the image, according to literary convention. For other Byzantine
writers as well speak of Mary combining the expression of a virgin and of a
mother. The description was applied by Leo VI, also in the ninth century,
to a mosaic in the Kauleas monastery.122 It appears, too, in a poem on a
painting of the Virgin which was written by Manuel Philes in the Palaeologan
period. 123
The commonplace that two or more emotions could be represented in one
figure has had a long history, and the tapas has taken many forms. It was
used by the writers of the ekphraseis both aptly and inaptly. But in itself
it does not constitute a falsehood.
In addition to repeating topai concerned with realism, Byzantine writers
inserted into their descriptions tags quoted verbatim from well-known classical

11' Chatzidakis, op. cit., 200, figs. 3 and 10.

12' napsEVOS ~i]TT)P, 'ITapSEvov &~a Kat ~T)TPIKOV 6pwaa, Kat 'ITpOS &~cpw TaS axiaEIS tv o~piCJ'r'i> axi]~an
~pl[O"'VT) TO 130\lA11 "a, Kat "116hEpov "ipOS T<;; <h'Ml '~vl3p{[ovO"a. 'YTIOKPiCnS apa Ti'jS avwSEV 'TImvoias i]
[wypa'l'OS TExVT), OlrrWS clKpll3ws els cpvalV Ti]v ~i"T)alv EO"'rT)CJEV. Kat yap oiove! Tfj "Ev CJ'ropyij TWV O"'ITAO:yxVWV
Ti]v O'+'IV 1Tpb, TO T'XS!V O"W1TaSw, ETrlCJ'rp''l'ovO"a, ala 6E TiiI o1ToSel Kat vmp'l'v,l '1'OV '1'OKOV ,\, aCJ)(E'rOV a"o
Kot &rapoxov ap"OLO~VT) KO'1'aO"'rT)~a SlaSEO"'''', 'ITapa'ITI,T)O"i",S 'l"PEI '1'0 0",,0 ax11"O'1'ILO",VOV. Homilia XVII,
ed. S. Aristarches (Constantinople, 1900), II, 299; ed. B. Laourdas (Salonica, 1959). 167; C. Mango,
The Homilies 0/ Photius (Cambridge, Mass., 1958), 290. I am indebted to Professor Mango for per-
mission to quote from his translation.
121 C. Mango, E. J. W. Hawkins, "The Apse Mosaics of St. Sophia at Istanbul," DOP, 19 (1965),
115ff., esp. 142f.
122 Logos 28, ed. Akakios (supra, note 89), 246; Frolow, "Deux eglises byzantines," 47.
123 Manuelis Philae carmina, ed. Miller, I (Paris, 1855). 317, no. CXXV.

294
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 1:3:3
authors. These, too, could be used either to give a misleading or an accurate
description of the object. An interesting example of the former is provided
by an inscription on a sixth-century floor mosaic in the north wing of the
transept of St. Demetrius in Nikopolis. The mosaic has a central field containing
trees, flowers, and birds. No other animals are represented, but the inscription
below this panel says that it shows the earth with "everything '" that
breathes and creeps." This somewhat inappropriate phrase is a quotation
from Homer 124
On other occasions, however, Homeric tags were used in contexts where
they may well have been relevant. We may take as an example the descrip-
tion by Mesarites of a mosaic in the Holy Apostles in Constantinople which
portrayed the Apostles on their way to Galilee, after the discovery of the
Resurrection. The writer records that the disciples follow "in a row," and,
borrowing from Homer, he compares them to a "linked golden cord."125 They
are led by St. Peter, who is depicted as taking "long strides," another Homeric
phrase;126 some disciples are represented as young and some as old men, but
"the youth does not outrun the old man, and the grown man does not spring
out before the aged one .... "127 This description perfectly fits the illustration
of the Apostles going to Galilee which survives in an eleventh-century Gospel
book in Paris (fig. 24),128 probably illuminated in Constantinople. The Apostles
are shown in a long file, as they run toward Christ; the elder Apostles,
represented with white hair and beards, are toward the head of the procession;
St. Peter leads, and the runners are taking long strides. Thus the two quota-
tions, which Mesarites takes from Homer, are seen to be perfectly apt; if the
mosaic in the Holy Apostles resembled the minature in the Paris Gospels,
his description was accurate. Of course, since the mosaics of the Holy Apostles
are lost, we will never be able to prove conclusively that the use of topoi by
Mesarites was accurate or inaccurate; we can only point to instances where
the evidence of other surviving Byzantine works of art tends to corroborate
or contradict what he says.
One of these Homeric quotations was used in a different context, but no
less appropriately, by Constantine the Rhodian in his tenth-century poem
on the Holy Apostles. In his description of the mosaic of the Betrayal, Con-
stantine records of Judas that" ... his feet are stretched out in haste, and he
takes long strides on his wicked path."129 The reference to "long strides" is

,2< "ITaVTa ..• oera TIViEl TE Kai epml' E. Kitzinger, "Studies on Late Antique and Early
Byzantine Floor Mosaics," DOP, 6 (1951), 95ff., esp. 100f., fig. 18.
125 Xpvcrea CTEIPO Tl, aAA'1AEVOETO). XXXII.9, ed. Downey; Iliad, VIII, v. 19. See also Proco-
pius of Caesarea on the suspension of the dome of St. Sophia in Constantinople by means of a golden
cord from heaven; De aedificiis, 1.1.46. A twelfth-century poet applied the simile of the golden cord
to a dynastic portrait of the Emperor Manuel Comnenus, with his father John and his son Alexius;
S. Lambros, Neos 'EAA., 8 (1911), 173, no. 318.
120 ~aKpa ~1~wVTa. XXXII.6, ed. Downey; Iliad, VII, v. 213, XIII, v. 809 (Ajax); Odyssey,
IX, v.450 (the ram of Cyclops), XI. v. 539 (Achilles) (references given by Downey).
127 6 VEas oli TIPOTPExEl TOV yepoVTos, 6 CtV,;p TOV TIap'1~'1K6TOS oli TIp01fT]oq; .... XXXII.2, ed. Downey.
128 Paris, Bibl. Nat., gr. 74, fol. 61v; Omont, Evangiles, pI. 56.
'" TICOES -re yap crmIioovaw !KTETa~~evol. J.laKpa ~1~WVTES e!s clToerSaAOV Tp{~OV, ed. Legrand, "Des-
cription," v. 896f.

295
 136 HENRY MAGUIRE

again apposite. In works of art Judas was often depicted striding toward
Christ in order to kiss him. One finds him represented thus, for example,
in an early sixth-century mosaic in Sant'Apollinare Nuovo in Ravenna (fig. 19),
as well as in post-iconoclastic works.1 30
A number of the ekphraseis take the form of, or are part of, sermons. It
was natural, then, that certain topoi characteristic of homiletic literature
should be incorporated into the ekphraseis. Again, we find that these topoi
could give rise both to accurate and to inaccurate descriptions, depending
upon their use. A good example of the misuse of such a topos is an inscription
in the church of Asinou in Cyprus; it is dated 1332-33. The painting to which it
refers is placed over the west doorway and shows the Virgin raising her arms
in the orant pose; enclosed in a medallion, which is painted over her chest,
is a bust of the infant Christ (fig. 25). The inscription around the painting
reads: "How is he who holds together all judgments held as a babe in a virgin's
arms."131 Since the fresco does not show the Virgin actually holding Christ,
the inscription seems unsuitable. It is, indeed, a repetition of a common literary
conceit, which is also to be found, for example, in Mesarites's ekphrasis on the
church of the Holy Apostles in Constantinople. In his description of a mosaic
showing the Nativity, Mesarites writes :132 " ... He who holds together all
things in his omnipotent hand is carried by a hand without strength .... "
The same sentiment was expressed in the ninth century by the Patriarch
Photius, in the sermon devoted to the mosaic of the Virgin and Child in St.
Sophia. 133 Here we can see that Mary was actually shown carrying her Child
(fig. 23). This topos, with its play upon the paradoxes of the Nativity, also
appeared in sermons which were not concerned with works of art.134
Sometimes the conventions of the sermons and biblical commentaries were
rendered pictorially in works of art. When this occurred, the writers of the
descriptions were justified in quoting from homilies. For example, Mesarites,
in his description of the mosaic of tl).e Walking on the Water, gives an inter-
esting description of Christ rescuing Peter from the waves: " ... Peter's right
hand is seized by both hands of the Pantocrator which draw him up completely
like another Adam out of the depth of Hades .... "135 The link between the
saving of Peter from the water and the redemption of man was also made
in the eleventh century by Theophylactus, in a commentary on St. Matthew's
gospel: "The ship is the earth, the waves are life troubled by bad spirits, and
the night is ignorance. In the fourth watch, at the end, that is, of the ages,

13. E.g., Diez and Demus, Byzantine Mosaics in Greece, fig. 98 (Daphni); O. Demus, The Mosaics
0/ Norman Sicily (London, 1949), fig. 70 (Monreale).
131 ,;, "ITWS 6 "ITCxVTWV ""VExiJs TWV KPI~6:Twv /lP''l'0Kp<n<iTCXI
TrapS,vlKaiS wAEvalS. "'. H. Buckler, "The
Church of Asinou, Cyprus, and its Frescoes," Archaeologia, 83 (1933). 336.
'32 6 ""vEx"'V TO "ITOVTa TraVT05vvo~'i' OpCXK\ "TrO XElpOS aV""KlooS l3"'O"Ta1.6~Evol .... XXIII. 2, ed.
Downey.
133 Homilia XVII, ed. Aristarches, 299; ed. Laourdas, 167.
'34 See, for example, a sermon ascribed by its title to St. Athanasius, but not written before the
fifth century, in PG, 28, col. 961A. Compare PG, 46, col. 1142D ("St. Gregory of Nyssa").
'" ... WI ,.fj,5E~I1i1 "n' a~'P0T'pWV TWV "IT"'VTOKPCXTOPIKWV EmIATJ~~EVTJS XElPWV K",i 6AOV ,.6v nhpov WI
&""ov'Aool! WS E~ ';'50v &vcryovO"wv ,.00 I3vSoO ..•. XXV.1S, ed. Downey.

296
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 137
Christ appears. For the first watch is the covenant with Abraham, the second
the law of Moses, the third the prophets, and the fourth the advent of the
Lord. For he himself saved those who were buffeted in the waves .... "136
Thus Mesarites's comparison of Adam and Peter had literary precedents, at
least with respect to the theme of redemption. But it also had a very specific
reference to works of art. Among the late twelfth-century mosaics of Monreale,
which were executed by Byzantine artists, there is a scene of Christ walking
on the waves (fig. 26). The Lord is shown stepping toward St. Peter, and
raising him by the right hand. Through the waves St. Peter's right leg is
seen bent beneath him, as if he were trying to rise to his feet; his left leg is
stretched out straight behind him. At Monreale, in the mosaic of the Walking
on the Water, the attitudes of Christ and St. Peter seem to have been similar
to those of Christ and Adam in the mosaic of the Anastasis. Unfortunately,
the Anastasis mosaic at Monreale has been badly restored, and we only know
its original form through an eighteenth-century engraving 137 The surviving
eleventh-century mosaic of this scene at the Nea Moni on Chios (fig. 27),
however, mirrors the composition of the principal figures in the Walking on
the Water at Monreale.
Frequently, Byzantine authors incorporated into their ekphraseis topoi
which were not merely short phrases or single conceits, but relatively long
passages of description. But even these extensive descriptions could be accu-
rate, in part if not in their entirety. There was a venerable tradition, for ex-
ample, of accounts of storms at sea. In describing the mosaic of Christ Walking
on the Water, Mesarites gives a vivid representation of the angry sea. First,
he expresses wonder that there can be waters depicted high up on the wall of
the church, as it were in the air. Then, he proceeds: "Gaze on this howling
sea; see the waves, how some are piled up high as mountains, as they roll
in the open sea, while others lie quiet, as they are drawn to shore on the coast,
as if, from reverence for the Lord who stands there, they gathered themselves
together and broke themselves off. Observe how the atmosphere about it is
dark, how it is misty, so to speak, and smoky; how the clouds are gathered,
how violently the air bears along the ship on the clashing of the waves, as
some northeast or arctic Boreal wind blows stormily."138 This passage follows
a standard pattern for descriptions of storms at sea in late classical and Byzan-
tine writers. A similar passage occurs in the eleventh-century Chronographia
of Michael Psellus. Although many words are the same, the context is com-
13. C'K<XCPOS, Ii yfi' KVIlCXTCX, 0 ~(05 .nro To;V ;roVTJpo;v ;rVElJIl<XTCilV TCXPCXTTOIlEV05, vV~, ,; OyvCilcricx. 'E v Tij
TET<Xp'TlJ oe CPVAcXKij, ;rp~ Ti;'> TEAE1 0T\Aov6T1 To;V cx!o,vwv, hrEO'T1'\ 6 Xp1<TT05. npo,TT\ lleV ycxp CPVACX1<';, ,; ;rpo<;
'A~pcxall 01cx.9';KT]· oEVripcx, 0 MwO'e~ VOIlOS' TPITT\, 01 ;rpocpfiTcx1' TETCxPTT\, ,; Toii Kvpiov ;rcxpovO'(a' cx\JT~
yap EO'WO'E TO~ KA.VOWVtlOIlEVOV5, ElO'EA.9WV KCX! YEVOllEV05 1lE.9' ';llo;V •... Enarratio in Evangelium M atthiae,
XIV.31-33; PG, 123, col. 304A.
137 M. del Giudice, nescrizione del real tempio e monasterio di Santa Maria Nuova, di Morreale
(Palenno, 1702), pI. 22, 7.
138 'Opa TfJv WPVWIlEVTJV Ta\rn]V .9«ACXTTCXV, cpa Ta KVIlCI'TO, 11'0;5 Ta IlEV KopvcpoiiVTCX1 10'cx KCXt 0PEC1V, 60'0
Sf] mpl TO mi\cxyos Kwcrr!lovO't, Ta Se YCXAllVt(;;atV, oO'cx Of] KOTOyOVTOt mpt TTjv c'xKT1iv, TC,. e;r'cxtm'j5 e<TTo;TO
S.a;r~v olov CX!OOVIlEVCX Kat ;rp05 EauTa O'V<TTEAAOllEva TE KCXt <XvcxKA.o,llEva, 11'&5 KCXTCxC'KOTOS 6 mpt Ta\rn]V
Cnlp, 11'~ OlltXAw0115 olov Kcxt Kcx;rv11P~, 1I'0;S O'VVVE<p';S, ;ro;s mptO'KEAES TO 1I'Aoiov Tij TWV KVIl<XTWV «At-E-
;rcxAhtjt-'tl cpopC;;, rupuKAVOCilVOs TtVOs i\ KCXt <X;rcxpKTlov ~op!ov ;rvtoVTOS 5vO'cx!s. XXV.6, ed. Downey.

297
 138 HENRY MAGUIRE

pletely different, for here Psellus is describing the storms which beset the
life of a Byzantine emperor. The emperor's existence, he says, is like "a sea
which for a short while is smooth and quiet, but at other times the waves
flow and shake, as now the Boreal wind disturbs it, now the arctic wind,
and now some other of the winds which stir up the waves, as I myself have
seen on many an occasion. "139 Similar accounts of storms at sea can be found
in the works of the fourth-century Cappadocian Church Fathers St. Gregory
of Nazianzus and St. Gregory of Nyssa. l40 And we can go back further, to a
description of a storm in one of the Fishermen's Letters by the sophist Alci-
phron which reads: "The sea, as you see, shudders, a mist has spread under the
sky, everything everywhere is clouded over, and the winds, dashing against
each other, proclaim that they will immediately stir up the sea. "141 Since the
description of the sea in Mesarites owes so much to literary convention, one
would not expect that he has here given us a very accurate account of the
mosaic of Christ Walking on the Water. But his description, conventional
though it is, may still contain elements of truth, particularly in the opening
sentence describing the waves. The sentence clearly is a topos, for Psellus also
opened his description by contrasting a calm with a stormy sea. But if we
look at the mosaic at Monreale, we find that here too the sea behind Christ
is calm, while in front of him it forms high waves (fig. 26). This is, in fact,
an illustration of Matthew 14: 32, which relates that when Christ and Peter
boarded the ship, the wind ceased. Here again, therefore, a literary conven-
tion may have conveyed a truth, if the mosaic in the Holy Apostles resembled
that at Monreale.
In these pages I have reviewed many of the topoi which occurred in the
Byzantine literature on art and have found some in the introductions to the
ekphraseis, others scattered through the descriptions in the form of tags
quoted from classical authors. A majority took the form of stereotyped al-
lusions to realism, often referring to .the depiction of emotion. There were also
long ready-made descriptions.
For the sake of brevity, I have omitted several varieties of topos, such as
conventional allusions to famous artists of antiquity,142 ways of describing
color, and references to precious materials, particularly gold. But sufficient
examples have been discussed to show that the conventions of the ekphraseis
139 6:AA''''C''TEP .96:)..aO'O'a flpaxv flEV KaTEerr6pEerra. Kal yaATjv'9, 'TO 5'&AAa 'TCCiTc flEV 'ITATjllflVPEi, 'TCCiTc
Bt Kal T1VCxCTcrETOI KV~aCi1, vOv !-lEV ~OpEOV 01O'TOperrrOVTOs, vVv S'anap1<TlOV, viiv 6'OAhOV TtVOS TWV tyEtPOVTWV
KAv5ci>VIcv, omp a&oos hrl 'ITcMciS 'WpOKElV. Chronographia, V1.27.
1<. Gregory of Nazianzus, Carmen de vita sua, PG, 37, col. 1038; Gregory of Nyssa, In Cantica
Canticorum, PG, 44, col. 869B.
'" T"v flev .9aAo:T'TaV, ws op<;is, 'PP{Kl1 KaTEXEl Kat 'TOV clipavov VTICfllflTjKEV 6:xA"s Kat 'ITaV'Ta 'ITavoaxo.9EV
O'VVVE'Pl1ACC, KCCI 0'1 &VEflC. 5e 'ITPes 6:t.ATjACVS apccO'O'oIlEVO. oO'cv c\'rrrw KVKl\O'E'V TO 'ITIAayes E'ITo:yyeA?CV'TCC'.
Alciphron, Epistulae, I.10.1ff. Another storm ekphrasis appears in Apollonius of Tyre, Historia, 11,
ed. A. Riese, Teubner (1871).
142 SO'me examples have been given by MangO', "Antique Statuary," 65f. See also Anth. Pal., IX,
nO'. 776 (Zeuxis), and XVI, nO'. 344 (Lysippus); PhilO'stratus, Apollonii vita, II.20 (Zeuxis, PO'lygnO'tus,
EuphranO'r); Psellus, Chronographid, III.14 (Pheidias, Polygnotus, Zeuxis); PrO'dromus in Carmina
inedita Theodori Prodromi et Stephani Physopalamitae, cd. C. Welz (Leipzig, 1910), 15 (Apelles,
Praxiteles); Constantine Manasses, cd. Sternbach, "Beitrage zur Kunstgeschichte," CO'I. 75, lines
21-23 (Pheidias, Praxiteles, Lysippus, Parrhasius).

298
 BYZANTINE DESCRIPTIONS OF WORKS OF ART 139
were in themselves neither true nor false; they could be used to give both
accurate and inaccurate descriptions.
It remains to account for the disparities in accuracy between the ek-
phraseis. One might perhaps expect that, as Byzantine art became further
removed in time and overall character from its late antique origins, the
ekphraseis, still bound strictly to classical tradition, would become less and
less reliable. This does not seem to have been the case. Neither in the ek-
phraseis, nor in the shorter inscriptions, was there any chronological develop-
ment toward a lesser or a greater accuracy in the employment of literary
formulae. We have found the apt use of tapai in the sixth-century works of
Procopius and Choricius of Gaza, in the tenth-century poem of Constantine the
Rhodian, and more than once in the long ekphrasis written by Mesarites
around the year 1200. On the other hand, the clear misuse of literary tapa!
has been found in a sixth-century inscription at Nikopolis, in a ninth-century
sermon by Photius, and in an early fourteenth-century inscription at Asinou.
The explanation for the variations in the accuracy of the ekphraseis must be
provided not by the dates of the authors, but by their relative skills. Ekphrasis
was a highly artificial and also a very demanding exercise. Frequently the
composer of an ekphrasis was describing an object which his audience could
already see for themselves. For example, in 867 the Patriarch Photius delivered
in the church of St. Sophia in Constantinople the homily mentioned above
in which he described the apse mosaic of Mary carrying her Child 143 The
twelfth-century Greek preacher Philagathus has left us a description of the
Palatine chapel in Palermo, in Sicily, which was also delivered in the very
building.144 In the case of the shorter, epigrammatic descriptions, we have
found that these were often attached to the works of art themselves, in the
form of inscriptions. In all these instances it is obvious that the purely
descriptive value of the ekphrasis was superfluous. And even in the case of
the ekphrasis on the Holy Apostles by Mesarites, which he seems to have
written while he was in Constantinople, one wonders how many of the readers
were completely ignorant of the appearance of the church. At the end of the
ekphrasis there is a long encomium of the Patriarch of Constantinople, John X,
which suggests that he was one intended reader, who must already have been
well acquainted with the building,14.5
1£ the authors of the ekphraseis described buildings and works of art which
were often familiar to their audience, this was no less true of the literary
tapai which they incorporated into their descriptions. The modern reader may
well be unfamiliar with the paraphrases and quotations, but the educated
Byzantine presumably knew them. In this connection it is interesting that
in his encomium Mesarites compares the rhetoric of the Patriarch to that of
Hermogenes of Tarsus, the second-century orator who codified the ekphrasis.
In many cases, therefore, the Byzantine orator was describing a work of art
14. Mango, The Homilies of Pho/ius, 279ff.
'" Cf. supra, note 116.
us XLIII, ed. Downey.

299
140 HENRY MAGUIRE
known to his hearers with a known set of quotations or formulae. The skill
must have come in matching the one to the other; an unskillful writer, like
John Phocas, produced inaccurate descriptions, but the clever author managed
to quote aptly, so that his exercises had the virtues of both precision and
literary erudition. The achievement was perhaps not unlike that of the Byzan-
tine composer of the Christos paschon, who made up his drama of the Passion
out of a patchwork of quotations from classical tragedies and the Bible. Our
conclusion must be that the literary traditions of the ekphraseis make it
necessary to interpret these documents with extreme caution, but they by no
means render all Byzantine descriptions of works of art invalid. 146

Harvard University

'" The sequel to this article will discuss the extent to which the emphasis on the depiction of
emotion (particularly sorrow) in Byzantine ekphraseis can be justified with reference to surviving works
of art.

300
The Poverty of Ecriture and the Craft of
Writing: Towards a Reappraisal of the
Prodromic Poems*
MARGARET ALEXIOU

It is the purpose of this paper to challenge some traditional

assumptions about authorship and style in twelfth-century Byzan-
tium, and to present a new, if tentative, interpretation of the
fourth Prodromic Poem, which may have important implications
for an understanding of literary perceptions of every-day life in
twelfth-century Constantinople, as well as for the ways in which
texts can be read.

1. The Author
All four Prodromic Poems present an apparent paradox, which
is by no means unique. Probably our richest 'low-style' source
of information from the twelfth century, they are consistently,
if not quite unanimously, attributed by the manuscript tradition
• This paper is presented as 'work in progr.:ss' in the fullest sense of the term: origi-
nally delivered to the Literary Palle! of the Seventeenth Spring Symposium of
Byzantine Studies ('Life and Death in Byzantium', University of Birmingham,
March 1983), it now incorporates, inevitably, much of the collective work done
during the course of the academic year 1983/4 by staff and postgraduate members
of the Byzantine Text Seminar (Centre for Byzantine Studies, University of
Birmingham), which is preparing for publication a translation. glossary and full
commentary of the four poems. Work on the fourth is nearing completion. In
the meantime the convener of the Seminar wishes to put on record some of her
ideas which first generated interest in the text, as well as our current findings.
Acknowledgements for specific items of research are made in footnotes (by
asterisked names) insofar as possible; however, this article could not have been
completed without team work. As always, the author retains sole responsibility
for errors.

301
to Theodore Prodromos. 1 Known prolific writer for the
Comnenian court of prose and verse texts of diverse genres and
styles, his undisputed compositions are written in extreme or
modified forms of learned Greek, with frequent use of several
classical metres; none is in the vulgar Greek, combined with
politikos stich os (fifteen-syllable accentual verse), of our four
poems. 2 The question arises, could Theodore Prodromos have
written them? Or does the use of popular language, in addition
to what editors and most critics assume to be inconsistencies,
repetitions and glaring contradictions, preclude such an established
and acclaimed writer from authorship of a work so far 'beneath
his dignity?3 What makes the picture more complex, Theodore
Prodromos is also considered a likely contestant, along with the
front-running Nikolaos Kallikles, for the authorship of the
Tim arion, a high-style Lucianic dialogue, which, as has been
argued elsewhere, ironically subverts certain aristocratic manners
and ideals, while affording a wealth of insight into twelfth-century
Constantinopolitan society.4 Is it conceivable that the same per-
son could have written two such seemingly different kinds of text?
. I. D.C. Hesseling and H. Pernot, Poemes prodromiques en grec vulgaire
(Amsterdam 1910) 10-22. The attribution in the titles to Poem 111 in mss CSA
to one 'Hilarion Ptochoprodromos', a monk, together with the name 'Hilarion' as
the narrating persona in the text (111.387), has misled some literal-minded Byzan-
tinists into postulating the 'real' existence of such a person as author of this poem,
see M. Hatzidakis, VV 4 (1897) 100-27, and S. Papadimitriu, ibid. 5 (1898) 92-130,
whose proliferation of a sub-species of Prodromic personae goes so far as to count
both 'Hilarion Ptochoprodromos' and '(Theodore) Ptochoprodromos' as authors
separate from the 'real' Theodore Prodromos. Hesseling - Pernot (1910) 91-130,
H.G. Beck, Geschichte der byzantinischen Volksliteratur (Munich 1971) 103-4, and
W. Horandner, Theodoros Prodromos: historische Gedichte (Vienna 1974) 65-7, rightly
consider the attribution of Poem III to 'Hilarion' to be a misunderstanding of the
copyists. Jacopone da Todi's use of the name 'Hilarion' in the satirical context of
an impoverished monk might usefully be explored, see P. Dronke, The Medieval Lyric
(Hutchinson 2 1978) 216. A similar, if more grotesque, bogus attribution is cited by
Beck (1971) 176 n.I, from the fifteenth-century Vienna manuscript of the unrhymed
poem, Synaxarion tou Gadarou, glossed by P. Lambricius as 'De quodam Gadaro
sanctitatis vitae claro'!
2. For the most recent summaries and discussions of Theodore Prodromos' works,
certain and disputed, see Horandner (1974) 37-78 and A. Kazhdan, Studies on Byzan-
tine literature of the eleventh and twelfth centuries (Cambridge 1984) 87-114.
3. The arguments on authorship have hitherto rested almost exclusively on sty",
language and metre. The details and principles are summarised in Appendix I. The
Birmingham Text Seminar, while devoting no more than an Appendix to the ques-
tion, would wish to keep the debate open.
4. Pseudo-Luciano, Timarione, 00. R. Romano (Naples 1974); see also M. Alexiou,
'Literary subversion and the aristocracy in twelfth-century Byzantium: a stylistic

302
 The question raises some fundamental problems of methdology
and approach. First, can authorship be determined according to
objective criteria, whether linguistic, metrical, stylistic,
biographical or historical? Or is there a danger of imposing
modern concepts of authorship and of linguistic or stylistic unity
upon twelfth-century texts? Second, is the dichotomy between
'high' and 'low' styles so absolute as has been assumed? Were
they perhaps determined by genre and context rather than by the
educational level of the individual author?
The first question has implications beyond the authorship of
the four Prodromic poems. Michel Foucault has argued convinc-
ingly, albeit in general and theoretical terms, that both the con-
cept and function of the 'author' have changed radically from
antiquity and the middle ages to the present day, more particularly
since the Renaissance. 5 In the case of a substantial number of
Byzantine texts, both learned and vernacular, which have been
transmitted anonymously or dubiously, sometimes in differing
versions, the attempt to fit an author to the text is demonstrably
the work of subsequent scribes and scholars. Medieval tradition
was content to ascribe authorship of any work which became
famous to the most prominent exponent of the genre, as with
those later hymn-writers who wove the acrostich ROMANOS
THE MELODIST into their kontakia as a tribute to the master's
genius. 6 Traditional attribution of a name to a work is no proof

analysis of the Timarion, chapters 6-10',8 (1982/3) 29-45, and B. Baldwin, Timarion.
Byzantine Texts in Translation (Detroit 1984).
5. M. Foucault, 'What is an author?', in Textual Strategies, ed. J.V. Harari (Cornell
1979) 141-60. The implications for dubiously transmitted Byzantine texts are noted
by Alexiou (1982/3) 30-31 n.4. ·G. Calofonos has cogently argued the relevance of
Foucault's case to dream-books, anonymously composed, but later authenticated by
attribution to specific persons. D. Lodge, 'Milan Kundera, and the idea of author
in modern criticism', Critical Quarterly 26.1/2 (1984) 105-21, has recently challenged
the rejection by R. Barthes and M. Foucault of the concept of 'author'.
6. Romanos the Melodist, Cantica dubia, edd. P. Maas and C.A. Trypanis (Berlin
1970); see also E. Wellesz, The Akathistos Hymn, in Monumenta Musicae Byzan-
tinae 9 (Copenhagen 1957). Problems of authorship are by no means restricted to
vernacular and religious texts: the date and authorship of the learned tragedy, Christos
Pasch on, probably twelfth century but amibuted to Gregory of Nazianzus, remain
open, see A. Tuilier, Gregoire de Nazianze: la Passion de Christ, tragidie, in Sources
chrefiennes no.149 (Paris 1969) for a SL .nmary of the evidence, and for a pro-fourth-
century case. Contra, see evidence cited and summarised by R. Macrides, 'Poetic justice
in the Patriarchate. Murder and Cannibalism in the Provinces,' Cupido Legum
(Frankfurt am Main 1985) 167.

303
of authorship; nor does lack of attribution necessarily indicate
oral composition, or a phase of oral transmission. 7 In the case
of Theodore Prodromos, the fact that subsequent tradition
accredits him with authorship of our poems may be less relevant
than the fact that the chronology of dateable elements in the poems
is not inconsistent with that of his life, if Kazhdan 's revised, and
well-substantiated, dates for his birth (around 1100) and death
(late 1160's or early 1170's), are accepted.BOne thing is certain:
neither the historian nor the literary critic can afford to wait un-
til the drear question of authorship is solved before analysing the
text, since, in the absence of new data, only close studies and
comparison with other contemporary material can yield results.
A further point relevant to the question of authorship needs
to be clarified at the outset: in dealing with a literary text, care
should be taken to distinguish between author and ego-
narrator. 9 This elementary observation needs re-stating in view
of persistent claims that, since each of the four poems assumes
a different narrating persona, they cannot be the work of the same
writer; or that Theodore Prodromos can be excluded on the
grounds that, at the probable date of composition of the fourth
poem, he would have been too old to be a student po Why
assume that twelfth-century writers were less astute than ourselves
in the structuring of their narratives and in the manipulation of
narrative perspectives? In each of the four poems, the narrator
adopts an exaggerated social position: (I) hen-pecked husband
married above his social class; (II) impoverished father of a family
of thirteen; (III) poor monk abused by abbot; (IV) destitute writer.
The four cases are literary, not literal.
The second question - how absolute is the dichotomy between
'high' and 'low' literature? - perhaps needs redefining, since
7. See M.J. Jeffreys, 'The literary emergence of vernacular Greek', Mosaic 8 (1974)
171-93, and H. and N. Eideneier, 'Leser oder Hiirerkreis? zur byzantinischen Dichtung
in der Volksprache', Ellenika 34 (1982/3) 119-50. No firm conclusions may be drawn.
8. Kazhdan (1984) 92-104 examines in detail the life and works of Theodore
Prodromos and 'Ptocho-Prodromos', noting the remarkable coincidences of historical
and social circumstance, which outweigh statistical analyses of language and metrics.
9. See G. Genette, Narrative Discourse (Blackwell 1980) 213-4. The relevance of
the distinction for the Timarion has already been argued by Alexiou (198213) 29-45.
For the Prodromic poems, Kazhdan is among the few to argue correctly that they
are 'genre exercises, and their supposed "authors" are no more than literary per-
sonae', (1984) 91.
10. The argument is posed, hypothetically, by Beck (1971) 104.

304
it presupposes that there i~ a valid literary distinction between
texts in differing stylistic and linguistic registers. It is more rele-
vant to examine how 'reality' is perceived and presented in fic-
tive terms. Some traditional questions should be discarded, or
inverted. Instead of asking, for example, 'who is the author?',
'What do we know of when, where or why he wrote the text?',
'What is genuine and what can be interpolated in the manuscript
tradition?', it is possible to enquire, 'How is the narrative struc-
tured, and from what perspectives?', 'What traditional topo; are
employed, and how are they handled?', 'What allusions to other
texts are made, and to what effects?', 'What historically dateable
elements exist, and how can they be related to the dominant trends,
or myths, of the time?' These questions dictate that, for the
present at least, the text is accepted as edited, with all its variant
readings, and that judgement is suspended regarding its naive and
discursive nature (according to present-day 'high' standards of
realism and logic). Close analysis and extensive research along
these lines can render this and other texts more, not less, accessi-
ble to modern readers.11

2. The Text
According to its editors, the text of the four poems is in a
deplorable condition, posing every conceivable problem of date,
authorship, authenticity, unity, relationship to other Prodromic
texts. 12 Yet no fewer than seven manuscripts survive, dating
from the late thirteenth or early fourteenth centuries to the fif-
teenth and sixteenth centuries. Poems I and II are extant in one
and two manuscripts respectively (G and GH), both described
as 'unreliable'; Poems III and IV survive in six and five (H, CSA,
gV and G, CSA, g).13 All contain interpolations, none is un-
contaminated, hence no stemma can be constructed, and therefore
(according to traditional principles) no reliable text can be
established. The editors frequently dismiss large sections as un-
sound, mainly on grounds of inappropriate style, content and
II. Our Text Seminar has striven to steer a narrow course between the Scylla of
circular arguments based on philological, metrical and historical minutiae. and the
,cductive Charybdis of structuralist and post-structuralist theories. which tend to
dismiss all history (and archaeology) as 'literary texts', thereby comfortably justify-
ing any lack of reference to historical time or context.
12. See Hesseling - Pernot (1910) 10-24.
13. Ibid. 84-103.

305
structure. 14 Therefore (the logic is inscrutable) the poems can-
not be attributed firmly, on the basis of the manuscript tradi-
tion, either to one person (let alone to Theodore Prodromos) or
to the twelfth century in toto. Inconclusive evidence should not,
however, permit circular argument. Either we follow the advice
of the textual critics and deny that reliable use can be made of
the text, or we avert the predatory but 'dead hand of classicism'
by accepting the text, with all its imperfections and variant
readings, as the only extant, and therefore relevant, form of four
remarkable poems, the fourth certainly composed between the
years 1160 and 1204, and probably dateable, at least in part, to
the early 1170's. The original, 'authorial' version may be
irrecoverable; ours has at least been validated by subsequent tl ,.ji-
tion. The earliest manuscript (G), containing Poems I, II and IV,
belongs to the turn of the thirteenth and fourteenth centuries,
proving, according to the editors, reliable for IV but not for I
and II. Does this sugeest G's use of earlier models, pointing to
a textual tradition established in the thirteenth century, or is it
a case of later contaminatio? Judging from the number and
distribution of extant manuscripts, Poems III and IV seem to have
enjoyed wider circulation than I and II, possibly passing through
a stage of oral transmission. 15 Speculation that the four poems
were originally composed, perhaps by Theodore Prodromos, in
learned Greek, and subsequently rendered into the vernacular by
a Prodromic school of poets, can be neither proved nor
disproved. 16 No trace of a learned original, or of a school of
poets, has so far come to light.

3. Narrative Structure
The text of Poem IV opens problematically, with a fulsome
proem addressed to the Emperor Manuel I in liturgical vein (see
Appendix II for translation of all four proems). Absent from the
14. Ibid.
i 5. The case is argued in detail by Eideneier (198213) 119-50, on the grounds that
some of the variant readings in the manuscripts can be explained as oral rather than
as purely scribal changes. A further possibility, proposed by R.A. Fletcher in rela-
tion to Digenes Akrites, should be investigated: that differences between extant ver-
sions derive from scribal mangling or aural mishearing rather than to 'genuine' oral
tradition, see Mandat%ros 8 (May 1976) 8-9. For a more general reassessment of
the problem, see F .H. Bauml, 'Medieval texts and the two theories of oral-formulaic
composition: a proposal for a third theory', New Literary History 16.1 (1984) 31-49.
16. H6randner (1974) 66.

306
oldest manuscript (G), it is found in brief form in g (fifteenth
century), and more fully in CSA (fifteenth, fourteenth and six-
teenth centuries).17 The presence of a proem is justified struc-
turally both by the closing epilogue, also addressed to Manuel
(lines 274-92), and by the existence in a majority of witnesses of
an opening and closing address to the patron in each of the three
other poems. 18 Further, even this supposedly later interpolation
can be shown to contain elements dateable to before the year
1176. 19 It should be noted that each of the four proems differs
markedly in tone and style from the rest of the poems, and that
each contains playfully personal allusions to the poverty of the
writer and wealth of his patron, as he begs indulgence for his
vulgar language; it would therefore be rash to dismiss the fourth
proem on grounds of style alone, although problems of textual
variation remain.
The narrator then proceeds (or opens) with his well-known
cautionary advice, couched in the familiar Byzantine topos of
a father's words of wisdom to his son: 20
Ever since I was a lad, my father used to tell me,
'Learn letters, boy, and there'll be no-one like you.
You see that man there, child, he used to walk on foot,
now he rides a fat mule with double leather straps.
While he was learning, he wore no shoes at all,
and now, look at him, he wears long pointed ones.
While he was learning, he never combed his hair,
now he is well groomed, with prize-curled locks.
While learning, he never glimpsed the bath-gates,
but now he takes a bath three times a week.
His belt-fold used to swell with lice the size of almonds,
now it is swell with Manuel's hyperpyra (= recently minted gold coins)
So be persuaded by your old man's fatherly advice,
and learn your grammar - there'll be no-one like you!'
So I learned my grammar, with a great deal of effort.
Now that I've become, supposedly, a craftsman,
I lack both bread and the means of its provision.
I spit on grammar and I say with tears,

17. Hesseling - Pernot (1910) 30-1 (1.1-26), 3S-41 (11.\ -25), 4S-50 (I I1.1-32).
IS. See Appendix II for a translation of the four proems.
19. See n.57.
20. Similar devices, possibly literary rather than autobiographical, are to be found,
for example, in Constantine Porphyrogenitus, D" Caerimoniis, ed. Reiske (CSHB,
Bonn IS29) 3, and Kekaumenos Strategikon, edd. B. Wassiliewski and V. Jornstedt
(Amsterdam 1965) 6.

307
 'Christ, down with letters, and with whoever wants them!
Cursed be the hour and cursed be the day
when they delivered me unto the school,
to learn my letters, as though I could live off them!
Just suppose I'd trained as a gold cloth craftsman,
like those who work gold filigree, and earn a living,
if I'd learned the despised craft of gold filigree,
I could not fail to open up my cupboard, and find it
filled with bread, copious wine and cooked tunny fish,
tunny sliced and dried, mackerel large and small.
Instead, as I now open mine, I see naught but empty space,
and I see files upon files stuffed with paper.
I open my bag to reach a crust of bread,
and I find yet another, smaller file.
I dig into my pocket, I reach down for my purse,
looking for a stamenon (= small change) - that too is stuffed with paper.
Having groped around in pockets and all corners,
I stand dejected and devoid of nerve,
fainting and losing heart from excessive hunger,
Rather than extreme hunger and distress,
filigree work is better than grammar and the text!' (lines 1-39)
In this passage, writing is perceived above all as a craft (techne),
while the lot of scholar as craftsman (grammatikos technites) is
compared with that of other artisans in the City. The term
technites, originally denoting 'one who fits together; builder;
carpenter' ,21 provides the keynote and unifying feature to the
whole poem. Learning the cnft of writing may mean poverty and
filth at first, but, according to the precepts of the older genera-
tion, ensures wealth, privilege and fashion in the end. Unfor-
tunately, as things turn out, the despised embroiderers of gold
cloth (and other artisans) fare well, while our versifier, in place
of bread or ready c~sh, finds only papers in his pockets. Here,
perhaps, is a strong indication, hitherto overlooked, that there
is a level of irony in the text: both paper and parchment (the word
charti can mean either) were valuable commodities in the twelfth
century. 22

21. G. Nagy, The Best of the Achaeans (Johns Hopkins 1979) 300.
22. Evidence for the scarcity of parchment and paper in the twelfth century is cited
by N.G. Wilson, 'Books and readers in Byzantium', in Byzantine Books and Bookmen
(Dumbarton Oaks Colloquium, Washington D.C. 1975) 2-4; partly repeated from
L.D. Reynolds and N.G. Wilson, Scribes and Scholars (Oxford 1968) 51-63: Balsamon
comments on a canon forbidding the palimpsesting of biblical texts; Michael Choniates
complains that the supply of books may fail because shiploads of parchment are sold
to the Italians; seasonal supply is indicated by Gregory II of Cyprus (1283-89), who

308
 Our writer's neighbour is a lowly cobbler, not even a proper
shoe-maker (petsiJtes, pseudotsangares); yet he begins each day
by ordering his wife to prepare breakfast of tripe (chordokoilitsa)
and Vlach cheese, washed down by four large mugs of wine (and
an extra), before he even begins the day's work. When meal time
comes, he instructs his wife to set the table with stew-pot and
dishes, putting away his cobbler's awl, strigil and waxed thread
(souglin, sphetlin, sphekiJmata). A three-course meal ensues, after
which all thought of work is abandoned, as the cobbler 'spim
out his food' (klathei ten mageirian), leaving our narrator drool-
ing, with nothing but iambs, spondees and pyrrhics to stave off
his 'measureless hunger' (ametron peinan). Envying his
neighbour's life-style, he tries his luck at cobbling, only to get
his right hand (his writing member) drilled by the cobbler's awl.
He ends up in hospital for a month (lines 40-89).23
Other trades are considered. The tailor, hawking his wares
around the City with appropriate cries, is lowly, but at least he
gets due respect and remuneration from the women, who hail
him as 'craftsman' (deuro, tech nita, deuro) (lines 90-96. If only
he were a baker, or a baker's boy, he could count on his fill of
early morning prophournia (flat loaves made from left-over,
unrisen dough, or pitta-bread). Only the other day he passed the
bakery to find the baker's wife chewing the end of a crisp white
loaf made from semolina flour (a baguette?), refusing him even
the promise of a nibble. Dashing into the next side-street, he finds
the yoguri seller (oxygalatas), calling out something like 'Take

notes that he cannot have a text of Demosthenes copied until spring, when people
begin to eat meat and parchment is available (before selective breeding only 8 pages
can be expected from each animal); John Tzetzes complains of lack of charti in the
capital in the twelfth century, commenting on Aristophanes Frogs 843: tous chartas,
'which may be taken as meaning parchment or as a generic term for paper and par-
chment', Wilson (1975) 2. There is no evidence that paper, supposedly introduced
by the Arabs from China via Samarkand after 768, was either cheaper, or more plen-
tiful, than parchment at the time in question. Prices are hard to calculate, but parch-
ment might have cost roughly 50 ff. per nomisma (*A.A.M. Bryer). A comparable
irony may underlie the impovecished father's complaints in Poem II, whose desiderata
among 'basic foodstuffs' include some spices and condiments which must have cost
a small fortune at the time (II.3645G, 37a45a H).
23. eis Ion xenOna lV.89, cf. IlI.334a (H,CSA). Ph. Koukoules, Byzantin6n Bios
kai Polilismos (Athens 1948-1955) 1.130, 131f, 141, adduces other evidence for the
word as 'hospital' as well as 'guest house'; see also M.J. Kyriakis, 'Poor poets and
starving literati in twelfth-<:entury Byzantium', B 44 (1974) 305-6.

309
a pot of frothy white stuff, ladies'. He is soon sold out, with
money in his purse. Even the cloth-dyers and cloth-carriers get
rewarded at the end of their working day by a big helping of hot-
pot (mertikon ek ta laparimaia) and a mug of wine. The humblest
sellers and hawkers of wares, including pepper-grinders, can sell
their wares to the womenfolk (lines 97-129).
The first narrative section is rounded off by a brief comparison
with the writer-craftsman's neighbour on the other side, a sieve-
maker (koskinas): through the narrow partition which separates
their dwellings he can see, hear and smell the hot crackle of
roasting meat and fish, plentifully laid out (keimena) on the char-
coal. He is refused a share, and told to eat his grdmmata. Faint
from hunger, he appeals once more to the Emperor's mercy, pray-
ing that he may keep the sceptre of imperial rule. 24 Among the
craftsmen of the City, everyone else has money in his purse, or
can exchange services for food and drink; only our versifier has
nothing to sell and nothing to eat but letters (lines 130-144).
The next section introduces an apparent digression (lines
145-62). Warned that his habitual swearing, enforced by poverty,
will result in eternal damnation, he explains to the Emperor
(kosmokrator) that he already, in his present life, suffers triple
Hell:
On account of my poverty I swear a lot,
and people tell me, 'Take care, don't speak too much,
or after death you will be condemned
to the sleepless worm, to Tartaros, to darkness'.
But I, my world-ruler, am damned to these three
torments here and now, before my death:
poverty I regard as the sleepless worm,
which ever eats irrto and devours me.
As for Tartaros, I suffer now my death of cold,
from the icy winter, with nothing to put on,
and without anything to wear I shiver greatly.
Again, as to darkness, lord, I call it dizziness,
which is always with me, emperor, when I have no bread,
for without anything to eat, I get dizzy and fall down.
Here, then, is unlit darkness, Tartaros, and worm.
The unsleeping worm (akoimetos skOlex) has eaten him down to
the bone; Tartaros has taken its shivering hold of his naked and
penniless state, without a stitch (clever triple pun on Tartaros,
24. See especially lines 141-4. The perfective aspect of krateses, in juxtaposition with
kratiste, suggests a veiled irony.

310
tourtourizo and tarteron): while Skotos has him in its grip, star-
ving and fainting from want of food (another triple pun on skotos,
skotasmos and skotizomai). He is as good as dead. Another timely
address, somewhat ambivalendy to Christ as 'saviour of mortals:
(Christos mou brotosostes) begs deliverance through imperial
benificence. 25 Our narrator, at least, has nothing worse to fear
in afterlife ... (lines 145-62).
Another apparent digression follows, as the narrator reports
to the 'mighty-ruling Emperor of the four climates of the earth'
(kratarcha ba~ileu tessaron ges klimaton) a recent event pertain-
ing to his latest visit to the parental home. He found the table
laid with the most delicious hunk of fat, smoked apakin (a kind
of salami sausage, probably similar to pastourma),26 but was
excluded from partaking of the meal, on the grounds that he could
buy his own food. At this juncture, a noise from the cellar drew
the company (who feared divine retribution, it seems) downstairs,
whereupon the narrator devours the apakin, taking care to set
the cat on the table as scapegoat, and joins the family downstairs.
They return, stone the cat to death, realising too late who was
the real culprit. At least the dishes needed no washing (lines
163-201). The point is that the father's initial promise of wealth
and privilege has been totally undermined.
Lack of bread means lack of appetisers (prosphan); lack of
both means lack of recollections, hence the narrative sequence
may be disturbed (lines 201-5). Addressing now not the Emperor,
or Christ, but Hunger, our narrator declares that the only true
philosopher, rhetor and calligrapher is the craftsman assured of
cash or bread. Despite his learning, from Homer to Libanios and
Oppian, he is confronted with nothing but Hunger's snarling fangs
and wrinkled folds. If, on the other hand, he had trained as a
baker, he might have enjoyed unrisen bread (prophournia),
satiety, and a good day, with a 'difference' - diaphoroteritsin
- to his credit, (lines 206-66).

25. See especially lines 160-2. The appeal, directed to Christ acting through Manuel's
beneficence, is couched in terms which can scarcely be refused.
26. The definitions given by E. Kriaras, Lexiko mesaionikes ellenikes demodous gram-
maleias (Thessaloniki 1968) and in the Hislorikon Lexikon tes Akademias Athenon
(Athens 1933- )are not precise; however, the fact that it is smoked (akropastonj with
two sides, i.e. with a kind of skin? (sympleuronj, and coated in fat (syllardon) sug-
gests some kind of smoked sausage. See below, n.35, on obscene connotations of
apakin and its ancient comic equivalents; also n.39 on the joys of stolen food.

311
 Our penniless scholar takes again to the City's streets, led by
his nostrils to the butcher's shop. Excessive - not to say sug-
gestive - flattery of the butcher's wife earns him the promise
of a meal, which proves, on second mouthful, to be belly-meat
(lapara), stuffed with shit, than which even his ink would have
been more palatable (lines 227-57).
His next appeal is to 'Kappa', his rotten old cloak, but also
the tenth letter of the Greek alphabet, and, more significantly,
a term used for a small chapel for private prayer and worship.27
The Vlach woman who wove you, he observes to his kappa, sealed
into the fate of your owner tears, groans and hunger. Even on
high days and holidays, he is excluded from Church on account
of his disreputable appearance, so he has recourse only to his
dilapidated home, where he snuggles into his kappa, whose charms
prove irksome: at midnight he is awoken by lice, which cause
him to scratch and scruntch until he is blood-red from head to
toe (lines 258-74). And so straight into the closing address to
Manuel I, 'rose from the purple offshoot of the Komnenoi,l
emperor of emperors and ruler of rulers'. The juxtaposition of
the words olokokkinon (,fully red'), used for his lice-bitten state,
and komnenoblaston apo porphyras rodon, used for Manuel's
imperial majesty, suggests a deliberate contrast between the nar-
rator's poverty and his patron's wealth, possibly also with allusive
reference to the complex tetraugust metaphor and play on the
figure four, so prominent in the proem, and to an associated pro-
phecy, current since the birth of the future Alexios II, which is
based on the four-letter word AlMA {'blood'}, according to which
the Comnenes would rule so long as the initial letters of the
emperors' names .spelled the word A-I-M-A.28 His final appeal

27. The many uses of the word capa (or cappa), from personal garment to clerical
vestment and small place of worship, are listed in Du Cange, G/ossarium ad scrip-
tores mediae et infimae Graecitatis (Leyden 1688, repro Paris 1943) S.v., 118-9. From
it derives the word 'chapel' (*Z.Gavrilovic). The play on meanings is important, since
it links the seemingly unconnected themes of cloak, exclusion from public worship
and resource to private prayer. For lice, see below, nn.28 and 38.
28. The sequence runs: Alexios I (1081-1118), Ioannes II (1118-43), Manuel I
(1143-80), Alexios II (1180-83), Andronikos I (1183-85); for the prophecy, see Niketas
Choniates 169.91-5, ed. l.A. Van Dieten (CFHB, Berlin 1975). lowe this informa-
tion to the kindness of M.F. Hendy. The figure four, developed in the closing ad-
dress of the fourth poem with reference to the four soldier saints, is taken up from
the proem, where the tetraugust metaphor is further linked with the motif of the Cross.
It is particularly relevant to Manuel's porphyrogenital state, since he is the fourth

312
to his patron takes the form of a parody of the Lord's Prayer
('deliver me from want, deliver me from penury,/ resolve, 0
emperor, the demands of my creditors'), and an invocation to
Manuel's four favourite military saints, Georgios, Demetrios,
Tyron and Stratelates, whose images were engraved onto Manuel's
coinage. 29 All in all, the closing appeal is a subtly playful com-
ment on the significance attached to the concept of blood in the
Comnene court, and a bold demand for money.
The narrative, seemingly haphazard, is carefully structured.
Between the opening proem and closing address, themes relating
to the City's craftsmen are skilfully interspersed with shorter
appeals:

PROEM (CSA, g)
Father's promise of wealth and privilege in return for scholar's training(1-14)
Scholar's lot compared with that of othp.r craftsmen (gold cloth embroiderer,
cobbler, tailor) and of artisans and salesmen (baker, yogurt-seller, dyer,
weigher and hawker of cloths, pepper-grinder, sieve-maker) (15-140)
Appeal to Manuel, with prayer for continued success (141-4)
Scholar's sufferings of Hell's torments in present life (145-69)
Appeal to Christ, through Emperor (160-2)
Report to Emperor of recent exclusion from paternal home (initial promise
belied) (163-201)
Detrimental effects of hunger on memory (202-5)
Appeal to Hunger to go where she belongs (the peasants) (206-57)
Bakers and butchers rich; scholar excluded even from their scraps(225-57)
Appeal to Kappa; scholar excluded from Church; return to Kappa - and
lice (258-74)
APPEAL TO MANUEL (as 'rose of purple') and request for
money (274-92)

son, addressed as such en tois porphyranthesi sou adelphois by Michael Rhetor. see
W. Regel, Fontes rerum byzantinarum (St. Petersburg/Leipzig 1892-) VIIl. The modes
of address range from standard titles to less usual metaphors, all of which can be
paralJeled in contemporary sources: for Manuel as both 'born in the purple' and 'of
Comnene lineage', see, for example, Regel (1892) I, 111.41-12 (0 porphyras blaste
and '0 porphyras augasma), VI.121.9-1O (blaste porphyras), 122.3-4 (0 kalon basileias
anthos) *l.F. Haldon).
29. See M.F. Hendy Coinage and money in the Byzantine Empire 1081-1261 (Dum-
barton Oaks, Washington D.C. 1969) 111-22, Plates 12-18. The four soldier saints

313
 The poem's structure foregrounds the underlying complaint
of the scholar-craftsman: the status of the writer in the twelfth
century has been devalued, despite the lavish promises of the
previous generation; and, what is more, reduced to a level below
that of the common artisan, or of the common street hawker.
He has nothing to eat (except for his papers, his metres, stolen
foods and shit); nothing to wear (except his Kappa); he is excluded
even from Church; while they all live a life of relative luxury.
References to the average diet, or income, of artisans, and to his
own penury, do not need to be taken literally in order for the
basis of his complaint to be understood 30 : the scholar, who can-
not sell his wares on the market, is obliged to beg from his patron
(or steal from his father) in order to survive. At a deeper level,
perhaps the whole poem can be interpreted as a humorous - but
subversive - comment on the Lord's Prayer: he has been denied
his 'daily bread' by all his 'fathers,.3l

invoked here balance the winged emperor and Cross motifs of the proem, also engraved
on coins (the former from mid- to late thirteenth century), see Hendy (1969) ibid.
For Manuel's munifiC'nce, a recurrent theme of the poem (Ic and I r), see Hendy,
Studies in Byzantine monetary economy (Cambndge 1984) 199; and for imperial
euergesia, see H. Hunger, Prooimion, Elemente der byzantinischen Kaiseridee in den
Arengen der Urkunden (Wien/Graz/Kiiln 1964) 139-43. Manuel as successful soldier
is a favourite Prodromic topos, see somces cited by Kazhdan (1984) 105-7. Wealth,
munificence and military prowess are thus subtly interlinked in an indirect request
for money. For possible parody of the Lord's Prayer, see n.31.
30. See Kyriakis (1974) 290-309 for a crudely literal interpretation of the author's
(sic) poverty. Kazhdan (1984) 104-5 reminds us that the 'autobiographical' informa-
tion in the vernacular poems shouid not be taken literally, and itemises what can be
reconstructed of Theodore Prodromos' actual lands and properties: he was no pauper.
31. Clear allusion to the Lord's Prayer is made in the closing address to the emperor,
see lines 284-6. However, both the formal structure and the persistent message of
the whole poem can be read as a commentary on the Lord's Prayer which reinforces
the narrator's appeals to the emperor for help. The extent and subtlety of the allu-
sions can best be demonstrated by comparing the text of the Lord's Prayer with parallel
passages in the poem:
I. n(llEp il~wv 6 tv roi<; oupavol<; 2. aYlUa91l'tW 'to ovo~a aoo 3. tA9E'tW ill3amAEia
aoo 4. YEvV1l9Tj'tw 'to 9EAT]~a aoo 5. w<; tv oupavi!> Kat btl 'tii~ yfj<; 6. 'tOY liprov
il~wv 'tOY E1tlOUmOV 00<; n~iv ail~EpoV 7. KatliqJE<; n~iv 'to. 6qJEATj~a-ra il~wv 8.
w<; Kat n~Ei<; c'1(piE~EV roi><; Oq>EIAt-ra<; Tt~WV 9. Kat ~i] EiaEvEyKEI<; Tt~a<; d<; 1tEtpaa~6v
10. ana puaal Tt~a<; a1tO 'tou 1tOV1lPOU. 'A~Tjv.
Line 1: The idea of the emperor as supreme father is projected through the passages
where the poet seeks shelter and expects to be heard, protected and comforted (Ia-lh,
la-Ip, Imm-lxx) and 141-2,160-2,275-7. The poet seeks refuge in the emperor (the
supreme father) from the hardships provoked by his actual father's advice (I-IS);
he also prays that the emperor might keep his eternal and almighty sway (141-4),

314
4. Writing and Other Crafts
Before examining some of the possible social and historical
aspects of our writer's complaint, some literary and linguistic
features implicit in the reference to crafts may be usefully sum-
marised. The sedentary craftsmen (as opposed to street vendors)
with whom comparison is made include the gold embroiderer,
cobbler and tailor. Each shares with the writer a comparable use
of the hand in the exercise of their trade; each has a named work-
ing implement (pen and ink, awl, strigil and thread, needle and
scissors); and each works with a processed material (paper, cloth
and gold, leather, cloth). Each seeks a living (the verbs gureuo
and zeta occur no less than seven and four times respectively
throughout the poem); but, while othels find their mouths, cup-
boards and pockets stuffed with goodies (there are thirteen in-
stances in this context of the verb gemein and its cognates), our
scholar finds nothing but empty space and papers, or shit and
lice (except for what he can steal). Furthermore, there is a signifi-
cant interweaving of words connoting the execution of craftsman-
ship, and the enjoyment of the fruits thereof, which again
highlights the writer's disadvantaged position. It may be tabulated
as follows:
1. grammatikos technites
na plexo stichous ('to weave verses') (76)
na grapso kallista, na laryngyso stichous ('to write most finely, to spout
forth ') (77)

thereby perhaps suggesting that unless his own appeals are heard the emperor may
be treated by the Almighty as he has been by his own father? Line 2: the poet ex-
presses his plea for money in terms of a plea for bread. Reference to bread, and to
his lack of it since he became a grammatikos, is constant (16-17, 23-7, 31-2, 79-82,
135-7,156-8,202-12): he is in dire and daily need! Lines 7-8: he is deep in debt (Itt)
and implores the emperor to satisfy the demands of his creditors (285-7). Lines 9-/0:
the command rusai is used to implore the emperor to save him from hunger and poverty
(160-2,285). In fact, the whole poem is a plea for deliverance not only from hunger
but from moral decline, in accordance with the Byzantine belief in man's constant
need to combat the bestial side of his dual bestial/spiritual nature in order to reach
the desired state of apatheia: it is the poet's immediate hunger which enforces his
obsession with food, his jealousy of other tradesmen (23-140), his encounters with
females o~ the str~e~s whose sexual appetites appear voracious but fulfilled (99-108,
227-57), hIS (unwIttIng) provocation of advice from others to give up the spiritual
struggle ~ papas-grammatikos and become entirely bestial, like the common crafsmen.
VIewed In the context of the Lord's Prayer, the fourth poem is far from lacking in
deSIgn and structure (*c. Galatariotou).

315
 Podas melron Ion slichon ('measuring out the feet of verses'), contrasted
with amelron peinan ('measureless hunger') (69)
tsiknono ('I grow faint'); skolizomai ('l get dizzy') and kolazomai ('I am
damned') (135, 158, 150)
phage grammala ('eat letters'); me tragei (Penia) ('(Penury) eats me); n'alroges
. . . ·10 melanin . .. para ten skalolen laparan ('(better) to have eaten ink
than shit-filled belly-meat') (137,152,256-7)
2. olher craftsmen
Isangares ('cobbler') as technites stichistes ('stitcher-craftsman') (74)
other lechnites as philosophos, rhelor, kalligraphos (because they eat and
earn money) (210)
pelsonei ('cobbles') (57)
klasei, klathei ('spin out', in sense of 'enjoy food') (54, 68)
kenteson ('sew') (96)
roukanizo ('chew', 'nibble at' - with enjoyment) (102, 105)
keimena ('that which lies before', usually used for 'texts', here for plentiful
fish or smoked meat (134, 181)
The metaphors of embroidering, weaving, sewing, writing and
eating are inter-linked, inherent in the concept of techne, consis-
tently underlining the writer's exclusion from the fruits of his
labour.

5. The (Pro)-creative Act: Food, Sex - and Literary Texts

Much reference has been made by commentators to our author's
'obsessional attention to what everyone has to eat'. 32 Less
caution has been shown in the interpretation of some metaphorical
terms as literally gastronomic. To give a few examples: the baker's
wife, who nibbles at the end of a crunchy white loaf (104-8); the
yogurt-seller, whose frothy white goods are there for the asking
(109-14); the fuller, the cloth-seller and the pepper-grinder, who
summon the c;heiromachisses ('good lady hand-workers') of the
City to size up their wares (114-29); the sieve-maker, whose hearth
is always crackling and sparkling with spitted meat and fish
(130-4); the tasty apakin, stolen by our writer but blamed on the
cat (192-201); the butcher's wife, with her tempting breast-meat
(mastarin) and hanging belly-fat (lapara) (233-9): all suggest the
promise or practice of sexual fulfilment among other craftsmen.
The systematic correlation between eating and sexual activity not
only adds to the humour, but also comically intensifies the
abjectness of the writer's plight: while others have flesh piled
32. C.A. Mango, Byzantium: the Empire of New Rome (New York 1980) 83,251.
The remark is a worn lopos of Prodromic commentarv.

316
before them (significantly, the word used is keimenon, 'that which
lies before' or 'text'), he is forced to suffer starvation (apart from
metrics, papers and scatophagy, or stolen foods).
It is significant that many of the tradesmen, and their attributes,
are comic topoi of prestigious lineage. The cobbler (AG skuteus),
for example, was traditionally regarded as the lowest of the low
banausoi, often bald, small and avaricious, from Plato (who op-
poses him to the Philosopher), Old and New Comedy, the Mimes
of Herodas, late prose-writers and Byzantine literary sources. 33
The connections between the crafts of writing, weaving, sewing
and patching are no less ancient. 34 Sausage-sellers, butchers,
bakers and fullers were among the banausoi, exploited as much
for their dubious associations as for the execution of their
trade. 35 Nor is Prodromic use of comic topoi limited to crafts-

33. See, for example, PI. Gorg.49Ia. The lowly status, dubious practices (and
products) of cobblers (in addition to the sale of extravagant shoes at exorbitant prices),
are documented comprehensively by W. Headlam, Herodas: the Mimes and Fragments
(ed. A.D. Knox, Cambridge 1922), pp.xlviii-xlix, 301-2, 304, 328, 336-7, 339. Lucian
i.636 chooses the cobbler as the typical poor man who is delighted to die; while Julian
Or.p.8IB stresses the ignobility of cobbling and other menial trades. Also traditional
is the coupling of cobbler with schoolmaster (grammatistesj as despicable, see Dio
Chrys.2.219, Plut. Mor.776B. Among Byzantine writers, Tzetzes expresses similar
sentiments: lamb. p.511, Kiessling. To these sources cited by Headlam may be added
one (out of many) remarkable parallels to our poem from the historical poems of
Theodore Prodromos, ed. Horandner (1974) 378-9 (XXXVIII.3544), where, in a poem
addressed to Anna Doukaina, he recalls his father's advice to 'learn his letters', since
he was too weak to be a soldier, and other trades were too base! As in our poem,
his efforts were rewarded with poverty. Headlam's editorial practices, both for
Herodas' Mimes and Aeschylus' Oresteia (ed. G. Thomson, Cambridge 1938), re-
main a model for scholarship and a mine of information, not least for Byzantinists.
Of particular value is his careful sifting of literary topoi which remained traditional
from antiquity throughout the Byzantine period, always viewed with reference to the
specific literary and historical context of each author. It is important to stress that
the use of topoi is not incompatible with historicity; nor should the process of documen-
ting them be confused with the nebulous and romantic concepts of 'Greekness', or
cultural continuity which is perpetuated by the unlettered Greek 'folk'.
34. See, for example, the literary and historical evidence cited by A.T.L. Bergren,
'Language and the female in early Greek thought', Arethusa: Semiotics and Classical
Studies 16.1/2 (1983) 69-96, and the etymological evidence adduced by Nagy (1979)
299-300.
35. Apakin seems to have taken over the metaphorical overtones of AG alias
('sausage'), used to denote the male member in Hipponax 48, Crates Com.l6, Ar.Eq.
161: the role of the allantopoles ('sausage-seller') in Aristophanes' Knights is sug-
gestive for sexual, homosexual, scatological and political innuendoes. For these and
other sources, see J. Henderson, The Maculate Muse: ob.orene language in Attic comedy
(Yale 1975) 20,66-70. The connection between alias ana apakin (in conjunction with
loukanikoj has been noted by R. Bancroft-Marcus, George Chortatsis: a critical study

317
men: the Pontos, as plentiful source of small fish and fat old
men;36 luxury in shoes as a sign of wealth and effeminacy, much
frowned upon by the Church;3? the lice-infested cloak of
Poverty;38 even the poor old cat as scapegoat: 39 all have an

(D. Phil, thesis, Oxford 1979) 242ff, who cites a number of remarkable parallels bet-
ween ancient and Cretan Renaissance comedy in the obscene connotations of certain
foodstuffs, animal and vegetable (see for apaki Chortatsis Panoria A 389, B 18,
Katzourbos glossary and Foskolos Fortounatos E 55,70, all of which pass without
editorial comment). That this, too, was a literary topos, rather than G!le to conscious
imitation (or oral tradition), is indicated by its presence in the commentaries to
Aristophanes (ad loc.) by Tzetzes. For enteron as "guts, reached through the vagina",
see Henderson (1975) 20,69, 125: cf. the taunt of the butcher's wife to our scholar-
narrator at line 255 grammatike philosophe, enterochordoplyla. Kreas ('meat') occurs
in Aristophanes both in homosexual contexts (Eq.428, 484 Sch., jr.130.3) and as a
slang term for 'female parts' (Ach.795, Lys.I062). As for bakers and their confec-
tions, different kinds of cake (plakous) were exploited in antiquity for obscene con-
notations, Henderson (1975) 144-5, duly annotated by Tzetzes. The katablaltas ('dyer',
'fuller') seems to be the medieval equivalent of AG knapheus, gnapheus, on whose
lowly status and dubious associations, see Headlam (1922) 211-2. Of uncertain meaning
is sekotes: either 'weigher of goods', cf. zygostates, as suggested by Hendy (1984) 589
and n.168, or 'carrter', 'porter', in accordance with the modern sense of the word;
if the latter, his status on the City's streets was, indeed, the lowest of the low, yet
even he, according to our text (114-8), was assured of his 'daily bread'. See Koukoules
(1948-55) 2, 185-6 for the status of bastagarios, notophoros, phortiaphoros, sekoles.
36. See, for example, Men. Samia 98-101, ed. A.W. Gomme and F.H. Sandbach,
Menander: a commentary (Oxford 1973) 555 (reference owed to the kindness of A.
Henrichs).
37. IV.5-6: for condemnation of male luxury in shoes, especially vehement in the
Church fathers, see sources cited by Headlam (1922) xlviii. Koukoules (1948) 4, 397,
407 -8, provides parallels to Byzantine fashions in footwear.
38. Cf. Ar. PI.537-47: the similarities are extremely close. According to Byzantine
dream books, to dream of a shepherd's cloak signified envy of another's wealth,
poverty, or long sickness (all of which our narrator complains of throughout the four
poems), see F. DrexI (ed.), 'Das Traumbuch des Patriarchen Nikephoros', Beitriige
zur Geschichte des christlichen Altertums (Bonn/Leipzig 1922) 103.64; 116.324;
117.325; 326 (*G. Calofonos). The association with dreams is suggested by our nar-
rator's drowsiness (nustazo 268) and sleep (koimoumai 269), wrapped in his kappa
(tyligomai268), following his exclusion from Church (ekklesia 262) and his enforced
retreat to his dilapidated home (palaiospiton, kainourgiochalasmenon 267). The col-
lapsing house and human disease (cf. 1.75-87, II.57a (H), IV .177, 180) as apocaplyp-
tic symbols are attested in Joseph Bryennios, ed. Eugenios Voulgaris ii (Leipzig 1768),
p.191, and paraphrased by Mango, Byzantium and its Image (Variorum, London,
1984), p.34: "for just as the death of a body is foreshadowed by sickness and gradual
disintegration, as the collapse of a house by cracking walls and fallin,' ..,laster, so
indeed is the end of the world indicated in advance by the disappearance of aU goodness
and virtue, the growth of wickedness and superstition, and by the fact that the Roman
Empire had contracted as never before". Once more, apocalyptic dreams and portents
are interlinked with the contrast between wealth (his patron's) and poverty (his own)
in order to suggest his dependence upon letters (kappa is the tenth letter of the Greek

318
honourable literary pedigree, from Aristophanes to prestigious
Byzantine writers.
As for obscene connotations of certain foods and cooking
methods, these too are based on traditional metaphors exploited
by comedy and satire: to eat (br6z6 = 'nibble at') for sexual
appetite, especially female;40 frothy milk, or soft white cheese,
for male ejaculation;41 grinding pepper, or roasting hunks of
meat and fish on sparkling red-hot coals for female and male

alphabet) and his need for cash: the letter kappa was inscribed as part of the five-
letter Cross monogram on the obverse of Manuel I's coins, see Hendy (1969), Plates
17 and 18, and p. 121 for documentation on the half tetarteron, mint of Thessalonica,
Type A (heavy standard):
M
~-n-K
J\
MIANOYHIJ\ ~IEcnoTHq nlOP<I>YPOrENNHTOC 01 KloMNHNOq
39. Cf. Babrius xxvii, Lucian i.603 (harpaktikoteroi ton galon), Ar. Vesp.363, Pax
1151, Thesm.559 ('it must have been the cat!'), Plu!. Mor.519D (kathaper opson,
gales paradramouses, airousin ek mesou) (sources cited by Headlam (1922) 35S. The
point was not lost on twelfth-century intellectuals, see, for example, Eustathios of
Thessaloniki on the joys of stolen foods, particularly among monks: Mtpo~ /)/; OUK
i;),.aXlCHov urrOA.U00EW~ EV KOOI1IKOi~ I1/;V 'to QJuYEiv, EV 110vuxoi~ o/; ou 'to
UOUVOE'tWC F.!rrEiv QJuYEiv (umo yup rraYKOlvDv Eon) UA.A.U 'to A.U8pOQJUYEiv, 6 'tUu'tov
I:O.I'<!J ,;otw~ QJuYEiv, Ku8ml KUt KA.orrii~ ()/)wp yA.UKEPOV ypaQJE'tU1 (*G. CaIofonos)
(0.229.94).
40. See Headlam (1922) 347-S on brozousi (Mimes VII. 63) as a form of bibroskousi.
cf. Hesychios: trozein' psithyrizein. synousiazousin. and Aeschylus, fro 44 era men
hagnos ouranos trosa; khthona. Dogs' appetite for leather was proverbial, as a habit
hard. if not impossible, to unlearn. Lucian iii.121: the sense in Herodas is that women
exceed even dogs in th.eir appetite for 'leather'.
41. "Eparete droubaniston oxygalan, gynaikes!" (IV .112). For droubaniston, see
Kriaras (I 96S), S.V. droubanizo. Droubani is glossed by G. Hadzidakis in BZ I (lS92)
99-100 for contemporary meanings. See also V. Tsiouni, Dieg. Paid., for form
draganizo (?Slavonic root), and F. Miklosich, Lexicon palaeoslovenico-graeco-Iatinum
emendatum auctum (AalenJ 977). Droubanistes is the name of a torrent on Mt. Athos,
according to a document of 1496, Actes de Dionysiou, ed. N. Oikonomides (Paris
1968) 39.S, IS4 (*A.W. Dunn). Oxygalon is probably yogurt, here beaten up to make
a refreshing and frothy drink, much like the modern Turkish ayran (drink made with
yogurt and water, mixed with snow or ice, rather than 'buttermilk'), still sold on
the streets of Istanbul, see Hendy (1984) 588-90. For obscene connotations of milk
and cheese, see Bancroft-Marcus (l97S) 253-4 on Chortatsis Pan.C38S: na sou gemiso
alhogalo ole Ie galautia sou (Yiannoulis to Frosyni). Tzctzes, commenting on Ar.
Ran. 1328, elaborates on the 'twelve positions' of the hetaira Kyrene, with a
characteristic attack on earlier scholia~ts which implies an acknowledged association
between milking goats and odd forms, f sexual intercourse: W.l. Koster (ed.), Scholia
in Aristophanem (Scripta Academica Groningiana, Gronigenl Amsterdam 1962) IV.3
(In Ranas 132S, 1340, pp.1074-5, IOSI. The topos is not unattested in Hebrew and

319
roles in sexual intercourse;42 scatophagy for homosexual and
auto-erotic practices. 43 The enjoyment (or deprivation) of food
and sex is expressed in an interchangeable system of metaphors,
which modern anthropologists are just beginning to
understand. 44

6. Text and Inter-textuality

The existence of comic stereotypes, both ancient and contem-
porary, suggests that our poet, although writing in 'vulgar' Greek,
was working within a known and established literary tradition,
adapting familiar topoi to his own purposes. The question arises,
are t'1e similarities, most strikingly with Aristophanic hllmour,
coincidental, or do they imply conscious exploitation? First, both
the number and closeness of the parallels, listed here cursorily
for the fourth poem only, would seem to preclude the former.
Second, there is no objective difficulty in presupposing a close
and direct knowledge of Aristophanes (and other comic writers)
on the part of our author, as well as a predisposition to compare
and contrast past with present mores, since any contemporary
of John Tzetzes or Eustathios, Bishop of Thessaloniki, especially
if he were intimate with the Comnenian court, would have en-

Byzantine religious literature, see 1. Grosdidier de Matons, Romanos Ie Melode,

voI.4.pp.166-71 (Sources Chretiennes 128, Paris 1967).
42. Grinding pepper as metaphorical for sexual intercourse is frequent in modern
folk tradition, see, for example, D.A. Petropoulos, Ellenika demotika tragoudia II:
Basike Bibliothekeno.47 (Athens 1959) 209-10 (·A. Kasdagli). The nearest equivalent
in Byzantine source~ is perhaps to be found in the satirical song, purportedly of the
tenth century but extant in a sixteenth-century ms. from the Marcian Library, Venice,
edited by G. Morgan, 'A Byzantine satirical song?', BZ 47 (1954) 292-7. Although
somewhat obscure, the obscenities connected with grinding, baking and trumpeting
are plain enough. For roasting meat or fish on red-hot sparkling charcoals (IV .130-4),
see Henderson (1975) 142-3 for the unequivocal connotations of Ar .Eq.1286 (kukon
tas escharas), glossed by the scholiast as ta cheile ton gunaikeion aidoion. Eschara
as 'labia' was a common term, see Eus!. Thess. 1523.28, 1539.33. These and further
particulars are cited by Henderson.
43. Henderson (1975) 192-3: eating dung meant low, animal behaviour, notably in
the contexts of homosexuality and auto-eroticism (see especially the sausage-seller's
threat to Kleon in Ar.Eq.295 koprophoreso se). Note the proverb concerning
scatophagy recorded by Tzetzes. (Chil.IO.306, 82) (·G. Calofonos).
44. The interconnections between eating and sexual activity appear to be universal;
the particular forms they take (including taboos) are culture specific, see E. Leach,
Social Anthropology (London 1982) 114-6, 196, 221.

320
joyed access to the latest literary discussions, as well as written
commentaries, on that subject. 45
Can such a stance be reconciled with our narrator's consistently
stated 'ignorance' and lack of education? No real problem exists
here, since, once again, the statements, as elicited from the four
proems, consist of literary devices rather than biographical facts
(see Appendix II).
Our poet tells us clearly not that he is uneducated, nor that
he is incapable of writing learned verse, but that he seeks a new
diversion, for the emperor, himself (and, implicitly, the reader).
By writing and speaking (graphOllalO) - the tautology is explicit
and significant - unadorned lines in 'political verse', he claims
it possible to be more truthful and outspoken than the 'fable-
mongers' (my thoplastes) , who merely flatter (another familiar
Byzantine topos in new form).46 Further, learned verse, quite
simply, does not payoff, as the scholar's fate in Poem IV, who
seeks pyrrhics, iambs and other abstruse metres in lieu of bread,
while others stuff thetnselves, seems to prove.
The use of vulgar Greek is appropriate to the context of beg-
ging, not least because the narrator promises to amuse rather than
to lament his lot (in fact he does both quite successfully); it is
also an apposite means of expression for the numerous comic
45. Wilson (19M3\ 190-6 provides ample evidence for the continued interest in
Aristophanes throughout the twelfth century. Of Tzetzes, he notes that 'while he
thought that some of Aristophanes' plays were excellent, he was unable to enjoy the
Frogs. He complains more than once (at lines 25 and 1144) that the poet must have
been drunk when he wrote, while on line 358 he remarks that the poet does little except
talk nonsense in this play. The obscenity of 422 irritated him. But it does not seem
to have occurred to him to reave the play out of his reading-programme' (194). Cur-
rent research in our Text Seminar is revealing many subtle allusions to Aristophanes
and other comic writers in undisputed works of Theodore Prodromos, in particular
in the historical poems edited by Horandner (1974).
46. See, for example Constantine Porph~rogenitus, De Caerim., praef., 5, 2-11; and
Kekaumenos, Strategikon, 75-6, 30-1. 1.• The opposition between what has been
seen and experienced by the writer as incontrovertibly true, on the one hand, and
what has been learned only from hearsay (akoe), or from other writings, as impli-
citly or potentially false, on the other, is one which runs right through Byzantine
writers from Prokopios to An-na Comnena and beyond: it suggests that, despite the
advanced state of literacy (at least among the elite), prime importance was attached
to direct visual evidence, as in oral or residually oral cultures, rather than to
documented (therefore 'written ') proof, as in societies reliant upon the printed word,
see W.J. Ong, Orality and Literacy (London 1982) 96-101. The new subtlety introduced
by the form of the topos in our poems is that narrator foregrounds his intention to
write fluently in the language used normally for speaking directly, and therefore, to
write truthfully.

321
topoi, permitting greater licence of speech than more conventional
forms of language would allow. The choice of linguistic register
is, then, deliberate, bold and original, rather than the result of
chance, or lack of skill. It could, of course, be argued that the
similarities between our text and comic tradition are due, not to
conscious borrowing, but to the survival, through oral trans-
mission of certain comic themes and metaphorical nuances.
However, our author is neither illiterate, nor a folk bard, but
a skilled writer who makes constant reference to his own presence
in the text. 47 The oral hypothesis cannot explain the use of in-
trinsically literary devices, although it cannot be denied that several
obscene connotations in the poems can be paralleled in comic
and satirical works from Aristophanes to the Renaissance, and
in popular tradition.
Those who argue that the text, as we have it, reflects varying
degrees of oral interference, particularly in the later stages of its
transmission, support their case with reference to the numerous
lines which can be paralleled in popular verse extant from the
fourteenth century and in modern folk songs, as well as to the
nature and degree of variation in the later manuscripts. 48 Such
a possibility cannot be excluded, although it remains to be
demonstrated that the variant readings are the result of oral rather
than aural transmission. Further, by no means all vestiges of
popular tradition can be eradicated as interpolations of later
scribes, since many such lines are attested in all the manuscripts
and are integral to the meaning and context (notably in the appeal
to Kappa).49 It is equally possible that the author was con-
47. For the intrusion of narrator in the text as a phenomenon first found in Byzan-
tine texts of the late.eleventh century, developed in the twelfth, see Kazhdan (1984)
192-4. Correctly, he situates the development in the context of social and intellectual
upheavals of the time. R.M. Beaton, in a paper delivered to the Eighteenth Spring
Symposium of Byzantine Studies (University of Oxford, April 1984), explores the
literary aspects of some twelfth-century narrative techniques, first discussed in the
context of the Prodromic poems by Alexiou in the draft paper (April 1983) which
formed the starting-point for this article; see also Alexiou (1983) 29-46 for a literary
and stylistic analysis of the Timarion.
48. See Appendix I for details of the discussion.
49. Jeffreys, in his careful review of Horandner (1974), notes the fondness in the
Prodromic poems for rhetorical devices which can also be found in folk songs, par-
ticularly for the 'framework of three', whereby the balancing and antithetical halves
of one line are rounded off by the first half of the following line, (1977) 106-7.
However, the parallels are by no means limited to rhetorical and structural devices,
as a d tailed study of lines 258-65 will show:

322
sciously exploiting formulaic lines familiar from folk songs, as
would be consistent with his choice of 'vulgar' language. If cor-
rect, the supposition that the author consciously and simulta-
neously alludes to both ancient literature and to popular tradi-
tion must suggest a deeper affinity, and closer proximinity, bet-
ween 'high' and 'low' literature in the twelfth century than has
hitherto been assumed.
Nor is that all: other levels of allusion in our text which can-
not be ignored, although more research is needed before they can
be documented, include imperial and ceremonial modes of address

(il Ka1t1tu 1l0U, 1tUAIV KU1t1tU Ilou, 1tUAUIOXup!3UAWIlEV11,

Ka1t1tU Ilou, 5V1:UV O'I\9EKEV Tt !3Auxu va of: q>aVEI (258-9).
Cf. appeals to objects in the folk tradition:
aq>Evt11 ll'aq>EvtuK11 Ilou, m:v'tE ~0A.tC; aq>Evt11 Petropoulos (1959) 1I.23
rUYlo Ilou, XpuooyaylO 1l0U, 1tUAI Xpuo6 Ilou YUylO Ibid. 129-30
!3AuvtillOU, 5V'tEC; o'aVe01:UIVU, Ill: 1tpO~EvoAOYououv
N.O. Polites, Eklogai apo ta tragoudia tou ellenikou laou (Athens 1914),
p.45.31
~Iuoilil, ltoAulilUOllio, KUAov KUlPOV liluollevo,
lilUOilil, 5tuv 01: liui~oUIlOUV, TtP9civ 01 OUIlltE9tpm ibid. 123.85.
Similar appeals are found in the verses of Michael Glykas (ed. E. Tsolaki, Thessaloniki
1959): ",ux1'\ 1l0U KUKOtuXEPE, Iliuv ~XUP11C; ropu (196),
and in the Erotopaignia (ed. Th. Siapkaras-Pitsillides, Thessaloniki 19): OKOUq>IU 1l0U
ltuYKAuoiliwt11 KUt ltuYKAaOlOWlleVTj (435).
(ii) Kut t1'\v AUIl1tP1'\v t1'\v KuplUK1'\v Ot1'\v £KKA.110IaV liv 1tuyw (263)
Cf. Kut IlUI AUIl1tP1'\, 1110. KEpYlUK1'\, IlIIlV 1ti0111l0C; TtIlEPU ... Petropoulos
1I.59
(also 11.63,11.78, II.122, II.129, O. Ioannou Para/ages (Athens 1975) p.52).
Kut 1110. YlOptlj, Ilill KupluKlj KUt IllaV AUIl1tPIIlV TtIlEPU Petropoulos
11.87 (also Politis 110.80, 122,85)
Kat Ilill YlOpt1'\, ilia Kupla Klj Ilill nUOXUA,la IlEYUA,11
t1'\v Elliu ltOV otoAi~ovtuV Otf]v £KKA11Ola va 1taYEI Oiankas, Epeirotika
aemotika tragoudia (Athens n.d.) 415-6.
Again, the line is found in vernacular verse of the fifteenth and sixteenth centuries,
for example Synaxarion tou timemenou Gadarou (ed. O. Wagner, Leipzig 1874) 5:
An' 51lcoc; taxu Ka1tOtE, AUIl1tPU TtllepU Tttov, and Gadarou, /ykou ki aloupous
diegesis oraia (ed. Wagner)25 : Kutllia Aall1tP1'\ tf]v KUPIUK1'\ tUxa Au1t1'\911K€ toy
(iii) 5AOUC;':WPEI it ~KKA11Ola KUt IlEV' ou,sl:v tXcbPEI
This formula, suggesting exclusion, occurs both in folk songs:
"OAU ta vt6mu ta nOUAla I\xouv q>O>At£c; Kat IlEVOUV Ioannou 283
"OAu ta &tVtpO to 1tpwi liPOOla Eiv(!l YEIlIOIl£VO ibid.285
OUAm tOv TtAlo tOY tllPiiv, 1tOV 1t0.&1 vu !3oOlM",EI Politis 163.128B
(also Petropoulos II .59, II.J 58); and in vernacular laments for historical and
natural disasters, see M.F. Herzfeld, 'New \ight on the 1480 Siege of Rhodes',
British Museum Quarterly 36 (1972) 69-73, and, in slightly different form, To
Anakalema tes Konstantinoupo/es (ed. E. Kriaras, Thessaloniki \956) 31.56-60.
(iv) Kat altO to OEio!loV to ltOAVV KOt TO 1tOAV to 5t&llo (265)

323
and religious and liturgical texts. 50 The most striking examples
of the former occur in the proems and formal addresses, while
religious allusions are commonest in the frequent episodic digres-
sions (see pages 10-14 above, and Appendix II).
In differing degrees, a quadruple system of literary allusion
may be postulated: to ancient and Byzantine literary tradition;
to orally transmitted verse; to ceremonial documents; to religious
and liturgical texts. The author's motives are a matter for con-
jecture only; but the artistic functions of his inter textual play in-
clude, on the one hand, his attempt to amuse and divert patron
and readers by introducing fresh areas of allusion in new com-
binations, appropriate to the linguistic register; and, on another,
his effective subversion of the 'truth' of what he seems to be saying

The motif of swaggering motion, usually in an erotic context, is common, cf.

\.10, 'V<1<; VIO<; K<1/..6<;, K<1A.O<; K<1\ liuo\.l<1'tapll<; Politis. 151.1 10
\.IE 01;10\.1<1 KU! \.IE: A.lryIO\.lU 't1] OKO,A.<1 V-aVE~<1(VEI ibid. I 13 .81.49
\.IU f.yW 'to VIO ltOU aYO,ltllO<1, 'tou K60\.loU lilw\.I<1'tapll<; ibid. I 15.82.14
KO\\.ITlOOU ytt \.IOU, K<1/..6YU:,<J\.IOP<PE liuo\.l<1'tapll Petropoulos 11.147
Further, the syntactic structure, introducing the result (usually negative) of excess,
can be paralleled in laments and Kleftic songs:
K1 (utO TOV \.I60xo 'rov ltOAl>, KI altO TTt \.IupwlilO, TOU Ioannou 288
Kl anD 'rov \.I60xo 'tOY ltEp06, KI altO 'r1] ,.1UpouliIO, 'rou Politis 214.204
KI alto 'rTtv ~EPll<po'vla TOU Kat altO T1] A.E~EVna 'rOU Petropoulos II .246
(Folk song parallels researched by *A. Kasdagli). More research is needed before firm
conclusions can be drawn; but :neanwhile such formulaic lines should not be dismissed
or attributed to later scribal interference.
50. Standard titles and modes of address in the poem include autokrator. autokrator
ton Romaion (adopted from Michael I, 811-3), despotes (introduced by Justinian I,
527-65), augoustos (after tenth century only used on imperial edicta and letters to
foreign powers), stephephoros, kratistos, tropaiouchos, anax, skeptouchos, kratarr::hes:
documentation for their use in the twelfth century is cited in books and articles refer-
red to by F. Dolgei, Byzantinische Diplomatik: 20 Aufsiitze zum Urkundenwesen
der Byzantiner (Ettal 1956); see also F. DOlger and J. Karayannopoulos, Byzantinische
Urkundenlehre (Munich 1968), Hunger (1964), Hendy (1969), Mango 'The Conciliar
Edict of 1166', DOP 17 (1963) 315-30. Less usual modes of address and metaphors
include the emperor as skepe('protection'), cf. sources cited in Regel (1892) IIIAI,
VLI03; as harbour/haven/sea of tranquility (lg-Im, loo-rr, cf. IlI.440), see Regel
(1892) IX.156; as refuge and haven (Ii, Ik,lrr,283), cf. Regel (1892) IlI.27, 41; as
bestower of wealth (lc,280), cf. Regel (1892)
I.IV.17-18; as ruler of rulers (lp,I60,163,276,284), cf. Mango (1963), Diilger-
Karayannopoulos (1968) 157, Regel (1892) I, III.40, VI.I03; as successful soldier
(Icc-Ill), cf. Mango (1963), Regel (1892) 1.1.4, 15, IlI.28-9, 40-1,43
VI.10I,103,VIII.15I,IX.156,162; as wise as Solomon (lxx), cf. Mango (1963) 28, Regel
(1892) I,VIII.I32, X.I66; as mimesis theou (lj, cf. Ill. 16), cf. Hunger (1964) 58-63.
See also note 28. (* J.F. Haldon).For liturgical and religious texts alluded to in poem
IV, see the Short Note by Z. Gavrilovic, in this issue below.

324
(or writing). Can we take seriously the exaggerated enkomia of
the four proems, with their obsessional re-turns to the emperor-
addressee's purple-sprouting tetrarchate, and to the author-
addressor's illness and poverty? How are we to react to the jux-
taposition, at the end of the fourth poem, of the poet's scarlet
and lice-bitten state with the emperor's 'rose-hued purple'? And
are the obsequious appeals to the emperor and others for mercy,
interspersed as they are with graphic accounts of penury, perhaps
intended to remind his patron that fortune is fickle? - just as
his own father's promise of wealth and prestige has not been
fulfilled, so, too, his glorious patron should remember that no
pinnacle of glory, however mighty, can :ast for ever.
The narrative framework is fictive, the allusive system com-
plex and skillful. Does this mean that our four poems cannot be
'historical'?

7. Text and context

Granted that our four poems cannot be dated on either tex-
tual or linguistic grounds, what evidence exists, within the text,
for their composition in a form essentially similar to 'our' ver-
sion, during the late twelfth century? The question is complex,
and fraught with pitfalls, since historically dateable elements do
not necessarily dictate a definable terminus ante quem, but may
reflect later re-workings or garbled scribal memories. 51
The evidence of material artefacts, especially coinage and cur-
rency, as well as references to historical events and figures, are,
however, likely to prove valuable indicators. Further, the under-
lying social complaints should be viewed in context, not dismissed
as literary conceits, however universal and ubiquitous a figure
the poverty-stricken academic, in comparison with dustmen, or
miners, may be.
The numismatic evidence points unequivocally to the years 1163
to 1204, probably to the early 1160's or early 1170's, with
indications of later re-workings in variant readings of some later
manuscripts. The main points may be briefly summarised by
reference to the text:

51. H. Gregoire's attempts to date Digenes Akrites from 'historical' elements in the
text have been questioned, see Beck (1971) 63-97 for a useful summary; over-reaction
to 'Gregoirism' should not, however, lead scholars to ignore the historical and dateable
evidence provided in literary texts.

325
 IV .13, cf. IIl.408.
1. Kat 'twpa 'to. tl1ttp1topa ytJ,lEt 'to. J,laVOTJAa'ta
The term manoelaton was not much used after 1204, the last
recorded instance being from the Ionian Isles in 1216; the terminus
post quem is 1143. 52 Most probably, a date some time after
Manuel's second coinage c.1163, from which time on large quan-
tities of the coin were struck, during the third coinage, is
indicated. 53 Further, each of the four denominations of
Manuel's first coinage depict Christ Emmanuel on the obverse
as a young and beardless figure, possibly with reference to the
emperor Manuel's young and beardless age of 17 or 18; this may
explain the use in the text of poems III and IV of the epithet
christomimetos, all with direct reference to Manuel. As for the
term hyperpyron, it is used, in common with elektron, as an epithet
for trachy, to distinguish the manoelaton from the billon trachy
(= stamen on) and from the copper tertarteron. Again, a twelfth-
century context is indicated.

2. Kat vo, 1tntoiv J,lOO, o'taJ,lEVOV et~ 'to. XOPOOKOtAl'tOa IV .51,

cf. IIi. 127,130.
The stamen on (or histamenon, Latin staminum), is known from
the eleventh century as a billon coin, referred to as staminum
in western sources, but as nomisma trachy in the majority of
Byzantine sources. 54 Our text provides one of the few twelfth-
century references to the coin as stamenon, with the contextual
implication that the term was in normal, everyday use, although
not preferred by more learned writers.

3. J,lE'to. J3EA.OVtv mp'tEpou Kat paJ,lJ,lam o'taJ,lEVOU IV. 91, cf.

III. 130
The tarteron (or tetarteron), a gold coin at the time of the Book
of the Prefect, became i~creasingly debased during the second
half of the eleventh century, and was finally restructured, becom-
ing a base metal denomination in the coinage reforms of Alexios
I in 1092. The implication here that it was a small coin of low

52. Hendy (1969) \9-20, 27.

53. Hendy, Catalogue of the Byzantine coins in the Dumbarton Oaks Collection
and in the Whittemore Collection IV : Alexios I to Michael VIII (1081-1261)
(Washington D.C.): in progress.
54. Hendy (1969) 5,6,28-9.

326
value points to the early 1200's at the latest. 55 This can be sup-
ported by the variant tornesakin for tarteron in one manuscript
(g) at IV.83 and 91, the former having replaced the tarteron as
a low-value billon coin during the thirteenth and fourteenth
centuries. 56 Numismatic evidence therefore points consistently to
a period from the 1160's to the early 1200's.
References to historically identifiable events and figures are
rare, except for the dubious proem to poem IV, where cumulative
evidence points to the years around 1175. 57 However, the
general complaint, that craftsmen and artisans were better off
than they deserved, receives excellent corroboration from other
sources which can be dated to the latter part of the twelfth cen-
tury, probably before the mid-1180's. Benjamin of Tudela, who
visited the imperial City in the year 1168; reports as follows:
They say that the city's daily income, what with rents from shops and markets
and what with the customs levied on merchants coming by sea and by land,
reaches 20,000 gold pieces. The Greeks who live there have a wealth of gold
and jewels. They walk about dressed in silk, with patterns of gold sewn or
embroidered onto their garments. They ride their horses like princes. Now
the land is very abundant in good fruits, in bread, in meat and wine, so that
no other can be compared to it in richness. It has men learned in all the books
of the Greeks. Its inhabitants, each and everyone, eat and drink beneath their
vine and their fig tree. 58
55. Ibid. 6-7, 23-5, 28-9.
56. Hendy (1984) 534-45. Sec Appendix II, Proem IV lines 1jj-lI.
57. Nur ad-din (1 kk), ruler of Aleppo, established Saladin in Egypt, and in 1171
he was free to turn his attention to the north, where he built up a strong coalition
of forces against the Selcuk ruler KIllc Arslan. A pact between the two in 1173 agreed
that Arslan was to take his responsibilities in the Holy War more seriously; Nur ad-
din died in 1174. Melias, or Mleh, brother of Theodore (Toros II, ruler of the Arme-
nian principality in Cilicia 1145-69) succeeded by virtue of his under-age son Ruben
11(1169-70); Mleh invaded the realm on Ruben's accession with the support ot Nur
ad-din, and seized control. Reversing the traditional policy of the dynasty for allying
with the Crusaders, Mleh made a pact with the Muslims. His position was undermin-
ed by the death of Nur ad-din in 1174, and he was murdered by his own vassals in
1175. Stefan Nemanja, the most vigorous of the Serbian princes to submit to Manuel
after an imperial show of force in 1172, was brought to Constantinople as a captive.
The fact that he was not heard of throughout the rest of Manuel's reign, but emerged
soon after as the successful leader of a breakaway Serbian state, argues a date prior
to 1181 for the composition of the proem, since after then he would not have been
used as a negative example for the prospective rebels; by the later thirteenth and four-
teenth centuries he would have been a forgotten figure. It should also be mentioned
that the proem seems to contain no reference to the unfortunate outcome of the Bat-
tle of Myriokephalon (1176). (0 A.E. Harvey)
58. Cited in A. Sharf, Byzantine Jewry from Justinian to the Fourth Crusade (London
1971) 136.

327
It may indeed be doubted, in the light of this passage, whether
scholars were in fact as poor as our narrator makes out; never-
theless, the prosperity of the City's craftsmen (and Comnene
resentment of it) can be supported from a substantial number
and range of twelfth century sources. 59 Once more, a date prior
to Manuel's death in 1180 can be postulated on the grounds that
urban riots subsequently became frequent due to imperial
weakness. 6o
As for food, drink, utensils, clothes and fashions, the evidence,
comprehensively listed by Koukoules, may not be so firmly
dateable as the numismatic, but it is certainly consistent with the
latter part of the twelfth century. In particular, the association
of the three types of bread with the three social classes (aristocracy,
middle classes, artisanate) can be corroborated in Tzetzes. 61 By
complaining that he has no bread at all, while it is the baker's
wife who nibbles the end of a loaf made from aphratitsin, the
'aristocrat's bread', our scholar perhaps implies not so much literal
starvation as lack of social position and status. Hence his need
for direct appeal to his patron.
These comments on the chronology of the poems are not in-
tended to be exhaustive, but simply to indicate possible ways for-
ward for establishing approximate dates with reference to social
and historical context rather than by means of circular linguistic
and metrical arguments.

8. Some concluding comments

Although much work remains to be done, several points emerge
from this preliminary survey. First, the vulgar language of the
four Prodromic Poems should not be allowed to obscure the

59. These are cited by Kazhdan (1984) 78-86, 144-6 and Hendy (1984) 577-88.
60. Hendy (1984) 587 n.
61. Chil.II.364.22 (Kiessling 404). For the division of other commodities into three
categories, starting always from the top, see Const. Porph. De Caerim. 1.470.10 and
the detailed analysis by Hendy (1984) 307-10 and 31On. Mesos for bread of the second
quality is attested in the dream-book of Daniel, Drexl (1922) 314. The case for inter-
preting to mesokatharon ... tlis meses as referring to social status is argued by
Koukoules (1948-55) 5.19. For a contrary view, see K. Amantos 'Glossikai
Paraterescis', EEBS 2 (1925) 78-86 and Glossika Meletemata (Athens 1964) 279-81.
The division of society according to a tripartite formula is further attested in Eustathios
of Thessaloniki and Niketas Choniates, see Kazhdan (1984) 143, who points out that
it departs in some respects from earlier Byzantine conventions and has parallels in
contempory western classifications.

328
artistry - or craft - of their composition. Poem IV, as has been
seen. is artfully constructed so as to conceal both the writer's
temerity in 'speaking plainly' to the emperor and the ambivalent
nature of his lavish praises. Second, within the conventional
framework of beggar poetry, several innovations are possible
precisely because the author has chosen to 'write fluently' in a
language normally reserved for 'speaking plainly', thereby
challengi,ng the assumed superiority of the medium favoured by
the 'fable-mongers', who know only how to winge (11.16) or cringe
(III. 15-19). Third, the affectation of ignorance is a literary
artifice, devised by the writer for his narrating personae, as the
quality and quantity of literary allusions to a wide range of other
texts (literary, ceremonial, religious and ,Jopular) suggests. Our
author is capable of a rich, subtle and original level of intertex-
tuality, which consistently exploits all texts to humorous effect.
Fourth, the numismatic, historical and social evidence c,onfirms
the mid-to late-twelfth century context of the four poems. 62 In other
words, the use of literary conventions and 'timeless' comic topoi
does not preclude the poems from being very much a product
of and for the times. For this reason alone, it is surely necessary
for literary and linguistic studies of this and other texts to be more
closely integrated with the results of historical, social and
economic research than has hitherto proved the case, at least
among a majority of Byzantinists, in order that a better understan-
ding be reached regarding the literary perceptions and distortions
of reality in Byzantine texts. 63
Perhaps most significantly, our author, although adopting a
'low' style, expounds an aristocratic viewpoint in his contempt
for the artisanate, a feeling which was expressed with increasing
frequency and intensity from the late eleventh century and
throughout the twelfth. Here, he touches upon what must be
62. The twelfth-century context has recently been questioned, largely on metrical
and linguistic grounds, by Eideneier (198213) 139·40 and n.1; it is affirmed, on
numismatic, historical and social grounds by Hendy (1984) 588-90 and no. 158.
63. Mango's view, formulated in an inaugural lecture, 'Byzantine literature as a distor-
ting mirror' (Oxford 1975), that despite the bulk of extant texts, not much historical,
social or even literary sense can be made of them, has provoked undue hostility, par-
ticularly among literary scholars. A constructive reappraisal of the problems and their
possible solutions may be found in Kazhdan (1984) 27ff, 43,105,158-9,194-5, who
stresses that a subtler reading of literary and other texts within their historical and
social context often reveals significant patterns of change, despite the rigid conven-
tions in forms of expression.

329
among the most fascinating key problems of Byzantine society
and economy at a crucial turning point, when events in the
medieval West were taking a markedly different course. It must
be left to others to extract from the four poems the full wealth
of historical and social information which they contain, particu-
larly on marital status and the position of women (poem I); means
of income, forms of state support and problems of balancing the
family budget (Poem II); monastic life and the financial, dietary
and other abuses among the hierarchy (poem III); and the poverty,
in contrast to the expected social status, of the grammatikos in
relation to common artisans, who were well off financially but
socially repressed from the time of the reforms of Alexios I. For
the present, it emerges that the grammatikos, while seeing himself
as a craftsman (technites), seems to be caught somewhere bet-
ween his aristocratic outlook (the product of his education) and
his actual income. He shared neither the privileges nor the status
of the aristocracy, yet at the same time lacked the means of secur-
ing a livelihood as a craftsman. He was therefore a 'declasse'
parasite; dependent upon his patron and excluded from the fruits
of his labour. Practice, the poem effectively tells us, was at odds
with theory so far as the value of a good education in twelfth-
century Byzantium was concerned. 64

64. Apaideusia was considered a misfortune, hence parents were responsible for the
education of their children. Gregory Nazianzenos had recommended that fathers should
teach their children rudimentary letters (Patr%gia Graeca 37.381); if unable to do
so, they should send them to an elementary school-master, a lowly figure who often
had a second job as a notary (taboul/arios), cf. the insulting appellation grammatike
notare of the butcher's wife in our poem (245). Complaints about parents' failure
to provide a satis,factory education for their children were not infrequent, see
Neophytos the Recluse, Hermeneia despotikon ent%n Cod.CoisI.Gr. 287,
fols.180b-181a, and Joseph Bryennios, ed. Voulgaris (1768) I.I08. On the status and
duties of the grammatikos. required to provide rudimentary literacy and knowledge
of basic scriptures and religious texts, se~ Browning, 'Enlightenment and repression
in Byzantium in the eleventh and twelfth centuries', Past and Present 69 (1975) 3-23,
'Byzantine world', BMGS 4 (1978) 46-8. Had the narrator of poem IV been no more
than an ordinary grammatikos, his complaints of poverty might not have aroused
sympathy. However, he displays knowledge of secular learning, e.g. ancient myths
(1 bbb), learned verse and metre (71-7,82), Homer, Libanios, Oppian and the classics
(213-:3), and philosophy (255), and is thereby qualified as a teacher of the thurathen
or es(){hen paideia, as well as of the enkuklios paideia, the second cycle of Byzantine
education, see K. Konstantinides, Higher education in Byzantium in the thirteenth
and early fourteenth centuries (Nicosia 1982) 16, 135-44 and Koukoules (1948-55)
1.66-7, 105. At the same time references to the priesthood (137-40, 173) are consis-
tent with the teacher of exothen paideia, or hiera grammata, see Konstantinides (1982)

330
 Finally, a great deal of material from the twelfth century
remains to be explored, of which these and other Prodromic texts
form but a small part, on the changing awareness of language
as a means of communication, as rhetoric, and the incipient
emergence of the vernacular. 65 The fact that such intellectual
developments were curtailed in the east, but not in the west, gives
added significance to Byzantine texts of this period for all
medievalists.
A last word: if the Prodromic poems prove a greater source
of fun and information than has hitherto been recognised, can
it be that it is Byzantinists, not the Byzantines themselves, who
lack a sense of humour?

1.7. In this way our narrator's plight is perhaps intended to illustrate the poverty
not of one particular teaching profession, but of all men forced to earn a living from
their education. ('e. Galatariotou).
65. On the emergence of the vernacular, it should be noted that prevailing attitudes
in the twelfth century favoured the ornate and high-flown style for written texts. and
no longer tolerated the excuses, traditional until the tenth century, for 'unadorned'
or 'low' style on grounds of comprehensibility. The climate of the times is exemplified
by the order of the Patriarch, Nikolaos Mouzalon (1147-51) for the destruction of
the Life of Saint Paraskevi the Younger because it was written idiotikos para tinos
choritou, see Browning, 'The language of Byzantine literature', Byzantina kai
Metabyzantina (Malibu 1978) 103-33 and Beck, Kirche und the%gische Literatur
(Munich 1959) 640. When a learned writer of the eleventh and twelfth centuries employs
the 'simple' language, he is careful to explain precisely why he does so, whether for
reasons of lack of education (Neophytos the Recluse) or because he is deliberately
'writing down' to his audience (Philip the Monk, the introductory works of Psellos,
texts by Tzetzes and Manasses intended for imperial court ladies), see Jeffreys, 'The
nature and origins of political verse', DOP 28 (1974) 162-3 ('C. Galatariotou). The
author of the Prodromic poems neither lacks education nor can claim to be address-
ing the lower classes, hence the choice of the vernacular points to a deliberate attempt
to use the popular language for amusement at high levels, cf. the prison poem by
Michael Glykas. Such a choice is unlikely to have been made by a person without
education and without standing in the imperial court, see Appendix I.

331
 Appendix I:
On the Question of Authorship

It is not intended to give an exhaustive treatment here to the complex question

of the authorship of the four Prodromic poems (the problem is probably insoluble),
but to present, more extensively than is possible in text or footnotes, a brief review
of some of the major contributions to date which aruge the case against Theodore
Prodromos on the basis of style, language and metre, and to suggest some alternative
modes of approach.
First, style. Hesse\ing - Pernot (1910) 21-3, draw attention to 'Ies fautes de gout,
les nombreuses exagerations, les contradictions et Ie nom meme de Ptocho-prodrome'.
Their lofty superiority has, predictably perhaps, been perpetuated by subsequent critics,
for example Mango (1980) 251, 'He tries to be funny by introducing scenes of slapstick
comedy ... , but the humour is spoilt by a tone of monotonous servility and tedious
repetitiveness'. It would indeed prove tedious to list all such dismissive strictures.
Unless founded on rigorous stylistic analysis and some awareness of Byzantine theory
and practice, judgements based on 'style' are likely to reveal more about our own
(literary and academic) preconceptions than abut the quality, date and authorship
of a particular text, as I. Sev~enko has argued in a seminal article, 'Levels of style
in Byzantine prose', JOB 31 (1981) 289-312. Repetition, redundancy, eulogy, hyper-
bole and exaggeration are all, in fact, acceptable hallmarks of the medieval popular
style, sec Ong (1982) 36-49 and M.T. Clanchy, From Memory to Written Record:
England 1066-1307 (Harvard 1979) 208-30. If our author was deliberately cultivating
an oral-popular style in the text (as distinct from the proems) of the four poems,
as is indicated by the frequent use of verbs of speaking, then the features decried
by modern critics are a sign of sophistication, not incompetence.
Second, language. As R. Browning has pointed out, the difficulty of distinguishing
language from style in Byzantine literature has proved a constant source of confu-
sion, the concept of 'register' as used in modern descriptive and socio-linguistics pro-
viding a better model, 'The language of Byzantine literature', Byzantina kai Metabyzan-
tina, ed. S. Vryonis (Malibu 1978) 103. With this principle in mind, let us review
the case against Theodore Prodromos. Beck (1971) 104, while carefully not excluding
the manuscript attribution, considers it more probable that the four poems are com-
posed by a later poet in the style, perhaps parodic, of Theodore Prodromos, reflec-
ting an imitation or rendering into the vernacular of the poet's learned verse. Horander
(1974) 66, is non-commital, but observes that 'Eine Umsetzung eines hochsprachlichen
Gedichtes in die Volksprache konnte wohl Werk Eines spiiteren Dichterlings sein,
eine Fortsetzung in der hier vorleigenden Art jedoch kaum', without explaining
precisely why, and in defiance of the evidence of other (admittedly'few) examples
of vernacular poems emanating from the Comnene court, see Beck (1971) 101-9.
Kyriakis (1974) 290-309, assumes, on the basis of linguistic differences, the existence
of no fewer than three separate authors - Theodore Prodromos, the author of the
Mangana poems and 'Ptochoprodromos' - understanding our poet's complaints so
literally that he populates almost an entire wing of the Mangana monastery with star-
ving literati, among them one 'Theodore', allegedly suffering from internal disorders
of the stomach, spleen and liver (surely the result of over-indulgence rather than defi-
ciency?)! Most recently, H. and N. Eideneier (1983/3) 119-50, conclude that the mixed
language of the four poems, and of other comparable texts, can be explained by the
influence of oral transmission (in the case of III and IV), and by the tendencies of
later scribes to correct and corrupt, therby implicitly rejecting both Theodore
Prodrvmos as author and the twelfth-century context.

332
 It will be noted that the case against Prodromos rests on assumed 'linguistic in-
consistencies', both between learned and vulgar Greek, and in the almost riotous range
of morphological, syntactic and lexical forms to be found in the vernacular poems.
First, if we observe Browning's timely reminder that register (chosen level of com-
munication, whether in speech or writing) is a more accurate concept than 'language'
(langue or parole?), then the difficulty concerning Prodromos writing both in learned
and in vernacular Greek disappears, as Kazhdan (1984) 91, perceives in calling the
poems 'genre exercises'. The second question, concerning inconsistencies within the
vernacular, is more complex, and has .been much debated. Jeffreys (1974) 176, is
categorical in his assumption that Byzantine intellectuals shared our own preconcep-
tions of linguistic homogeneity, finding it 'difficult to accept the picture of genera-
tions of Byzantine intellectuals who experiment in popular language yet fail to carry
their experiments through to the logical conclusion of a completely vernacular poem.
The education of such men was directed entirely to the elimination of mistakes from
their writing, towards the preservation of a uniform linguistic level. If they decided
tq experiment with the vernacular, surely at least one man could have been found
in several ceniuries to impose a similar uniformity on his demotic writing?' (italics
mine). In order to explain the lack of uniformity in the written vernacular poems
(which, it is assumed, cannot have been characteristic of everyday speech), Jeffreys
advances the important but controversial hypothesis of a literary 'Kunstsprache',
dependent upon a lost oral tradition in politikos stich os, which admits variant linguistic
forms for metrical and oral-formulaic reasons.
While not wishing for the present to preclude the possibility, I would like to pro-
pose, tentatively at least, several arguments towards a different explanation, namely
that the 'inconsistencies' encountered are precisely those to be expected during the
tumultuous and painful process of adapting the spoken to the written word, see B.
Stock, Implications of literacy (Princeton 1980) and Ong (1982) for parallel
developments in the medieval west. So far as the forms of Greek actually spoken
in the Byzantine capital are concerned, Browning remarks, with judicious caution,
that no firm conclusions can be drawn, but that the probability is that there existed
'a common tongue in which a great many alternative forms, belonging historically
to different dialects, were acceptable. Men from all over the Greek world mingled
in Constantinople as they did nowhere else', Medieval and Modern Greek (2 ed.,
Cambridge 1983) 82. Nor can the destabilising influence of non-Greek tongues in
the capital, noted by Tzetzes, be ignored, see G. Moravcsik, 'Barbarische Sprachreste
in der Theogonie des Johannes Tzetzes' BN}7 (1928/9) 352 (Turklc, Arabic, Hebrew,
Latin and Russian languages are mentioned). Further, not one of the early vernacular
works reflects dialectal differences, nor can they be dated with any certainty on
linguistic grounds, nor are they to be related directly to chronological developments
in the Greek language, see Browning (1978) 125-{). On the other hand, notions of
linguistic unity and consistency have been an obsession among scholars, Greek and
non-Greek, from the struggle for the independence of the Greek state to the present
day, most particularly at the turn of the nineteenth and twentieth centuries, as is evi-
dent from the predominant concern of Prodromic scholarship with linguistic ques-
tions; see also D. Tziovas, The Nationism of the Demoticists and its impact on their
literary theory (1888-1930. (Unpubl. Ph.D. thesis., University of Birmingham, 1984).
Given the present state of the evidence, it seems safer - and simpler - to attribute
the profusion of linguistic forms in our text to the formidable difficulties of record-
ing in writing a spoken form of the language which c~nnot have been standardised,
simply because it had never been written, than to postulate the existence of a
'Kunstsprache' or a subsequent phase of oral transmission and/or scribal interference.
It should not be forgotten that the earliest vernacular texts are courtly, not 'folk',
in genre, tone and context, see Beck (1971) 3-6, nor that the complex task of writing

333
in the spoken language (so greatly emphasised in the proems, appropriately perhaps
in a more even and archaising language) is unlikely to have been undertaken by a
'Dichterling'. Once more, the interaction between Latin and the vernaculars in the
medieval west can provide an instructive pariillel, although the cases cannot be regarded
as identical, due to the greater inter-penetration of learned and popular forms in Greek,
see M. Alexiou, 'Diglossia in Greece', Standard languages, spoken and written, ed.
W. Haas, Mont Follick Series 5 (Manchester 1982) 156-92, and Ong, Inter/aces 0/
the Word (Cornel\ 1977) 28-34. In general terms, the survival of Latin as the language
of school education and the medium of much formal oral communication, long after
it had ceased to be a spoken language, can be explained by its being the only mutu-
al1y intel\igible language amidst the 'swarming, oral vernaculars which often had dif-
ferent, mutual\y unintel\igible forms among populations perhaps only 50 miles apart',
Ong (1982) 112-3. Further, where a more or less standardised form becomes established
in writing, as a 'grapholect', its rules for 'correct' grammar and usage are popularly
interpreted as the grammar and usage of other dialects, ibid. 108; hence, in the case
of Byzantine Greek, interference between learned and vernacular registers is to be
expected, especially during the early stages of developing the vernacular.
Third, metre. Fol\owing the suggestion of Jeffreys (1977) 105-7, H. and N. Eideneier,
'Zum ftinfzehnsyl\aben der Ptochoprodromos', Aphieriima stan Lino Politi
(Thessaloniki 1979) 1-7, apply rigorous metrical tests, based on H6rander's four distinct
stress patterns in fifteen-syl\able verse, and conclude that Theodore Prodromos can
virtually, if not decisively, be excluded as author. H6randner, 'Zur Frage der Metrik
frtiher Volksprachlicher' Texts: Kann Theodoros Prodromos der Verfasser
volksprachlicher Gedichte sein?' JOB 32.3 (1982) 375-81, accepts the validity of metrical
and linguistic data in discounting Prodromos' authorship of the four vernacular poems,
but not necessarily of the earlier vernacular poem edited by A. Maiuri, 'Una nuova
poesia di Teodor Prodromo in greco volgare', BZ 23 (1914120) 397-407, and con-
cludes that the distinction between 'volksprachlich - hochsprachlich' may prove less
significant than that between 'volkstiimlich - gelehrt'. The subtlety of the last distinc-
tion is elusive, more especially as the four poems can be shown to contain all four
elements - archaising language (proems) / evolved vernacular language; learned,
ceremonial and religious al\usions / popular style.
Is metre any more 'objective' than language and style as a precise tool of measure-
ment for date and authentiticy of texts? So far as metrical regularity is concerned,
R.F.J. Henrichsen pointed out as long ago as 1838 that considerable flexibility has
always been allowed in the position of the stress accent in the fifteen-syllable line,
Ueber die so-genannten politische Verse bei den Griechen (Leipzig 1839, tr. P.
Fredrichsen). There is no reason to retract one's position of ten years ago, that
'a careful examination of ... voluminous statistics shows that the metre always allowed
considerable flexibility in the position of the stress-accent, ... and that no hard-
and-fast rules can be laid down. It is therefore unlikely that metrical analyses alone
can determine with certainty ... the date of specific works. Preference for a par-
ticular form of accentuation may be due to a number of variable and unknown fac-
tors: individual taste, tendency either to archaise or to regularise, or simply a desire
to vary the monotony of the rhythm', M. Alexiou and D.W. Holton, 'The origins
and development of 'politikos stichos': a select critical bibliography', Mandat%ros
9 (1976) 22-34.
Once assumptions of inherent unity and organic growth in an author's work (whether
stylistic, linguistic or metrical) are discarded, or seen in proper perspectivl roman-
tic preconceptions traditional from the early nineteenth century, then the case against
Prodromos, 'far from proven' as Kazhdan has shown (1984) 91, begins to disintegrate.
It is too soon to draw conclusions on the case for Prodromos, postulated by A.
Papadopoulos-Kerameus, 'Eis kai monos Theodoros Prodromos', Annales de la societe

334
his/orique e/ philologique d'Odessa 7 (1899) 585-602, and cautiously supported by
Maiuri (1914120) 397-407. Three factors deserve closer investigation: first, literary
and historical allusions in the four poems which can be linked with what is known
of prodromos' life, work and times; second, similarities and differences in literary
and stylistic terms between the four poems and Prodromos' undisputed works, in
particular the historical poems and the Maiuri poem (our Text Seminar has already
eSlablished several remarkable parallels); and third, a wider study of the growing flex-
ibility of genre, style and register in the twelfth century, alongside an incipient awareness
of the complexity of language as a means of communication (as opposed to a crea-
lion of God). Kazhdan's study (1984) richly documented from a wide range of rele-
vant sources, provides the framework, and goes a long way towards researchng the
first and third of these three factors (see especially on language as a concept, p.11l).
Viewed in this wider context, the authorship question need not be merely a search
for the author, but can be linked with the most significant intellectual developments
of the twelfth century. Meanwhile, the artfulness of 'he fourth poem has, it is hoped,
been established.

335
 Appendix n:
The Four Proems

I t is hoped that the following provisional translation of the four proems will help
to draw attention to some of their more unusual common features, in particular the
more archaising language and complex syntax (difficult to render in translation), the
importunate yet intimate tone of address, the persistent playfulness of concepts of
writing/speaking, prose/verse, political/politikos (stichos), truth/fiction, and on the
name Prodromos. Only the fourth proem has been worked on systematically by the
Text Seminar (*C- Galatariotou). No attempt has been made to render the fifteen-
syllable verse; but names, epithets and titles have been translated as consistently as
possible.
Proem I (MS G)
From Prodromos, kyros-Theodoros, to the Emperor Mavroiannes (John II)
What am I to dedicate to you, lord, crowned lord, (de"pota)
what kind of recompense or favour can I bring
of equal value to your own bright benefactions, (exisomenen)
all ·dnds of which were rendered to me from your might. (kratous)
5 Already, some time ago, but a short while since,
I had nothing, unfortunate as I was, to offer you,
appropriate to your might and to your godliness, (chrestotetl)
as also to your pride and to your graciousness, (charitoll)
except some verses, once more 'political' and unmeasured,(politikous ametrous)
\0 restrained, playful, yet in no way shameful, (paizontas)
for old men too can play, albeit more wisely so. (paizousl)

Do not, therefore, exclude them, send them not away, rather

receive them as condiments, although they have no smell, (kodimenta)
and hear with loving mercy what I write in my misfortune.(akouson, grapho)
15 Even though I seem, lord, to laugh and play at once, (gelon, paizon)
yet I have infinite grief and heaviest affliction,
an ailment troublesome, and suffering - yes suffering! (arrostema, pathos)
Hearing of suffering, think not I have a rupture,
nor any graver of the more secret (troubles),
20 no obvious distress in belly or in tooth,
neither heart disease nor lung trouble,
nor eye disease, nor waterworks, nor breathing apparatus,
but only the great audacity of a hostile wife,
who, producing problems and plausible language,
25 it seems fluently gives forth to me with richness. (eulogos, ploutarchos)
And I will make manifest this woman's spitefulness,
yet, lord, I fear those more brazen fellows,
lest they should hear me, and go to my home (akousosl)
and write reports about me unexpectedly. (pittakososin)
30 I would far rather, lord, they buried me alive,
and put me in the earth, and dug me into it,
than she should learn of what has just been written. (ton arti graphomenjjn)

(The narrative then commences, closed by a brief epilogue (268-74), in which the
speaker warns his patron that unless he takes swift action 'you will lose your
Prodromos, your finest congratulator').

336
proem II (MS H). From the same similar (verses).
My most revered master, master, glory, honour and my pride, (authenta)
the poor man, who is destitute, encircled,
encompassed on all sides by myriad misfortunes
and besieged by woes uncountable,
5 wishes to speak out his affairs unto his lord; (eipein)
if it were master and servant like yourself and me,
(and if) he were to sit and take soundings, speak and write(psi!lapha, lege, graphe)
'political' metres/je>cs and political writings/citizen affairs, (metriasmata,
politographiasj
and frequent spoutings-forth and high-sounding words,
10 yet from necessity they would reduce (him) to loquaciousness; (periltolexian)
then let him write and speak whatever moves to pity, (graphe, lali?)
whatever moves to mercy and compassion;
for he who writes fluently and speaks verbosely (legon tsourouchias)
seems to be a simple soul who does so from e.<travagance.
15 But as for me, I have departed somewhat from my susceptibility;(eupatheian)
instead of moans and groans, laments and tears,
I write words which give pleasure, words of delight, (remata)
because, by your love and on your life,
I have a sore-vexed soul, a heart aggrieved,
20 eyes much mourning and entrails on fire,
and a parched rumbling from eating dry food. (gourgollron. xeropiragian)

(MS 0). From the same to the Sebastokrator.

My most revered master, glory and my pride,

a poor man, who is destitute, encircled,
encompassed on all sides by myriad misfortunes
and fenced around by woes uncountable,
5 I wish to speak out my affairs unto my lord;
and if it were master like yourself and speaker like me,
and he were to sit and take soundings, speak and write
'political' metres and political writings,
and many spoutings forth and noisome words,
10 he would reduce himself to the language of prose. (pezolexian)
I write and speak those things which move to pity,
which move to mercy and compassion;
whoever writes extravagantly and speaks verbosely (legon serphetias)
seems to be a simple soul who does it from extravagance.
15 As for me, I have been diverted from the straight path, (ek tes eutheias)
and always with many moans and groans and lamentations
I write the pleasure of words and the joy of letters, (rematiin charmonen
grammatiin terposunen)
yet I do it not from joy, nor from simpleness of soul,
but rather in the name of the sore affliction now besetting me,
20 in the name of the despairing long perambulation (pezoporian)
- 0 woe - to the palace and up to the church,
let me tell you how things are, let me set it out,
take care, only take care you do not bury me.

(The narrative section in MS H is rounded off with an epilogue (101-17), addressed

by 'Ptochoprodromaton' to his patron, complaining that he cannot live off moun-

337
tain greens, locusts or his landed property, and begging for succour if his continued
praises are required. MS G breaks off at line 105).

Proem III (MSS H, CVA, S, g). Verses of the grammatikas, kyros Theodoros
Prodromos. Of Ptochoprodromos the second book against abbots (g); Of
Ptochoprodromos to the emperor lord Manuel Komnenos, the Purple-born (V); Other
verses of Hilarion, the monk Ptochoprodromos, to the most revered emperor lord
Manuel the Purple-born and Komnenos (S); Verses of Hilarion the monk,
Ptochoprodromos, to the emperor Manuel the Purple-born (A).

HI a Hardly escaping, lord, from the hands of the foes,

b I came as far as the shelter of your sceptre's might,
c for even as the bright trumpet of your achievements
d has not urged silence, so let it sound forth praises,
e and let tears from the heart cleanse the faults,
f but it has revived entire the lifeless corpse here prostrate,
g and I have taken refuge in your great majesty,
h well founded upon the firmly rooted rock,
i which happens, Manuel, to be Christ the crown-giver.
j So do not send me off as one debased,
k who runs towards your shelter with tears, (prostreehanta)
I but (receive me) as you are, a Christ-like emperor. (Christomimetos)

So accept and dissolve my manifold griefs,

for they need no great expenditure or cares,
5 provided you look on them with lenience and tolerance.
So then it is necessary to learn the sick man's pains,
then in this way to payoff the cure with craft. (enteehnos apodounUl)
And marvel at such great daring as the ant's,
how he escaped entirely from his wretched hole (mu6xias)
10 and at a run perhaps darted against the strong wild beasts, (treehein)
following fearlessly in the tracks of the lions,
not at all possessed by the power of their claws.
Think on me, crowned lord, as the ant,
against the strength and the poverty of words, (ton logon . .. aklemosunen)
15 while for the lions, count the rhetors, along with the philosophers,
who are brave in versifying and in writing (stiehizein, graphein)
and in concocting imperial victory pieces; (sungraphas ... ekplattein)
yet they write wisely and according to reason, (graphousi ... kala logon)
wise men and rhetors as they are, while I write otherwise,
20 (for 1 am an unlettered man, a new wearer of rags, (agrammatos, neos rakendutes)
monk of the meanest, lowest of the low)
rather (I write) succinctly, monastically, simply and most easily.
Not at all do 1 write you fables of old stories, (muthous palaion istorion)
such as cost the mind dear, hard to get the sense of,
25 rather (I write) simple things and clear, familiar to all
who run the lonely course in cenobetic life (treehausin ... dromon)
and endure, lord, what 1 am first to write you;
for in the monastery everything accrues to Philotheos
which the word/account comes in a short while to check. (logos)
30 So then lend to me your ears for a while,
and I will make all clear to you in deed, king. (anax)

338
(All MSS dose with a long address to Manuel (420-47), appealing for succour in the
form of food and money).

Proem IV (MSS G, CSA, g). From the same to the emperor (G); Ptochoprodromos
(g): Verses of Theodore Ptochoprodromos to the emperor lord Manuel Komnenos
(the Purple-sprung A) (CSA).

gl a Hardly daring, emperor, crowned lord, (basileu, despota stephephore)

b sceptre-wielding scion of the Komnenoi, most powerful world-ruler,
(kosmokrator)
c I come under the shelter of your golden wings (chruseon pterugon)
d a wretch, entreating, begging, pleading for
e the attention of your grace's ears;
f for you are the serene haven of all in need,
g and so pay me the attention of your most impeial ears, (akoos tas basilikotatas)
h and do not turn me, a wretched slave, away from you.

la (CSA) Hardly daring, emperor, crowned lord,

b sceptre-wielding scion of the Komnenoi, most powerful world-ruler,
c I come under the shelter of YOCir golden wings
d a wretch, entreating, begging, pleading for
e the attention of your grace'e ears,
f so that I may speak of my circumstances in greater detail.
gg Having only just coursed over the sea of life (paradramonj
h which so overwhelmed and nearly sank me,
i I raced to the world-saving harbour (prosedramon)
jj of your Christ-like imperiousness; (christomimetou ... basileias)
k for you are the waveless harbour of all who tak.e refuge in you.
I1 Having just left the sea of salty worries,
m I reached the sea of benevolence;
m
n for you are the source of compassion and joy of joys,
o and the sea of godliness and charity,
p you are the might of regal leaders, the sole king of toparchs,
P
qq in the name of the quadruple kingship, the form of the four Augusti
r of the new golden Rome, the Rome of Constantine,
s in the ancestry of your might, yea indeed and of the purple,
t your might shall be proclaimed over the four parts of the world.
u For he, who in address is quadruply august
v indeed co-signifies the four parts of the Cross;
V
w
w for just as the imprint of the Cross happens to bear four parts, (tupos,
tetrameres)
Xx so you too rule over the world's four parts
Уy by means of the quadruply august, for me, the purple-born.
zz Yet you, quadruply august, without the tetrarchy, (tetraugouste)
aa wield mightily the sceptre of single rule,
bb aided by the most God-ruling trinity.
bb
cc For if some of those in power have evil thoughts
dd + were to look askance and envy us, +
dd
ee by a mere gesture on your part, just a nod,
ff like smoke they will disperse before your face,
gg fearing lest they be burned, monarch, single ruler, (monarcha monokrator)
hh
hh lest they be turned to ashes by the brightness of your face,
ii even as wax melts when faced with fire;

339
 jj as happened a short while ago to Dantonis and Melias,
k k and Nouradinos with them, out of limitless stupidity,
II as well as to a certain Neeman, who had erstwhile revolted. (0 prin apostatesas)
mm But let me return once more unto my story:
nn having just been saved from the difficult worries of life
00 which were overwhelming and drowning me,
pp I came quickly, as you see, as if from the waves,
qq to the serene might of the empire,
rr to the haven and refuge of all in need.
ss But pay me the attention of your ears, turn not away from me,
lt your wretched slave and menial servant, who is in your debt; (dou/on,
oiketen, katechreiomenon)
uu for if I began to speak the words of fable-mongers,
uu (muthoplaslon)
vv you, emperor, would easily have solved them all,
ww for in you lOok residence the providence of God,
xxX Solomon's ineffable and theosophic wisdom.
X
yy But let me start acquainting you with my circumstances,
УУ
zz Iand how much loss I suffered, thrice-miserable,
ZZ

1 aaa from learning letters and from reading books

bb and from the silly myths of ancient fable-mongers.
b bb (muthourgon,
leromulhourgias)
ccc Here, then, I will tell you all from the beginning.

340
 Poetic Justice in the Patriarchate.
Murder and Cannibalism in the Provinces*
by

RUTH MACRIDES

The wide range of occasions and variety of subjects for which the verse form
was employed in Byzantium is documented by the Marcianus graecus 524, a mis-
cellany ofthe early fourteenth century, containing verse and several prose works
of the tenth to the thirteenth centuries. l Although the greater number of verse
pieces are epigrams, there are a few larger works. Prominent among these are
the Katomyomachia, a parodic tragedy attributed to Theodore Prodromos by
this manuscript, 2 an unpublished poem 'on teeth', 3 a lament on a fire in Constan-
tinople (1197) by Constantine Stilbes,' and another unpublished piece, the
subject of this paper, 165 verses attributed by the heading to the protekdikos
Andronikos,5 an otherwise unknown official of the Great Church. These verses
tell the story of a woman who had committed murder and cannibalism, and carne
before the tribunal of the patriarchate to confess.
An unusual story told in an unusual way, the piece illustrates well the problems
of interpretation and evaluation with which students of Byzantine texts are con-
fronted.1t contains the names of places, institutions, functions and events which
are in part, at least, identifiable. But at the same time the unique, extreme, and
sensational nature of the case, combined with the language and verse form of the
text, make it difficult to take at face value. The interpretation which follows the
text and translation attempts to show that questions of form and content are
interrelated: an answer to the question which the historian immediately asks . .:.
'Is it a real case?' or, to put it in the most pedestrian terms, 'Should it be in
Grurnel's Regestes?' -lies in an analysis of the structure, language, and imagery
of the text.

• Earlier drafts of this paper were read at a colloquium on Byzantine legal history (at Frankfurt-
Bad Homburg, in July 1983), and at seminars at the universities of Queen's Belfast, St. Andrews,
and Birmingham. I am indebted to the participants of those meetings for the interest, comments
and suggestions they expressed. Thanks are also due to the British Academy for a grant which
enabled me to consult at the Marciana the text which is the subject of this study.
1 A description of the manuscript and the partial or, in cases, complete transcription of its contents

can be found in SP.LAMPROS, "0 Mapxl(xvo~ XW01~ 524', Nco, 'E).).'7vOJ.JVl7jJWV 8 (1911) 3-59, 123-192.
2 LAMPROS, op. cit., 11; H. HUNGER, Der byzantinische Katz-Mause-Krieg (Graz, Cologne,
Vienna 1968) 13,25.
3 LAMPROS, op. cit., 12.
• LAMPROS, op. cit., 12-16.
5 Ibidem, 11.

341
138 Ruth Macrides

Tou XUpOU 'Avopovixou -rou 1tpwn:xOixou YCyovo'We;

EV 'tfj -rou 0cou IlcyaATJ ExXAT)Oi~
xai aVTi OT)IlCU~lla'toe; -rou'toue; ycypa<po'toe;

:Axouolla xalVov xai 1tepa 'tpay<pOiae;

oiov1tcp DUX TJVCYXC 1tW1to'tC xPOvoe;
4r 'tTJV 'tf]llepov o'tabdoa 1tpO xpl'tT)piou
'tAf]llwv llovaXTJ Mapia XaAOUlleVT)
5 bella'tOe; We; C<pT)OC Klpuppalw'tou
'tOle; woi 1taprnCIlWc 'tou ouvcopiou,
o xai xa'tco'taAa~c 1tav'twv xapoiae;
o xai 1tapaoxeuaocv EX paboue; o'tevclV,
iAlYYlav, opbae; 'tC 'tae; 'tpixae; <pepClV
10 xai mxpiac; mAT)OCV axpOWlltVOue;

xai oaxpuclV mCloc 1tav-rae; abpowe;

xa'tT)<plav 'tC xai o'tcva(clV xai 'tpeIlClV.
OU yap ne; av EV'tauba opUoe; Tl1te'tpT)e;
~uo' EX baAanT)e;, we; moe;, YAauxfje; ACyCl,
15 we; OUIl1tabfj VOUV IlTJ <pepclV aVT)llepWe;

aU' illlcpoe; ~UIl1tav'tCe; aVbpW1tOl <pUOle;

xai 1tav'tee; EO'ttva~av 0IlPpOUe; oaxpuwv
xai 1tav't'CC; WA6AU~CXV. W <'iclVou 1taboue;,
1tWe; av ne; d1tTJ, 1tWe; a1tcxyydATJ AOY<P;
20 1tOOOUe; puaxcxe; EX xopu<pfje; uo'a'twv

noocxe; o£ 1tT)yite; cunopf]ocl oaxpuwv

bpT)vwv OOUPIlOV a~wv bpT)v<pOiae;
IlUOe; lleA<POWV a~wv IlcA<poiae;;
'tie; av IlOv<JlM10ClCV we; llov<Jl0iav;
25 'tie; EX'tPCXY<Jl<'if]oCtcv we; 't'pcxy<pOiav;

fl rie; ypa<pij Owocle 'towu'tov <'ipalla;,

'toooU'tov oyxov oUIl<Popae;, 'tie; pao'taocl;
TJA1C <ppi~ov IlTJ <paclVc de; f}eav
• 'tfje; <'ipalla'touPYou <'ipaIlCX'tOe; ~cvo'tponou
30 otovncp DUX c<'ipaaev aUoe; EV Pi<p.

oupav£ vuv PPov-rT)Oov avwbcv Ileya

XAUClV a1teipywv 1tav'tae; axof)e; ~tVT)e;
oiav1tcp OUX TJXOUOCV avbpwnoe; naACX1'
I yaoTTjp buya'tpoe; 1lT)'tpOe; EyvWabT) 'ta<poe;.

13-14 cf.ll.16.34-35; 22.126; Od.19.163

342
 Poetic Justice in the Patriarchate 139

By Kyr Andronikos who was protekdikos

in the Great Church of God
and who wrote these (verses) in place of a semeioma

An unprecedented report and an ultra-tragic one

such as time never before produced,
today, standing before the tribunal,
a wretched nun called Maria,
from the Kibyrraiot theme, as she said,
conveyed to the ears of the council,
(a report) which made the hearts of all bleed
and caused them to groan from their very depths
brought on dizziness and made hairs stand on end
and fIlled those listening with bitterness
and caused all to weep together,
to be downcast, to moan and to tremble.
For there was none there of wood or stone,
nor of the blue sea, as the verse says,
so as not to bear, with agitation, a sympathetic mind
but all men are gentle in nature
and all heaved showers of tears
and all wailed aloud. 0 terrible misfortune,
how might one recount it, how can one report it in speech?
How many rushing streams from a peak,
how many sources of tears can he provide,
wailing a dirge worthy of a lament,
singing a song worthy of singing?
Who. might sing it as a monody?
Who might express it as a tragedy?
Or who could commit such a drama to writing?
Such a burden of misfortune, who will endure?
Sun, shiver, do not shine on the sight
of the performer of a strange kind of drama,
such as no other ever performed in life.
Sky, thunder now on high loudly
preventing all from hearing the strange story
such as man never heard of old:
the daughter's belly was the mother's tomb.

343
140 Ruth Macrides

35 WYTl xavoiioa xai xpul\laoa 'r0 ~p{tlJ.a·

(J(paX'rpla xai 'rpWX'rpla IlTJ'IT)P 1tal~(OtJ
aUwv n: vexpwv 1tall<paYOe; xai 0TJP(WV.
CtT)P btevMOT)n 1taXOe; CtXAUOe;
-rT)V ~palla'rOtJPYoV XpU1t'rWV Ctopaa\~
40 We; IlTJ~' OpiXoOal 1ta1al -rT)V 1ta1~ox'r6vov,
IlTJ~' cXV OeiXoOal (-rf)V) IlTJ'rpo<payov nAamV,
IlTJ~' 01t'raVeoOal 1tiXm, <peii, -rf)v 1tall<payov
We; 1lT) ~lappfJ~wo1 1tav'ree; ev~(XWe;
'ra<pov <pavetoav 1tiXmv T)V£WYlltVOV.
45 iw 'raAa1Va xai 1l6VT) 1taVaOA(a,
o'rpe<pw yap de; OE 'r0 1tp6ow1tov 'rOU MyotJ,
-rT)v ~palla'rotJPYov <Tle; ~eVT)e; IlOtJootJpyiae;.
1tWe; 1tav'ra'roAllov 'r6AIlTJV CtVC-rATJe; 1l6VT);
'itwe; 'roue; vexpoue; c<payee; we; xai IlTJ'repa;
50 1tWe; xa'rC-rpw~ae; <papllaxw~TJ OT)pia
oaupae;, o<pele;, IlUe; oi>v xeAwvale;, ~a'rpaxoue;
'rOlaU'ra xwpTJoaoa, 1tWe; oux eppaYTJe;;
a'rATJ'ra 'rAiXoa, 1tWe; <pepe1e;, 'rATlIlOV, Atye,
1tWe; 1tai~a xa'reo<pa~ae; wo1tep apviov;
55 1tWe; tl1telleVae; xal xpeotJPYTloa1, 'raAav;
1tw<; 5' OU)( Ulyyiaoa<; eOTIwIlEvTJ;
tie; d()e, tie; T]xouoe ,[010U,[OV ()palla;
IlUOO1 1taAal0l 1tAao'roypa<poum ~eva
ibtep xcXv T)ATJOeUOV OU()EV 1tpOe; 'r6()e.
60 ATJpOU01 1tai()ae; xa'ra1t(velv 'rov Kp6vov
0IlWe; 5' aOlVete; auOle; CtvIX51Mval
&la&OXae; &TJAouv'ree; oilllXl 1tpaYlla'rwv
ooa xp6voe; 1te<puxe 'tij <pop~ <ptpelV.
cXv yap xp6vou xUTJlla XlXl <pOelpel xp6voe;
65 CtVn()1C~OUe; OIlO1IX 'roie; e<pOapllevOle;
eX~AlXo'ravwv '[e ,[IXU,[IX 'tij yevvq. 1taAlV,
we; xal ()oxetv exetvlX '[a npWTJv IltvelV.
xpeoupY1IXV neA01tOe; iO'r6PTJoe ne;,
4v aU' ou xpeoupyov XlXl xpew'[pwx'IT)V C<pTJ.
70 avOpW1to<payoue; '[oue; KUxAW1tae; <pTJoi ne;,
ou IlTJ'rpo<payotJe; o\iLe xal 1tal()ox'r6voue;,
~evox'[6votJe; 1l6vov ()E XlXl 'rpwx'[ae; ~evwv.

53 a'rA'l'ru diXou: A. Ag. 408

344
 Poetic Justice in the Patriarchate 141

o Earth opening up and hiding the drama:

a mother, slayer and devourer of her child
omnivorous, a devourer of other dead people and animals.
Air, put on the thickness of mist,
hiding the performer in obscurity,
so that the child-slayer may not be seen by children,
so that the mother-eating creation may not be perceived,
so that, 0 grief, the omnivorous one may not be seen by all,
so that every one will not tear her open justly,
appearing to all an open tomb.
o wretch, alone all-wretched one,
I turn the attention of the discourse to you,
the performer of a strange poetry.
How did you endure alone an all-daring act?
How did you eat the dead, even your mother?
How did you eat up poisonous animals,
lizards, snakes, mice as well as tortoises, frogs?
Having contained such things, how is it you did not burst?
Daring a deed undareable, how can you bear it, miserable one,
How did you slay your child, like a lamb? [speak.
How did you persist also in the butchering, 0 sorry one?
How is it you did not faint while feasting?
Who has seen, who has heard such a drama?
Old myths fabricate strange fictions
which, even if they were true, would be nothing to this.
They relate nonsense, that Kronos swallowed his children
but sent them up again, unharmed,
signifying, I think, the succession of things,
which it is in the nature of time to produce in motion.
For if time destroys the offspring of time
(he is) giving in turn the same to those who have been destroyed
and causing these to grow again in their offspring,
so that those things of old seem to remain.
Someone told the story of the butchering of Pelops
but he did not say he was a butcher and a flesh eater.
Someone says the Cyclopes were cannibals
but not mother-eaters or child-killers,
only killers of strangers and eaters of strangers.

345
 9"17£
toI-OOI 'IA ffl 'snqdasor 'P 06-98
n;I ~Ot-onI sallpnr 98-18
~'('f)~Q.0-3 ;3t\UQ.~ 8 0-0 "3J..udl~l.uX '~lL
,ut\~rlOUYlL3lL t\('f)x,}rldudl '13Q~1 ,uxd,}o
~uy~rI ~l. t\~l.0-u ;uj('f)dl.~l.uX Ul.t\'}lL IJ
t\('f)!duQ. t\~l.> ;ud'}x !UX ,~d0-o ;XlII\Ox~
<
~ut\9r1 ud~rI IJ ';3A.udl~ 'o.0Y2 '13dl2 gOI
~;uQ.1}d<:> ;oA.'}dlorl<:> '~Q('f)lduQ. IJ
~UY9j ~l. lUX GOlL ';3dG3dl~ dGlL Ql. GOlL GOlL
~;toA.9Y '!Ol. '13)1}t\J..<l.l.O !l. 'uouo'}dQ uA.d~ ~
~~dx ~l. wGoQ<;' '!OO 'uj('f)dl.~l.uX ;~lL
~0.00 t\!Od3X 3xuQ.:JdulL t\udmX'}rI t\~l. ;!l. 001
't\U}dUl.('f)O 'QdlL Ul.t\1}lL 30.3d9A.Uj~
'3~y luo'}dQ Q.D Ul.GUl. 'U"(l.:Jt\~ ;<:,>lL ;<:,>lL
,url!!dQ Ql. ;OQ13 ,l.UX plX (1.D ;('f)Ti2 ,Q 13Y:YY
~3Q9l. lXlOGOX~ t\3130}OlL<;l ;}l. ;<:,>lL !UX
'j"XllL1}Q."XlX ru.":YlL 3xug:Jgd3lL<;l ,Q."1}lL !"XlX g6
;('f)X93~ "C?oDld3lL !UX i1. "0:JYlL 13d:Jch
,t\od~dl t\('f)l.",}lL t\~l. Xll."1}lL ;~XUlLU yyo.O
13X~ X0-0 '"orl\J.yl. '"XlrlA.13Q:yd"XllL ~Q t\QO Q1.
'"Xlt\~dch 13l.l.I,.LYlLX~ "UD!!lL t\Gt\ GOl. ldX:Jrl !UX
0.0rl9dl. plX ;UX!dch 3o(\')dI:J.ylLX~ ;Xll.",}lL !UX 06
'o.od:Jrly. 'Q..(}0-3 t\3j13Q~ 'o.O!dA.~ ;(1.01.
3l.9l. t\0l.X10 ;13 t\3tj13dl.~ ;pt\3rlOo.Q ;(1.01.
';o.od1}gdug '(1.0l. !UX 3jUy~l.UX Ol.GOl. !UX
't\0t\9rl !UX "XlQ!UlL 3judlo~ ;}l. lJ}Udga.
'"XlX1}l. "url~t\A. ~l.XlX ~rl t\oxd2 t\Q1. t\<:,>dUylL gg
;U!g X~ d3lLD<!> t\3judlo:Jl.XlX "XlQ!UlL t\~1.
t\('f)"(:JQ. !"XlX 3l. t\(\')nQ. ~rl '3l. t\~X~ 't\(\,)X~
't\0t\9dl 'QQlXllL o.Ot\:JQ.d"XllL t\<:'>Q.t\3lL 1i.lYYOlL !UX
t\(\')t\~l.O 1i.lYYOlL "XlQ!UlL 3X<l.Q.:J1. 3!!Q.ch3I.
'U:Jdx ;U!Q1 ;Qdl.url GOl.0-U d1i.lA. lqdx Og
u:Jdx XllQ! t\3tj1rl3lL:Jl.UX !dWXlJ.. 0-0
uQ.1}d<:> ;UJ.O!XlddOdl.url o.Ot\pX~ '!XllL <;'
'o.O!QlXllL G0l. 1Xl ;3xd1}0 GOl.0-Xl d1i.lA. ;3xd'}0
;Xl!Q1 ;Xlxd1}o !dWXlA. lXltjlri:JlLXldXllL 0-0
lXlOnQ. t\O!Xlt\UXo.W t\Ql. t\wXlch '}Q!XllL t\~l. gL
't\OX9l. t\Ql. !dwuA. 30<l.Q.:Jl.XlX !;((:IO ,yy~
;0A.9Y ;<:> ;XlQ!XllL Xll3QI:J. W !XlX 31Xlcho~
Sap!l:JllW q1ncr ZvI
 Poetic Justice in the Patriarchate 143

Medea also slew her children, as the story goes,

but she did not sacrifice the offspring to her belly.
They say that the Mycenaean sacrificed his daughter
but not that he passed his own flesh on to his belly
for his flesh was the flesh of his child.
His son was a matricide
but he did not dispatch his own flesh to his stomach
for his flesh was the flesh of his own mother.
Jephthah killed his child groaning greatly
and mourning greatly the murder of a virgin child,
willy-nilly, both consenting and not,
his daughter he slew, as if under constraint,
fulfIlling the vow, probably not in accordance with his will.
A Jewish woman killed her only child
and this astounded even the barbarians,
it turned the enemies to compassion then,
it rendered the savages at once gentle,
and fIlled all with fear and trembling
and to this day amazes every heart and mind.
But your example, wretch, does not exist.
It bears everything collectively
and more besides, extravagantly excessive,
it has surpassed everything altogether at once.
How can one bear to hear it?
But you tell the drama as it was.
How did you endure to do such things, how, speak,
confess everything for your salvation.
Who placed the knife in your hands?
How did you eat the flesh with your teeth?
Having committed the deeds, why do you shrink from words?
Where did you find the fire and where the wood?
Or did you savagely eat the flesh raw?
Did you eat snakes whole or only parts?
Did you cut off the tails and heads of the creatures
or did you eat all their parts?
Poisonous flesh full of poison
how could you devour, how is it you did not die at once?

347
144 Ruth Macrides

110 flovll, flovaXf], ,aih' hOAfllloaC; ,On:;

il xai ouvcpyouC; dxcc; iov d06mc;
oinvcc; aTIrxOljlav iWOll ~rAll;
AaACt, ~ovaxf], ~llOCV anoxpuTILr ~Ol.
vuv xaAov i:onv i:xxaAuljJal ,0 opa~a
115 OTIWC; l1p(a~~ov T]~rpt;t oixT)C; <puyT.lC;
xai ,Tjv aTIapahll'oV aioxUvTJV ,On:.
aUa ,i <PT)~l; ,ov AOyW~OV i:x)..aTIT)v,
xpnTje; xal1w,WC; de; ,pay<poov hpaTIllV'
AUTIllC; ,0 TIal1oc;. aUa TIaAlV o,prn,rov
120 TIPOC; vuooav mhTjv LOU A6you ,ae; T]viae;.
~oVT) ,a oCtva ,au,' hOA~T]Oae; ,Mc
il xui ouvcpyoue; dxce; avl1pwTIOX,OVOue;;
"Eoxov ouvcpyoue; a<popf],oue; Tij ~it;t·
coxov ouvcpyoue; oue; AfyCle; ~po,o<pMpoue;,
125 txl1poue; ouvcpyoue; coxov de; xaxoupyiav.
~OUAOlO ~al1dv ouo~cvde; ouvcpya,ae;;
XCtflwV tTIaxl1Tje; xai Al~Oe; ~apue; o<p60pa,
oi xai xa1:T]vayxaoav tv TIoUfj ~it;t
,Olaum ,oAflfioal ~c xai TIpa~al ,O,c
130 <pcuyouoav txl1poue; ai~oxapde; ~ap~apoue;
5r xai ouaxdh:ioav tv 'tOTI01C; tpTJlliac;
Olxac; nvvuaav a~E'tpf]'twv o<paA~(hwv.
aU' 6 OTIOPCUC; OUOl,O 'twv (l(aviwv,
t~ wv aVEp)..aa'lloav dOll xaxiac;
135 Ol' WV i:nnor<PPlloE xaxa ,Q ~i<p,
xai ~na ,au,a ~cwnYEe; l1dae; Olxlle;,
WV flia TIav,WC; xai flaxalpa ~appapwv
xa,a,pexouaa ,wv U1tEUWVWV OlXT.l
o<panouoa, mr(ouoa, AlilOiC; TIOUaXle;
140 iiUale; ,c TIOlvaie; UTIO~aAAOuoa ~eva\e;'
we; dl1c xa,rO(pa~cv xa~c xai ~6vov
flTj opafla,oupyoe; Wpal1llV xaxwv ,oowV'
avO' wv uTIeo'rllv ~aonyae; TIoAu'rp6TIoue;
OUAWV ~lyrv'rwv xai o<ppaydolle; do60ou
145 xai oupfla&llv (f]oaoa rnvoe; xai flOVlle;,

119 cf.Arist. EN 1105b21 133 Matth. 13.25, 38-39

144 OUAWV cod. 145 ~iia(wa cod.

348
 Poetic Justice in the Patriarchate 145

Alone, nun, did you dare those things then?

Or did you have accomplices, knowing about the poison,
who cut off the poisonous parts?
Speak, nun, do not hide anything from me.
It is good to reveal the drama now
to escape the disgrace on the day of judgement,
and the inescapable shame at that time.
But what am I saying? r have taken leave of my senses,
appointed a judge, a singer of laments r have become
(through) the emotion of grief. But it is again time
to turn the reins to the very starting point of the story.
Did you dare those dreadful deeds alone at that time
or did you have accomplices also in murder?
r had accomplices unbearable in their force.
r had accomplices whom you call man-destroyers.
Enemies I had as accomplices in evil-doing.
Do you wish to know my hostile collaborators?
Heavy Winter and exceedingly severe Famine,
who forced me with great violence
to venture and to execute such deeds at that time,
fleeing bloodthirsty enemies,
and confined in desert places,
paying the penalty for countless errors.
But may the sower of tares perish
from which all kinds of evil shot up
through which evils crept into life
and, because of these, the scourges of divine justice,
of which one surely is the barbarian knife
which ravages those answerable for judgement
slaying, oppressing and, often to famines
and other outlandish penalties subjecting.
If only it had slain me, too,
r would not have been the performer of so many evils
for which I underwent scourges of many kinds,
the joining of my gums and the sealing of my mouth,
living by the line of the nose alone,

349
 146 Ruth Macrides

eeou 1tpovoi~ xed XpiOCl qHAaya~4>

'Wu xal1taAlv ~pi:l"aV"roe; oc\1tV4> rQ ~i:v4>
:Appcq.naioue; 1tarOae;, we; rpaq111 A{:YCl,
xal j.lCXPl rou vuv tlvvvoa, IPcu, rae; oixae;,
150 1tpaYj.laOlv xallPaoj.laolv appT]rOrp01tOle;'
oue; xarclPayov 1tavrae; a1talrOUj.lCVT)
au(wvrae; ou ~avovrae; a1tOOlOOVal,
xai1tcp Acyouoa 1tarO' a1toO(pa~al 1l0VOV
aUoue; oe vcxpoue; tv ralP<+> pcppwxcval
155 OlO 1tapclOfjA~ov exoixwv olaxpioCl.
TOlaur' elPT]ocv T) 1l0vaXTJ Mapia.
eoo~c roivuv avcvcyxci'v, we; ~Cj.lle;,
a1tavra raura rij XpiOCl roD oco1torou
rare; axoare; n: rfje; ayiae; auvooou
160 OCOj.lci'v exovoale; xal AVClV e~ouoiav
xal ocolloAurov auro~cv ri}v Mapiav
we; eOtlV dxoe; a1tolPav~fjval raXCl'
MOle; yap eotl OCOIlOe; ex rfje; auv600u
OCOIlOUIlCVOle; a1taOl rQ ecQ IPOP<+>
165 xavrcu~cv cupiOXOUOl AVOW OlPaAlla.-WV.

147-148 Exodus 16.15, 31, 33, 35; Nehemiah 9.20-21

14 7 o£i1tv~: ii1tv~ cod. 149 "t\ VVUCI~ cod.

165 cupioxouoo: cod.

The story presented above is told by two speakers: the narrator, who identifies
himself as a 'judge' (Xpl'ti]e;, 1.118) and has the larger part (11.1-122, 156-165),
and the nun Maria who has 33 lines in all (ll. 123-155). The story begins and ends
in the narrator's present (.-fjj.lcpov, 1. 3) and in the same location, before the tribu-
nal (XPl'ti]Plov, 1. 3; auvCOPlOV, 1. 6) in the patriarchate (ll.158-159). The narrator
relates the story in two ways: by retelling the 'report' (axouoj.la, 1.1) Maria had
brought to the tribunal, and by replaying or reenacting the scene at the tribunal
(ll. 45 fT.), both his own part asjudge and that of Maria as bearer of the report. The
reenactment takes us back to a time earlier in the day at the tribunal, while the
story Maria tells reaches back to some unspecified time in the past (ton:, 1. 129)
and takes place in a location outside Constantinople, probably the Kibyrraiot
theme from which Maria came (1. 5).

350
 Poetic Justice in the Patriarchate 147

by the providence of God and his benevolent judgement

who once more nourishes with a strange meal
Abraham's children, as the Scripture says,
and to this day, alas, (I am) paying the penalty
in occurrences and apparitions horrible,
required, all those whom 1 devoured,
to return alive, not dead,
and although I said I killed my child only
I ate others dead in the tomb
whence I have come to the judgement of the ekdikoi.
The nun Maria said these things.
Accordingly, as is just, it seemed right to refer
all this to the judgement of the despotes
and to the ears of the holy synod
who have the power to bind and to loose
and (it seemed right) Maria at once redeemed
to declare, with speed, as is only fitting,
for a bond from the synod is an absolution
for all those bound to God by fear
and finding thence absolution from sins.

The arrangement of the text is as follows: 11.1-44: narration of the report

brought before the tribunal; 11.45-155: replay of the scene at the tribunal
(11.45-122: series of questions addressed by the narrator to Maria; II. 123-155:
Maria's reply); 11.156-165: verdict on the case.

01.1-44] The disclosure of Maria's 'report' by the narrator comes only at the
end of the opening section and is made in stages, with intervals. The narrator
builds up to the revelation of Maria's deeds by first describing the emotions and
reactions her story had aroused in her audience (11. 7-18), then expressing the
difficulty oftelling the story (ll. 19-27), and finally invoking the sun, sky, earth,
air (11.28-44). The themes expressed in the opening lines are common to laments
which are found in a variety of literary forms - funeral orations, inscriptions,

351
148 Ruth Macrides

homilies and liturgical hymns. 6 Moreover, these lines share not only the topoi of
threnoi but also their stylistic features - alliteration, assonance, anaphora -
which create the effect or sound of a lament. 7
Lamentation is the expression of grief over a disaster, death of a loved one or
some other tragic event. 8 Here the object of the lament is not a dead person but
the tragedy which Maria's story embodies. The narrator not only declares her
story to be a tragedy in so many words, in these lines and throughout, 9 but
demonstrates the intrinsically tragic nature of the story by describing the effect it
had on its audience (II. 7-18). Their emotions - fear and pity - evident from their
trembling, sighing, weeping, hairs standing on end, are the very ones which trag-
ic poetry, in accordance with Aristotle's definition, aims at arousing. 10 Even the
verb used by Aristotle in this connection (1tCi~TJ 1t(xpaoxruci(c1V: Poet. 1456b) is
included in the description of the audience's reaction (1.8).
The narrator, then, in this first section, reveals to his audience that he has cho-
sen the form of a lament to retell Maria's tragic story. He conveys this informa-
tion implicitly, in his treatment of the subject, using the topoi, language, and the
metre l1 of the tragic lament, and also explicitly, later, when he refers to himself
as a tpaY4106~ (I. 118), a' singer of a lament'. 12 The story is a tragedy, 'performed'13
by Maria (1. 29), and sung by the 'judge'.
The 'strange drama' Maria performed - the eating of the dead - earns for her,
in the opening section, the metaphor of a tomb (1. 34, 44). But more than this, as

6 M. ALEXIOU, The ritual lament in Greek tradition, (Cambridge 1974), 137,142, 145, 153,156,
161-177; M. ALEXIOU and P. DRONKE, 'The lament of Jephtha's daughter: themes, traditions,
originality', Studi Medievali 12 (1971) 819-863, esp. 846-851.
7 See the discussion by ALEX IOU, The ritual lament, 151-160; also W. B. STANFORD, Greek

tragedy and the emotions: an introductory study (London 1984),93-95. For alliteration: II. 4,22,
2328,29,40,48,53,54,58,61,62,66,81,82,86,98,99, 103, 105, 110, 113, 118, 121, 129,158;
assonance:II.3,22,23,31,32,34,36,37,48,53,60,63,67,68-69,70-72,76,78,85,86-87,91,92,
93,94,95, 104, 105, 106, 108, 110, 112, 113, 114, 115, 116, 119, 122, 127, 130, 132, 139, 140, 141,
142-143,144,149,150,152,153,155,158,160; anaphora: II. 7-8,10-11,13-14,17-18,19,24-25,
41-42,48-50,54-56,57,77,88-89,90-91,95-96, 103, 123-124; homoioteleuton: 8-9,22-23,24-
25,117-118,122-125,146-147,160-161.
8 D. HADZIS, 'Was bedeutet "Monodie" in der byzantinischen Literatur?', Byzantinische
Beitrage, ed. ].IRMSCHER (Berlin 1964) 177-185; ALEXIOU, The ritual lament, 102-103 n.6.
9 ,paytpo(a: 1.1; opalla: II. 26, 29, 35, 57, 97, 114.

10 Poet. 1449b. STANFORD, Greek tragedy and the emotions, 21-48 and passim.
11 The author's dodecasyllables, the Byzantine iambic trimeter, always have a stress on the
(short) paroxyton, and a caesura after the fIfth or seventh syllable. Some verses have eleven syl-
lables (I. 93). For the rules which regulate this metre see P. MAAS, 'Der byzantinische Zwolfsilber',
BZ 12 (1903) 278-323.
12 See, also, the use of the word in the miracles of St. Demetrios: P. LEMERLE, Les plus anciens
recueils des miracles de Saint Demetrius (Paris 1979) I, 147 § 134.
13 opalla,oupy6~ is used by the judge of Maria (11.29, 39, 47) and by her of herself (1.142).

352
 Poetic Justice in the Patriarchate 149

a devourer of hwnan, as well as animal, dead, she is a mtjl<piiyoC; (I. 37, 42), an
epithet applied to Hades, who 'swallows the dead', in a wide range ofliterary
forms, secular as well as religious. 14 In the text the word appears only in the
opening lines but the analogy is exploited later (1. 152).
01. 45-122] The strange and unique aspect of Maria's deeds, announced from
the beginning (II. 1-2), and reiterated throughout by the word ~i:voC; in various
forms/ 5 and word-play on f.lOVOC;/6 is 'proven' in this section, in the reenactment
ofthe scene before the tribunal. In these verses continuity from the opening lines
is maintained by tragic diction and perpetuation of the style of the lament. The
judge's questions, statements and appeals to feeling are reminiscent of a similar
approach in laments,17 and the tragic element is conveyed through the use of
specifically 'tragic' words and expressions - iw, <pet), 'tiiA(lW(l, cX'tA'Tl't(l 'tACtO(l - as
well as the figures of speech common to tragedy.
The judge's role in the recreated scene before the tribunal consists of a series
of questions addressed to Maria (I. 46), concerning 'how' she committed her
deeds - eating the dead, killing and eating her child, as well as lizards, snakes,
mice, tortoises and frogs. In the middle of his interrogation he digresses to seek a
parallel for her behaviour in mythology and history (11.58-95). A catalogue of can-
nibals and murderers is presented but, in the entire list, not one case can be
found which equals hers. The judge also deduces from the mythological
examples that destructive behaviour is cyclical and repetitive. This is the mean-
ing, he states (II. 62-67), of the story of Kronos who swallowed his children but
regurgitated them whole and well. In a word-play on Kronos - chronos the judge
explains that time conceives and time destroys, handing down the same legacy to
those it has destroyed. The truth of this statement is demonstrated by the other

1< Hab. 2.5; Is. 5.14, and commentary by Theodoret, bishop of Cyrus: PG 81, 257 A; the Epita-

phios Threnos: Tp,,;)(J/Ov KaraVVKW(OV (Athens s.d.) 473-47; ProcJus, On the Crucifixion, and On
the Sunday after Easter: F.J. LEROY, L 'homiletique de Proclus de Constantinople, Studi e Testi
247 (1967) 209, 237; Germanos, On the Entombment: PG 98, 285B; Christos Paschon, ed.
A. TUlLIER, Gregoire de Nazianze: La passion du Christ, tragedie, Sources chretiennes no. 149
(Paris 1969) 934; Pseudo-Prodromos, verse encomium, ed. E. MILLER, Recueil des historiens des
croisades. Historiens grecs II (Paris 1881) 265, 158-162. The stomach of a gluttonous friend is
compared with that of Hades in Theodore Daphnopates Correspondance, edd. and transl.
J. DARROUZES and L. G. WESTERINK (Paris 1978) 183,8; 7t(q.l(p&yo~ is applied to the Lykos river
whose stomach is full of the dead: Theodoros Prodromos, Historische Gedichte, ed. W. HORAND-
NER (Vienna 1978) XIX, 31-40; to the cat who devours mice, in the Katomyomachia: HUNGER,
Katz- Miiuse- Krieg (as in n. 2) 43,215; to fire, in Stilbes' lament on the fire of 1197: LAMPROS (as in
n.1) 14,43; to time, in an epitaph, also from the Marc. gr. 524: LAMPROS, op.cit., 124,36.
15 II. 29, 32,47, 58, 72, 140.
16 n.48, 86, 105, no, 121, 153: the author plays on !lOVT] , one, only, alone, and !lovaXtl, nun.
17 ALEXIOU, The ritual lament, 137, 144.

353
 150 Ruth Macrides

examples in the list: they are, with one exception, Kronos' descendants - Pelops,
the Cyclopes, Agamemnon, and his son Orestes, named indirectly in the play on
1lT)1"popp(Xio1"T)~ (I. 78). The inclusion of Medea (II. 73-74) breaks the cycle of
father-son reproductive-destructive behaviour but is an obvious addition to the
list as an example of a woman who killed her children. The mythological
examples, then, show Maria's case to be both outside the cycle, unparalleled,
'such as time never before produced' (II. 2, 59, 92), and also within it, for she
devoured the mother who had produced her and destroyed the child she herself
had produced.
To the mythological examples the judge adds an Old Testament case, the
judge Jephthah who sacrificed his own child, but under constraint, in fulfilment
of a vow (ll. 81-85). The final example (ll. 86-91) bridges the stories of old and the
judge's own time, bringing the catalogue up to date, for it was a case which still, in
his time, amazed people (1.91).
The story of the Jewish woman who 'killed her only child' is one whose ultim-
ate source is Josephus' Jewish War. The Jewish author tells the story 'terrible to
relate and unbelievable to hear', in the context of the suffering which the siege of
Jerusalem (66-70 AD) caused. During the famine at the time ofthe siege, a Jew-
ish woman, named Maria, killed her infant son, roasted and ate half of him.
'Rebels' (Josephus uses this term to describe Jews who fought against the
Romans) smelled the roasted meat and demanded to be given the food. When
Maria showed them the remains of the baby, they were fIlled with horror and
fear. Josephus comments that although they had previously acted with great
cruelty, the sight of Maria's deed left them cowardly and weak. 18
The story of the Jewish Maria was transmitted directly from Josephus by
Eusebius 19 and by later authors who for the most part used intermediate sources.
Theodoret, bishop of Cyrus in the early fifth century, referred to it in his com-
mentary on Isaiah 5.13, where he remarked, 'he described that grievous drama of
the woman who ate her son because of famine'.20 Byzantine chroniclers also
mention the story. George the Monk gave a paraphrase, commenting, 'thus, the
infant who emerged well from the womb returned again wrongly to the same
wretched belly'. His account and this comment were repeated by Kedrenos. 21

18 Bellum Judaicum, ed. S.A.NABER (Leipzig 1896) VI 99-1Ol.

19 E.H. III, vi, 20-28. For knowledge of Josephus' Jewish War in Byzantine authors, see the dis-
cussion by R. FISHMAN-DuKER, 'The Second Temple Period in Byzantine Chronicles', Byzantion
47 (1977), 136-138.
20 PG 81, 256-257.
21 George the Monk, ed. C. DE BOOR (Leipzig 1904), corr. P, WIRTH (Stuttgart 1978) II, 385,10-
386,16. G. Cedrenus, ed. I. BEKKER (Bonn 1838) I, 375,19-376,10.

354
 Poetic Justice in the Patriarchate 151

Zonaras related an abbreviated version ofthe story in his Epitome, as did Glykas
in his world chronicle. 22
Josephus' case and the one before the judge contain two points of remarkable
similarity which the judge does not mention: the name of the woman, Maria, and
the fact that she ate, as well as killed, her child. To accuse the judge of imperfect
knowledge of the story or deliberate suppression of the facts does not seem
appropriate since he himself states that it was well-known in his day and still
created a strong impression. Rather, it is precisely because it was a cause celebre
that a brief and elliptical reference would suffice to recall it to mind. In fact, one
can attribute to the judge the use of ellipsis in his account of the story in general,
for it would not otherwise be explicable why a mother who killed her only child
should make enemies compassionate and savages gentle (11. 88-89). In any case,
the point which the judge wishes to make with this example - to conclude from
the amount of space he devotes to it - is the emotional reaction which the case
created both in its time and in his.
At the end of this long digression, the judge returns to his questioning of Maria,
urging her to tell her story, to 'confess everything' for her' salvation' O. 99). What
Maria did in the past (rMr, I. 110) is to be revealed in the present (vDv, I. 114) so
that she can avoid disgrace (U'pia~pov, I. 115)23 on the future day of judgement
('!:o'!:r, I. 116).
[II. 123-155] Maria's reply to the judge's questions, her confession, is in a style
similar to the preceding verses, drawing on the themes and imagery of those
lines. The confession concentrates on the circumstances of Maria's deeds: winter
and famine forced her to act as she did, at a time when she was confined to
deserted areas, fleeing the enemy barbarians. But the background to this story
(ll. 133-150) shows that Maria is a victim, caught in a cycle of cause and effect out-
side her control. 24 The devil, who is at the origin of this destructive cycle, planted
the seeds of evil which led to sin; for this man was punished by God in the form
of the barbarians who caused famine, which made Maria commit her deeds, for
which she, in turn, was punished. Her punishment - the sealing of her gums and
the branding or sealing of her mouth (I. 144) - is symbolic in view of the earlier
analogy of Maria with a tomb (ll. 34, 44), sealed in order to prevent more dead
from entering. Her life-line was the air she breathed (I. 145). But God, who sends
punishment to man for his sins, is also a benevolent judge. He kept Maria alive

22 Ioannes Zonaras, ed. M. P1NDER (Bonn 1841) I, 538,21-539,10. M. Glykas, ed. L BEKKER
(Bonn 1836) 442,8-443,18.
23 For ilpicqJ~o~, a word used of a degrading procession of those in disgrace, see P. KOUKOULES,
Bv(aVTlvwv Bior; xaz IJoAmOj.lOr; III (Athens 1949) 184-208.
2, Compare ll. 141 ff. with E. Med. 1 ff.

355
 152 Ruth Macrides

nourishing her, like the children of Abraham, in the wilderness, with a miracul-
ous meal from heaven. Alive, she still suffers for what she did (11.149-150).
Maria gives, in closing, the reason for her presence before the tribunal: she
killed her child and ate other dead people, all of whom she must give back
(anocn06vcn) alive, not dead (11.151-155). This phrase recalls the Kronos myth
(11. 60-61: aOlvd~ ... avaol06vcn) but also the analogy of Maria with 'omnivorous
Hades' (11.37, 42). The return of the dead by Hades is a central theme of the
Resurrection, expressed in hymns and homilies: 'the omnivorous Hades who ate
You, vomited the dead he had swallowed since eternity'.25 The image is also
found in literary and visual treatment of the Last Judgement. The description in
the Book of Revelations (20.13) of the sea and hell delivering up their dead, fmds
its visual counterpart in scenes ofthe Second Coming where personifications of
the Earth and Sea - female figures seated on imaginary creatures - are depicted
'throwing up' the dead, who emerge from the mouths of the creatures. 26
Thus, the reason for Maria's confession is conveyed through allusion to the
analogy with Hades' return of the dead, an image which has literary and visual
associations of the Resurrection and the Last Judgement. On the future day of
Judgement Maria will have to account for her deeds, to return those she ate, the
dead she contains in her stomach -tomb. Her confession now makes possible red-
emption and the renewal oflife, both for herself and those she ate, on that future
day before that other tribunal. 27 For the penitent, it is possible to break out of the
repetitive cycle of sin and suffering to attain incorruptibility and eternal life.

25 The Epitaphios Threnos, still chanted today on Good Friday, contains this image: see n. 14
above and the discussion by M. ALEX IOU, 'The lament of the Virgin in Byzantine literature and
modern Greek folk-song', BMGS 1 (1975) 119. It is also found in the homilies of Proclus: see the
discussion by LEROY (as in n. 14 above) 169 and n.24.
26 For an early description of the scene, see Ephraem the Syrian, PG la, 903. The earliestrepre-
sentation known to me, Ayvah Kilise, Cappadocia, datable to the fIrst quarter of the tenth century,
differs from those cited below in its depiction of the scene. The funerary programme in the north
chapel includes, on the west waJllunette, a representation of the Earth and Sea giving up their
dead: A woman is seated next to a tomb-like enclosure which is piled with heads; the inscription to
her right reads, 'The earth [giving up] the dead'. On the right side of the lunette only fIshes' tails
are visible. For a description, see N. and M. THIERRY, 'Ayvah Kilise ou pigeonnier de GiiJlii Dere,
Eglise inedite de Cappadoce', Cahiers archliologiques 15 (1965) 144. Dr. Lyn Radley kindly put
photographs and the unpublished inscription at my disposal. For examples from churches of the
eleventh to the fourteenth centuries, see K. PAPADOPOULOS, Die Wandmalereien in der Kirche
IIavayia TWV XaA.Kiwv in Thessaloniki (Vienna 1966) 57, 63-73; D. C. WINFIELD, Asinou, a guide
(Nicosia 1969) 23; for Mileseva (c. 1236) and Gracanica (c. 1320) see D. MILOSEVIC, Das Jungste
Gericht (Recklinghausen 1963); P. MIJOVIC, 'La personnification de la Mer dans Ie Jugement
dernier a Gracanica', XaplOT17PIOV ci, AvaoTaolOv K. ·OpA.avoov IV (Athens 1967/8) 208-219.
27 Maria's confession brings together actions and their consequences in the past, present and

future which correspond to sin, suffering and redemption. For the treatment of the corresponding
planes of time, place and situation in laments (hymns), see ALEXIOU, The ritual lament, 142-144;
EADEM, BMGS 1 (1975), 116-118.

356
 Poetic Justice in the Patriarchate 153

UI. 156-165] In the concluding 10 verses the judge gives the decision on the
case. He refers the matter to the judgement of the patriarch (1. 158) and the holy
synod 28 who have the power to bind and to loose, that is, to bind the sinner with
penances (OfOJ.L0i) and to loose (AUW) the bonds or, absolve the penitent. 29 But at
the same time, the judge, in anticipation of the synod's decision, declares Maria
'redeemed' (OfOJ.L0AU't"O<';, 1. 161),30 freed from the bonds of sin. He explains this
'verdict' with an oxymoron which also includes a pun on Aum<; (solution,31 absolu-
tion): 'for a bond from the synod is an (ab)solution for all those bound to God by
fear' 01.163-164). The sinner who in fear of God confesses his sin, will be
redeemed or absolved (1. 165) through the penances assigned to him by the
church. Maria, then, will ultimately find absolution through the performance of
the penances which she will receive from the patriarch and synod.
The story's likeness to a tragedy is maintained to the end. In the concluding
verses, the parallel with drama is provided by the double-entendre of the words
OCOJ.Lo<; and AU01<; - technical terms in canon law (penance, absolution) and in tra-
gedy (the plot and its resolution).32 The legal solution is the same as the poetic
solution. Through fear and pity the catharsis of redemption is achieved. 33
The above analysis of the language and imagery of the verses has indicated an
attempt by the author to sustain an analogy between the extraordinary story he
relates and a tragedy. The tragic element is conveyed not so much by the direct
borrowing of verses from ancient drama, as by the imitation or reproduction of
the general effect or sound of Greek drama through the use of the verse form,
poetic figures of speech, and certain representative words and phrases.

28 For the expression, ml~ aXOnl, ... avnq>fpw (II. 157, 159) which is used in legal contexts of

'referral' to a higher authority, see, e. g., a chrysobull of Manuel I Komnenos (1153); ]. and
P. ZEPOS, Jus Graecoromanum I (Athens 1931, repro Aalen 1962) 379,8-9. On the synod and its
functions see J. HAJJAR, Le synode permanent ( oiivooo, eV01]f-lOvoa) dans l'Eglise byzantine des
origines au Xle siecle (Rome 1962) 21-79,115-136,150-178.
29 Matth. 18.18. On the authority of bishops to bind with penances (epitimia) and release the
penitent from the bonds, see Theodore Balsamon, commentary on canon 74 of Basil:
G. A. RHALLES and M. POTLES, EiivraYf-la .wv ltdwv xal iepwv xavovwv IV (Athens 1854) 236-237.
30 OCOf.LOAu'toe; is attested only for the Christos Paschon (ed. TUlLIER [as in n. 14]2529): LAMPE,
Patristic Greek Lexicon (Oxford 1961); see also below n.87.
31 For lysis, a technical term for an interpretation of, or solution to, a case by the authority to
whom it has been referred, see F. DOLGER and]. KARAYANNOPULOS, Byzantinische Urkundenlehre
(Munich 1968) 82-87.
32 Arist Poet. 1445b: ocme;, AUOIC;.
33 Arist.Poet. 1449b, 1445b (i] ow'tTJPia 'tTic; xaMpocwd. A document of 1059 issued by the
patriarch Constantine Lichoudes refers to the cleansing (imoxai)aipw) of the foulness (ii'to1tov) of
murder through repentance: RHALLES-POTLES, V 48; V. GRUMEL, Les Regestes des actes du
patriarcat de Constantinople I.III (Paris 1947) N.887.

357
 154 Ruth Macrides

While dramatic elements are present in the poem, its structure derives from
an altogether different genre, the legal document known as a semeioma, a record
ofthe proceedings and verdict on a case, issued by civil and ecclesiastical courts. 3.
The records of murder cases before ecclesiastical courts provide the most appro-
priate material for comparison with the verse case. Although the patriarchate
itself has left no documents of this nature,35 they are available from two thir-
teenth-century courts, those of Demetrios Chomatenos, archbishop of Ochrid,
and of Ioannes Apokaukos, archbishop of Naupaktos. 36
The semeiomata issued by these authorities open with the identification of the
person who has come before the court, including his or her place of origin. Two
examples from the Chomatenos collection are representative: 'A woman from
the theme of Kolonia, by the name of Zoe, came to the church and stood before
the archbishop, then, prostrating herself, confessed her sin';37 'Theodore Dem-
nites, originally from the east, now living in the theme of Acheloos, came to the
holy church of God, stood today before our holy lord the archbishop of all Bulga-
ria, and confessed his sin, thence moving those listening to pity and tears'. 38 This
opening is followed by an account of the confession, in the third person,39 intro-
duced by the words 'he/she said that .. .'. After the confession is reported, it is
summed up with the expression, 'he/she said those things', a transitional phrase
leading to the judge's decision. The 'verdict' states the category of killing com-
mitted - involuntary or wilful, premeditated or not40 - and specifies the nature
and duration of the penances to be performed in expiation of the sin. 41 A final
34 DOLGER- KARAYANNOPULOS (as in n.31), 82, 85-87; J. DARROuziis. Recherches sur les
'Oq)(pIXla de l'eglise byzantine (Paris 1970) 482-508.
35 Although the document issued by Lichoudes (see n.33) is called a semeioma and deals with a
murder case, it does not transmit the confession of the person who came before the tribunal. See
DARRouziis, 'Otptplxla, 489, on the diplomatic of the document.
36 References to the decisions of these archbishops on murder cases are most readily available
in the table prepared by M. V. STRAZZERI, 'Drei Formulare aus dem Handbuch eines Provinz-
bistums', Fontes Minores ill (1979) 342-351. Full references will be given to individual pieces
cited below. See now too the discussion of an Apokaukos case in M.TH. FOG EN, 'Ein ganz gewohn-
licher Mord', Rechtshistorisches Journal 3 (1984) 71-81.
37 J. PITRA, Analecta Sacra et Classica Spicilegio Solesmensi Parata VI (Rome 1891, repro
1967) no. 129, col. 529.
38 PITRA, op. cit., no. 118, col. 503.
39 The one exception known to me is the first-person account in a semeioma from Apokaukos:
A. PAPADOPOULOS-KERAMEUS, • "Iw&vvT)~ 'A1t6xauxo~ xai N\xiita~ XWV\&t'T)~', Ttooapaxovra-
tTTJpl\ rofi xa07Jyrrrofi K.E. K6vrov (Athens 1909) 379-382.
40 For categories of killing, see Constantine Porphyrogennetos'legislation: ZEPOS, Jus Graeco-

romanum I, 230-231; F. DOLGER, Regesten der Kaiserurkunden des ostromischen Reiches,

N. 676; also the discussion by S. N. TROIANOS, '0 «IIOIvaAIO\» rou 'ExAoyaMov (Frankfurt am Main
1980) 6-10.
41 For the relationship of the penances to the category of killing committed see STRAZZERI
(as in n. 36) 334-339.

358
 Poetic Justice in the Patriarchate 155

statement is often attached to the document and is addressed to civil officials,

warning them, under pain of excommunication, not to harm the said person phy-
sically or materially.42
This schematic rendering of the structure of a semeioma shows that the poem
on the case of Maria contains the same three basic elements as the legal docu-
ment; opening identification, confession, decision. It is in the opening section
that the poem's similarity with the semeioma is most evident, for it conveys the
same kind of information - including a description of the emotional reaction of
the audience - and in the same language. However, after this introduction and
before the confession, the poem adds information and material not found in any
semeioma: the section in which the judge himself reveals the substance of the
confession, in the form of a tragic lament (ll.18-44) and the judge's address to
the penitent, consisting of a long series of questions, with a digression on compa-
rative cases (11. 45-122). These two elements, which are additional to the usual
semeioma structure, combine to give the judge a far larger and more dominant
role than the person who has come before him to confess. Another difference lies
in the fact that the confession is communicated in direct speech rather than in
the more usual third-person narrative of the legal documents. Likewise, the
concluding section, giving the decision on the case, is introduced by the summing
up phrase ,00au,' e<pT}OCV (1. 156) as in semeiomata, but differs from them in its
lack of reference to the category of killing committed, and to the penances
assigned to the confessed killer.
The poem contains remarkable similarities with the legal text known as a
semeioma, both in arrangement and in detail. Yet it also conveys at once more
and less information than the legal document. The 'extra' information is trans-
mitted both directly, in those lines which are literally' additions' to the usual con-
tent of a semeioma, and indirectly, through figures of speech, dramatic language
and the verse form. Taken as a whole, these additions are of a rhetorical nature.
The poem is a combination of rhetorical and legal language and forms, a tragic
lament and a record of a case at one and the same time.
The case, like the form in which it is written, is composed of elements which
can be identified. Furthermore, these elements - the situations, places, insti-
tutions and functions which make up the story - can be located in time, to a late
eleventh/twelfth-century context. To begin with, the practice of confessing
before a tribunal in the Great Church is well-known, in particular with regard to

42 For an example, see the document of 1059 issued by Lichoudes (as in n. 33) 49. Chomatenos
and Apokaukos vary in their practice with regard to the inclusion of this warning. Chomatenos in·
c1udes it only when the person has confessed to an involuntary killing (PITRA, no. 116, col. 499-
502; no. 120, col. 509-512; no. 131, col. 533-536), while Apokaukos attaches the clause in almost
every case (see STRAZZERI, op. cit., Table, with references).

359
 156 Ruth Macrides

murder cases. Imperial novels of the tenth and twelfth centuries refer to the con-
fession of murderers in the context of granting asylum in the church to these
people. Constantine VII's novel which extended the privilege, mentions confes-
sion and penances, as well as civil punishments. 43 Two eleventh-century patriar-
chal documents, concerned with the assignment of ecclesiastical censures to con-
fessed murderers, indicate that the church considered the killer who managed to
reach the holy precinct immune from physical and material punishments." A
novel of Manuel I reveals that by the twelfth century confession of murderers in
Hagia Sophia was a common practice and had even got out of hand, for murderers
from the provinces were coming to the church because of its reputation for secur-
ing the safety ofthe killer.· 5 Finally, the entire procedure, from the arrival ofthe
murderer at the church until his departure with semeioma in hand, is described
in a formula which survives in manuscripts of the fifteenth century. 46
The poem's more detailed and technical references can also be identified. The
ekdikoi (I. 155), whose judgement Maria sought, are the clerics who constituted
the ekdikeion, a tribunal which met in Hagia Sophia and whose activity in cases
of killing is attested from the twelfth century on. 47 According to a late twelfth-
century text, the head of the ekdikeion, the protekdikos, was responsible for
'punishing and censuring, in a more compassionate manner, and as the lawofthe
church wishes, those who have soiled their hands with human blood'. 48 The parti-
culars of his function are given in the formula mentioned above, a text which out-
lines procedure for the confession of a murderer at Hagia Sophia, and for the
composition of a semeioma. This source states that the protekdikos instructed
the penitent on the manner of his confession, listened to the confession and, on
the basis of that statement, came to a decision with regard to the kind of killing
which had been committed. Accordingly, he assigned the penances appropriate
to the case, describing and listing them in written form. This document was given
to the penitent to serve as a reminder of his penances and as a protection against
civil officials or others who might attempt to harm him: 9
From the above description it emerges that the role of the judge of the poem is
identical with that of a protekdikos. Even a trace of the protekdikos' instructions

'3 ZEPOS, Jus Graecoromanum, I, 232-235; Constantine's novel was reexamined by Manuel I
in his own legislation on the subject: see n. 45 below.
.. RHALLES-POTLES V, 48-49; 49-50; GRUMEL, Regestes, N.887, N.888 .
• 5 R. MACRIDES, in Fontes Minores VI (1984) 156-162 (text), 190-204 (commentary) .
• 6 A. PAVLOV, 'Grecheskaia zapiso tserknovnom sudie nad ubiitsami, pribiegaiushchimi pod
zashchitu tserkvi', Viz. Vrem. 4 (1897) 155-159; DARROUZES, 'Orprpixla, 330-331.
.7 DARROUZES, 'O<p<p ix1a , 323-332, esp. 328; STRAZZERI (as in n.36) 332-333 .
• 8 George Tornikes, oration for the patriarch George Xiphilinos, ed. DARROUZES, 'Orp<pix1a,
534,19-22 .
• 9 PAVLOV (as in n.46), 158-159.

360
 Poetic Justice in the Patriarchate 157

to the penitent, not to lie or hide anything in confession, 50 is present in the poem
in 1. 113, where the judge urges Maria to speak up and reveal everything. Another
reference in the poem explicable in terms of procedure is the mention of the
speedy decision on the case (l. 162). This is in conformity with the practice of
reaching a decision and recording it for the penitent on the same day as the con-
fession. A sign of this one-day procedure can be seen in the word 'today' at the
beginning of semeiomata 51 and of the poem (1. 3).
To turn to the object of the judge's attention, the penitent Maria. She states
that she came from the Kibyrraiot theme which, it can be assumed, was also the
scene of her deeds. This theme in south-western Asia Minor, created in the
eighth century, is mentioned in sources until the mid-twelfth century. 52 The
deprivation Maria describes in her confession, the famine brought about by the
'barbarian knife' (l. 137), can be identified with the situation in Asia Minor after
the Byzantine defeat at Manzikert in 1071 and into the twelfth century, when
most of Asia Minor was overrun by the Turks. 53 A document of 1143 refers to
the restoration of order in a region of the theme, an area once prosperous, well-
populated and well-stocked which had been reduced to nothing by barbarian
invasion. 54 Another point of correspondence between the poem's story and
twelfth-century reality is Maria's coming from south-western Asia Minor to
the capital to confess at the patriarchate. This is in keeping with the central role
of the Great Church as a place of refuge for provincial murderers, to which
Manuel I's novel of 1166 attests. 55
Although there is a great deal of evidence for murder in Byzantium and the
fate of those who took human life, less is known about cannibalism, as one might
expect. It occurred rarely or was rarely reported but it was not unheard of in times

50 Idem., 158: [n/){xoxc,cxl napa ,ou npw,cx/)(xou ,ov ,ponov ,f)~ t~ollol.oyf)ocw~, xal onw~
DUX O",c\ACI tjlcu/)o~ rinciv, xal xal.utjlal xal ttnoxputjlal ,ilv ai"t(av "tOu <povou.
51 E.g. Chomatenos, ed. PITRA, no. 118, co1.503; no.131 col. 533; Apokaukos (ed. PAPADO-
pouLOs-KERAMEUS, as in n. 39) 379; ed. S. PETRIDES, 'Jean Apokaukos,lettres etautres documents
inedita', Bulletin de l'Institut arch€ologique russe 14 (Sofia 1909) XIV, 27; XV, 34.
52 M. GREGORIOU-IoANNlDOU, 'To vaunxo Mila "(Giv KI!3uppau;nwv: I;ull!3ol.il 0"(0 npo!3l.lllla

,f), iopuocw~ ,ou', Bv(avnva 11 (1982) 203-221; H.AHRWEILER, 'L'histoire et la geographie de

la region de Smyrne entre les deux occupations turques (1081-1317) particulierement au XIII'
siecle', Travaux et Memoires I (1965) 129 and n. 30 for references.
53 S. VRYONIS, Jr., The Decline of Medieval Hellenism in Asia Minor and the Process of Islami·
zationfrom the Eleventh through the Fifteenth Century (Berkeley 1971) 103 ff. The area was also
exposed to invasion by Arabs much earlier but for reasons discussed below, this date is a less likely
possibility for the reference in the text.
5, F. MIKLOSICH - J. MULLER, Acta et Diplomata graeca medii aevi sacra et profana (Vienna
1860-1890) IV, 324,11-19. Both texts use the same expression, 'barbarian knife' to indicate the
cause of the destruction. For the date of the document see AHRWEILER (as in n.52), 129.
55 MACRlDES (as in n.45) 158,40 -166,147, and commentary, 203-204.

361
 158 Ruth Macrides

offamine. An instance is related by Kekaumenos in connection with an attempt

by Samuel of Bulgaria to force the inhabitants of Larissa to surrender. The tzar
caused extreme hardship by preventing the people of that town from gathering
their harvest for three consecutive years. The ensuing famine force~ them to eat
unclean meat (IlUtpWV ... XpEWV), including dogs and asses. They even ate hides
which they had sorted out from among the excrement; also, reports Kekaume-
nos, 'one woman, whose husband had died, ate his thigh'.56
The poem's story, then, relates practices and situations which can be identi-
fied and located in time and place. Yet these identifications do not prove that the
poem reflects a genuine case. In fact, although each part of the story is plausible
in itself, the sum total seems excessive and unbelievable. A nun who ate her
mother and other dead people, including her own child whom she had first killed,
makes quite a story, even without the catalogue of animals Maria is also said to
have devoured (11.50-52) and about which the judge is particularly curious
(11.105-113). However, it is not only excesses in content which create problems
of interpretation. The irregularities in form, discussed above, also generate
questions about the purpose of the poem. If the case were a real one we might
expect it to have been transmitted as a normal prose semeioma and not in the
'hybrid' of genres which it is. The role of the tragi-legal form is, then, of central
importance in determining the truth of the story and the function of the piece.
Could it be an indication that the work is a satire of the procedure of confession at
the patriarchate, in the form of a parody of a semeioma?57 Another piece in the
same manuscript, the Katomyomachia attributed to Prodromos, is precisely
this, a satire in the form of a parody of a tragedy. In that work comic effect is pro-
duced through the surprising contradiction between the expectations which the
tragic form arouses and the actual story it tells of mice, the tragic victims of an
omnivorous (TtIXIl<Payoc;) cat. 58
Is there, then, a discrepancy between the tragi-legal form of the poem and the
facts ofthe case? Any attempt to determine the precise relationship ofform and
content must take into consideration the elements in the work which constitute a
departure from the normal semeioma, that is, those lines which contain informa-
tion 'additional' to the usual semeioma and those rhetorical devices which also
indirectly add to the normal legal document.

56 G.G. LITAVRIN, Soveti iRasskaziKekavmena (Moscow 1972) 250,14 - 252,16. For the date of
the incident see the commentary by P. LEMERLE, 'Prolegomimes a une edition critique etcommen-
tee des «Conseils etRecits» de Kekaum{mos', Academie Royale de Belgique, Memo ires LIV(1960)
25-27.
57 For a definition of parody, and its use in satire see H. ElDENElER, Spanos, eine byzantinische
Satire in der Form einer Parodie (Berlin-New York 1977) 29-56.
58 HUNGER, Katz-Miiuse-Krieg (as in n.2), 51-65.

362
 Poetic Justice in the Patriarchate 159

On the surface, the 'additions' could be read as subversive in purpose, for in

ll. 13-122 the judge seems to overshadow the penitent in importance, not only
because he has the greater number of lines but also because he preempts her by
revealing the contents of her confession. To be sure, this is a distortion of the
normal semeioma arrangement, but is its purpose subversive?
In the opening lines the judge tells Maria's story, using the topoi and language
of a tragic lament. He offers an explanation for his adoption of this form of expres-
sion, an explanation which is also an argument. He begins, as do most semeio-
mata on murder cases, by narrating the effect which the story had 0..'1 the ekdikoi.
From this description, two emotions can be recognized - fear and pity - but ofthe
two, it is compassion which is singled out by the judge. He underlines and empha-
sises that this was the dominant emotion by citing the Homeric verses (ll. 13-14)
and elaborating on them: those who heard Maria's story were not made of wood,
stone, or the blue sea - that is, they were human and, therefore, compassionate. 59
He also demonstrates that compassion is the overwhelming emotion produced
by her story. This he does implicitly in his choice of the lament as the form in
which to relate her deeds, for lamentation is the expression of grief and com-
passion. 60 He makes this choice in a self-conscious manner, asking what form
might be the most appropriate to convey her story - a threnos, a song, a monody
or a tragedy - all the while making use of a topos of the threnos or monody.6J His
choice is confirmed by the lines which follow in which he invokes the sun, the sky,
the earth, and the air, another tapas of the lament, and it is in the course of this
invocation that he reveals Maria's story.
These lines, then, set the tone of the piece and provide an argument for the
use of the lament: if compassion is the most appropriate emotion in reaction to
the tragedy which Maria's story is, then the lament is the most apt form for
communicating that tragedy. But more than this, the introductory lines show
that to use the lament is to express a defence of Maria. The lament literally
becomes the vehicle of the judge's defence in his invocation of the natural
elements. In those lines, a commonplace of the lament is charged with a new

59 These Homeric verses are among the most commonly quoted in Byzantine literature:
H. HUNGER, 'On the imitation (f1(f1TjOI~) of antiquity in Byzantine literature', DOP 23-24 (1969-
1970) 29-30. However, the conceit is found in a more developed form, as it is here, in the context of
man's.compassionate nature, in an excerpt from the twelfth-century romance of Constantine
Manasses which was part of the collection of 'maxims' made by Makarios Chrysokephalos, metro-
politan of Philadelphia in the mid-fourteenth century: R. HERCHER, Erotici Scriptores Graeci
(Leipzig 1859) I 48-54. This suggests that the author of the poem could have had intermediate
sources for his citations and mythological examples.
60 Menander Rhetor, edd. and trans!. D.A. RUSSELL and N.G. WILSON (Oxford 1981) 202,18-19;
346-347.
6J ALEX IOU, The ritual lament (as in n.6), 161-162.

363
160 Ruth Macrides

function. The sun, sky and air are summoned to darken the sky, to thunder, to
wrap the world in mist, but not as an expression of grief, as is usual. 62 Rather, here
they are invoked to protect Maria by hiding her from those who would tear her
apart. 63 Yet another departure from the topos is indicated by the invocation of
the earth (1. 35) who (unlike the sky, sun and air) is not instructed to hide the
drama, for she is already performing this function. The change in construction,
the use of the participle instead of the imperative, breaks the parallelism in
the invocation, in this way drawing attention to the parallel nature of the earth
and the 1tCq.l(payo~ Maria: both hide within them the dead whom they have de-
voured. 64 The topos of the invocation has thus been adapted to suit the occasion,
and this invocation strengthens the imagery and contributes to the case for
Maria.
In the following section, in the address to Maria (II. 45-122), the judge conti-
nues his defence for her case through the examples he cites and the questions he
asks. Again in these lines, although he appears to be overshadowing her by virtue
ofthe large role he has, he is in fact foreshadowing and supporting her confession.
The judge's defence is really an extension of the argument he made in the
opening section: if the case is a tragedy - and nowhere does he refer to Maria's
deeds as anything else - then she, like a character in a drama, is a victim of forces
outside her control, not fully responsible for what she did. He builds up this argu-
ment by means of the parallel cases he presents. Although no one example
equals hers, there is something of each example in her case (II. 92-96). The
mythological examples show that an inescapable cycle of destruction exists. In
her confession Maria reveals that she is caught up in such a cycle; evil leads to
suffering and more evil. The historical examples also have something of her
story in them. The Old Testament judge Jephthah sacrificed his own daughter,
caught in a tragic predicament, under the necessity of acting and yet knowing
that his action must involve destruction. The judge brings out the poetic-tragic
nature of Jephthah's situation with the words axwv E:XWV 1"£ (1. 83), but also its
legal nature, for the phrase is reminiscent of the 'legal' term E:xouO\oaxo60\o~,
the person who commits a killing which is neither completely involuntary nor

62 Eadem, 158-159, 169-170; ALEXIOU and DRONKE, The lament of Jephtha's daughter (as in
n. 6) 846-851, with examples from Pindar on. See also, Proclus, On the CrucifIxion, ed. LEROY (as in
n. 14) 209; Prodromos, ed. HORANDNER (as in n. 14) XLV,1-5; Stilbes' lament on the flTe (LAMPROS,
as in n.1) 13,32-34; A. PAPADOPOULOS-KERAMEUS, 'epiivo~ ,ii~ Kwvotavnvou1t6AEW~', BZ 12
(1903) 269.
63 11.43-44 have overtones of Christ's tearing open the 'omnivorous stomach of Hades': Proclus,
On the Sunday after Easter, ed. LEROY (as in n. 14), 237-238: ,ou ~wtiipo~ i]IlWV IlLappiJ~av,o~
,f)v 1tttllIPayov wu ~oou yao'tEpa ...
64 For the earth as womb·tomb and the womb as 'grave' see the discussion by P.E. SLATER, The
Glory of Hera: Greek Mythology and the Greek Family (Boston 1971) 77-78, 86-89.

364
 Poetic Justice in the Patriarchate 161

completely intentional. 65 This expression could also be applied to the state in

which Maria committed her deeds.
Finally, the example of the Jewish woman provides the fullest and most appro-
priate parallel of all. If that case was as famous as the judge claims, the similari-
ties would have been immediately apparent to his audience: both women killed
and ate their child, in a time offamine caused by war. For Josephus, the source of
the story, and for all subsequent writers who retold it, the tale exemplifies the
extremes to which human beings are driven by the cruel hardships of war. By his
citation of this example, the judge alludes to the specific context of Maria's
actions, the conditions which forced her to commit her deeds. He furthermore
implies by reference to that case that the same kind of reaction is appropriate in
the case before him. His argument runs - if the story of that Maria of long ago
moved people to compassion and still amazes men today, how much more sym-
pathy does this Maria deserve, whose situation, although similar, was far worse. 66
Having argued through these examples that Maria was compelled by circum-
stances outside her control to commit her deeds, the judge then demonstrates
just how appalling those circumstances were. This he does by means of the ques-
tions he poses, in particular his queries concerning the poisonous animals she
ate, the lizards, snakes, mice, tortoises and frogs. He inquires anxiously into the
manner of their consumption, and his concern seems inappropriate, if not comic,
because of the discrepancy between those creatures and the human flesh she
devoured. Yet, the eating of these animals, 'among the creeping things that
creep on earth' (Lev. 11.29-30) was prohibited by canon law, except in cases of
great need. 67 And it is in association with extreme desperation in times of famine
that sources report the eating of 'unclean meat'.68 In those sources, too, it is often
the stage before the final resort to human flesh. 69 Therefore, the introduction of
this catalogue of creatures by the judge is in itself an indication that conditions of
famine obtained. Moreover, the judge's questions about the preparation and con-

65 See Balsamon's commentary on canon 13 of Basil: RHALLES-POTLES IV 133, and the scholion
to the Basilika at 60.39.3 (ed. HEIMBACH).
66 For synkrisis see Nicholas, Progymnasmata, Rhetores Craeci, ed.J.FELTEN (Leipzig 1913)
60,6-10.
67 canon 63 of the Holy Apostles: RHALLES-POTLES II 81; canon 67 ofTroullo, RHALLES-POTLES
II 462-463; canon 86 of Basil, RHALLES-POTLES IV257; E. PATLAGEAN, 'Byzance, Ie barbare, l'here·
tique et la loi universelle', Ni Juit, ni Cree: Entretiens sur Ie racisme, so us la direction de Leon
Poliakov, Eeoles des Hautes Etudes en Sciences sociales (1978), repro Structure sociale, fa mille,
chretiente a Byzance (London 1981) 81-87.
6B Kekaumenos (as in n. 56); Anna Komnena, ed. LEIB III (Paris 1945) XI, IX.1.26-30; Niketas
Choniates, ed.J.L. VAN DIETEN (Berlin-New York 1975) 164,54-56; Prodromos, ed. HORANDNER
(as in n. 14) XIX,24-26.
69 See, also, descriptions of the Russian famine of 1601-1611: R.E.F. SMITH and D. CHRISTIAN,
Bread and Salt: A social and economic history offood and drink in Russia (Cambridge 1984) 109.

365
 162 Ruth Macrides

sumption of snakes whose flesh was widely believed to be poisonous,70 are the
best illustration ofthe extremes to which Maria was pushed: even fear of poison-
ing herself did not deter her. This fact, in turn, detracts from the abomination of
eating human flesh. The judge's fear is coupled with pity.
Up to this point the judge has only alluded to the conditions or circumstances
of Maria's acts. His final question, concerning accomplices (ll. 110-111) puts a
name to those conditions. Winter and Famine are implicated in Maria's confes-
sion as those who 'forced' her to commit her deeds but, as is also revealed, the
sower of tares', the devil, is ultimately responsible. The devil is often named in
semeiomata also as the' accomplice' without whom the penitent would not have
committed his sin. 71 Here, too, the function of the accomplice-question is to
emphasize Maria's diminished responsibility.
The closing statement of the judge (ll.118-1l9) draws attention to the dual
nature of his role and can be said to summarise the twin aspects ofthe work, the
legal and the poetic. In his address to Maria, the tragic nature of the case has
transformed him, a judge, into a tragodos, a word used of the singer of a lament
but also of a tragic chorus member. His questioning of Maria is a role appropriate
to the tragodos as well as the judge, and the question-answer form is found in the
structure of the tragedy as well as the lament. 72
The question ofthe relationship ofform and content has been answered by the
judge in ll. 13-122. He has made a case for the form he employs in an argument
which reconciles irregularities in form with excesses in content; content dictates
form. Can we believe the persuasive case he makes?
A phrase in a semeioma issued by Chomatenos brings us closer to the answer.
A soldier who killed a fiscal official in the heat of anger confessed before the arch-
bishop's tribunal, 'moving those who heard him to pity and tears, since he had

70 Snakes' flesh was used as an antidote in ancient medicine and is referred to as such in Byzan-
tine orations: Michael 0 tou Anchialou, oration for Manuel I: R. BROWNING: 'A new source on
Byzantine-Hungarian Relations in the twelfth century', Balkan Studies 2 (1961) 189,77-78;
205,77, repro in Studies on Byzantine History, Literature and Education (London 1977); Michael
Choniates, oration for Isaac II, ed. SP. LAMPROS, M1xaiJA. 'AxoJ.llvarov roD Xwvlarov Ta Ew(6-
,"eva I (Athens 1879/1880, repro Groningen 1968) 239,1-4, The heads and tails of snakes figure in
Greek folk songs as sources of poison: M. ALEXIOU, Sons, Wives and Mothers: Reality and Fantasy
in some Modern Greek Ballads, Journal of Modern Greek Studies I (1983) 102,19 -103,25. Signifi-
cantly, in those songs, it is the middle part of the snake, minus head and tail, which is considered
most poisonous.
71 See the formula n£pi !povcw~, ed. PAVLOV (as in n.46), 159: ouv£py(q: 'tOU novl1pou oa(flo-
vo~ n£plcn£O£v ei~ i:yxAl1fla !povou; Chomatenos, ed. PITRA, no. 118, col. 503: 'tou nAl1flfl£ATlfla-
'to~, ou ycyov£v aAwolflo~ i:~ i:nl1pEia<; 'tou novl1Pou; also, FOGEN, 'Ein ganz gewohnlicher Mord' (as
in n. 36) 80-81.
72 ALEXIOU, The ritual lament, 85, 162,

366
 Poetic Justice in the Patriarchate 163

coIl1IIl.itted deeds worthy of a lament'. 73 This statement suggests that the idea for
the treatment ofthe Maria case was not so very far removed from the everyday
language of the ecclesiastical courts. Not content to describe his case as worthy of
a lament, this judge wrote one.
His treatment of the case may make good reading, but we may wonder whe-
ther it makes good law. The judge in the Maria case does what one would expect
of a judge; he presents an argument. But his argument is not based on the law.
Instead, he relies on word-play, synkrisis, and figures of speech to build and sup-
port his case for a compassionate attitude toward Maria. Furthermore, the argu-
ment is put forth not only in the lines he speaks but in every line of the work. He
puts words in Maria's mouth, producing her confession in the language and style
which befit the rest of the piece," and not as Maria, a provincial, would have
delivered it. Finally, the argument is conveyed not only by the judge and by
Maria; it is in the verse form itself. In short, the work is highly rhetorical.
A comparison with the decisions of other Byzantine judges shows that these
methods were also employed in more ordinary cases. Prose legal documents
exhibit varying degrees of rhetorical usage: poetic language, puns, and the inter-
jection of emotive words are the rule rather than the exception. 75 Confessions by
penitent killers before ecclesiastical tribunals, and statements by parties in liti-
gation before civil courts are transmitted in the literary, not the spoken, language
and style. One judge, commenting on the contrast between the case as it came
out of the plaintiffs mouth and as it was recorded in the semeioma, stated that
the former was 'shapeless and completely inadmissible to the courts and to the
laws', whereas 'the atticising idiom bestowed on it a better appearance and
shape'.76 That rhetorical devices, far from being an extravagance in a legal context,
were fundamental to the decision-making process and were highly rated as such,
has been demonstrated for, and from, the decisions of Eustathios Rhomaios,
the eleventh-century judge of the Hippodrome." A semeioma of Apokaukos on a

73 PITRA, no. 118, col. 503.

" The confession is a kind of ethopoiia: Aphthonius, Progymnasmata, ed. H. RABE (Leipzig
1926) 34-35. HUNGER, Die hochsprachliche profane Literatur I, 108-116; D.A. RUSSELL, Greek
Declamation (Cambridge 1983) 11-12.
75 Chomatenos, ed. PITRA, no. 131, col. 534: <pi)«VCl TOUTOV i:~ ciVCA1tiatUlV, <pcu, to PtAo~ to
1tCXtPIXOV' XCX! cxui)T]flCPOV ci~ ~OT]V 1tCXPCX1tEfl1tU tOV <pihcxtov tOV i:yxapolOv ... ; no.118, col. 503:
~ClVoV tl XPf)flcx i)uflo, ... (prooimion); Apokaukos: N.A. BEES, 'Unedierte Schriftstiicke aus der
Kanzlei des Johannes Apokaukos', Byzantinisch-neugriechische Jahrbiicher 21 (1976) 75,19-21:
XCX! ElO£V 6 ijAIO, XU! tv talC; ~fl£ti:pCXI~ ~fli:pulC; tOV BaaiA£lOv TOUTOV 1tCXtEpa tCXVOXtOVOv ... ; see
also the description of a land-dispute, attributed to Michael Psellos: K.N. SATHAS, Bibliotheca
graeca medii aevi V (Paris 1876, repro Athens 1972) 197-203.
76 Ed. SATHAS (as in n.75) 197-198. This statement makes a play on the word 1tpOaUl1tOV

(appearance / plea in court).

" D. SIMON, Rechtsfindung am byzantinischen Reichsgericht (Frankfurt am Main 1973) 7 -32.

367
 164 Ruth Macrides

divorce case shows too that the judge's citation of poetry and imitation of poetic
style in the description of the case were not external decorative additions but an
integral part of his argument - in fact, its vehicle. 78
What differentiates the judge in the Maria case from his colleagues is not his
use of rhetorical devices to describe and shape his case, but the degree to which
he employs them. He makes active all the normally passive elements of a
semeioma, putting them to work to argue the case. Prose becomes verse to con-
vey the intrinsically tragic nature ofthe case; a formula becomes a lament; judge
and penitent deliver their statements in person.
Rhetoric exaggerates but it does not necessarily lie. The judge's innovative
use of topoi and mixing of genres, as well as the contextual associations of his
examples and images, work to provide a strong argument for a true case pre-
sented in a way which is true to its essence. The evidence which the structure,
language and imagery give is sufficient to render what at first seems incredible,
all-too-real. The judge did not need to make up the story; truth is stranger than
fiction.
The judge who narrates the story of Maria can be identified with the protekdi-
kos Andronikos to whom the heading attributes authorship of the verses. The
protekdikos' function of hearing confessions of murderers, and writing the
semeioma which was a record of the case, is compatible with the role which the
judge of the text has. Furthermore, the literary skills needed to write apiece of
this kind would have been within the reach of a protekdikos of the Great Church.
Although the protekdikos Andronikos is not otherwise known, the career of
Michael 0 tou Anchialou, who was protekdikos and hypatos ton philosophon
before he became patriarch (1170-1178), gives an indication of the level oflearn-
ing which a protekdikos could possess. 79 The late eleventh-twelfth century is also
most likely the period of Andronikos' activity and therefore probably also the
date of Maria's case. This date is indicated by the complex of information which
survives from the twelfth century concerning the protekdikos and murder cases
in Hagia Sophia, by traits which the verses exhibit as literature, and by the con-
tent of the manuscript in which the verses have been preserved, there being no
verse piece earlier than the eleventh century. If the case of Maria was indeed one
which the protekdikos Andronikos himself heard - and there seems no reason to
doubt this - then the breakdown of order which occurred in south-western Asia
Minor in the late eleventh century and into the twelfth can be identified with the
situation Maria experienced and described.

78 M.TH. FOGEN, 'Rechtsprechung mit Aristophanes', Rechtshistorisches Journal 1 (1982)74-82.

79 Michael Choniates, oration for the patriarch Michael, ed. LAMPROS (as in n.70) I 82,23-
83,13.

368
 Poetic Justice in the Patriarchate 165

However, the story does not end with the entry of the case in the Regestes.
The question of the function of the verses and their intended recipient or audi-
ence still remains to be answered. The thin line which exists between literary
and legal text is all the more difficult to draw in this case because the work is both
a record of a case and a lament on a tragedy. The genres overlap to a degree
which is unprecedented. Nor is the verse form itself necessarily a determining
factor in assigning the piece a literary rather than legal function; the verses, it
has been shown, convey 'technical', i.e. legal, as well as literary information. so
An indication of the purpose of the piece and therefore, by implication, its
intended audience, is provided by the heading, which states that the verses were
written 'in place of a semeioma'. This brief statement, made by the author him-
self or a copyist, implies that although the verses do not constitute a semeioma,
they could have had a legal function. The semeioma, a record ofthe proceedings
ofthe case, was given to the penitent on the same day as his, or her, confession,
while a fuller version was kept in the archives. The protekdikos Andronikos
makes a point of stating that he reached a decision on the case quickly, declaring
Maria 'redeemed' at once, even though he referred the case to the judgement of
the patriarch and synod. The protekdikos' verses do not contain the list of
penances which are always found at the end of the semeioma - for these have still
to be assigned by patriarch and synod - but they do include the 'decision', in a
sense: Maria has received the compassion and pardon of the ekdikeion. The
verses, then, might have been given to Maria as a record of her presence before
the tribunal. The penances vital to her salvation would have been given to her
later, in a separate document (a semeioma). In the meantime, the protekdikos'
verses could serve as a record and, more important, as protection. The 'warning'
to civil officials and others, often found at the end of the semeioma, is not present
in the verses. However, the protekdikos' argument for a clement attitude toward
Maria might have fulfilled the same function, directed at those who would have
harmed her because of her deeds. Maria had already suffered punishments; whe-
ther at the hands of civil officials or others is not clear. She was still 'paying the
penalty' at the time of her confession before the tribunal. Like all those who
confessed in Hagia Sophia to murder, she could not have been indifferent to the
physical protection which the church could provide, even if this was not the stat-
ed reason for her presence there. One might well wonder, however, whether the

so It seems, however, that verses written at the bottom of a legal text could seriously jeopardise
the validity of that document: Ioannes Tzetzes, Epistulae, ed. P. A. M. LEONE (Leipzig 1972)
no.47, 68,8-14. For a definition of literature as a 'work not bound by a technical aim' see
H.-G. BECK, Das byzantinische Jahrtausend (Munich 1978) 109; also, A. P. KAZHDAN, People and
Power in Byzantium (Washington, D. C. 1982) 95.

369
 166 Ruth Macrides

protekdikos' argument, as it is expressed and developed, would have found an

appreciative audience in those from whom Maria might have sought protection.
The message in the concluding verses gives an indication of a more plausible
audience for the protekdikos' work. The ending expresses the saving power
which the synod's 'bonds' have on those who fear God. The synod's authority to
perform this vital act is stressed. The protekdikos could well have addressed his
verses to the bishops who composed the synod, a group of individuals who were in
a position to appreciate the work as law and as literature. As has been shown, the
verses contain an argument which far exceeds that found in any semeioma given
to a confessed killer. The document which penitents received was almost formu-
laic; it rarely included a discussion of the considerations which must have contri-
buted to the judge's decision on the category of killing and the duration and kinds
of penances to be imposed. As the heading to the verses states, the protekdikos'
piece is not a semeioma - no seimeioma issued by a protekdikos ofHagia Sophia
on a murder case has in fact survived. However, it may have been the fuller
account of the case given to the patriarch and synod, the recommendation for a
clement decision made by the protekdikos to that higher authority to whom the
case was referred because of its extreme and unusual nature. 81 H so, Andronikos'
verses would be the only record of a murder case which came before the ekdi-
keion to have been transmitted to us by the head of that tribunal. This sugges-
tion helps to explain the very strong element of argument in the piece, as well as
its legal structure.
Whether or not the verses had for the synod the legal function which has been
argued here, their value as literature remains. The story has been told in a way
that renders it drama. Even though the structure of the story is a legal one, this in
no way impairs a reading of the piece as a tragedy, since each part of the structure
is capable of being invested with dramatic purpose. The judge's opening narra-
tive and long address to Maria is like a monologue by a tragic chorus member who
reveals the story, laments the tragedy, searches the past for parallels,82 and ques-
tions the protagonist to elicit her story. The ending, too, brings its catharsis of
emotions which have been aroused.

81 It is not clear from the surviving evidence to what extent the protekdikos had the authority to
assign penances without reference to the patriarch and synod, even in cases more ordinary than
Maria's. Although the formula published by Pavlov shows the protekdikos in complete control of the
entire procedure, in a document of 1401 the patriarch himself decides on the appropriate penances,
charging the protekdikos with the drawing up of the semeioma: MIKLOSICH- M tiLLER n, 533-535, esp.
534;). DARROUZES. Les regestes des actes du Patriarchat de Constantinople (paris 1977) N.3230.
However, it may be that in the 1401 case also, the patriarch assigned the penances because of the seri-
ous nature of the case (murder of three men) and not because this was the normal procedure.
82 See Euripides, Medea 1282-1283, where the chorus fmds parallels for her actions;
Sophocles, Oedipus Tyrannus 1194.

370
 Poetic Justice in the Patriarchate 167

The drama which the protekdikos has written corresponds to Byzantine theo-
ries on the purpose of tragedy, as expressed in treatises on the subject and in
other works. The stress is on the representation of emotions, eliciting tears, and
lamenting misfortune. 83 The judge's drama is also comparable to a literary
drama, the Christos Paschon, which is a compilation of ancient Greek tragedies
as well as biblical sources. 8' In this work, whose authorship has been attributed to
Gregory Nazianzenos but also, and again more recently, to twelfth-century
authors,85 action is replaced by description of action, the high point is represent-
ed by a lament, and grief is transformed into joy through the Resurrection. 86 The
protekdikos' work contains the same elements and achieves effects in the same
way.87 Both dramas are real Christian tragedies for which the authors have used
the language of ancient Greek tragedy to express, and even to enhance, the story.
The protekdikos' tragedy demonstrates an eternal truth; it moves from the
personal to the universal, ending on a note of hope for man's salvation. In choos-
ing this form to tell a real tragedy of his time, he has made of it a timeless story. If
the protekdikos intended his verses for a private audience, to be read as drama,
he is the only known Byzantine judge to have exploited a case which came before
him for a wider public. 88

83 Eustathios of Thessalonike, llcpi imoxp[ocw,>, PG 136, 373BC; R. BROWNING, 'A Byzantine

treatise on tragedy', Acta Universitatis Carolinae (Prague 1963) 67 -81, repro in Studies (as in n.
70). For a discussion of these and other sources see R. DosTALovA, 'Die byzantinische Theorie des
Dramas und die Tragodie Christos Paschon', JOB 32/3 (Vienna 1982), XVI. Internationaler Byzan-
tinistenkongreJ3, Akten II/3, 73-82. On dramatic forms in Byzantine literature see HUNGER, Die
hochsprachliche profane Literatur, II 142-148; BECK, Das byzantinische Jahrtausend, 112-115.
8. TUlLIER, Gregoire de Nazianze (as in n. 14). For appraisals of the work see 5.5. AVERINCEV,
Vizantijskie eksperimenty s zanrovoj formoj klassiceskoj greceskoj tragedii, Problemy poetiki i
istorii literatury (Saransk 1973) 255-270. HUNGER, op. cit., II 102-104; ALEXIOU, The ritual
lament, 64-65.
85 HUNGER, Katz-Miiuse-Krieg, 49; IDEM, Gnomon 43 (1971) 127 f.;]. GROSDIDIER DE MATONS,
'Une edition recente du XplOtO~ Diioxwv', Travaux et Memoires 5 (1973) 363-372; DOSTALOVA,
op. cit., 73 tr.
86 DOSTALQVA, op. cit., 79-80; AVERINCEV, op. cit., 269-270; KAZHDAN, People and Power, 104.
87 Certain key words in the protekdikos' drama appear also in the Christos Paschon: OPUflU-
toupyo~ (ed. TUlLIER, I. 1887); 1tUfl<piiyo~ (I. 934); OCOflOAUtO~ (I. 2529) and n. 30 above; also I. 46:
atpr<pW yap d~ of: 1:0 1tPOOW1tOV tOU AOYOU (I. 1419).
88 For this audience, see KAzHDAN, People and Power, 103; M. MULLETT, 'Aristocracy and
Patronage in the literary circles of Comnenian Constantinople', The Byzantine Aristocracy, IX to
Xli Centuries, ed. M. ANGOLD, British Archaelogical Reports, International Series 221 (Oxford
1984) 173-20l. An interesting comparison for the protekdikos 'presentation of the case is provid-
ed by the example of the sixteenth-century French judge who exploited one of his own criminal
cases to write a work (vernacular) which combined the features of a legal text and a literary tale. He
used classical, biblical and more recent examples as comparisons for the case which he described
as a tragedy. In his dedication to a bishop he claimed that it was 'an argument so beautiful and so
strange' that it should bring the man recreation and release from worries: N. Z. DAVIS, The Return
of Martin Guerre (Cambridge, Massachusetts 1983) 94-125.

371
 168 Ruth Macrides

Finally, whatever function or functions the verses were intended to serve, they
show, in common with much Byzantine literature particularly ofthe twelfth cen-
tury, a creative use of the classical tradition. 89 The judge has put old conventions
to new use and displays versatility and flexibility in the way he employs images
and mixes genres. Classical learning did not create a distance between him and
contemporary reality. If anything, it helped him, a Constantinopolitan, to iden-
tify with the fate of a provincial woman. It was through and with this learning that
the protekdikos Andronikos, like his colleagues, other products of twelfth-cen-
tury education, was able to do justice to a human being whose experience and
situation were so far removed from his own. Was this not humanism in the fullest
sense of the word?

89 For a discussion of these features in twelfth-century literature: H. HUNGER, 'Die byzanti-

nische Literatur der Komnenenzeit: Versuch einer Neubewertung', Anzeiger der phil.-hist.
Klasse der Osterreichischen Akademie der Wissenschaften (1968) 59-76; M. ALEXIOU, 'A critical
reappraisal of Eustathios Makrembolites' "Hysmine and Hysminias"', BMGS 3 (1977) 23-43;
M. MULLETT, 'The classical tradition in the Byzantine letter', Byzantium and the Classical
Tradition, edd. M. MULLETT and R. SCOTT (Birmingham 1981) 75-93; A. KAZHDAN and S. FRANK-
LIN, Studies on Byzantine literature of the Eleventh and Twelfth Centuries (Cambridge 1984);
P. MAGDALlNO, 'The literary perception of everyday life in Byzantium: some general con-
siderations and the case of John Apokaukos', Byzantinoslavica (forthcoming).

372
BMGS 13 (1989) 183-218

Honour among Romaioi: the fmmework

of social values in the world of
Digenes Akrites and Kekaumenos 1

PAUL MAGDALINO

Medieval Greek writers of biography and encomium, Hke their

ancient models, conventionally began by extolling their subject's
ytyOc; (birth, kin, ancestry) and 1tCl'tpiC; (native land or city) as
the two basic co-ordinates of a social existence, which both con-
ferred an honourable start in life, and gained glory from a life
honourably lived. This was not just a rhetorical convention. In
a world where neither settlement and communication patterns nor

1. This paper builds on ideas which I have expressed in four other studies of Byzan-
tine aristocratic society and political ideology: 'Aspects of Twelfth-Century Byzan-
tine Kaiserkritik', Speculum 58 (1983) 326-46; 'Byzantine Snobbery' and 'The Byzan-.
tine Aristocratic Gikos', chapters 4 and 6 of The Byzantine Aristocracy, IX to XIII:
Centuries, ed. M. Angold (British Archaeological Reports, International Series 221)
(Oxford 1984); 'Nationalism and Hellenism in Byzantium' (forthcoming). In the in-
terests of economy, I have chosen not to reproduce arguments and citations made
in those articles. For the same reason, I have chosen not to qualify and document
all the statements made below. It is sufficient to record, at the outset, my obvious
debt to two groups of studies.
First, discussions of social values in modern Greek and Mediterranean society, and
notably: Honour and Shame. The Values of Mediterranean Society, ed. J.G. Peristiany
(London 1965); J.K. Campbell, Honour, Family and Patronage (Oxford 1964); J.
du Boulay, Portrait of a Greek Mountain Vii/age (Oxford 1974).
Second, recent studies of Late Antique and Byzantine society which have, albeit
in most cases unintentionally, pointed to the need for a study of honour and shame
in Byzantium: E. Patlagean, Pauvrete economique et pauvrete sociale aByzance, 4e-7e
sii!c1es (Paris/La Haye 1977): P. Brown, The Making of Late Antiquity (Cambridge,
Mass. 1978); idem, The Cult of the Saints (Chicago/London 1981): H.-G. Beck, Dos
byzantinische Jahrtausend (Munich 1978); C. Mango, Byzantium. The Empire of New
Rome (London 1980); A. Kazhdan and G. Constable, People and Power in Byzan-
tium (Washington D.C. 1982); A. Kazhdan and A.W. Epstein, Change in Byzantine
Culture in the Eleventh and Twelfth Centuries (Berkeley/Los Angeles/London 1985);
and, last but not least, a number of contributions to The Byzantine Aristocracy, ed.
Angold (see above).
The uninitiated reader may also find much of value in W.H. I\1cNeill, The Metamor-
phosis of Greece since World War II (Chicago 1978).

373
PAUL MA(iDALINO

economic and social priorities have changed radically until re-

cent times, the kin group and the community had much the same
significance for the medieval subjects of the East Roman emperor
that they had for the citizens of the ancient polis and have had
for the inhabitants of the modern Greek villages studied by social
anthropologists. The patris was still, for most people, a small,
introverted, self-regulating agricultural community whose
members all knew each other well, wanted to know more about
each other's business, and took a pride in their unique collective
identity. The genos of relatives (ClUYYEVEt<;) sharing a common
great-grandparent was still, within the patris, the group in which
most peop\t: instinctively invested their loyalty and trust, and with
which one of the basic qualifications for honourable status -
£uytv£la (good birth, nobility) was associated. 2 Although the in-
stitutional system through which honour was pursued and
recognised was no longer that of the city-state where the notions
of patris and genos had been forged, these notions were adapted
to the realities of a large, monarchical, territorial state - or, to
put it more accurately, that state never grew out of the thought
world of the ancient polis. Constantinople, the New Rome, was
more than the imperial capital: it was the 'ruling city', the 'com-
mon fatherland' (KOlV"il 1taTpi<;) of all who belonged to the
'Roman race' (ytvo<; 'Pwllaiwv). The words genos and eugeneia
were loaded with all the rejection of aliens, barbarians and slaves
which had characterised citizenship in the ancient world. To be
well born was, in principle, to be free-born and native, and, ideal-
ly, to belong to a genos which was good because it was indigenous
to, or adopted by, the New Rome.
Yet despite - indeed because of - their great elasticity, the
terms genos and patris are not wholly adequate to describe the
social framework of the medieval Greek code of honour and
shame. In extending to cover a wide range of meanings, they
smoothed over, but did not resolve, the basic contradictions at

2. For purposes of marriage and inheritance, kinship was considered to extend at

least to the sixth degree, and sometimes to the eighth: see Peira 49.3: ed. I. and P.
Zepos. Jus Graecorcomanum (Athens 1931, repro Aalen 1962) IV 197. Cf. in general
K .G. Pitsakis, To KWJ..VIlU yri/.wv J..6ycp (]vyyeveiuc; t{3t56Ilov {3u()lloiJ t~ uflluroC; (]ro
Bv(uVTlvO t5iKUIO (Athens 1985).

374
 HONOUR AMONG ROMAIOI

the heart of Byzantine society, between the universalism and the

parochialism of its political ideology, between the ecumenism and
the exclusivism of its culture, and between the civic ideal of nobili-
ty inherited from the ancient world and the courtly, chivalric ethos
which matured in the twelfth century. Moreover, the intrinsic
value of genos and patris was explicitly denied by Christian ethics,
which reached their logical conclusion with the complete inver-
sion of the honourable citizen into the homeless, kinless figure
of the Holy Fool - the man without honour. 3 As hagiographers
never tired of pointing out, a saint's real homeland was the
heavenly Jerusalem, and he or she inevitably renounced the
prestige of honourable birth. Christians commemorated not the
saint's birthday, but the day when the saint had moved on from
earthly exile to the 'eternal dwellings' in the Kingdom of Heaven
which were the believer's true home.
The images of home and kingdom in which .:ternity was con-
ceived reveal the importance of another pair of social units which
complemented patris and genos as co-ordinates of honour and
shame: the household (oh<o~) and the state (1toAltEia, or, in its
monarchical form, l}acnMia). In a sense, the oikos was contained
within the genos and formed its basic structure, the nuclear family
of father, mother and unmarried children. But the household of
the honourable man contained a relationship of another dimen-
sion: the relationship of master to slave and servant. Discussions
of daily life, social structures and demographic patterns in Byzan-
tium all-too often forget the existence of the slaves (aouAOl) who
formed part of even quite modest households, and the legally free
dependents, the servants (U1tllPEtat, OiKEtat), familiars (OiKEtOl),
or simply 'men' (av9pW1tOl), who with varying ties of obligation,
swelled the retinue (A,aoc;) that was vital to the important man's
public persona.
The distinction between servant and served was fundamental
to Byzantine conceptions of honour. In principle, one role was
thoroughly honourable and the other thoroughly dishonourable.
However, the principle was complicated by its projection upwards

3. Cf. R. Browning, 'The 'Low Level' Saint's Life in the Early Byzantine World'
in The Byzantine Saint, ed. S. Hackel (London 1981) 117-27, esp. 126-7.

375
PAUL MAGDALINO

and outwards into the constitution of state and church. There

can be no doubt that the Byzantines believed in and experienced
the state as an impersonal, public affair - a politeia. The fiscal
system, the standing army and the law courts formed an institu-
tional ensemble which was mote than the sum of the personal
ties that held it together. The dignities and salaries bestowed by
the emperor were perceived as rights which men deserved by their
merits and forfeited only by committing.capital crimes. The hier-
archy of dignitaries could flatter themselves that they were not
imperial but public servants, not courtiers but senators, whose
individual and collective honour was guaranteed by a traditional
order ('t'a~lC;) and established procedure (Ka't'ao't'aOlC;) to which
the emperor was as committed as they were·. The structure of the
hierarchy, and the elaborate ceremonial by which it was made
manifest were, in effect, the Byzantine constitution. 4
And yet there is a sense in which the public, 'constitutional'
nature of the state and the honour of the free citizen existed only
in the minds and words of intellectuals steeped in ancient history
and fluent in the language of ancient political thought. Not only
the reality but also the formality of imperial power tended to cast
the emperor in the patrimonial role of a head of a universal
household. As an eleventh-century legal text puts it, 'before the
power of the emperor the paterfamilias is nothing, but the subor-
dinates of the latter are his fellow subordinates of the former'. S
Another passage in the same source shows that no effective distinc-
tion was made between the judicial and the executive authority
of imperial officials, or between the functionaries whom they com-
manded and the litigants whom they judged. 6 1t is equally dif-
ficult to distinguish between the public office and the private
patronage which they exercised. To illustrate this point, we need
look no further than the reign of Justinian and the roles of John
4. For Byzantine attitudes, see the preface of the De cerimoniis, ed. J. Reiske
(CSHB), I 3-5 led. A. Vogt (paris 1935, rpr. 1967), I 1-2]; the Kletorologion of
Philotheos, ed. N. Oikonomides, Les listes de preseance byzantines des IXe et Xe
siec/es (Paris 1972) 83; see also the sources cited in n.32 below. For modern com-
ment, cf. P. Speck, Kaiser Konstantin VI (Munich 1978) I 87ff; Kazhdan and
Constable, People and Power, 146-7.
5. Peira 66.2 (Zepos, IV 244).
6. Peira 51.27 (Zepos, IV 218).

376
HONOUR AMONG ROMAIOI

the Cappadocian and Belisarios in executing the imperial pro-

gramme of that most statist of emperors. 7
To a large extent, therefore, the relationships which set the tone
of Byzantine public morality were those of the extended
household, in which the honour of perfect freedom was com-
plemented and modified by the honour of loyal service. The point
at which servant status became honourable was never properly
defined, and the very idea of honourable servitude was fundamen-
tally at odds, not only with the values enshrined in Roman Law
and classical literature, but also with the irrepressible individualism
which has rightly been identified as one of the distinguishing
characteristics of Byzantine society. Nevertheless, the idea can-
not be dismissed as nothing more than an artificial sweetener for
the bitter pill of imperial absolutism. It was consecrated by its
application to man's relationship with God and the saints, and,
in this, formed an integral part of religious as welJ as political
orthodoxy. Byzantines drew a subtle and profound analogy be-
tween the imperial court and the court of heaven, constantly
enriching their conception of each one with their perception of
the other. This was, in fact, much more than an analogy; it was
an attempt to integrate, or at least harmonise, heaven and earth
by interpreting the earthly order as an icon of the divine order,
as an image of a divine prototype from which the copy was
separated only by its material and technical shortcomings. The
Communion of Saints was conceived in terms of imperial taxis
and katastasis. Angels were likened to eunuchs, and commitment
to the monastic life was likened to enrolment in imperial service. 8
Popular and elite culture were at one in idealising the patronage
and intercession of the saints in terms of the extended household
relationships which permeated the imperial administration. A
saint's church was his oikos, where he was master and dispensed
personal favours face to face by virtue of the trust in which his

7. John the Lydian, De magistratibus. II 20-1, III 62; Procopius, Wars. III 11.7,
17.1-2, IV 4.8, VII 1.20-1.
8. E.g. La Vie ancienne de S. Symeon Stylite Ie Jeune. ed. P. Van den Ven (Brussels
1962-70) I 193-4, " 219-209; N.F. Marcos, Los Thaumata de Sofronio (Madrid 1975)
264-9; P. Lemer1e, Les plus anciens recueils des miracles de S. Demetrius (Paris 1979)
I 96ff., I 59ff. , 234ff.

377
PAUL MAGDALINO

sovereign held him. Hagiographical accounts of the personal con-

frontations between saints and those who came to them for
miraculous benefits, or upon whom they inflicted miraculous
punishments, evoke the audiences held by officials in the recep-
tion halls of their residences, as well as the atmosphere in the
Late Roman law court. The episodes are highly charged with the
expectations of honour and shame with which patron and client
confronted each other: the saint expects, and receives, public
honour from the supplicants and servants who attend on him in
his oikos, where he presides from his tomb or his column; the
supplicant humbles himself by making his request, and can ex-
pect to be publicly shamed if the heavenly patron's discerning
eye catches him in an act of bad faith.9
Of course, the idea of the heavenly court did also imply a con-
trast with earthly practice, and involved the inversion of worldly
values referred to earlier. To enlist in divine service was to re-
nounce imperial service and its rewards; to honour God and his
saints was to abase oneself, and not to stand on the merits and
the dignity which meant honour in this life. None were more con-
scious than the saints themselves that the committed Christian
was faced with a stark choice between the service of God and
the service of Mammon. Yet on the whole the assimilation of
the two service systems helped to reconcile Christian Orthodoxy
with the continuing existence of a conventional culture of honour
and shame. It allowed most men of the world, and also most holy
men, to venerate the paradox of the Holy Fool without having
to enact it.
The foregoing considerations may serve as a frame for the por-
trait of honour and shame in Byzantine society which can be
drawn from the evidence of two near-contemporary texts: the
Grottaferrata version of the verse romance Digenes Akrites,9

9. Ed and tr. J. Mavrogordato (Oxford 1956). For a recent analysis, with a full
bibliography of secondary literature, see C. Galatariotou, 'Structural Oppositions
in the Grottaferrata Digenes Akritas', BMGS'II (1987) 29·68. My own conclusions
are generally in accord with this study, although 1 place greater emphasis on the for-
mal importance of religious and patriotic values, and on the equation of honour with
'face'.

378
 HONOUR AMONG ROMAIOJ

and the Precepts and Tales of Kekaumenos. lo These two texts

recommend themselves for several reasons. First, both are directly
concerned with the winning and preservation of honour in this
world. Second, they complement each other neatly, the one be-
ing an idealised, romanticised tale of a hero, and the other a sober,
pragmatic discourse full of cautionary maxims and examples.
Third, neither is obviously a piece of official propaganda. II
Fourth, in neither is the perspective distorted by religious or
classical conventions, and although both owe a certain debt to
the 'inner' as weII as the 'outer' learning, this is no greater than
one would expect of the normaIIy literate Byzantine of honourable
status. 12 FinalJy, both date from the late eleventh or the very
early twelfth century - the period which marks the definitive
transition from the 'imperial centuries' of Byzantium to the

10. Pending publication of the long awaited edition, translation and commentary
by Charlotte Roueche, see G.G. Litavrin, Sovety i rasskazy Kekavmena (Moscow
1972). The most accessible edition, however, is still that by B. Wassiliewsky and V.
J ernstedt, Cecaumeni Strategicon (St. Petersburg 1896, repr. Amsterdam 1965). See
also P. Lemerle, Protegomenes a une edition critique et commentee des 'Conseils et
recits' de Kekaumenos (Academic royale de Belgique, Classe des lettres, Memoires,
54) (Brussels 1960); H.-G. Beck, Vademecum des byzantinischen Aristokraten (Graz
1956).
11. That is, they were not commissioned by, or written to gain favour with, the
political establishment. Although, as will be shown below, they reOect the official
Ideology of 'king and country', they also take a less enthusiastic view of imperial
centralisation than is to found in most Byzantine literalUre: see I. Sevcenko,
'Constantinople viewed from the Eastern Provinces in the Middle Byzantine Period',
Harvard Ukrainian Studies 3-4 (1979-80) [repr. in idem, Ideology, Leiters and Culture
in the Byzantine World (London 1982)/ 726ff., 732ff.
12. Their stylistic level is below that of the educated KOIvij of this period, and the
works with which they show familiarity - the Fathers, saints' lives, chronicles and
(in the case of Digenes) the Leukippe and Kleitophon of Achilles Tatius - could
all be found in small, and not particularly highbrow, private libraries. See P. Lemerle,
'La testament d 'Eustathios Boilas (avril 1059)" in idem, Cinq etudes sur Ie Xle siec/e
byzantin (Paris 1977) 24-5; Sp. Vryonis, 'The Will of a Provincial Magnate, Eustathios
Boilas (1059)" DOP 11 (\ 957) 263-77, [repr. in idem, Byzantium: its Internal History
and Relations with the Muslim World (London 1971)/ P. Gautier, 'La Diataxis de
Michel Attaliate', REB 39 (1981) 92-6,122-6. Kekaumeilos neither pretends to, nor
assumes in others, great learning or eloquence, but he shows that reading was a stan-
dard leisure activity among military men and provincial aristocrats: ed. Wassiliewsky
and Jernstedt, 19,47,60,64 (ed. Litavrin, 154-6,212,240,248). Cf. R. Browning,
'Literacy in the Byzantine World', BMGS 4 (1978) 39-54; C. Roueche, 'Byzantine
Readers and Writers: Story Telling in the Eleventh Century', The Greek Novel A.D.
/-/985 ed. R. Beaton (London 1987) 123-33, esp. I 24ff.

379
PAUL MAGDALlNO

nascent Neohellenism of the Comnenian and Palaiologan era.

They were products of the same aristocratic milieu as Alexios I
Komnenos (1081-1118), the emperor who revived the medieval
Greek empire in the form in which it continued to exist until 1453
and lived on in folk memory. Yet they predate the westernisa-
tion of Byzantine society which began in Alexios' reign and left
no aspect of Byzantine culture untouched.
I n examining each text, we shaIl be concerned to identify:
(a) The qualities in which honour consists.
(b) The means by which honour is manifested, acquired, increased,
preserved, or lost.
(c) The social groups on whose behalf and at whose expense
honour is defended or pursued.
(d) The social group in whose eyes honour is sought.
Digenes Akrites is a tale of love and war, sex and violence.
As we should expect, honour in the heroic world it portrays is
predominantly male honour, and consists very largely in the glory
(cS6~a) won by the display of 'manliness' (uvopda). All the male
characters, as well as one female, the Amazon Maximo, possess
andreia to a greater or lesser degree, and the hero, Basil Digenes
Akrites, possesses it in superabundance. It governs most of his
actions and is manifest in the extraordinary feats of heroism
(uvopayaeiat) which he performs. Having discovered his talent
at the age of twelve, when he kills two bears anti a lion, he misses
no opportunity to take on overwhelming odds. He abducts the
jealously guarded daughter of a powerful general (cHpan1Y6<;),
taunts him into giving chase, and then, single-handed, defeats
him and his whole entourage. When the emperor visits the region
and invites Digenes to meet him, D. dictates the conditions, and
the emperor knows better than to object. Pressing home his ad-
vantage in the relationship, D. gives two demonstrations of his
superhuman strength, taming a wild horse, and killing a lion that
has given the emperor a fright. Appointed by the emperor to be
lord of the frontier marches, D. leads a nomadic existence on
the borders with his bride and a small retinue. Here he has to
protect his wife from the advances of a serpent, a lion, and a
group of 'drovers' (U1tEA.Il"Cat), who prove particularly persistent
because they are famous heroes in their own right, with a high

380
 HONOUR AMONG ROMAIOI

reputation for andreia to uphold. Needless to say, however, D.

vanquishes them all, and goes on to vanquish their kinswoman
Maximo, whom they incite against him, along with her whole
army.
Just as warriors in this society gain honour by the conspicuous
display of andreia, so they lose it when they fail to display the
andreia appropriate to the situation. Such failure lies not so much
in the fact of defeat as in the intention to cheat, or, worse, in
the exposure of that intention. Shame (aloxuvll) is a subjective
state of mind which prompts a fighter to 'do the decent thing',
or, failing that, drives him to indecent desperation in the effort
to win. The soldiers sent by the strategos after the abduction of
his daughter would rather die than live with the shame of having
been defeated by a single adversary (IV. 643-4). The three ape/alai
are initially ashamed to fight against one man, and although, as
D. points out, they conveniently forget their shame in the heat
of the battle, it returns to infuriate them all the more when they
take stock of their humiliation (VI. 201,212,231, 314ff). D.,
by contrast, can fight against Maximo without feeling ashamed,
because although she is a woman she is famous for her andreia
(VI. 749-52).
A superficial reading of the poem's heroic episodes can give
the impression that the honour which D. gains through his ex-
ceptional andreia is an amoral, asocial quality which the hero
possesses by and for himself. Certainly D., like the hero in a
Western, and like the typical Byzantine holy man (whose
asceticism was commonly portrayed as the exercise of andreia
against demons and bodily passions),13 is cast as a loner living
on the margin of society, neither serving, nor, as he tells the
defeated ape/alai, wanting to be served (VI. 288f). His sense of
honour sometimes seems to be pure inflated egoism - the egoism
of the 'macho' male who overreacts to any slight, however
unintentional, and insists on justifying his actions even when he
knows himself to be in the wrong. Thus D. warns the emperor
that he is likely to kill any imperial soldiers who do not watch

13. On hagiographical echoes in Digenes, cf. E. Trapp, 'Hagiographische Elemente

im Digemis-Epos', AB 94 (1976) 275-87; Kazhdan and Epstein, 117.

381
PAUL MMiOALINO

what they say in his presence (IV. 996-1000), and after commit-
ting adultery with Maximo, he takes out his guilt-feelings by slay-
ing the wicked temptress (IV. 834fO.
However, it is helpful to recall that in traditional modern Greek
society, egoism (tyWlOJ.!OC;) is not necessarily a denial of social
responsibility.14 On closer inspection, the ideal of andreia in
Digenes can be seen to conform to the expectations of social
groups with which the hero identifies. It is, for one thing, intimate-
ly connected with the Christian Orthodoxy of 'Romania', the
Byzantine Empire. whose cause D. champions and advances as
'lord of the marches' (VII. I 99fO. The hero's superhuman strength
is a gift from God (1. 4, III. 340), which he exercises with the
continual help of God and the military saints (I. 17ff, VI. 691-701).
Andreia without divine assistance is not enough. This is made
clear in the story of D. 's father, the Saracen Amir, who, although
a paragon of andreia, is defeated in single combat by the youngest
of· three Christian brothers, who pray to God before the contest
and praise Him afterwards (1..142ff, 200ff). Much is made of
the Amir's conversion to Christianity and of the way in which
he persuades not only his mother but also his entire kin to follow
his example (I. 306, II. 41, III. 160fO.
In other respects, of course,-Christianity, like the Empire, is
a very shadowy presence. Bishops and monks are as conspicuous
by their absence as are judges and tax collectors, while military
governors (strategoi) appear as figures of largely independent
means and status. In romanticising the story of the Amir, and
in making D. the son of a mixed marriage, the first three books
- the 'Song of the Amir' - show more than a sneaking fascina-
tion with the world of Islam.ls The Orthodox prejudice which

14. Du Boulay, Portrait ofa Greek Mountain Vii/age, 75ff.

15. A. Pertusi, 'Tra storia e leggenda; akritai e ghazi sulla frontiera orientale di
Bizanzio', XIVe Congres International des Etudes Byzantines. Rapports II (Bucharest
1971) 27-71. For the idea that the first three books, which most faithfully reflect the
administrative realities of the frontier in the tenth and early eleventh centuries, form
the original core of the work, see H.-G. Beck, Geschichte der byzantinischen
Volksliteratur (Munich 1971) 71 ff; N. Oikonomides, 'L' "epopee" de Digenes et la
frontiere orientale de Byzance aux Xe et Xle siecles·. TM 7 (1979) 377ff. From the
perspective of the present study. howev.er. the common features of the two parts are
more striking than the differences.

382
 HONOUR AMONG ROMAIOI

the poem expresses is therefore all the more remarkable. The Amir
is a 'good guy' not only because he converts, but also because
he is 'not black as Aeth'iops are, but fair and lovely' (I. 32), and,
it so happens, is not of Arab stock (I. 283). This presumably helps
his future mother-in-la~ to get over her initial apprehension on
learning that her daughter is to marry a Saracen (II. 23-5). Similar
prejudice against foreigners and infidels comes out in the descrip-
tion of Maximo's henchman Melimitzes (VI. 499), and in the
general description of D.'s achievements against the Arabs, who,
as descendants of the slave-girl Hagar, are contrasted with the
'honourable and well-born Romaioi' (VII. 208-10). It could be
that these passages were added to give a semblance of propriety
to a work whose sensationalism and exoticism might otherwise
have rendered it suspect. But it is equally likely that they repre-
sent a sense of national and religious solidarity which transcends
the poem's apparent, and perhaps studied, indifference to the
structures of Church and State. Either way, the declaration of
allegiance to Orthodoxy is significant, as is the way in which the
mirror-image of this allegiance is projected on to the Muslims:
the Amir's mother upbraids him for having 'denied kindred, faith
(rriotH;) and country (patris). You have become a reproach in
all Syria, while we have become abominated by every man, as
deniers of the faith and transgressors of the law' (II. 54-7 and
cf. II. 9, 226).
The association of 'nation' and 'faith' with kinship alerts us,
however, to the fact that while the poem makes only summary,
if honourable, mention of the exploits which D., performs on
behalf of the basileia of Orthodox Romaioi, the exploits which
it relates in full are all concerned with the honour of his genos
and oikos. D. sets out on his first adventure accompanied by his
father and his maternal uncle, and when his father advises him
to wait until he is older, he replies, 'I want fame now, to make
my genos illustrious. I assure you, my benefactor, that you will
have in me a brave great servant, to help and fight with you in
all your battles' (IV. 96-9). He goes on to make good this promise,
for by abducting the strategos' daughter, he not only demonstrates
individual prowess, but brings off a marriage alliance which his
father has proposed several times without success (IV. 308-9),

383
PAUL MMiDAl.INO

and thus vindicates his family's honour. Even the tours deforce
with which he impresses the emperor are not entirely gratuitous,
for they secure his appointment to a role as 'marcher lord', which
allows him to leave the parental home and set up his own indepen-
dent household. As we have seen, the remaining exploits which
the poem describes are all performed at the expense of adver-
saries who seek to violate this new family unit by laying hands
on the hero's bride.
The decisive importance of oikos and genos in the social world
of Digenes becomes even clearer when we consider that andreia
is only one of the constituent elements in the hero's honour. While
it is highlighted as the quality which makes him unique, it is often
associated with attributes which are equally integral to his role,
even though they put him on a par with other people, including
his wife. Paramount among these associated attributes are good
birth and wealth. D. 's father is introduced as 'one of the well
born, exceedingly rich, having the utmost share of prudence and
bravery' (I. 31-32). He makes war on the Byzantines 'with wealth
and load of bravery exalted' (I. 44). D. himself 'sprang from a
race of well born Romans, and was illustrious for his braveries'
(IV. 63-4). He is frequently referred to as 'the well born youth'.
The ancestry and wealth of his bride are fabled beyond descrip-
tion (1 V. 264-6), and although he wins her by his andreia, it is
important for the formalisation of their relationship that he is
her social equal, as she acknowledges when they meet: 'You are
one of the well born and the very wealth, and a kinsman of ours
through the Doukai. But my father the strategos has look-outs
posted for you, for he has heard all about your heroic deeds' (IV.
324-7). When, after the fight, D. asks for her father's blessing,
he points out that he is neither ignoble nor cowardly (IV. 679),
and when the strategos repairs his own honour by offering a
generous dowry (IV. 704ff, esp. 724-5), the hero goes one bet-
ter, declaring that he has no need of it and will happily donate
it to his brothers-in-law (IV. 744fO.
Eugeneia may imply the prestige of an illustrious pedigree, but
what it actually denotes throughout the poem is something more
tangible: the solidarity of a powerful kin group. A character's
genos is evoked less in terms of ascending genealogy than in terms

384
 HONOUR AMONG ROMAIOI

of a body of close relatives (syngeneis) , among whom, besides

parents and siblings, uncles and grandparents are specified. Since
no lineage is taken back more than two generations, we can only
guess that syngeneia extends to second cousins. But whatever the
effective limits of the genos, cohesion within them is considerable.
When the Amir's bride and brothers-in-law take him home to
Cappadocia, a host of syngeneis and others turns out to greet
them (II. 30). When the Amir returns to Syria in response to his
mother's appeal, they all turn out to see him off (II. 297), and
at the other end, his mother welcomes him at the head of an
equivalent crowd of kinsfolk (III. 116-8). Just as his apostasy
has brought disgrace on them as well her, so they follow suit when
he persuades her to convert to Christianity and follow him back
to Cappadocia (III. 228ff) to a warm welcome by his wife's
relatives (III. 316-8, 325-6). Here he allots them quarters in the
house which evidently he has received from and, to some extent,
shares with his in-laws (III. 329-38). Similar patterns of syngeneia
can be seen in the household of the strategos whose daughter D.
abducts. The girl sees her elopement as a denial of father, brothers,
and syngeneis (IV. 512-3), and the couple assume that her relatives
will join her father in chasing after them (IV. 337,453,462,468).
Such is the emphasis on eugeneia and syngeneia, especially in
Books I-IV, that one is strongly tempted to conclude that of the
social groups with which this study is concerned, the genos is the
one for whose sake and in whose eyes honour is upheld. It is a
much more concrete reality than either the shadowy ideal of the
basileia, or the notion of patris, which, insofar as it applies, serves
to locate characters in terms of the provincial area, rather than
the local community, from which they come (I. 264-5, III. 106,
V. 18, VII. 2, 107fO. Friends, neighbours, and retainers all ap-
pear in association with relatives as people to be taken into ac-
count, but never by themselves (I. 297; V. 142, 148, 230; VIII.
7ff, 33). In some ways, the genos seems to overshadow the oikos.
Slaves and servants are all present as part of the decor, but never
emerge as individuals whose roles and relationships are worth
exploring. As we have seen, the oikoi in which D. and his bride
grow up house extended, not nuclear, kinship groups.16

385
PAUL MAGDAlINO

But before we conclude that genos, as opposed to oikos, is the

definitive aspect of family structure in the world of Digenes, we
should look more closely at the role of the nuclear family. Firstly,
we should note that the family relationships most clearly
highlighted are those between parents and children, brothers and
sisters, husbands and wives. We should also note, in this con-
text, that while syngeneia carries obligations, it by no means in-
volves complete trust. The fact that D. 's mother and the slralegos
are related does not facilitate marriage negotiations between the
two families; the ape/alai turn for help to their kinswoman,
Maximo, but conceal from her their real motive in asking (VI.
375ff,418ff).
Secondly, we should recall that when D. marries, he moves away
from the embrace of the extended family. The house he even-
tually builds by the Euphrates is intended for himself and his wife
and the children for which they hope in vain; his mother moves
there only after his father has died (VII. 152ff).
Thirdly, we must not forget to examine the third component
in the heroic ideal of honour, the element of wealth. The display
of~ealth is one of the poem's favourite themes, which receives
as least as much coverage as all the manifestations of andreia
and eugeneia, and it is, on the whole, presented as a function
of the oikos. We shall have more to say about the elaborate
descriptions of the fine clothes, richly caparisoned horses, and
sumptuous buildings. Here it is appropriate to point out that these
descriptions are especially concentrated in the section (Book VI)
which deals with the hero's courtship and marriage, and that they
are particularly eloquent on the subject of the things that con-
stitute an oikos: its physical structure and the slaves who run it
- and whose absence from the poem's human relationships is
thus explained by their importance as property.
Weare again brought back to the oikos when we return to the
question of the ways in which honour is lost. We have seen that
the honour upheld by andreia is compromised by failure of an-
dreia. But the honour upheld by good birth and wealth is not
16. For historical examples of aristocratic households containing more than one con-
jugal unit, see my 'Byzantine Aristocratic Oikos'. 97, 98 and nn. 56, 58, 60; Lemerle,
Cinq etudes. 23.

386
 HONOUR AMONG ROMAIOI

correspondingly compromised by the failure of these commodities

- at least for the characters in Digenes, for whom the question
of low birth or poverly /lever arises. Excepl where andreia is at
stake, they are in danger of losing their honour only when they
fail to fulfil their roles as husbands, wives, fathers, mothers, sons
and daughters. A husband is supposed to protect his wife and
remain faithful to her, while she is supposed to follow him
wherever he goes, even if this means betraying her own family's
trust (IV. 113-33). A father's role is to educate his son in andreia
(IV. 70ff), and to keep his daughter secluded, or, if he has to
part with her, to give her a generous dowry. Mothers are there
to shame their sons into doing their family duty; a son does not
dare to incur a mother's curse (I. 70ff, 110-12, 138,259-60; II.
98, 228ff). He will obey and serve his parents, venerate their
memory, and hunt down any man who wrongs his sister - in-
cluding her husband. She, if unmarried, must above all be modest,
neither speaking nor showing herself to strangers.
In the main, the characters in the poem acquit themselves well
in all these roles, for although D. fails miserably as a faithful
husband, his infidelities cause him no loss of face. But there are
two episodes where the protagonists, for all their good intentions,
find themselves in shameful positions. One is the point when the
Amir, having decided to obey his mother's command to return
to Syria, and having persuaded his bride to come with him, is
found out by her brothers. They berate him for being up to no
good, and he, 'unable to reply, fell fully silent, being replete with
shame and fear and grief: shamed by detection, frightened as a
stranger, grieved at the thought of parting from his love' (II.
171-4). The other episode in question is the scene in D.'s abduc-
tion of his bride when, in reply to his challenge to declare her
intentions, she insists that she has already done so by the mere
fact of talking and showing herself to him. 'Unseen by other eyes
I kept myself - for save my kinsmen and my familiars, no one
has ever seen my countenance - observing the strict order becom-
ing to (t<i~lV tilv 1tpt1tououv) a maid. Now I have gone too far,
transgressed the bounds, and become shameless (avmoT]s) all for
love of you' (IV. 497-501). D. has enslaved her mind and made
her shameless (avuiaxuvtos) - for him she renounces relatives,

387
PAUL MAGDALlNO

parents and brothers (l V. 509ft). As she continues to chide hersel f

for her 'great shamelessness' (IV. 522), the poet draws a moral
for both sexes: however great a regard for taxis one may have,
Love (Eros) upsets it; 'that's why the lover has no sense of
decorum (El)'ra~i<l), no shame before his kin, no fear of what the
neighbours say, but is all shameless and a slave of Love' (IV.
528-30; cf. I. 285).
Shame, then, besets, or should beset, the honourable person
when, in a conflict of loyalties, he or she yields to one side and
is found out by the other, whose taxis has been violated. Honour
is accordingly satisfied when taxis is restored. In the case of the
Amir, it is restored - or found never to have been infringed -
when his wife points out to her brothers that he is acting by the
same code of filial piety, the same fear of a mother's curse, that
had made them come to her rescue (II. 222ff). In the case of the
abduction, taxis is reintroduced when the ritual of marriage con-
tracts and wedding preparations set in motion; even so, there is
a tense moment of transition when D. feels embarrassed to visit
the home which he has so recently violated, while the Girl is
similarly reluctant to go to his house without her attendants and
accoutrements (IV. 736ff, 808ff).
Although there is a basic difference between male honour,
characterised by andreia, and female honour, characterised by
modesty, both are invested in the taxis of the oikos. The remark
which the Girl utters as D. carries her off sums up the social values
of their world: 'for order true is called nobility' (IV. 565: it 'ta~t~
yap it uAn9ilc; EUYEVEt<l K<lAEt'tat). This principle underlies not
only the rites of marriage but also those of death. In describing
the palace which D. builds by the Euphrates,17 the poet con-
cludes with a mention of the church, dedicated to St. Theodore,
which D. erects in the central courtyard, and in which he buries
'his most honourable (1t<lvtV'tlIlOV) father, bringing the body from
Cappadocia, duly with shining stones the tomb adorning' (VII.
106-8). There follows a lengthy account of how D., on learning
that his father was on his death bed, hastened to Cappadocia,

17. On this. see L.-A. Hunt. 'Comnenian Aristocratic Palace Decoration: Descrip-
tions and Islamic Connections', The Byzantine Aristocracy, ed. Angold, 342-5.

388
 HONOUR AMONG ROMAIDI

where he duly uttered a ritual lament, honourably (EV'tlIlWC;) per-

formed the funeral rites, and then returned with his mother and
the body to inter it honourably in its final resting place at the
heart of his own oikos.
I t is not fortuitous that both the account of D. 's marriage and
the account of his father's death are closely linked to elaborate
descriptions of conspicuous wealth. In these, luxury and finery
are emphasised not just because they have snob appeal, but also
because they have a ceremonial function that is integral to the
narrative. The same is true of other such passages. The golden
throne on which the Amir sits, and the host of attendants who
surround him, when he receives his future brothers-in-law, evoke
the infidel status and power which will shortly be laid low and
which he will eventually renounce (I. 94-6). The fine Byzantine
clothes into which he changes on his final return from Syria sym-
bolise the triumphant conclusion of his conversion (III. 247fO.
The change of clothes which D. makes after his first manly ex-
ploit marks another rite de passage, that from childhood to
manhood, besides giving his arrival at his next destination the
character of a triumphal adventus (IV. 219fO.18
What makes the descriptions of wealth in Digenes particularly
interesting is the fact that, with the possible exception of the
description of the oikos, they correspond to written genres that
we would be inclined to classify as documentary rather than
literary.19 Through their enumeration of precious objects in
18. This is just one of the ways in which O. seems to take on imperial attributes.
On adventus ceremonial, cf. S. MacCormack, 'Change and Continuity in Late Anti-
quity, the Ceremonial of Adventus', Historia 21 (1972) 721-52; eadem, Art and
Ceremony in Late Antiquity (Berkeley/Los Angeles 1981) 17-89.
In all the other versions of the poem, O. 's next destination is his own home. In
the Grottferrata text, however, he rides straight on to his meeting with the daughter
of the strategos which will lead to her abduction. This clearly unsatisfactory in view
of the fact that O. is still only twelve, and his exploits are supposed to be well known
to the Girl's family (IV. 321ff). On the complex question of the textual transmission
at this point, see S. MacAlister, 'Oigenis Akritas: the First Scene with the Apelatai',
B 54 (1984) 568-9, for the plausible suggestion of a book division at an earlier stage
of the Grottaferrata version. However, this version's present narrative sequence may
well be intentional. Among other things, it has the narrative function of involving
O.'s father, and explaining the hero's ignorance of the marriage proposal that has
been made on his behalf.
19. The description of the oikos seems to belong to the rhetorical genre of ekphrasis,

389
PAUL MAGDALINO

technical, 'low-style' vocabulary, they echo, firstly, the Book of

Ceremonies which the emperor Constantine Porphyrogenitus
(9\3-959) compiled in order to document the means 'by which
the imperial power is conducted in order (taxis) and harmony'. 20
Secondly, they are reminiscent of the wills and inventories in which
private individuals, churches and monasteries lovingly catalogued
their valuables. 21 The combination of parallels is suggestive. It
suggests, on the one hand, that Digenes is appropriating the taxis
of the imperial court to the domain of the private oikos. On the
other hand, it shows that this was itself a model - if not the
model - of eutaxia. Doubtless there is a strong element of wish'ful
thinking to the fact that in Digenes, as in the patriography of
Constantinople, 'Ie grand absent, c'est l'empire'22 - that D. can
treat the emperor as an equal, that he can refuse to serve or be
served outside his own home, that Love is the only ruler capable
of taxing the taxis of the oikos and t~aring a person away from
the family. It is but a short step from this world to the idyllic,
neo-c1assical world of Eros the King. 23 D. 's oikos is as mythical
as his andreia. Yet it clearly represents an ideal to which Byzan-
tines aspired, and which, in the 'Wild East' of tenth and eleventh-
century Anatolia, had some claim to credibility. The problem was

but ir wuld have been inspired by inventories of buildings of the kind which were
evidently made for purposes of transfer or valuation: see, e.g .• The Byzantine
Aristocracy. cd. Angold. 254-66. In any case. we should be careful not to impose
too rigid a distinction between literary and documentary descriptions, both of which
served ritual purposes; for ekphrasis. see R. Macrides and P. Magdalino, 'The Ar·
chitecture of Ekphrasis: Construction and Context of Paul the Silentiary's Poem on
H agia Sophia'. BMGS 12 (1988) 47-82.
20. Ed. Reiske, 5 (Vogt. 12); see also, in particular, the descriptions of the trium-
phal processions of Theophilos and Basill (ed. Reiske, 498-508), and of state recep-
tions given by Constantine VII and Romanos I (ibid 570-98).
21. Cf. M. Hendy. Studies in the Byzantine Monetary Economy. c. 3(}()-1450
(Cambridge 1985) 20Mf., esp. 217.
22. G. Dagron. Constantinople imaginaire. Etudes sur Ie recueil des Patria (Paris
1984) 19.
23. See the twelfth-century (?) romance Hysmine and Hysminias of Eusthathios
Makrembo1ites, ed. R. Hercher, Erotici scriptores graeci (Leipzig 1859) 161-286; cf.
C. Cupane, '"EpwC; PUOlA.&UC;. La figura di Eros nel romanzo bizantino d'amore',
A tti dell'Accademia di Scienze, Lettere e Arti di Palermo. 4th series 33 (19734) 244·97.

390
 HONOUR AMONG ROMAIOI

how to achieve it if one did not live in that environment or possess

superhuman strength. This is where Kekaumenos comes inY
K. assumes, no less than the author of Digenes, that the
honourable man will possess andreia, wealth and good birth, and
will seek to increase his honour by pursuing the glory they bring.
Andreia is essential to the military profession which the author
clearly has in his blood and regards as the most honourable public
career open to men of his status. A general's aim is to win vic-
tories (9/134) for which he will receive rewards and honours from
the emperor (25/166). He must strive to surpass all 'in counsel,
bravery, prudence, wisdom and virtue'; he must not fear death
'on behalf of the fatherland (patris) and the emperor', but should
rather be afraid of leading a shameful and blameworthy life
(16/148). He is to value andreia in others. (16/148). The display
of andreia is just as important as the reality and can even enhance
this (21/158). When on campaign, the general is to show himself
'lofty and imposing to all who see him' (20/156). In his spare
time, he should improve his military expertise by wide reading,
'that all may admire your bravery, prudent counsel, knowledge
and eloquence' (19/155-6). A subordinate officer must similarly
strive to excel, 'that he may be renowned' (ovol.woT6~) (35/188).
Even the gentleman living privately on his estate may have to play
a military role, either in quelling a revolt, or as the unwilling leader
of one. 'If enlisted (otpatElJOd<;), be brave in battle, even if it
means you will die. Remember this is why you enrolled, and that
no-one is immortal' (541226).
K. takes it for granted that if the people for whom he is writing
do not hold remunerative office, they will be able to fall back
on their inherited, landed wealth (36/188), and, if they manage
it sensibly, will not have to borrow (37/190), take risks (39f1194ff)
or stoop to base profiteering (471212,511220,521224,591236-8).
They will not be miserly and their beasts of burden will be well
fed (491216). They will take a competitive pride in having a well-
stocked, well-staffed, and well-ordered household (431202,
471210,501218,511220,561230). It goes without saying that they

24. References are to the page numbers of the Wassiliewsky-Jernstedt and Litavrin
editions respectively (see above, n.IO).

391
PAUL MA(jLJALlNO

are not ill born: the thought that his addressee might be one of
'the lowly' (1"WV KCl1W) is one that K. barely contemplates and
quickly dismisses (3/120,5/126). It is true that he equates high
social staWs less with eugeneia than with rank and office and with
the epithets 'glorious' (£v80~os) and 'splendid' (m:ptcpavtls). He
insists that men are born equal, that the man of low origin may
have something to say and may even become emperor (11-21140,
98-9/286). In this, K. clearly stands apart from the new cult of
nobility which was coming into vogue, and into force, at the time
that he was writing, and identifies with a more traditional Byzan-
tine attitude. 25 Yet whether that attitude was as egalitarian or
meritocratic in practice as it sounds in theory is highly debatable.
I f it fails to stress good birth, this may be because it took good
birth for granted. There is reason to see the new insistence on
nobility as a reaction against the unprecedented social mobility
of the mid eleventh century, which had upset a previously stable
and largely hereditary transmission of rank and office. 26 This
mobility mainly affected the court and the capital. K. registers
it in the short sections where he deals with service in the
bureaucracy and expresses his contempt for 'city-slickers'
(7tOAl1"tKOi) Off/118ff, 8-9/132, 20/156). Otherwise, he gives the
impression that it hardly touched the world of provincial 'gen-
try' (apxovu;s) in which he seems to have thought that his ad-
dressee would be most at homeY In this world, people faced
social unrest, but not social competition, from their inferiors.
They and their paers knew each other too well to need to adver-
tise their ancestry: what mattered were the visible assets of doxa
and periphaneia that distinguished one well born man from
another.
The low profile given to eugeneia does, nevertheless, reveal a
different conception of the pursuit of honour from that which
we encountered in Digenes. D. pursues honour in a power vacuum,
but K. offers the benefit of his experience to people who operate

25. Kazhdan and Epstein. Change in Byzantine Culture, 104.

26. Michael Pselios, ed. K. Sathas, MMU/WVIKr, BIPAIO(},jKI/ (Venice/Paris 1872-94)
IV 430-1, V 108-9. On the background, see Hendy, Studies, 507ff.
27. See M. Angold, •Archons and Dynasts: Local Aristocracies and the Cities of
t he Later Byzantine Empire', The Byzantine Aristocracy, ed. Angold, 236-53.

392
 HONOUR AMONG ROMAIOI

within a towering power and patronage structure. Their level

within that structure is decisive. In the social classification of mid-
dle Byzantine legislation, they are, above all, the 'powerful'
(O\)Vu'toi) as opposed to the 'poor' (7ttVll'tf:~).28 Af]dreia, wealth
and eugeneia are all important qualifications for 'powerful' status,
but they are subsumed by the exercise of authority which this en-
tails, whether the dynatos has a formal appointment, or whether,
as K. more than once envisages, he is simply a local archon with
strong local influence (40fflI98ff, 56ff1232ff, 64ffl248ff).
'Powerful' status can be enhanced through the exercise of
authority to enrich oneself, to win victories, and'generally to assert
oneself. But it can just as easily be lost by the same means; in-
deed K. implies that it is almost certain to be lost if one takes
it for granted and pursues, or hoards, the attributes of status for
their own sake. In other words, greed for the raw materials of
honour may well result in dishonour. This is not to say that honour
cannot be increased. 'If you have a title, honour it that it may
honour you; if you do not have one, strive that people may honour
you as much as, or more than, title-holders' (421200). But it can-
not be increased where it does not exist already: 'Some people,
having no honour from God or title from the emperor, strive
through bribes and gifts to be honoured by the provincial officials'
(591238). In any case, a man's first preoccupation must be to
prevent the loss of the honour he already possesses.
If it can be said of K. that, 'the whole sum of his wisdom is
suspicion' ,29 it can equally be said that the beginning of wisdom

28. On both the legislation and the classification, the final word remains to be said,
despite the detailed discussions by P. Lemerle, The Agrarian History of Byzantium
(Galway 1979) Itranslation of thl" author's article in Revue lIistorique, 219-20 (1958)1
85ff, and R. Morris, 'The Powerful and the Poor in Tenth-Century Byzantium: Law
and Reality', Past and Present 73 (1976) 3-27. In particular, the social reality of the
division between 'power' and 'poverty' - which also applied in the Latin West -
cannot be dismissed too hastily, nor (pace Morris, 21-2) can the relevance of the nuances
introduced by Constantine VII's Novel of 947 (Zepos, JOR I 215-7), including the
old legal definition of an absolute and unambiguous pauper as someone with a for-
tune of less than 50 nomismata. This, it may be noted, was the minimum sum which
the emperor Michael 111 had given to his favourite charioteers whose children he had
sponsored at baptism: Pseudo-Symeon, (CSHB) 659. The point was, surely, to raise
them to 'solvent' (EU1tOPO~) status.
29. Beck, Vademecum, 8; Lemerle, Proiegomenes, 96-8.

393
PAUL MAGDALINO

in his philosophy is the fear of dishonour. While he is quite sparing

with the vocabulary of honour (ttllit) and praise (E1t(HVOe;), that
of dishonour (attllia), shame (atOXuVT\, tv'tP01tit), shamelessness
(avatoXlJV'tia), insult (u!}pte;), blame (1l01lQ>it, \IIoyoe;), reproach
(OVW)oc.;), and mockery (yO..we;) is to be found on almost every
page. Fundamental to all these concepts is that ofaXP£tw(Jl)vll,
'ruin' or 'disgrace', which denotes not only the subjective feel-
ing of shame and the fact of exposure to insult, mockery and
reproach, but also an absolute loss of social position, commu-
nity respect, and self respect. 30 An examination of the passages
where the term occurs will take us to the heart of K.'s system
of values.
K. advises the man who serves an official (apxwv)31 to serve
him:
not as a mere man or official. but as if he were emperor or God. And if
he is an incompetent nonentity, but you are full of knowledge, wisdom and
resourcefulness, don't despise him lest he ruin (UXPEIWOEI) you. And don't
say, 'He wouldn't dare harm me' - [ have seen many ruined (UXPEIW9tv-
,0<;) in this way (5/124-6).
If you are a notary, let not your lord be ruined (UXPEIW9;;) by your greed,
but rather strive that he may be praised through your action and effort. For
he took you on for his honour and ease, not for shame and condemnation
(8/132).
The general who extorts 'shameful profit' from the inhabitants
of a frontier province will thereby lose what rightfully accrues
to him, 'and you will be indicted and ruined (aXP£lweit01]), and
the provincials will make the date of your dismissal an annual
holiday, and you will creep furtively away' (19/154). If the peo-
ple subject to his authority are shamelessly insubordinate, he
should deal with them prudently, not overbearingly, 'for that will
only make them more shameless still, and you will ruin the region
and yourself be ruined (aXP£tWeiie;)' (211158).

30. K.·s predilection for the word was noted by Lemerle, ibid. 8. Cr. for comparable
usage, Digenes 1.243; Kletorologion of Philotheos, ed. Oikonomides, Listes de pre-
seance, 83 line 24.
3 I. I.e. nol, in this case, a local notable with unofficial power.

394
 HONOUR AMONG ROMAIOI

In all these instances, achreiosyne implies a fall from rank and

office. The concept again turns up in the context of imperial of-
ficialdom in the section where K. offers advice to the emperor.
Much of this advice concerns the proper dispensation of court
titles and pensions, which K., in common with other contemporary
writers, clearly considered to be the essence of statecraft. 32 'Do
not raise foreigners (tevlKOUC;) who are not of their own coun-
try's royal family to great dignities or entrust them with impor-
tant commands, for if you do this you will certainly disgrace
(aXPElwm:lC;) yourself and your Roman officials' (95/278).
Yet achreiosyne can equally affect the man who lives privately
on his estate, and it can affect him not only when he falls foul
of the authorities, but also when he loses the respect of friends,
relatives and acquaintances. It is worth looking in detail at the
contexts in which such 'ruin' is envisaged.
Manage your estate well, K. advises, because otherwise in time
of need you will not have the wherewithal to feed your retainers,
and will be reduced to borrowing. This will fill you with shame,
and if you are lucky enough to find a lender, he will ask for in-
terest and a pledge and a repayment date, all of which you will
confirm in writing. If, when the date expires, the loan and in-
terest are unpaid, he will send repeated demands and finally
threats.
Then, finding you in the square, or in the praetorium,33 or in church on a
feast day, and wanting to show himself off to the people as being rich and
better than you, he will say, 'Man, what do lowe you that you hold on to
my property? Do you want me to ruin (iva UXPf:lWOW) you?' He says this
to ruin you, and already knows that he has ruined you.
And if you are silent, all will say, some to themselves, others out loud,
'The debtor has no nerve', and, 'It's an evil day for the debtor'.
If you answer him back, then all will laugh and say to you, 'Aren't you
ashamed to be so shameless to this man who has done you a good turn?'
And he, turning to them, will say 'Gentlemen, did you ever see such
shamelessness? '

32. See Michael Psellos, Chronographia, ed. E. Renauld, 2nd cd. (paris 1967) 1132,
II 83; Anna Comnena, Alexiad, cd. B. Leib, 2nd cd. (Paris 1967) I 114-5; John
Zonaras, CSHB, III 14-5, 766-7.
33. The governor's residence; cf. p.41/200

395
PAUL MAGDALINO

You will then feel cut to the quick, and will either have to part with a very
valuable property that you have inherited or gained at great expense, or you
will have to sign a second agreement promising higher interest. The whole
sorry affair will end with you, who were once so blessed, being dragged into
court (36-7/190-2).

Don't attempt to prove your family fortunes by becoming a col-

lector of extraordinary duesY Of course, your wife and friends
will tell you, out of inexperience, take this or that job and you
will have enough to keep yourself, your household and your re-
tainers. But don't listen to them. Whether you try to deal justly
or unjustly, there is no way, in the nature of the job, that you
can win. Instead of being what you ought to be, a mediator and
peacemaker to whom people listen with respect, you will be
shunned.
For one of the emperor's men (Iivepw1to~ ~aOlA.IK6~) will come and ask you
for a requisitioned beast of burden, but you'll be afraid to ask this man
because he is a friend, that man because he is a relative, another because
he is a dynast 35 ...
And you will tell the emperor's man, 'I haven't found anything'. He will
then arrest you and have you whipped, and ruin (~xpElwaEI) you in your
own town. And where you should have been honoured you will be
dishonoured, and you will have this to your reproach for generation after
generation.
I f, on the other hand, you choose not to spare someone, he may
obey, but you will alienate your friends, and if just one of them
has a plausible grievance against you, he will take you to the court
of the provincial judge, where you will be required to answer the
charge on oath. Whether you refuse or accept, you are undone
(39-40/194-8).
If you have a friend in another area and he is passing through
your town, don't invite him to stay at your house, but let him
lodge elsewhere. If he stays with you, imagine what people will
say. Your wife and daughters and daughters-in-law will not be

34. IJ1lEIOEA.flElv r.lc; 1rr.f'IOoorcpaKTopiac; (39/194): the word suggests not ordinary
taxation. hul eXa(tilln of extraordinary payments and corvee,~, such as the requisi-
lioning of animals mentioned in this passage. Cf. 47/210, where TO tit TCf'a.TTCIV Kat
Cf,OUO\(l~EIV flqH'lT(t)v is positively recommended.
35. I.e, a local archon with a preponderant share of de facto authority: cf. the ex-
ample on pp.33-4/184-6; Angold. Archons and Dynasts. 241ff.

396
 HONOUR AMONG ROMAIOI

able to go about their business; if they do, your friend will ogle
them, and when he is alone with his familiars (oikeiOl), will mimic
and make fun of them. Then he will set at nought your servant
staff, your table, the order (taxis) of your household, and he will
enquire concerning your possessions whether you have this or that
thing. If he gets the chance, he will make a pass at your wife and
'pollute' her; if not, he will boast of it all the same.
You must keep a close watch on your wife and daughters. There
was a case within living memory of a rich and distinguished
dignitary with a beautiful and virtuous wife. The emperor lusted
after her, and sent her husband off to the pr~)Vinces for three
years, but no amount of gifts and promises could make her yield
to the imperial advances. But when the husband r~turned to court,
he struck up a friendship with a handsome young man who
claimed to be his wife's relative. So the young man was invited
home, and the result was that he succeeded where the emperor
had failed.
When the event became known. her husband and relatives were seized with
depression and grief. or rather with achreiosyne. while the young man boasted
as if he had performed a Herculean labour. And so what the emperor and
promises of honour and wealth failed to achieve was accomplished by
familiarity and a friend. (42-41202-4)
Achreiosyne, then, is the complete loss of face in a 'face to
face' society where a man is as worthy as his reputation, and
reputations are made and broken by 'what people say'. K. clearly
shows that the most powerful moral deterrent in this society is
the fear that 'people will talk' and 'people will laugh '. The worst
consequence of military defeat is being 'without face'
(a1tp6(J{J)1to~) and 'without speech' (a1tappll(Jia(J'W~) among
relatives and peers (251266-8). It is unwise to tease the local idiot,
because
he will insult you and maybe even pull your beard. and just think what shame
this will cause you; if you let it pass, everyone will laugh, but if you hit him,
you will be blamed and jeered (63/246).
If you want to give alms. do not do so at dinner time. for people will think
you are giving because you are drunk. Give. rather. in the morning (491216).
If you are holding a sacred book and reading. and someone approaches you,
close it and speak to him. If you ignore him and carryon reading, you will
be considered a feckless and pretentious hypocrite (641248).

397
PAUL MAGDALlNO

Yet K. stops well short of what might seem to be the logical

conclusion: that the way to avoid achreiosyne is to keep up ap-
pearances. Only in his advice to military commanders does he
recommend deliberately cultivating an image. Otherwise he does
not distinguish between the private and the social, the inner and
the outer man. While he constantly advises precautions to avoid
embarrassing situations, he never suggests that it is possible to
bluff one's way through. This is partly, no doubt, because he
has little confidence that bluff will work: people will either know
the truth or suspect the worst, and they do not forgive or forget.
But K. also sees honour and shame as a matter of self respect
(471212). The honourable man gets into trouble because his per-
sonal sense of shame prevents him from defying social conven-
tions. The creditor who accuses him of shamelessness knows this
full well, and is playing on his aischyne to pull him still further.
into debt. If anything, the creditor is the truly shameless one.
K., far from commending his behaviour, assumes that when the
honourable man is in a position to lend money, he will not behave
in this way. On the contrary, he will himself run the risk of be-
ing humiliated by a shameless debtor, who will contrive to make
him seem mercenary and ungrateful! (47ffl212ff). In other words,
honour does not meet shamelessness with shamelessness, even
though the man with a_sense of shame is vulnerable to the man
who has none (3/122, 21/158, 56ff1232fO. Similarly, while
honour is damaged by hybris and oneidos, it is not restored or
enhanced by heaping insult or censure on others.
Thus although aischyne is socially and externally induced, it
is privately and internally experienced, and has as much to do
with conscience as with reputation; it is, in terms of a modern
conceptual distinction which works better in English than in
Greek, guilt as well as shame. 'For though we escape the notice
of men, we cannot escape the Eye that never sleeps' (421202).
K.'s prescription for avoiding 'ruin' is based on a deep religious
belief which is so thoroughly integrated with his prosaic, unsen-
timental philosophy of suspicion that its sincerity may not im-
mediately be apparent. By Byzantine standards it is singularly
unobtrusive. Miracles and the finer points of Orthodoxy do not
come into it; the Persons of the Trinity are hardly differentiated;

398
 HONOUR AMONG ROMAIOI

there is barely a mention of the Mother of God and the saints,

or, for that matter, of demons. The Scriptures are recommen-
ded as much for their practical tips as for their doctrinal message
(19/154). Yet God is constantly invoked, and not just as a
mechanism to explain what is lacking in the structure of human
society, but as the heart and soul of that structure, a tangible,
personal presence. The rich man, in his role as benefactor, is the
poor man's God (3/120). Men must serve an earthly lord as they
serve God (5/124). The emperor is God on earth (931276). 'If
you enter the priesthood to become a metropolitan or bishop,
don't accept until, through fasting and vigils, it is revealed to
you from on high and you receive full confirmation from God'
(51/220). A man may overlook insults to himself, but must avenge
insults to holy icons and the DivinitY, with his own life if need
be (451208). Above all, he must not neglect the daily offices of
Matins, Hours, Vespers and Compline (38/192-4). 'For these are
for the sustinence of our earthly life, and by these services we
make ourselves known to be familiar slaves (&OUA.Ot OlKdol) of
God.' Merely to acknowledge Him is not enough, for this even
infidels and demons do. It is essential to talk to God in the in-
timacy of midnight prayer, for this seals a relationship according
to which He is bound to reward keen and loyal commitment. What
is more, the reward will come in this life. K. leaves his reader
in no doubt that what is pleasing to God will also earn mens'
gratitude, and that the man who does God's will is sure to enhance
his social reputation (3/122, 6/128,121142, 15/146,45-6/208-10,
521222-4, 551228, 561230).
There is an apparent contradiction between K. 's philosophy
of suspicion and his Christian belief, which corresponds to a cer-
tain contrast betwen the astringent freshness of his examplesand
the complacent banality of his maxims. 36 However, he manages
to harmonise the two on the principle that God helps those who
help themselves as rational beings (991286, 1001290), and that
one can be good to people without trusting them (44-51206-8).
The secret of success as K. sees it can be summed up in terms
36. For similar maxims, cf. the parainetical poem known as Spaneas, dating from
about the same time: G. Danezis, Spaneas: Vorlage, Quellen, Versionen (Miscellanea
Byzantina Monacensia 3 I) (Munich 1987).

399
PAUL MAGDALINO

of three qualities ~hich are necessary both for the pursuit of

worldly ambition and for the fulfilment of divine commandments:
justice (OlKUlOCJUVTJ), zeal (CJ7touoit) and prudence (<PpOVTJCJl<;).
A man in any position of authority will himself deal justly and
will not stand by while injustice is done. If he is corrupt, fraudulent
or extortionate, or allows his subordinates to be so, he is sure
to be undone. But the converse does not automatically follow:
good intentions, and well-intentioned actions, are no guarantee
against dishonour unless they are backed by constant vigilance
and application, and executed with discretion and forethought.
K. only once uses the word 'discriminating' (OlUKPl'tlK6<;: 15/146),
but it perhaps expresses the quality that he values most highly.
This is the ability to think ahead to the long-term objective, and
to see the reality beneath the superficial appearance. In Chris-
tian terms, it is the ability to choose between good and evil; in
terms of social values, it is the ability to choose between the right
method and the wrong method of advancing one's honour in a
situation where one is apt to make the wrong choice for the right
reason. In effect, K. is warning the man of honour not to act
too impulsively on his honourable instincts: do not automatically
rush into battle, get rich, spend money, take offence, offer
hospitality, humiliate people, punish insubordination, favour
friends, despise foreign.ers or commoners, be afraid of gossip,
or be afraid of being considered a 'spoilsport'. Think of the cir-
cumstances and the consequences; remember someone is out to
get you and the world does not deal kindly with failure. Above all,
Let your strength be not in your money but in your mind, or rather in God,
for through Him we live and move and have our being (591236).
It is a great thing for a man to have power of servants, but stronger and
greater is the wise man even if he lacks wealth ... I don't despise wealth,
but I value prudence more highly (50-11218·20)
I praise the brave man (avopEiov) but the cunning schemer (1tOAU!30UAO<;
KUl1tOVllP6<;) surpasses all, for he is useful not only to himself but also to
many others (62/244).
In discussing the ways in which honour is gained and lost, we
have noticed significant differences between K. and Digenes. Yet
in considering the social group on behalf of which honour is
sought, we are struck by an immediate similarity: for K., as for

400
 HONOUR AMONG ROMAIOI

the legendary borderer, honour is invested in\the oikos. This has

already become clear from the three cases of achre~osyne which
we discussed at length. It is concern for the welfare of his
household that drive a man to the humiliation of borrowing, or
tempts him to become a fiscal agent. When he invites a friend
to his house, he puts his social reputation at risk because he ex-
poses his oikos to violation, which can occur even if the friend
fails to cuckold him.
It has been said that K. takes a low view of friendship;3? it
might be more accurate to say that he has a very exalted notion
of the oikos as a pure and sacred place, within which human rela-
tionships exist on a level removed from the worldly norm,
because there personal honour and group honour are identical.
The equation between the honour of the individual and the
mystery of his oikos is further developed in the section follow-
ing the cautionary tale of seduction. Her<i!·K.first warns his reader
not to tell anyone his secret (J.I,UOl"1lPtoV)38. 'for then you are his
slave and he will harm and insult you greatly', and then advises
him not to let inquisitive persons into his home,
for the inquisitive person will pry into your affairs, and nosing out the defects
of your household, will record them like a secretary, and will bring them
out when the time comes to abuse you (63/244-6).
As in Digenes, K.'s man of honour cannot venerate his own
parents too highly. They will be honoured by his good deeds and
dishonoured by his shameful acts (541226, 55/228). He will feel
dishonoured if someone praises him at his father's expense
(541226). He will avenge any insult to his brother, 'because the
ins.ult extends to your father and your mother', and he is accursed
if he fails to rush to the aid of a brother in danger (671244). He
must never forget the mother who suckled him. 'Worthy of
memory and praise are they who after their parents have departed
this life win God's favour for them through donations and alms'
(631244-6).

37. Kazhdan and Constable, People and Power, 26, 28-9. On the whole question
of Byzantine friendship, see now M.E. Mullett, 'Byzantium: a Friendly Society?',
Past and Present 118 (1988) 3-24.
38. Here K. may consciously be echoing the communion prayer of the Orthodox
liturgy: ou ~1'l yap -wic; hBpoiC; !lOU .0 ~U!l.T\plOv drrw.

401
PAUL MAGDALINO

Naturally enough, the recipient of K. 's good advice expects

similar respect from his own children, though he must respect
them in turn, and discipline them with words, not with blows
(471212, 53-41226). 'Pray that ~ou do not·have a son or son-in-
law or brother for an enemy' (51/220). 'A shameless daughter
has wronged not only herself but also her parents and all those
to whom she is related by birth. Keep your daughters locked up
out of sight like convicts, so you do not Iget bitten as if by an
asp' (511220). Even the most trusty male servant or slave should
not be allowed to make a daughter's acquaintance (551228-30).
These last remarks remind us of the obsessive jealousy with
which Digenes' father-in-law guarded his daughter, and they sug-
gest that such behaviour was not eccentric. 39 K. clearly sees the
women as the weak spot in a man's oikos. The death of his wife
will make it more vulnerable, for in losing her he will 'lose the
half or the whole of his life', and may well be tempted to remarry,
which will certainly destroy the harmony of his household
(55-6/230-1).
I n other respects, too, the oikos delineated by K. resembles
more than it differs from that which we encountered in Digenes.
Although its slaves and servants are slightly more human and more
in evidence, and we are more conscious of retainers who are not
menials, they are just as much part of the property, along with
fine gems and fat animals (491216-8, 61/242), and less in view,
perhaps, than we might expect. 40 Solidarity of the genos around
the oikos is less in evidence than in Digenes: there are not the
same attendant syngeneis participating in family events. The
family envisaged l:!y K. seems on the whole much more basic.
Yet there are details which blur the simple contrast. In both texts,
there is a tendency to treat relatives as the equivalent of friends

39. Cf. also Leo VI, Novel 48; Kazhdan and Epstein, Change in Byzantine Culture,
100.
40. Given the active roles in which they appear in some other texts. For example:
Peira 17.14, 25.25, 28.6 (cf.42.17), 30.40; E. Kurtz, Zwei griechische Texte iiber die
hei/ige Theophano, die Gemiihlin Kaisers Leo VI, Memoires de l'Academie Imperiale
des Sciences de S. Petersbourg, VIlle serie, Classe historico-philologique 3/2 (St.
Petersburg 1898) 15, 19; Chr. Angelidi, '~OUA,Ol OliJV KCIlvolavlIVOU1tOA,l1 lOV 100
aki>va. 'H Ilaplupia lOU Biou lOU 'Ooiou BaOiAtiou lOU Neou', IVIJIJ£IKra 6 (1985)
33-58.

402
 HONOUR AMONG ROMAWI

(K. 45/208; see above, 149). K. assumes that a household is likely

to include married sons and their wives (421202), that a shameless
daughter will shame her whole kin (511220), and that God will
forget the man who forgets his relatives (52/222).
If K. differs in his portrayal of the role and importance of the
oikos, he does so inasmuch as he sets the oikos much more clearly
in the context of an ensemble of neighbouring oikoi with which
it jostles for position. In other words, the social group in whose
eyes honour is sought is not merely the oikos or the genos, nor
even the sum of two rival oikoi or gene, but a local community
- a patris. The settings in which points are scored or lost in the
struggle for 'face' are public, community settings: the square or
market place «Xyopei), the church, the governor's residence, or,
if it comes to the worst, the law court. The patris thus emerges
in its own right and in a way which is conspicuously missing in
the world of Digenes. Whether this reflects the social difference
between the great military magnate and the lesser-ranking archon,
or the geographical difference between the 'ranchland' of Eastern
Anatolia and the towns of the Balkans (from which most of K. 's
examples are drawn)41, or the difference between fiction and
fact, or a combination of all three, is impossible to say. At all
events, K. shows that for Byzantines who were not of D. 's world,
the social framework of honour and shame had three main com-
ponents: the oikos, and the structures of state and local com-
munity which pressed rival oikoi together.
It now remains to examine the workings of this tripartite rela-
tionship and to determine what each of the partners derived from
it. At first sight, it looks as though the oikos, being the smallest
unit, was the most vulnerable, in constant danger of being crushed
between the pressures of the state and local public opinion. Both
state and community make demands on the local archon, and
neither offers immunity from the other. And yet the man who
keeps his head can use his intermediate position to save and even
win honour by mediating between the authorities and the locals
(40ffl198 ff, 66ff1252 fO. I n any case, he can neutralise the latter
by playing on their divisions, which are sure to surface sooner

41. Cf. Hendy, Studies, chapters 1-2, esp. 85ff, 100ff, 131 ff.

403
PAUL MAGDALINO

or later (58/236). In the face of the state, the oikos was on balance
better equipped to survive than the local patris, not least because
the sanctity of the honourable household was respected by the
imperial legal tradition, which, however, no longer recognised
civic immunities and autonomies. 42 In its dealings with the local
community, the state was at a great advantage. It provided, in
the law-court and the praetorium, the forum where local rivalries
were brought to a head and resolved. 43 The power and honours
which the state dispensed were far superior to anything local opi-
nion could offer, for they not only raised the local aristocrat in
the scale of small-town reputations, but also lifted him out of
this scale - to court in Constantinople, or to military command
in another region. Moreover, the state had a monopoly of patriotic
propaganda, which K., like Digenes, echoes distinctly. As we have
seen, he declares that it is honourable to die in battle 'for king
and country' (16/148).44 Patris here is ambiguous, but it almost

42. See Synopsis Major Basilicorum, K.I1.45 (Zepos, JGR V 323): 'A man's house
is his castle': Peira 17.14 (Zepos. JGR IV 63): the case of a betrothal contract broken
by the fiance. who claimed to have made it when under age and.misled by his bailiff
(baiou/os). The emperor at the time (probably Romanos III) rejected the claim on
the grounds that 'minors who deceive others have no excuse. What greater deceit
could there be than to enter the house of a noble lady and a senatorial gentleman,
and to see his daughter. speak with her and relax in his house, promising marriage?'.
Cf. also 49.5 (p.198-9).
The laws in tlTe Justinianic Corpus pertaining to civic councils, munera and privileges
had all been abolished by Leo VI (Novel 46), and the earliest firm evidence for Byzan-
tine imperial charters of privilege to Greek towns (as opposed to nominally subject
Italian cities like Venice) dates from the end of the twelfth century: see Michael
Choniates, Til.l:w(6f.J.Eva (Athens 1879-80), II 54; Niketas Choniates, Historia, ed.
J.L. Van Dieten, CFHB (Berlin/New York 1975) I 599; Geoffroy de Villehardouin,
La Conquete de Constantinople, ed. E. Faral (Paris 1939) 88; G.L.F. Tafel and G.M.
Thomas, Urkunden zur iilteren Handels-und Staatsgeschichte der Republik Venedig,
(Vienna 1856) II 17-9; cf. M. Angold, 'The Shaping of the Medieval Byzantine "City"',
BF 10 (1985) 22. However. for1'hcoming work by Haris Kalliga shows that Monem-
vasia may always have been something of an exception, and a hagiographical text
recently studied by Kaz.hdan may indicate that chrysobulls granting municipal tax-
exemptions were already being granted before WOO: A. Kazhdan, 'An Unnoticed Men-
tion of a Chrysobull ascribed to Constantine the Great', ·Aqllt(,wf.J.a arov NfKO
l:fJof'WVO (Rethymno 1986) I \35-8.
43. E.g. Peira 61.6: court case of two dignitaries who came to blows after calling
each other names.
44. For 'God and country' propaganda in the Byzantine imperial tradition, see M.
McCormick, Eternal Victory (Cambridge 1986) 237-52; for the tenth century, see also

404
 HONOUR AMONG ROMAIOI

certainly means the 'fatherland', 'Romania', rather than the native

town. The man who stands up for the interests of local citizens
against the abuses of central government does so 'for the good
of the territory (xwpa) and on behalf of the public ('"co K01V6v),
in the cause of justice rather than of patriotism (421200). A
foreigner is not a person from another province, but an ethnikos,
a person of different 'nationality'. If there is a local revolt, 'fight
on behalf of the emperor and the general peace', and if this proves
impossible, 'write to the emperor and strive, as far as possible,
to render service so that you and your children and your men
may be honoured.' In any case, 'keep faith with the emperor in
Constantinople and you will not be disappointed in your hope',
because 'the emperor in Constanti.nople always wins' (641248,
741268).
Thus in the final analysis, K. leaves us with a model of Byzan-
tine society which we have already glimpsed in our discussion of
Digenes: a vertical structure on an axis which culminates, below
God, in the emperor, and rests on the oikos. If the basileia is
supreme, the oikos is basic, and equally central. An educated
Constantinopolitan might have been inclined to see the system
as working from the top downwards. We may be more inclined
to see it working upwards from below. K. certainly allows room
for this perspective: 'Do not be harsh with your household, but
let those in it fear you ... But you too should fear your
superiors ... ' (60/240).
Whether K. can be said to take a 'd~scending' or an 'ascend-
ing' view of his society, it is clear that he saw it as held together
by vertical ties of service. He saw no security in ties between
equals. A man's honour was, after all, most at risk from the loose
talk and merciless scrutiny of his social peers, and from the
shameless disrespect of insubordinate inferiors. 'I have never liked
having a companion, and I have never, except under constraint,
shared quarters with a man who was my equal.' (611242). Such
an attitude did not preclude friendship, for friends are very much
a part of K. 's social landscape. But it implied that friendship
worked best when it stemmed from, rather than encroached upon,
the ties of dependence which began and ended in the oikos. For
K., as for other, more intellectualising Byzantine writers, the taxis

405
PAUL MAGDALINO

which constituted the universe and human society within it was,

above all, an order of precedence, of reciprocity in inequality.
I n the words which Theodore Prodromos put into the mouth of
his fictional character Bryaxes, King of Pissa: 45
I f all lived together in one condition, and none was a slave but each was
free, there would be no rule, no measure, no standard in life, no structure
to the universe, no good order (eutaxia). Ev<trything would be destroyed and
corrupted. But since natlnal reason rules all things, it is necessary that there
be slaves and masters, for how could cities be settled if man did not become
dependent on man?
Byzantine society as K. portrays it was a chain of such relation-
ships, in which each man was, in varying degrees, both master
and servant, and found honour in the fulfilment of both roles.
Yet it is important to note that K. does not idealise the princi-
ple of hierarchy at the expense of the individual. For him, con-
crete, personal relationships come first. Government service is
personal service, to an emperor or an official, and one must above
all respect the honour of the man rather than the system one
serves.
Don't deceive him in any matter. .. if you do and are found out, you will
be ridiculed by all. .. [f he commits a fault to do with the Fisc, don't de-
nounce it, even if it bothers you, hut keep his secret with all care. [f you
betray it, everyone will flee from you as from a snake. Respect his wife just
as you would a lady, a ffiother or a sister (5/126).
One should even shelter fugitives from justice if their crimes are
not too horrible (501218).46
K. 's inability, or reluctance, to conceptualise the state in th~
abstract means that he cannot idealise servitude, or recommend
it to those who are in a position to avoid it. In the section of

G. Dagron, 'Byzance et Ie modele islamique au Xe siecle', Comptes rendus de

I'Academie des Inscriptions et Belles Lettres ([983) 224-32. The possibility of an [slamic
stimulus is especially interesting in view of the Muslim ideology portrayed in the 'Song
of the Amir' (above, ).
45. Rodanthe and Dosikles, V[I. 364-72: ed. Hercher, Erotici Scriptores Graeci II
397. The exaltation of taxis seems to owe something to Gregory of Nazianzos, Ora-
tion 32.7-12: ed. C. Moreschini and tr. P. Gallay,.Sources Chretiennes 318 (paris
1985) 98-113. However, the justification of slavery by natural reason is more reminis-
cent of Aristotle, Politics I. 2-6.
46. For the regulation of 'everyday crime' through 'self-help', see R.J. Macrides,
'Killing, Asylum and the Law in Byzantium', Speculum 63 (1988) 509-38, esp. 533ff.

406
 HONOUR AMONG ROMAIOI

his work where he offers advice to toparchs - petty princelings

beyond the empire's borders who were imperial clients but not
imperial subjecls - he comes dose to endorsing the ideal of
perfect independence represented by Digenes Akrites.
Don't let wealth or titles or fine imperial promist;s sway you into giving your
territory to the emperor and receiving money and lands in return, even if
you are to receive fourfold, but keep your territory, however small and in-
significant it may be. For it is better for you to be his independent friend
than his dependent sla}'c. You will be noble, worthy, praised and glorified
in the eyes of the emperor and of all other men as long as you and your
children and their descendants are in your own territory and jurisdiction.
(76/298)

This contrasts markedly with the 'normal' Byzantine attitude,

a good example of which can be found in Procopius' account
of the lustinianic conquest of Vandal North Africa. A subor-
dinate of Belisarios, the Herulian Hun Pharas, persuades the
fugitive Vandal king Gelimer to surrender in these words:
And yet why should it not be better in every way to serve as a pauper among
the Romans than to be a potentate ... among the Moors? But of course,
it seems to you the very height of disgract: even to be a fellow slave of
Belisarios. Are not we, who also are born of noble families, proud that we
now serve the cmperor?47
This was the spirit which inspired Justinian's programme of im-
perial expansion, and sustained the Byzantine state through the
following centuries of crisis, contraction and reconstruction. But
in Digenes Akrites and in Kekaumenos we see, around 1100, the
literary stirrings of a different spirit: a yearning among the Byzan-
tine 'powerful' to disconnect the chain of servitude at the top,
or at least to recast its political links in a different mould from
the oikos. In the long run, this yearning was to prove fatal to
the cohesion of Byzantine society, because it was never translated
into a systematic effort to put the relationship between ruler and
ruled on a contractual, 'feudal' basis. In the short term, Byzan-
tium survived its second major crisis, that of the eleventh century,
not because it became more feudansed, but because successive
rulers, starting with Alexios I Komnenos, created a system
whereby the ties of douleia which bound the leading aristocratic

47. Procopius, Wars IV.6, 20-2, tr. Dewing II 259-61.

407
PAUL MAGDALINO

oikoi to the imperial oikos became strengthened, and sweetened,

by ties of syngeneia. In other words, the extended imperial oikos
became more closely identified with the imperial genos. The result
was undoubtedly to put a new premium on genos and eugeneia.
But it was also to extend the oikos as a model of imperial govern-
ment, both in its service and in its kinship dimension. And as
the imperial genos spawned more and more oikoi, their individual
honour came to coincide less and less with their collective honour.
The consequences for the superstructure of the social framework
of honour and shame in medieval Greek society were catastrophic.
But they showed that the infrastructure had never been more
healthy.

Department of Medieval History,

University of St. Andrews

408
 Chapter 10 Aristocracy and Patronage in the literary circles of
Comnenian Constantinople

Margaret Mullett

When questions of literature and aristocracy arise an automatic

response is to add the two together and make patronage. That this is
not the only right answer to the sum is shown by two other papers in
this volume, that of Alexander Kazhdan, which by charting the some-
times imperceptible shifts in the topoi of encomia, history and
parainesis beautifully documents changes in aristocratic values under
the Comneni, and that of Paul Magdalino, which divertingly lays bare
the social illusions, aspirations and fantasy-life of (who else?)
writers. But the question still poses itself: who paid for Comnen-
ian literature, or, as I should prefer to phrase it, what was the
social base of Comnenian literature?

It is no small task to attempt to answer this question. It

involves the production and reception of literary works of all kinds
from around the accession of Alexius Comnenus in 1081 to well into
the thirteenth century, for neither the death of Manuel I in 1180 nor
the captur l of the city in 1204 really marks a cut-off point in
literature. By concentrating on Constantinople very little litera-
ture is excluded, at least until the end of the twelfth centJry when
Michael Choniates and his associates in Greece appear to have set up
a literary alternative to Constantinople;2 previously works written
in the provinces seem above all designed for the capital. And above
all, it involves all the big issues, concepts and methodologies of
the social study of literature at any time in any society: concepts
of public, patronage, performance, orality, transmission and style,

I came to the question when I tried to set my study of letter-

writing of the period against the whole of contemporary literary
society. I looked for works of this kind and failed to find them.
In some quartess I found a flat denial that there was a literary
public at all, and when 1 looked for the producers of literature
they were sometimes seen as an isolated clas~ writing in a dream-
world of their own or the classics' making, sometimes as fully
integrated memberg of society drawn from the high bureaucracy of
church and state. Literary society tends to be described in terms
of analogies drawn from other societies, surely a dead giveaway: a
mandarin caste which went in for coteries offers uS a rather curious
mixture of Indian, Chinese and French cuisine.

In signalling a lack of works on the problem I do not of course

wish to suggest that Comnenian literature has been neglected: it is
clearly a very exciting period. Few would now agree with Robert
Browning's characterisagion of it as 'an age of uncreative erudition,
of sterile good taste'. Herbert Hunger rehabilitated it as long ago

409
 as 1968. 7 It is clearly a period of experiment and individualism, of
parody and wit and fun, of the revival of old genres and the dis-
appearance of others. It is becoming a fashionable home for the
redating of works otherwise hard to date: the Christos Paschon,
Digenes Akrites and the Belisarius Romance have, with varged success,
been proposed for membership of the twelffs-century club. Hunger~f
work on rhetoric,9 Browning on education, Darrouzes on the church
have all advanced our understanding. And we now know a lot more
about Comnenian society in general; we know about the ruling class in
the eleventh and twelfth centuries, about Comnenian government and
about the relations of ci~ies and aristocracies after the so-called
eleventh-century crisis. 1 Work has progressed on the use of the
vernacular, on levels of style, on the social status of writers and
on their perception of their social position. 13 And above all work
has begun on the truly literary af~reciation of Byzantine literature
for its own sake and its own time.

But of the functioning of literary society, and detailed study

of its composition and organisation, there is so far nothing. What
we are offered instead are 'literary circles', a commonplace of what
has been written on the period, which we might well do without - were
it not so prevalent in the scholarly literature. Since Chalandon,15
people ~ave happily said that X belonged to the literary circle of Y
or attended the salon of Z. Robert Browning, more cautiously, states
that we know some of the literary circles of the earlier twelfth
century,l6 and proceeds to sketch in a few. But in other fields this
potentially facile term has been seen for the delusion that it often
is. Michael Innes neatly parodies its use in Romantic literary
criticism: of a totally spurious writer he has a character say,
'There's quite a lot about him. It seems that he had a circle too.
Or several circles. He revolve9 in several circles. Whereas Shelley
revolved only in his own,.1 Anyone who seeks to use the term
loosely should look at the magnificent de~olition job done on the
circle of Julia Damna by Glenn Bowersock,l~ in which he shows that
the view which sees the empress as 'gathering around her a group of
intellectual luminaries who satisfied her deep instincts for philo-
sophy and rhetoric' is based on three contemporary references, that
the received list goes back to 1879 and no further, that Galen may
well have been dead before the circle got under way and that the one
member apart from Philostratus who is attested in antiquity is
usually missed off the lists. 'The custom', says Bowersock has been
to fill out the literary circles of Anti~uity 'with persons not
explicitly attested as belonging to them,.l

But Philostratus did say that he had been a member of her

kyklos, and similarly the Byzantines of the twelfth century had a
word for literary circle, in fact several. The word is theatron,
which was made part of our vocabulary first by Professor Hunger. 28 I
shall start by looking at some literary evidence for these theatra
before turning to social factors and more general issues.

The word theatron used for a literary gathering has a long

pedigree. In particular it was used during the Byzantine centuries
for the meetings of groups of friends to read aloud and comment on
literary texts written, perhaps from a distance, by absent members of
the circle. Examples are attested in the writings of Libanius, who
specifically mentions the reading out of a letter, Synesius and

410
 Procopius of Gaza in the early period, in Psellos, Nicephorus Grego-
ras, John VI Cantacuzenus, who records the amount of applause given
to any particular offering, Manuel II Palaiologos, whose circle
listened to s~eeches. poems and letters, Demetrius Kydones and John
Khortasmenos. 1 The emphasis in these gatherings was on the perform-
ance to a learned and literary audience of new works of art, a
preview rather than a seminar. For the Middle Byzantine period the
evidence has always appeared less strong, but with the recent stream
of editions of letters, speeches and other rhetorical works it has
become possible to fill in the gaps a little. But we are far from a
complete picture.

For one thing, it seems that writers were affected by the knowl-
edge of the potential performance of their work and occasionally
include images which derive from it. For example Theophylact of
Ochrid, writing to Romaios uses the imagery of performance: "You
have deafened my ears with your account and with your applause" and
in a letter to John Doukas he expresses the wish for the time to
write, so that he could lead the chorus and fill out the log ike
choreia. 22 Neither reference, only two of many, need mean any more
than its core of sense, but both images would have had more signifi-
cance for the recipient if he were envisaging it in relation to oral
performance. Again ambiguous, but possibly indicative is the
description by Eustathios Makrembolites of his novel Hysmine and
Hysminias 23 as a drama: where should a drama be performed but in a
theatron? But~ more useful is a series of references in the
letters and speeches of George Tornikes I and Michael Italikos to
something which they refer to sometimes as theatron, sometime3 as
kyklos. In letter 17 to Nicephorus Bryennios the historian, Michael
Italikos sends greetings from Constantiniple to somewhere in the
provinces and describes the effect the other's letter had had on his
log ikon theatron:

When your letter was brought into the logikon

theatron it gave forth your voice and your song,
with such literary grace, such a gift of the
Muse, such rhetoric that I cannot describe it.
How it sang, how it filled us with joy •... were
it not for the form, the regularity of rhythm and
the suitability of the language we should all
have been carried away with enthusiasm~ both the
reader of the letter and the audience. 4

In a speech we learn the identity of the leading light of at least

one of Michael's circles: Eirene Doukaina, the mother of Anna
Comnena and of the emperor. He compares himself to cicadas or a
swallow: just as their song responds to the sun in springtime 2~0
does his eloquence respond to the war~th of her presence as judge.

One of the activities of the kyklos is revealed in speech XV,

where he tells the audience that he ~lled upon by the Basilissa
Eirene to improvise a speech, which he does, ending with the remark
that he mixed philosophy with his rhetoric beca~~e the peri ten
basilida kyklos was oude pantapasin aphilosophos. We learn of
other members of the group in his monody on Andronikos Comnenus,
where ~, describes the dead man as an ast@r in the circle of his
mother, and in George Tornikes's funeral oration on Anna Comnena,

411
 where the connection of Anna and Nicephorus Bryennios with Eirene
Doukaina in the encouragement of the Muses is evoked, albeit in very
general terms.2~ Here at last is some proof of a literary gathering
under the aegis of the Empress Mother, involving the reading aloud of
letters and the improvising of speeches. But of its composition,
beyond Anna, Nicephorus, Andronikos, Michael Italikos and perhaps
George Tornikes, ~e know nothing.

But these literary gatherings do not always appear to ~ave such

a convener, at least explicitly. In a puzzling passage 2 Nicholas
o Kataphloron, teacher of Gregory Antiochos, describes a syllogos
composed of a dozen or so sophistai who emulate the ancients in their
literary productions, which they perform boisterously, tumbling or
dancing with words. They are so productive and their works so beau-
tiful that they become known and people invite them to their homes
where they make speeches and sing songs, at table and elsewhere.
From this both their hosts and the city derive honour. This appears
to be another example, slightly different from Eirene's, of a real
literary circle. In fact we should not be surprised at variations in
this embryonic pattern, for it seems likely that our theatra were
only one of many kinds of gatherings where reading aloud or declama-
tion occurre~b from the sewing circles of ladies complete with spiq-
tual father- through the reading aloud of saints" lives at meals.~l
the performance of speeches to the emperor in front of a kyklos or
choros 32 to the receptio~3(dying out at this time) by congregations
of new hymns and sermons.-

So far our literary evidence leads us. What ~e next need to do

is to consider ~ays and means of adding the missing social dimension
to these tantalising references and to see whether we can people
Eirene Doukaina"s ghostly log ikon theatron. I shall start by trying
to imagine the location of such gatherings, the milieux in which they
might flourish.

The fi rst of these is the court. We are fortunate in that we

know a reasonahle amollnt abo'lt the coLirts of the three Comnenian
emperors, Alex!"s, John and ~anuel. We are told about the physical
framework of the court in the t~) major palaces of the Hlachernai and
the Great Palace; 34 we kll"~ a r"asonahle amount about the personnel
of the COllrt throngh t~le very f·lll synodal lists especially of 1094,
11 ='7, J 166 and 1 J 70, enough tn (elll is~ the extensive system of 'clan
government" practised by the Co.nlleni. 3 We can watch the COllrt out-
side Constantinople, reconstruct a certain amount of ceremonial and
we kno'" thAt ~ntertAinment "'n6 r',volutioni,Jed through the adoption of
tuurnaments. 3 We nre ~~ll informed of the tone of the court:
Alexius's monasterY-C0lJrt, 7 John"s austere high-command, Manuel'~
anakainesis and flashy glamour, complete ~ith suburban love-nests. 3
Courtlife is "in short" well documented, that is to say writers were
aware nf ~hat went on in the cOllrt, and the Cnmnenlls family itsel f
WRS no stranger to literature, although it should be said that this
Was of varied quality. ~lexius composed verses and polemics, AnnR
Comnena. Nicephorus and a Sebastokrator Isaac each mAde a contribu-
tion. 39 But, unlike the Palaenlngan period, when, ironically, scho-
lars have poInted to falling imperial patronage RS a factor in liter-
arv.' decline , 0 it is so far impossible to anchor onE 41Of these float-
ing theatra in the capacious courts of any emperor. Indeed beyond
the specifically functional deme poetry of Theodore Prodromos it is

412
 hard to find any author writing for the court alone and certainly not
entirely financed on a long-term basis like some poets laureate.
Piece-work seems to be more the rule than sustained patronage, again
possibly with the exception of Theodore Prodromos. Certainly Nicho-
las Kallikles wrote a great deal of poetry geared to, and presumably
commissioned by, a court readership. But he was already ensconced in
court circles and on the pay-roll as archiatros, so I think it un-
helpful to label him a 'court poet,.42 Dedications and commissions
are few: Euthymios Zigabenos's Panoply, Stephen Physopalamites's
Alphabet, John Kamateros's astrological poem and Michael Glykas's
proverbs are among the works offered to reigning emperors. 43 Even if
we turn up any more, the evidence would still be so slight as to be
insignificant.

A second possibility is to look in the academic institutions of

Constantinople for coteries of friends meeting for the purposes we
have been examining. It is often difficult to decide in a rhetorical
text whether it is referring to a teacher with his kyklos of pupils
or to a literary circle proper. Michael Italikos writes a letter
inviting a friend to a philosophical ~~nquet offering separate
courses of Plato, Aristotle and so on. Is this a school or a
theatron? John Tzetzes, in a ,passage spotted by Nigel Wilson, refers
to a~5 advanced classical seminar attended by people with texts in
hand. School or theatron? Professor Browning in a preliminary
attempt to outline some of the literary circles of the twelfth cen-
tury marks out what he calls the patriarchal School as the home of
theology, grammar, rhetoric and belles-lettres and notes Michael
Italikos as a ~eading figure, distinguishing it from the one around
Anna Comnena. 4 While it is true that the teachers in patriarchal
jobs (maistor ton rhetoron, didaskalos tou psalteros, didaskalos
tou apostolou, oikoumenikos didaskalos) did write in these genres (it
was after all their bread and butter) they range widely outside them,
and no literary rigure of note with the possible exception of Nice-
phorus Basilakes 4 can be linked only to other patriarchal academics.
Other teachers are heard of only in the context of grammar or schedo-
graphy, but they appear to be rather dreary school-masters and fall
outside the scope of the present inquiry.4~ S~ even if the Patri-
archal School existed as a school, it would be no solution either.

Finally we must examine the most likely possibility from our

literary evidence, that the theatra and kykloi of the twelfth century
were in effect salons, under the kindly guidance of noble, often
imperial, often widowed ladies. This is clearly how Chalandon saw
what he described as literary circles, but it carries with it the
flavour of nineteenth-century Paris, a rather Proustian perspective.
I think we should take care. For the Comnenian period five noble
ladies have been canvassed as candidates for the role of patroness of
literati and grande dame of a literary salon. These are the ex-
basilissa Maria, Anna Dalassena, Eirene Doukaina, Anna Comnena and
Eirene the Sebastokratorissa. Without further information on salons
and theatra as su~~ we must deal at the same time with the not
inevitably connected roles of literary patroness and patroness of a
salon.

The ex-basilissa Maria,50 who was married successively to

Michael VII Doukas and Nicephorus Botaneiates, played a crucial part
in the arrival of the Comneni to power in lOBI. Her 'alternative

413
 court' at the Mangana after the coronation of Eirene, Alexius's wife,
has been seen by scholars like Robert Browning and Lowell Clucas as
the forum for the 'enlightened' philosophical group which they see
Alexios as repressing at the beginning of his reign. 51 Two manu-
scripts and an enamel with her portrait survive, and another manu-
script belonging to her is recorded; Eustratios of Nicaea dedicated a
meteorological treatise to her and Theophylact wrote one speech, at
least one letter and two biblical commentaries for her. 52 But there
is no evidence to suggest a theatron.

Anna Dalassena, the mother of Alexius Comnenus was equally a

figure of power - a very austere figure - in the early days of the
Comneni, and we rely upon Anna's account of her rule to furnish us
with certain indications that her entourage, though primarily
clerical, may well have included some literati; she provided Alexius
with the panoplist Zigabenos. 53 But the evidence again is simply not
there.

Eirene Doukaina, as we have seen, is quite another case. ' She

was the daughter of Andronikos Doukas, married Alexius on the insist-
ence of the Doukas family and the patriarch and established her
control over Alexius, especially through his illnesses, until he was
quite dependent on her - except over the succession. She founded th~
convent of the Kekharitomen2 in Constantinople and ended her days
there at the latest in 1133. 5 We are now on much firmer ground. ~5
know that she commissioned Nicephorus Bryennios's Hyle Historias,
that she was in correspondence with Nicholas Kataskepenos (the author
of the Life of Cyril Phileotes) and possibly with Theophylact of
Ochrid, 56--rhat she commissioned poems from Nicholas Kallikles and
possibly from Manuel Straboromanos. He wrote a speech of consolation
for her on the death of her brother Michael the protostrator; Theo-
dore Prodromos w~ote her verses of consolation of the death of her
son Andronikos. 5 , We now know also that she did have a circle,
whether in the Kekharitomene or elsewhere, which was called by con-
temporaries a theatron.

Anna Comnena is different. Her learning and her interest in

letters are quite clear from her Alexiad as well as from George
Tornikes's long funeral oration on her, but an important feature of
her work revealed there and commented on by Professor Browning is her
encouragement of a group of Aristotelian commentators who included
Michael of Ephesus, Eustratios of Nicaea and possibly James of
Venice. It may well be that she persuaded them to undertake the work
and divided up their tasks. But evidence of an independent literary
salon 0S8 her own, as distinct from that of her mother is so far
lacking.

The final example is perhaps the most famous: in every work

which touches on twelfth-century civilisatio~9a word is spared for
the 'circle of the Sebastokratorissa Eirene'. Recentl
extensive researches of Elizabeth and Michael Jeffreys6
o with the
the full
importance of this figure is becoming apparent. She was the wife of
Andronikos Comnenus, John II's second son, who died in 1142 while
bringing the corpse of his brother Alexius back to Constantinople.
She led a rather chequered career after the accession of Manuel, but
despite imprisonment in the Great Palace, the Princes Islands, the
Blachernai and the Pantokrator and exile in Sofia she was a focus for

414
 the works of some of the most remarkable writers of the 1140s and
appears also to have commissioned some manuscripts. Theodore Prodro-
mos wrote four occasional poems, an astrological poem and a grammar
for her; she commissioned Tzetzes to write Resiod and Homer and his
Theogony for her; Constantine Manasses wrote encomia for her and
dedicated the Synopsis historike to her. Manganeios Prodromos wrote
poems for her, and the monk Iakovos wrote forty-five letters of
spiritual counsel for her. Yet she cannot be seen as the literary
inspiration of all the works of these writers; some of the highly
esoteric satire a~parody of the novelists, for example, would seem
to be out of keeping with the standard of education which needed
Tzetzes's Theogony or Prodromos's grammar. 61 And there is no indica-
tion I am aware of that these writers met together under her aegis.

So at this point with the exception of Eirene Doukaina we appear

to have drawn a blank. If instead of chasing these circles round the
palaces, schoolrooms and prj vate houses of Constantinople we look
rather at the mechanics and processes of literature in the twelfth
century, we may see the significance of these theatra. In the

COMPOS
C O M P O S II TTl
I O0NN

Writer
W riter
N

N
IO

IO
PT

n
ON

c.
PT
RECEPTI ON
CE

~ ~2
E
I
OPNT

Cl
RE

EC

~. "
N

rj
n ;;,
T IE

R
N

IO

ERPE C

Co- ] "
X trt
V'
IO

';1;
PT

rt W
tt:
PT

......
CE

~.

oC
CE

C
E
RE

3
R

:l ::
RE

,(,1
A u d i e nc
AtJd n c c'
^ Pdtccn
Pa t r o n Readership
R eadership

RECE PTlON
RECEPTI ON

diagram they clearly come under the head of audience, a much under-
studied aspect of the reception of Byzantine literature, though in
its general sense audience analysis ha~ recently become a fashionable
aspect of modern literary criticism. 6 Students of oral literature
in Byzantium tend tc lose interest when they have decided that their
chosen work was not orally composed, just orally delivered,63 and
students of the written word are understandably more interested in
manuscripts. The written word of course has primacy in Byzantium,
the spoken word lacking both tg~ symbolic value attributed to it in
ancient Athens by Averintsev and the stylistic adaptat~ons of
Western medieval literature to suit it to aural reception. 5 But
orality is still importagg in Byzantium, for entertainment after the
death of the theatre, for publicity and the sharing of a rare
commodity and for the editorial purposes suggested by Aug,bach,
without much enthusiasm, for literary readings in Antiquity. It
has even been suggested by Hunger that the theatra were consolatign
prizes for the low status and lack of material wealth of literati. 8
And there may be a new life for orality in the twelfth century in the
wake of the eleventh-century revival of rhetoric. 69

415
 And if, turning to the category of readership in the reception
process, we draw extreme conclusions from recent work, the importance
of audience and theatra grows correspondingly. If (and here I brush
to one side the Brotests of Browning, Patlagean and to some extent
Marlia Mundell)7 the cost of b~oks is thought to be beyond the
pocket of a minor civil se'2ant, 1 if chronicles are thought of as
light reading for the court, if the verna5~lar experiments are seen
as a literary game of the learned classes, if Df§enes Akrites was
written for a learned Constantinopolitan audience, and above all if
lower-style metaphrases were used for reading other people's books
while you yourself wrote in high style,lS it is conceivable that the
readership of Byzantine literature was no wider than its audience, an
audience comprising the sum of all contemporary theatra with the
addition 0; country members (exiled bishops, for example, panting to
get back) 6 and superannuated members (there is some indication that
historiography tended t9 be written late in life or while authors
were out of circulation). 7 It may well be that audience and reader-
ship were simply two sides of the same coin.

Patronage comes at an earlier stage of the process than either

audience or readership. We have already encountered some of the
problems connected with the concept. As usual many of them arise out
of definitions. The word 'patronage' is made to 'grve many different
functions by literary historians of any period, and the twelfth
century is no exception. The range spreads from Eirene Doukaina's
suggestion to Nicephorus Bryennios that he might write the
Hyle Historias, through he, obtaining of a job as didaskalos to~
iatron for Michael Italikos 9 to the payment of such poets as Theo-
dore Prodromos and John Tzetzes, who clearly depended on commissions
as their major source of income, and the apparent maintenance of
Manganeios Prodromos as household poet.

Sometimes it is impossible to tie down these graduations: occa-

sionally a dedication of a work may give no idea whether it is a
purely hack job paid for on the nail, like many of the epigrams
written for works of art at the time ~presumably you commissioned
your icon painter, your silversmith and your poet at much the same
time) or a piece of educational work commissioned and paid for, or a
genuine dedication like perhaps Theodore Prodromos's Boem 'On the
Great and the Small', dedicated to Michael Italikos. 8 Or, with a
given body of information it may be difficult to decide whether one
person acted in the role of patron to another (even where it is
explicitly stated, it has been suggested recently, we should be just
as cogfcious in a literary work of the persona of patron as of
poet), as I have tried to show in a recent study of the fi~lations
of Theophylact of Ochrid and Maria the ex-basilissa. His
basilikos logos of the 1080s in which he praises her learning, her
philanthropia, but above all her generosity, and a letter of the
1090s in which he describes her as ten emen meta ge theon antilepsin
make us suspect that his rapid advancement throughout the 1080s may
have had something to do with a word in the right place during her
period of influence with the emperor. But the only two works known
to have been dedicated to her were written during Theophylact's
archiepiscopate and give us no further clue of their relationship.

416
 This problem of definition is not exclusive to literature,
though the patronage of literature does seem to pose particular
problems. One wonders whether we cannot learn a great deal from the
experience of students in other disciplines. For one thing, I may
quibble at accepting a dedication as evidence for patronage, but
fifteen thirteenth-century manuscripts can be assigned to the patron-
age of one noble lady (who does admittedly have the right surn~me) by
what looks to the historian-outsider like wishful thinking. 8J And
what would an art historian make of a classicist's recent dictum that
'there did not exist a Roman code of literary patronage' because 'the
patrons did not conceive g£ themselves as playing a role in the
encouragement of letters'? We surely should not price ourselves
out of the market by setting impossible standards. And definitions
of personal patronage, attrg§tive as they are and applicable as they
are to Byzantine society, are probably too restricting for our
present purposes; they demand reciprocal exchange of different goods
and services, vnequal status of the partners and a personal link of
some duration. 80

7
For what does seem clear in the twg fth century is that writers,
like perhaps manuscript illuminators, were promiscuous in their
reception of patronage: Theodore Prodromos in claiming gg was never
the servant of many masters appears to be an exception. If one
tries to draw links between major writers in the twelfth century and
their patrons and patronesses, it becomes easier to group the patrons
around the writers than the writers around the patrons. 89 It has
always been assumed that literary circles radiated from their
patronesses, and therefore it was only necessary to establish all the
authors who wrote for a particular lady and a circle would be estab-
lished. But it does not seem to have worked like that. Nicholas
Kallikles wrote for Eirene Doukaina, the Sebastokratorissa Eirene,
John Arbanites and his wife Anna, the wife of Gregory Kamateros, the
widow of Roger th~ sebastos and the husband of Dokeian~ as well as
for John Comnenus; 0 Theodore Prodromos wrote for Eirene Doukaina,
then for the court of John, then for the family of Anna Comnena,
Nicephorus Katakalon Euphorbenos, the Kontostephanoi, Alexius Ariste-
nos, Theodore Stypeiotes his ex-pupil, Zoe Botaneiatissa, Constantine
Alopos, Andronikos Kamateros, the Sebastokrator Isaac as well as for
the Sebastokratorissa Eirene; the 'innumerable occasional poems' he
is supposed to have written for her boil down in fact to four. 91
John Tzetzes wrote for Bertha von Sulzbach as well as for the Sebas-
tokratorissa Eirene, and when she died while he was in mid-work, he
replaced her with a new patro~ for his Iliad Allegories, whom he
found in Constantine Kotertzes. 2

Having considered the question of what patronage for our pur-

poses is or is not, there is another question to address: how in-
fluential was patronage in twelfth-century Byzantium? I suspect art
historians would be surprised if I were to suggest that the patron
did not have ~ say in the appearance, and indeed the quality, of a
work of art,9 yet Byzantinists to my knowledge have only recently
started to follow classicists in making the same assumptions for
literature. 94 Or, to put it another way: are we to look to changes
and new developments in literature as evidence of the tastes and
preferences of patrons? Did a twelfth-century nobleman get up one
morning and say: How shall I spend that small windfall? Shall I
found a monastery, or shall I commission a psalter, or do I need some

417
 more light reading, in which case let us have an adventure story this
time? Claims almost of this kind have recently been made for patron-
age; the twelfth-century romances, the development of politikos
stikhos, the popularity of Digenes Akrites are all attractively
explained by the demands of aristocratic patrons whose education (for
whatever reason) was not up to the stand§rds of the literary circles
they moved and were expected to move in. 5

Then again, there has been a great deal of discussion of a new

professional class of writers in the twelfth century, closer to
artisans than to the court, who would be more susceptible to the
pressures of patronage than the Beamtenliteraten of an earlier age. 96
There were freelance writers and kept poets and pensioners, but in
general I am quite sure than the origins of writers were quite dispa-
rate and they were employed in various occupations. Professionalism
in writing may have been born in the twelfth-century - but it failed
to catch on - or caught on only at a literary level, in an increased
sense of corporate awareness of what it was to be a writer: this
perhaps m~1ht be the explanation of the begging poems of a
Prodromos. It seems to me that a great deal of what is new in
twelfth-century literature can be explained by patronage, but a great
deal can only be explained within individual writers or within a
corporate set of values, perhaps encouraged by the theatra. Noble
ladies with Western connexions might have commissioned the twelfth-
century romances for reasons of their own, but only the asteiotes of
a group alive to rhetorical kompsa would have fully appreciated th~
contemporary allusions and parody contained in them. The satirg boom
(although the Philopatris is now placed in the tent~ century),9 the
literary parody in Skhede muos and Katomyomakhia, 9 the taste for
mock debates (Demetrios Tornikes put NiketaB Choniates up to defend-
ing the superiority of winter over summer),l 0 riddles, poems thick
with quotations for one's friends to identify and the new schedo-
graphiol not so much a way of teaching grammar, more a tortuous
game: all these are symptoms of the same phenomenon. Three of
the four writers of novels incidentally produced a schedographical
work, and the last produced a collec~ion of riddles. 102 And, as
Alexander Kazhdan has shown, there may be no need to seek outside
Byzantium at this time for the origins of a taste for romance. I03

Perhaps the mistake we have been making is to look for institu-

tions rather than individuals and relationships between individuals.
Professor Browning wrote: 'The more one examines Byzantine litera-
ture, the more one becomes convinced that it never appears sP~8-
taneously: it needs a salon, patronage, institutionalised forms'. 4
To some extent this is clearly true, but the arbitrary arrangement of
such forms by modern scholars leads to distortions. To place Michael
Italikos in the circle of the Patriarchal Academy while his associate
Theodore Prodromos is assigned to the circle of the Sebastokratorissa
Eirene seems ludicrous. I wonder whether a more writer-centred
analysis might not provide more helpful results. Rather than rely on
the flimsy evidence for theatra and clumsy groupings together of
proteges, we should perhaps turn to the writers themselves and try to
work out the links among them; only then can we see if they fall into
any clear grouping.

This is clearly not the place for such an ambitious exer8!se.

What is needed is nothing less than a complete network analysis 1 of

418
 the writers of the period, together with their literary connections.
Only this will give us the dynamics of literary society. We need to
establish connections among writers and their public, connexions of
family, education, acquaintance, patronage and friendship, and then
we can really begin to see what made literary society function, and
we will find that we have done in fact what Bowersock counselled uS
against: we shall have begun to flesh out our ghostly theatra with
real groupings of real writers with real careers - and will then no
longer need to have recourse to such question-begging terms as
'literary circle'.

It must be admitted that a great deal of groundwork still has to

be done before such an analysis is possible. It involves the whole
process of establishing exactly what was written and what survives
over the whole period, together with complex problems of authorship
and date, particularly with the rhetorical and poetic material. And
t~e stysgnt, despite the publication of the new series of Hand-
bucher, still has to complete this collection of material for
himself. Yet the material is extraordinarily useful for our enter-
prise: clearly not every literary work will yield some kind of
information of a social kind, but some kinds of sources are more
useful than others. The social importance of rhetoric has recently
been emphasised: Lemerle speaks of the 'cultural qualification' of
rhetoric; 107 Sevcenko sees' rhetoric as 'a distinctive badge either
of membership of the upper class or associati08 with it', 'without
rhetoric no-ona was able to succeed in life,.l 8 It becomes clear
that literacyl 9 is no indicator at all in our task of defining
literary society: what is at stake is 'rhetoricity'; the ability to
understand and derive entertainment llO or instruction from, whether
visually or aurally,lll WOfk~ written at the lowest in middle style,
and for most in high style. 1 This is the minimum entrance fee f.n
membership of the society. Not all its members need have understood
all the nuances, or been able to practise them themselves, but a
minimum appreciation - and therefore education - was necessary. The
Comnenian· period gives us an unrivalled opportunity, for while the
number of significant writers is fairly limited, there is a great
mass of rhetorical material (not all published) whi~h will allow us
to construct sequences and indicate connections. ll ] Funeral ora-
tions, while heavily rhetorical and ceremonial, at least indicate the
connection of speaker and the dead and may throw considerable light
on the career of the dead; monodies afforded more opportunity for
personal reaction, but were traditionally offered to those who died
young and may thus autoTatically exclude members of literary
society;114 basilikoi 10gOi l ) and inaugural lectures l16 quite apart
from their topical value allow us to reconstruct sequences of offi-
cials and individual careers. And it is a period when writers prac-
tised in several genres: to look only at the historians many were as
well or better known for work other than history: Zonaras for his
canonical work, Glykas for his exegesis, Kinnamos for his progymnas-
mata and so on. We are surely dealing with a profoundly interconnec-
ted literary society, not a series of disconnected publics. With
this evidence we have a good chance of drawing up fairly complete
tables of active writers and the genres in which they practised,
their families, education and careers and a parallel register of
passive literati, or the people with whom they had literary rela-
tionships. We need to arrive at criteria for deciding that X taught
Y and for estimating the importance of such teaching links. We may

419
 then produce network diagrams of the interrelationships of patronage,
kinship, teaching and acquaintanceship, which will show us isolated
figures and those with complex links, and we can begin to set this
picture against our general understanding of social changes over the
period.

In addition we need to gain some idea of the quality of rela-

tion-ships among our literati. Certain basic relationships may be
hard to fathom: ¥upils and spiritual sons are notoriously difficult
to disentangle. II And when we reach the difficult ground of
intimacy, we will find that far and away the most useful genre is the
letter. It brings its own difficulties with it: quite apart from
the pro~tgm of whether there were close relationships in Byzantine
society, those who have worked with letters are well aware of the
illusory charge of intimacy the genre brings with it: a kindly,
diplomatic and charitable man like Peter the Venerable serT~ to be on
terms of close friendship with everyone in Christendom'. And it
is always hard on literf~~ who have left no letter-collection, or one
inadequately collected.

But like no other source letters show up just like a stain on a

microscope slide the social and ceremonial relations ~f individuals.
I believe that it is possi~~~, by a minute analysis l I of such vari-
ables as forms of address, length, level of style, subject-matter
and general tone, to establish gradations of intimacy over a single
correspondence. Of course, there are the barriers of protocol and of
literary form; they have to be negotiated with care, not jumped, but
the student who learns to find his way through them may also have
access to a three-dimensional view of literary society.

Having built up this picture of literary society we should then

be able to determine what made it work, and answer many of the
questions we have been posing above. How important was the family?
Precisely who was educated? In what offices were the ablest intel-
lects to be found? How useful to a brilliant career in administra-
tion was membership of literary society? What proportion of writers
were employed as writers? And to what 'extent were writers otherwise
employed thought of as writers rather than officials or clerics?
Were patrons members of theatra? How far down did a rhetorical
education go? To what extent did regional loyalties establish them-
selves towards the end of the period? What effect did politics have
on literature? Did literary circles become more closed towards the
end of the period? How far did the Fourth Crusade disrupt patterns
of literary and social behaviour? And indeed, what was the social
basis of Comnenian literature?

This is clearly not the only way forward, and it necessarily en-
tails many others. Detailed combing of the sources is bound to turn
up a much richer documentation of theatra and patrons. Continued
investigation into manuscripts and the relationship of works within
manuscripts, together with work on their ownership may well make us
rethink our views on readership, just as an awar~~ess of recent work
on the strength of orality in the Medieval West may make us more
alive to the effect of aural reception of Byzantine literature. I
have suggested that we need to examine very closely with the help of
students of other forms of patronage the precise nature of Comnenian
patronage. We cannot do this without further work on what Kazhdan

420
 calls 'social localization'; his devastatingly exhaustive
'Attaleiates,124 is an example to all, while Paul Magdalino has
alerted us to the equally important question of what writers thought
was their social position. If above I appear to have dismissed too
lightly the Stilstufen approach, I should echo here the words of
Sevcenko: 'Relating levels of style to social and educational levels
of the writers and their reading or listening public is among the
most challenging tasks of our discipline. As yet we have only begun
to look for soiutions but we do see the problems.,12J

Social localization has so far been practised largely on wri-

ters, for the obvious reason that they provide uS with the informa-
tion. A great deal of the groundwork has been done, and not just in
individual cases: a great deal of counting has gone on in recent
years. Lemerle's all-too attractive figures for the tenth cen-
tury,126 Kazdan's figures in People and Power, Beck's in Der byzan-
tinische Jahrtausend, Sevcenko's admirable analysis of fourteenth-
century literatil~7 all attempt to quantify literary society. There
are obvious dangers in working with autobiographical material, which
they have mostly skirted: Sevcenko warns us of the tendency of
twelfth-century writers to play down their social origins, as if, as
quite recently, a working-class background was de rigueur for aspir-
ing literati. Magdalino war'ns us of the opposite: they are closer
to the people, he suggests, than they care to admit. 128 We shou12
beware of any too easy identification of intellectuals with the
aristocracy, even the aristocracy of taste, while in a highly mobi~2
society it seems unhelpful to place too much emphasis on origir.s:
the symposium seemed to agree very well that aristocracy depended not
so much on where you came from as where you got to. The Jeffreys in
their headcount of writers given at the symposium and not published
here, modelled very much on §eveenko's for the fourteenth century,
emphasised this common ground by basing their assessment of social
position on the rank at which a writer ended his career.

Their results, based on 122 writers of 198 works, tally very

closely with my own researches based on a sample of 100 'positive'
literati. 129 Of these, an overwhelming proportion, two-thirds of the
sample, are ecclesiastical; bishops alone account for over one-third,
and together with patriarchs make up one-half. Teachers (including
higher education) come next, closely followed by civil officials,
then ecclesiastical officials and royalty. Freelance writers and
'kept poets' are a tiny minority, at about the level of soldiers,
doctors and secretaries. There are several reasons why the propor-
tion of ecclesiastics may be exaggerated: for one thing their career
is easy to identify, and so they find their way easily into any
sample, second, appointment to a see or the tonsure could be the acme
or the close of many varied careers;130 finally, certain writings are
determined by career, and everyone knows that the largest body of
Comnenian literature is religious polemic. But 55% of Sevcenko's
sample were ecclesiastic, 'so it is possible that my figures are not
distorted after all.

Writer-centred analysis is all very well, but to get a balanced

picture of literary society, we need to go further and carry out a
similar study of the receivers as well as the producers of litera-
ture. If all dedicatees, commissioners, addressees of speeches and
letters, or people mentioned in poems are added together, they should

421
 constitute the public for 'Comnenian literature'. And producers and
receivers, 'positive' and 'negative', 'active' and 'passive' literati
should all together constitute Comnenian literary society.

Again it is premature to attempt a final headcount of receivers:

far greater problems of prosopography arise where we have not the
benefit of autobiographical information, however misleading. But
there are certain pointers from a pilot study. Taking a similar
sample of 100 'negative' literati, the same careers are represented,
but with marked shifts of emphasis. The largest single category is
now civil officials, followed by patriarchs (swollen of course by the
traditional Lazarus Saturday speech), royalty, not including empe-
rors, then a considerable drop before bishops, at less than half and
monks at less than a quarter of their 'positive' showing. On the
other hand, the showing of ecclesiastical officials is very much the
same, though low, and soldiers, doctors, lower clergy and grammatikoi
are notably more strongly represented than in the 'positive' listing
and have not been counted twice. The low monastic showing may well
be a result of lack of opportunity (to receive personal letters, or
to commission works of art), or of survival ('positive' monastic
literati may have had their works preserved for pious reasons, but
letters to obscure monks may not have been prestigious enough for
authors to wish to preserve). Clearly speculation could be rife on
this and other issues, but it would be safer to wait for hard re-
sults. What is clear, surely, is that the same categories of occupa-
tion or ra~k are represented in both counts: there is no question of
'rupture,l 1 between the producers and receivers of literature in the
Comnenian age.

So with this clearer picture, however imprecise, of the composi-

tion of Comnenian literary society, we may perhaps return to the
questions that WE posed at the beginning. We discarded 'who paid for
Comnenian literature?' as an insufficiently subtle tool for measuring
the complex issues of patronage, commission and professionalism
raised by -it. In our understanding of the social base of Byzantine
literature, patrcnage is not all, nor is it all to our understanding
of the processes of reception, not in any assessment of the contribu-
tion of the Comnenian aristocracy to tbe literature of the period.
We have seen that the composition of literary society in the twelfth
century, while varied in origin, was remarkably homogeneous and
intricately interconnected. While I am aware that there is no such
thing as a single homogeneous reading or listening public,l32 Com-
nenian conditions are suitable for the application of Goldmann's
approach i~ Le dieu cache, where there is far more than a minimal
consr~~us,l- 3 or to put it another way, where the horizon of expecta-
tion is common to the whole of a work's readers 'in the historical
mo~ent of its appearanc~' .135 In a far more real sense than th~t of
the 'implied reader',136 'inscribed reader,137 or 'narratee,lJ8 of
modern literary criticism, Comnenian literature, its producers and
receivers, were part of a single social organism.

It only remains to ask how aristocratic was this literary

society. And by now we have learnt to be wary of easy answers. It
is noticeable how often scholars refer to terms like 'elite,139 or
'upper class,140 rather than 'aristocracy' while treating of litera-
ture. And one really needs to have digested this present volume,
with its sophisticated discussions of definitions, before tackling

422
this final problem. It might seem an easy solution to take each
individual member of literary society as we have defined i t and look
him or her up in Kazhdan's Sostav,141 the Debrett of the Comnenian
world, but this might seem t~too much stress on aristocracy of
birth rather than that of wealth or rank or profession or taste.
Other techniques might be equally onesided. We certainly cannot opt
out by assuming that anyone who has made it to our 'literary society'
has automatically made it to the aristocracy: that would be to fall
into the intellectuals' own trap. By the same token it is useless to
adopt the standards of Comnenian ideology and measure literary
society like synodal lists by closeness of kinship to the emperor.

At any rate we can now discard most of the foreign analogies

which are so often used by Byzantine literary society. At this
period at least, literary society is hardly predominantly mandarin if
by that is m2ant the civil bureaucracy and the patriarchal archontes.
It cannot be a caste society, since higher clergy predominate and
thus rule out ~hereditary principle 142 - in any case mobility
seems the keynote of this period. And a literary society so intri-
cately inter-connected with very few sub-groups is hardly a matter of
coteries, though one should not rule out the effects of literary
friendships. I suspect also that twelfth-century literature is a far
more complex matter than the ~owbrow tastes of the court satisfied by
highly educated but essentially artis,an wri ters. (Here parallels
with the visual arts may be less than helpful). It was more aristo-
cratic than we might have imagined in the sense that a wider range of
aristocratic society is represented: it found room for the less-
than-top-df~~er Kekaumenos, whose rhetoricity should not be under-
estimated, as well as for smart military families. The civil
aristocracy, even if we believe in a sharp distinction, did not have
a monopoly. Robert Browning in his fundame.ntal article on 11 teracy
writes as if soldiers were unlikely to have had a rhetorical train-
ing144 - but we know from the letters of teachers to their ex-pupils
that some did. 145 On the other hand, a great deal of twelfth-
century literary activity was carried on independently of the court,
and writers were not just its voice. To that extent literary society
was less aristocratic than it has been pictured.

We might settle then for this rather unsatisfactory hypothesis

that Comnenian literary society was both more and less aristocratic
than has been thought - but of course no hypothesis is any substitute
for the hard work which remains to be done.

423
 NOTES

1. This mirrors the conviction of historians, e.g. M. Angold,

A Byzantine Government in Exile (Oxford, 1975), p. 141 ff., and
M.F. Hendy, Coinage and Money in the Byzantine Empire 1081-1261
(Washington D.C., 1969), p. 320, that there was no economic or
social cut-off at 1204: the importance of the period of the
late twelfth and early thirteenth centuries was emphasised
again and again in Paul Magdalino's discussion group at the
Birmingham symposium in 1983.

2. For the 'Helladic connexion' which bound together the influen-

tial group of twelfth-century clerics which comprised the Tor-
nikai, Euthymios Malakes, Michael Choniates and John Apokaukos,
see J. Darrouzes, 'Notes sur Euthyme Tornikes, Euthyme Malakes
et Georges Tornikes', REB, 23(1965), pp. 148-167, esp. p. 162.
The cut-off surely comeS-With the last of these literati to be
educated in Constantinople: George Bardanes, a pupil of
Michael Choniates at Athens, appears to be the first of a new
province-centred group.

3. P. Lemerle, Prolegomenes a une edition critique et commentee

des 'Conseils et Recits' de Kekaumenos (Academie royale de Bel-
gique, classe des lettres, memoires, 54, 1, Brussels, 1960), p.
95: 'litterature sans public et sans problemes'; R. Jenkins,
'The Hellenistic Origins of Byzantine Literature', DOP,
17 (1963), p. 40; 'no secular literature was written for a Wide
public, since no such public existed'.

4. E.g. C. Mango, Byzantine Literature as a Distorting Mirror

(Oxford, 1975).

5. E.g. H.-G. Beck, Das 1iterarische Schaffen der Byzantiner.

Wege zu seinem Verstandnis (Osterreichische Akademie der Wis-
senschaften, phil-hist. K1asse; Sitzungsberichte, 294, 4.
Abhandlung, Vienna, 1974).

6. R. Browning, 'Enlightenment and Repression in Byzantium in the

Eleventh and Twelfth Centuries', Past and Present, 69(1975),
p.5.

7. H. Hunger, 'Die byzantinische Literatur der Komnenenzeit: Ver-

such einer Neubewertung', Anzeiger der philologische-historis-
chen Klasse der osterreichischen Akademie, 105 (1968), pp. 59-
76.

8. For the Christos Paschon, see M. Alexiou, The Ritual Lament in

Greek Tradition (Cambridge, 1974), p. 64; for an early date,
see A. Tuilier, Gregoire de Nazianze, La Passion de Christ,
tragedie (Sources chretiennes, 149, Paris, 1969), pp. 11-74.
Hunger in /On the Imitation (MIMESIS) of Antiquity in Byzantine
Literature', DOP, 23/24 (1969-70), p. 34 ff., takes it for
granted that i~s a twelfth-century work. For Digenes, see N.
Oikonomides, 'L'epopee de Digenes et la frontiere orientale de
Byzance au Xe et XI e siecle', TH, 7 (1979), pp. 375-97; E.
Jeffreys, 'The Comnenian Background to the "romans d'anti-
quite": !, 50 (1980), pp. 455-86; R. Beaton, 'Was Digenes

424
 Akrites an oral poem?', BMGS, 7(1981), pp. 7-27; idem, 'Digenes
Akrites and Modern Greek folksong: a reassessment', B, 51
(1981), 23-43. On the Belisarius romance, see the PhD thesis
of Maria Fotina, A Critical Edition of the Medieval Greek Poems
on Belisarius (London, 1973).

9. E.g. H. Hunger, Prooimion: E1emente der byzantinischen Kaiser-

idee in den Arengen der Urkunden (Wiener byzantinische Studien,
1, Vienna, 1964); 'Aspekte der griechischen Rhetorik von Gor-
gias bis zum Untergang von Byzanz', Sitzungsberichte der oster-
reichischen Akademie der Wissenschaften, phil.-hist. Klasse,
277, Abh. 3 (Vienna, 1972); 'The Classical Tradition in Byzan-
tine Literature: the Importance of Rhetoric', Byzantium and
the Classical Tradition, ed. M. Mullett and R. Scott (Birming-
ham, 1981), pp. 35-47.

10. E.g. R. Browning, 'The Patriarchal School at Constantinople in

the Twelfth Century', B 32(1962), pp. 167-201; B 33(1963), pp.
11-40; 'Byzantine Scholarship', Past and Present~ 28(1964), pp.
3-22.

11. Eo g •J. Dar r 0 u z e s, -:Do..:. . :C-;,U. ::m.. ,e"'n. .,t. ::s.. . .::i_n__ec..;d...,i';-t:..:s=----:d';-'_e':-c;.-c':-l:..e.:...s.:...i,,0,,1707'<'g..::i..::e."..::b,:..y.;:z..::a:::;n;.:t.;:i:::;n:,;:e
(Archives de I' orient chretien, 10, Paris, 1966); Recherches
sur les offikia de l'eglise byzantine (Archives de l'orient
chretien, 11, Paris, 1970).

12. E.g. A.P. Kazhdan, Social'ny sostav gospodstvujushchego klassa

Vizantii XI-XII vv (Moscow, 1974); A. Hohlweg, Beitrage zur
Verwaltungsgeschichte des ostromischen Reiches unter den Komne-
nen (Miscellanea Byzantina Monacensia, 1, Munieh, 1965); A.P.
Kazhdan, review of P. Lemerle, Cinq etudes sur Ie Xl e siecle
byzantin Paris, 1977) in ~ 49 (1979), pp. 491-503, at p. 498ff.

13. See E.M. and M.J. Jeffreys, Popular Literature in Late Byzan-
tium (London, 1983), and 'On the Rise of Vernacular Literatures
~the Middle Ages', 'The Literary Emergence of Vernacular
Greek'. Mosaic, 8/4, (1975) pp. 171-93; for Stilstufen, H.
Hunger, 'Stilstufen in der byzantinischen Geschichtsschreibung
des 12. Jahrhunderts: Anna Komnene und Michael Glykas', Byzan-
tine Studies Etudes byzantines, 5( 1978), 139-170 and his
Anonyme Metaphrase zu Anna Komnene Alexias XII-XIII. Ein Beit-
rag zur Erschliessung byzantinischer Umgangssprache (Wiener
Byzantinische Studien, 15, Vienna, 1981). For social standing
and author's self-perception, see A.P. Kazhdan, 'Der Mensch in
der byzantinischen Literaturgeschichte', JBBG, 28 (1979), pp.
1-21, and Magdalino above Chapter 4, pp.58~

14. The inaugural lecture of C. Mango, see above n. 4, acted as a

powerful stimulus to scholars who could not accept the dispar-
aging judgements of that lecture and of R. Jenkins in Diony-
sius Solomos (Cambridge, 1940), p. 57. Already Hunger's MIME-
SIS, see n. 8 and G.L. Kustas, Studies in Byzantine Rhetoric
(Analekta Blatadon, 17, Thessalonike, 1973), had proposed an
alternative aesthetic. Margaret Alexiou's work on narrative
technique (in 'A Critical Reappraisal of Eustathius Makrem-
bolites' Hysmine and Hysminias', BMGS, 3 (1977), 23-43) and on
irony (see her study of the Timarion-In BMGS 8(1983), pp.29-45)

425
 are designed to show the literary qualities of Byzantine liter-
ature, and several contributions to Byzantium and the Classi-
cal Tradition were designed to show development in Byzantine
literature and its independence of classical models. With the
literary discussion at the Birmingham Symposium 1983 and the
appearance of A.P. Kazhdan and S. Franklin, Studies on Byzan-
tine Literature of the Eleventh and Twelfth Centuries (Cam-
bridge, 1984) the wind of change is quite apparent.

15. F. Chalandon, Jean II et Manuel I Comnene, 11.1, Paris, 1912,

repro New York, 1971), e.g. pp. 14, 213.

16. R. Browning, 'An Unpublished Funeral Oration on Anna Comnena',

Proceedings of the Cambridge Philological Society, 188 (n.s.8,
1962), p.7.

17. M. Innes, The Ampersand Papers (Harmondsworth, 1980), p. 19.

18. G. Bowersock, Greek Sophists in the Roman Empire (Oxford,

1969), ch. 8, pp. 101-9. Other classical circles under sus-
picion at the moment are those of Scipio Aemilianus, see J.E.G.
Zetzel, 'Cicero and the Scipionic Circle', Harvard Studies in
Classical Philology, 76(1972), pp. 173-9, and of Symmachus, see
A. Cameron, 'The Date and Identity of Macrobius', JRS,
56(1966), pp. 33-38. I am grateful to Oliver Nicholson and
Marie Taylor Davis for suggesting these parallels and discus-
sing them with me.

19. Bowersock, op. cit., p. 101.

20. H. Hunger, Reich der neuen Mitte (Graz, Vienna and Cologue,
1965), p. 341.

21. See Synesius, ep. 101, ed. R. Hercher, Epistolographi Graeci

(Paris, 1873), p. 698; Manuel II, epp. 9,24, 27, 32, 34, 44,
61, ed. G.T. Dennis (Washington D.C., 1977), pp. 25, 37,87,
95, 119, 171 and see Hunger, loco cit.

22. Theophylact, epp. Vat Il, Meurs VIII, Vat V, Migne, PG 126,
cols 309, 372-3, 317.

23. Ed. G.A. Hirschig, Scriptores Erotici Graeci, II (Paris, 1859),

pp. 161-286 at p. 161. Margaret Alexiou first suggested this
to me; I find a similar note in Hunger, Anzeiger phil-hist.
Klasse der osterreichischen Akademie der Wissenscheften, 105
(1968), pp. 73-4.

24. Ed. P. Gautier, Michel Italikos, Lettres et Discours (Paris,

1972), p. 154.

25. ibid. , no. 15, pp. 1.46-7.

26. ibid. , p. 147.

27. ibid. , no. 11, p. 86.

426
 28. Ed. J. Darrouzes, Georges et Demetrios Tornikes, Lettres et
Discours (Paris, 1970), no. 14, p. 255 11. 11, 38 ff.

29. Esc. Gr. Y-11-10, fol. 324 r ff. I am extremely grateful to

Paul Magdalino for offering me his transcript and to him and
Roger Scott for discussing the passage with me.

30. Niketas Stethatos, Bios kai Politeia tou en agiois patros @m6n
Symeon tou neou Theologou presbyterou egoumenou mones tou agiou
Mamantos tib Xenokerkiou, ch. 100, 109, ed. 1. Hausherr, (OC
12, Rome, 1928), pp. 138, 142; Vita Basilii Junioris, ch. 2~
AASS, March III, pp. 20-32. lowe these references (and much
else) to Rosemary Morris. See above, R. Morris, Chapter 7,
pp.ll2ff.

31. cf. Wilson, Byzantine Books and Bookmen, p. 13.

32. E.g. Theophylact of Ochrid, Paideia Basilike, ed. P. Gautier,

Theophylacti Achridensis Orationes, Tractatus, Carmina (Thessa-
lonike, 1980), p. 183, line 18.

33. See the interesting suggestions on the relationship between

aurally received literature, patronage and art in H. Maguire,
Art and Eloquence in Byzantium (Princeton, 1981), pp. 109-111.

34. Benjamin of Tudela, ed. tr. M.N.· Adler (London, 1907), p.

13ff.; Odo of Deuil, ed. tr. G.J. McLeod (Columbia, 1948), p.
65 ff; see also S. Runciman, 'Blachernae Palace and its Decora-
tion', Studies in Memory of D. Talbot Rice (Edinburgh, 1975),
pp. 277-83; P. Magdalino, 'Manuel Komnenos and the Great
Palace', BMGS, 4 (1978), pp. 101-114; see above, L.-A. Hunt,
Chapter 8,~ 138 ff.

35. See L. Stiernon's brilliant series, 'Notes de prosopographie et

de titulature byzantine' REB, 17(1959), pp. 90-126; 19(1961),
pp. 273-83; 21(1963), pp. 173-98; 23(1965), pp. 223-43;
24(1966), pp. 89-96; N. Oikonomides, 'L'evolution de l'organi-
sation administrative de l'empire byzantin au Xr e siecle 1025-
1118', TM 6(1976), pp. 125-152 at pp. 151-2. r have elsewhere,
BS, 45-C1984), n. 57 expressed reservations about the use of
the term.

36. E.g. in N. Kataskepenos, Bios kai Politeia kai Merike Thaumat6n

Diegesis tou Osiou Patros emon Kyrillou tou Phileotou, ch. 47,
ed. E. Sargologos (Subsidia Hagiographica, 39, Brussels, 1964),
pp. 225-35.

37. On ceremonial see W. Horandner, Theodoros Prodromos, Histori-

sche Gedichte (Wiener byzantinische Studien, 11, Vienna, 1974),
pp. 79-88; on jousting, S. Lampros, 'Ekphrasis ton xylokon-
tarion tou krataiou kai hagiou hemon authentou kal basileos',
Neos Hellenomnemon, 5 (1908), pp. 3-18.

38. Contrast Anna's picture of Alexius and Theodore Prodromos's of

John with Niketas Choniates's less idealised view of Manuel's
life-style. For suburban life see Chalandon, Les Comnenes,
11.1, pp. 236 ff; for anakainesis C. Mango, 'The Conciliar

427
 Edict of 1166', DOP, 17(1963), pp. 315-30.

39. For Alexius, see his Ekdotheis par autou pros Armenious doxa-
zontas kakos mian physin epi Christa, ed. A. Papadopoulos-
Kerameus, Analekta Ierosolymitika, I (St Petersburg 1891), pp.
116-23; Mousai, ed. P. Maas, 'Die Musen des Kaisers Alexios',
BZ, 22(1~pp. 48-69. On the identity of the Sebastokrator
see Horandner, 'Poesie profane au XI e siecle et la connaissance
des auteurs anciens', TM, 6(1976), n. 55. For editions, see
J.J. Rizzo, Isaac Sebastokrator, Peri tes kakon hypostaseos
(Beitrage zur klassischen Philologie, 42, Meisenheim am Glan,
1971), and J. Dornseiff, Isaak Sebastokrator, Zehn Aporien
uber die Vorsehung', (Beitrage zur klassischen Philologie, 19,
Meisenheim am Glan, 1966).

40. I. Sevcenko, 'Society and Intellectual Life in the Fourteenth

Century', Actes du XIVe Congres international Des ~tudes byzan-
tines I (Bucharest, 1974), pp. 69-92 at pp. 79-83.

41. This is not to quarrel with the view that there was 'a consis-
tent dynastic policy to display imperial magnificence to the
full', see P. Magdalino and R. Nelson, 'The Emperor in Byzan-
tine Art of the Twelfth Century', Byzantinische Forschungen, 8
(1982), pp. 123-83, at p. 170, an excellent and exhaustive
study which underlines the relationship between official art
and official literature, together with the special case of
Manuel I, without portraying either a eourt art or a court
literature.

42. Ed. L. Sternbach, 'Nicolai Calliclis Carmina', Rosprawy Akade-

mii Umiejetnosci Wydziat Filologiczny (Cracow, 1904), pp. 335-
92. The new edition, Nicola Calliele, Carmi by R. Romano
(Naples, 1980) had not reached me when I completed this paper.

43. Euthymios Zigabenos, PanopIia Dogmatike, Minge, PG 130; Stephen

Physopalamites, ed. C. Welz, Analecta Byzantina, Carmina
Inedita (Diss., Leipzig, 1910), pp. 54-9; John Kamateros,
Eisagoge Astronomias, ed. F. Wiegl (Leipzig and Berlin, 1908);
Michael Glykas, ed. E. Legrand, Bibliotheque de grec vulgaire,
I, pp. 18-37. Obvious additions are Ptoehoprodromos poems I
and IV, ed. D.C. Hesseling and H. Pernot,
Poemes prodromiques en grec vulgaire (Amsterdam, 1910), pp. 30-
7; 72-83; Manganeios Prodromos, poems III-XII, ed. S. Bernardi-
nello (Universita di Padova, Studi biz. e neogreci, 4, Padua,
1972), pp. 36-80.

44. No. 18, ed. Gautier, p. 155-9.

45. See N. Wilson, 'Books and Readers in Byzantium', Byzantine

Books and Bookmen (Washington, D.C., 1975), p. 6.

46. Browning, Proceedings of the Cambridge Philosophical Society,

188 (1962), p. 7.

47. For Basilakes see the studies of A. Garzya, Storia e interpre-

tazione di testi bizantini (London, 1974), esp. VIII, IX, X.
XI, XII.

428
 48. See R. Browning, '0 Markianos Ellenikos Kodikos XI.3 kai e
Byzantine Schedographia', Parnassos, n.s. 14(1973), pp. 506-19.

49. Note for example that the introduction of a recent work on

Roman literary patronage views literary patronage and patronage
of a salon as mutually exclusive and successive phases of
support for the arts: 'In the nineteenth and early twentieth
century patronage was frequently replaced by literary circles
gathered round a central artistic or literary figure like
Gertrude Stein', B.K. Gold, ed. Literary and Artistic Patronage
in Ancient Rome (Austin, Texas, L982), p. xi.

50. For greater detail see my forthcoming 'The "Disgrace" of the

Ex-basilissa Maria', BS, 45, (1984).

51. Browning, Past and Present, 69(1975), 3-23; L. Clucas, The

Trial of John Italos (Byzantina Monacensia, 26, Munich, 1981~

52. See my 'Maria', ~, 45( 1984), notes 5-7.

53. Anna Comnena, Alexiad, III. viii.3, ed. B. Leib, I (Paris,

1967), p. 126; see S. Runciman's suggestive 'The End of Anna
Dalassena', Annuaire de ~'Institut de philologie et d'histoire
orientale et slave, 9(1949), pp. 517-24.

54. Zonaras (Bonn), III, p. 747; P. Gautier, 'Vobituaire du typi-

kon du Pantokrator', REB, 27(1969), pp. 235-62 at 245-7.

55. Nicephorus Bryennios, Hyle Historias, Prooimion, ed. P. Gautier

(Brussels, 1973), p. 71; Anna Comnena, Alexiad, Prooimion, ed.
Leib, I, pp. 5-6. A certain amount of caution is advisable in
view of the frequency of such claims in historical introduc-
tions.

56. Nicholas Kataskepenos, epp. ed. M. Gedeon, Archeion ekklekia-

stikes historias, 1(1911), pp. 60-2; 70-2. Theophylact's ep.
Meurs LIl to the despoina who visited him in his illness is
taken by D. Polemis, The Doukai, a Contribution to Byzantine
Prosopography (University of London Historical Studies, 22,
London, 1968), p. 73 to be Eirene. We await Gautier's new
edition.

57. Nicholas Kalltkles, poems 2, 7, ed. Sternbach, pp. 319, 341-2;

Manuel Straboromanos, consolatio, ed. P. Gautier, 'Le dossier
de Manuel Straboromanos', REB, 23(1965), pp. 168-204 at 195-
201; Theodore Prodromos, ed:-Horandner, pp. 185-8.

58. Browning, art. cit. (n. 46) 6-10.

59. Chalandon, Les Comn~nes, 11.2, pp. 213; Papadimitriu, Feodor

Prodrom (Odessa, 1905), pp. 24-7.

60. E.M. Jeffreys, B 50(1980), 455-86; 'The Sevastokratorissa

Eirene as Li terary Patroness: the Monk Iakovos', JOB 32/1
(1982), pp. 63-71 (= Atken des XVI. internationaler Byzantinis-
tenkongress, 11/3); -see-below E.M. Jeffreys, Chapter II, pp.
202-10.

429
 61. See M.J. Jeffreys, 'The Nature and Origins of the Political
Verse', DOP, 28(1974), pp. 141-95, esp. 148-76.

62. See S. Suleiman and I. Crosman, The Reader in the Text, Essays
in Audience and Interpretation (Princeton, 1980), for an antho-
logy of audience-centred criticism. Western mediaevalists have
long been aware of the problem, e.g. D. Whitelock, The Audience
of Beowulf (Oxford, 1957) and in general seem happy to deduce
audience from internal evidence; see J. Coleman, English Liter-
ature in History 1350-1400. Medieval Readers and Writers (Lon-
don, 1981), pp. 18-26, e.g. p. 20: >If Chaucer, as is commonly
believed, wrote for the court circle whose literary tastes were
French, why did he write in English?'. Cf. A. Kazhdan and G.
Constable, People and Power in Byzantium. An Introduction to
Modern Byzantine Studies (Washington, D.C., 1982), p. 102.

63. An exception is the intelligent and subtle treatment by R.

Beaton in his Folk Poetry of Modern Greece (Cambridge, 1980),
esp. pp. 179-92 and his studies of Digenes Akrites, see note 8.

64. Kazhdan, People and Power, p. 102 ff.; S.S. Averintsev, Poetika
rannevizantijskoj literatury (Moscow, 1977), pp. 183 209. I
must thank Natalia Teteriatnikov for her help with this book.

65. R. Crosby, 'Oral Delivery in the Middle Ages', Speculum,

11(1936), pp. 88-110; M. Curschmann, 'Oral Poetry in Medieval
English, French and German Literature: some notes on recent
research', Speculum, 42(1967), pp. 36-52.

66. On the persistence of mimes, see F. Tinnefeld, 'Zum profanen

Mimos in Byzanz nach dem Verdikt des Trullianums (691)', Byzan-
tina, 6(1974), pp. 323-43. It is usually stated tha~
draIDatia of the twelfth century, e.g. Theodore Prodromos, Epi
Apodemou te Philia, Minge, PG 133, cols. 1321-32, were not
intended for performance - bu~there is no evidence either way.
At Edinburgh the Jeffreys suggested that some of the
epithalamia of the period resemble masques.

67. Publicity: cf the remarks of T.P. Wiseman, '-Pete nobiles

amicos'. Poets and Patrons in Late Roman Rome', ed. Gold,
Literary and Artistic Patronage, pp. 28-49 at p.38. For him
what Roman patrons offered poets was an assured audience, a
claque, and if worst came to worst effective bouncers. Shar-
ing: cf. M.T. Clanchy, From Memory to Written Record. Eng-
land 1066-1307 (London 1979), p. 198: 'Books were scarce and it
was ordinary good manners to share their contents among a group
by reading aloud'. Editing: E. Auerbach, Literatursprache und
Publikum in der lateinischen Spatantike und im Mittelalter
(Bern, 1958).

68. H. Hunger, 'State and Society in Byzantium', Proceedings of the

Royal Irish Academy, 82.c.8 (1982), p. 208; he clearly sees the
theatra as more like music-halls than salons.

69. The eleventh-century revival of rhetoric passes unnoticed (un-

less in passing on p.304?) in G.A. Kennedy, Greek Rhetoric
under Christian Emperors (A History of Rhetoric, 3, Princeton,

430
 1983), but is taken for granted in Kazhdan, JOB 28(1979), p.
15. Most useful work has been done by J. Lefurt, 'Rhetorique
et politique: trois discours de Jean Mauropous en 1047', TM, 6
(1976), pp. 265-303, to whom I am also indebted for a copy of
his paper to the Athens congress, 'Rhetorique et symbo1ique:
l'obscurite dans les discours a l'empereur au Xl-XII siecles'.
For further remarks see my 'Maria', ~, 45(1984), n. 48.

70. R. Browning, 'Literacy in the Byzantine World', BMGS, 4(1978),

pp. 39-54; E. Patlagean, 'Discours ecrit, discours parle.
Niveaux de culture a Byzance aux Vllle-Xl e siecles',
Annales ESC, 34(1979), 264-75; M. Mundell, 'Patrons and Scribes
indicated in Syriac manuscripts 411-800 A.D.', JOB 32 (1982),
pp. 3-12.

71. See C. Mango, 'Books in the Byzantine Empire A.D. 750-850',

~B~y~z~a~n~t~i~n_e~B~o~o~k_s~'_a~n~d~B~o~o~km~e~n,
pp. 29-45 at p. 39 ; N. Wilson,
ibid., p. 3.

72. H. Hunger, Die hochsprachliche profane Literatur der Byzan-

tiner, I (Munich, 1978), pp. 257-78).

73. M.J. Jeffreys, 'The Vernacular eis~erioi for Agnes of France',

Byzantine Papers, ed. E. and M. Jeffreys and A. Moffatt (Byzan-
tina Australensia, I, Canberra, 1981), pp. 101-115; H.C. Eeck,
'Der Leserkreis der byzantinischen Volksliteratur im Licht der
handschrift1ichen Uberlieferung', Byzantine Books and Bockmen,
pp. 47-68 at p. 58: 'Eine Affare literarischer Kreise'.

74. See above, n. 8.

75. I. Sev{enko, 'Additional Remarks to the Report on the Levels of

Style', JaB 32/1 (1982), pp. 228-9.

76. On the literary implications of the appointment of literati to

provincial bishoprics and the influx into Constantinople of
scholazontes fleeing from the Turks see my 'CJ.assical Tradition
in the Byzantine Letter', Byzantium and the Classical Tradi-
tion, pp. 75-93 at p. 91.

77. e.g. Psellos, Anna Comnena, Zonaras, Kinnamos in our period.

78. For example J .E.G. Zetzel, 'The Poetics of Patronage in the

Late First Century B.C.', ed. Gold, Literary and Artistic
Patronage, pp. 87-102 at p. 88, reserves the term 'artistic
patronage' for what he calls the 'Renaissance type'.

79. Michael Italikos, nos. 5, 33, ed. Gautier, pp. 97, 209, and see
Gautier's discussion, pp. 20-1.

80. Theodore Prodromos, Peri tou megalou kai tou mikrou, kai tou
pol lou kai tou oligou. Oti ou ton pros ti eisin, alIa tou
posou, kai enantia, ed. P. Tannery, Annuaire de l'association
pour l'encouragement des etudes grecques en France, 21 (1887),
pp. 104 19.

81. Zetzel, loco cit.

431
 82. See n. 50 above.

83. H. Buchthal and H. Belting, Patronage in Thirteenth-century

Constantinople: an Atelier of Late Byzantine Book Illumination
and Calligraphy (Washington D.C., 1978), pp. 99-100; see the
reviews of A. Cutler in Speculum, 56(1981), pp. 100-5 and of G.
Vikan, in Art Bulletin, 63(1981), pp. 325-8. I have always
benefited enormously from Robin Cormack's historian's view of
the proJuction of art objects, (see his paper in this volume,
chapter 9 pp. 158ff), and I should also like to thank Tony
Cutler for the stimulus of discussion with him in the legendary
'favourable atmosphere' of Dumbarton Oaks in the spring of
1983.

84. P. White, 'Amici tia and the Profession of Poetry in Early

Imperial Rome', JRS, 68 (1978), pp. 74-92 at pp. 82-84.

85. As I hope to argue in a forthcoming paper on philia. White is

surely wrong in suggesting that amicitia is not patronage and
right in emphasising the social (rather than strictly literary)
nature of Early Imperial patronage of writers. In his more
developed paper, 'Positions for Poets in Early Imperial Rome',
ed. Gold, Literary and Artistic Patronage, pp. 50-66, literary
skill appears as an important >good or service' to exchange.
But the patronage of literature in twelfth-century Byzantium
was not necessarily organised under a similar model of personal
patronage.

86. R. Saller, Personal Patronage under the Early Empire (Cam-

bridge, 1982), p. I, based on J. Boissevain, 'Patronage in
Sicily', Man, n.s. 1(1966), p. 18; many definitions are offered
in the excellent E. Gellner and J. Waterbury, Patrons and
Clients in Mediterranean Societies (London, 1977).

87. Or at least miniaturists, see Buchthal and Bel ting, op. cit., p.
92; cf. Wilson on calligraphers, Byzantine Books and Bookmen,
p. 9.

88. See Kazhdan, Studies, ch. 3. I am greatly indebted to Profes-

sor Kazhdan for the opportunity to read this work in typescript.

89. Paul Magdalino suggests that this is also true for the
fourteenth century.

90. See Sternbach's edition (abo"e, no. 42).

91. See Horandner Theodoros Prodromos. The four poems are nos. 44-
7, pp. 405-34.

92. For references see Jeffreys, DOP, 28 (1974), pp. 150 ff.

93. See R. Cormack, 'Painting after Iconoclasm', Iconoclasm, ed.

A.A.M. Bryer and J. Herrin (Birmingham, 1977), pp. 147-163; A.
Cutler, 'Art in Byzantine Society: Motive forces of Byzantine
Patronage', (: ~kten Des SVI. internationaler Byzantinistenkon-
gress, 1/2). JOB 31/2 (1981), 759-87.

432
 94. Much of the concern of the second part of Gold, Literary and
Artistic Patronage is to show that patrons did not dictate
poetry, e.g. p. 101: 'it is not the poets who are the clients
but the patrons'. Interestingly, E.W. Leach in the same volume
is quite happy to posit patronal direction in the visual arts.

95. E.M. Jeffreys, B 50 (1980), pp. 45-86; JOB, 32(1982), pp. 63-
71; M.J. Jeffreys, DOP 28 (1974), pp. 41-95; Byzantine Papers
(CanberrA, 1981), pp.-rol-115.

96. See Kazhdan, e.g. People and Power, p. 102; Sevtenko, art. cit.
(n. 40), pp. 74-5; see above P. Magdalino, Chapter 4. For
Beamtenliteraten see G. Weiss, Ostromische Beamte im Spiegel
der Schriften des Michael Psellos (Miscellanea Monacensia, 16,
Munich, 1973).

97. Note the caution of Hunger art. cit. (n. 68) p. 208. Begging-
formulae appear in a wide variety of writings and seem charac-
teristic of twelfth-century poetry in East and West. On the
genre in the West, see N. Eliason, 'Two old English Scop
poems', Publications of the Modern Language Association of
America, 81(1966), pp. 185-92; 'Deor - A Begging-Poem?',
Medieval Literature and Criticism. Studies in Memory of G.N.
Garmonsway, ed. D.A. Pearsall and W.A. Waldron (London, 1969),
pp. 55-61.

98. On the satire boom (which awaits exhaustive study) see Beck iil
Byzantine Books and Bookmen, p. 51 ff; C. Robinson, Lucian and
his Influence in Europe (London, 1979), pp. 66-81; his study
could have been augmented by a mere glance at the works of
Theodore Prodromos listed by Horandner, op.cit. pp. 49-51.

99. See Sevcenko, JOB, 82(1982), p. 297; H. Hunger, Der byzan-

tinische Katz-M~e-Krieg (Vienna, 1968).

100. See J. L • Van Die ten , .:cN..:::iC;k:c:e:..:t:c:a:-=s'-;;-iC:..:h:..:o:..:n:-=i:..:a=;:t,"e:..:s~._;-;E;;-r:-:l~arui'-"t=-e~r==u"'in",g"e..:::n=:-:-:z:.;:u;-::d:.;:e:::n

Reden und Briefen nebst einer Biographi~ (Berlin and New York,
1971)., pp. 138-9.

101. A. Garzya, 'Literarische und rhetorische Polemiken der

Komnenenzeit', BS, 34(1973), 1-14; G. Schiro, 'La schedograiia
a Bisanzio nei-Sec. XI-XII e 1a scuola dei S5 XL Martiri',
Bollettino della Badia di Grottaferrata, 3-4 (1949-50), pp. 11-
29; Browning, Parnassos, n.s. 14 (1973), pp. 506-19; Lemer1e,
Cinq Etudes, pp. 235-41.

102. For Constantine Manasses and Niketas Eugenianos, see R. Brown-

ing, 'II Cod ice Marciano Gr. XI.31 e la schedografia bizan-
tina', Miscellanea marciana di studi bessarionei (Medioevo e
Umanesimo, 24, Padua, 1976), 21-34, at p. 26; for schedographi-
cal works ascribed to Theodore Prodromos, see Horandner, op.
cit., pp. 62-3; M. Treu, Eustathii Macrembolitae quae feruntur
aenIgmata (Progr. Friedr. Gymnas., Breslau, 1893).

103. Kazhdan, People and Power, p. 108; pp. 112-3.

104. Browning, art. cit. (n. 46) p. 8.

433
 105. cf. J. Boissevain and J.C. Mitchell, Network Analysis. Studies
in Human Interaction (The Hague and Paris, 1973).

106. Curiously in some ways the problem with Hunger's magisterial

Hochsprachliche profane Literatur der Byzantiner is not as
Kazhdan suggests, JOB 28 (1979), 1-21, that it is not a
literary history as such, which it is not, but that it does not
attempt to be a thorough-going Handbuch. It is selectively
descriptive rather than exhaustively analytical.

107. Patlagean, Annales ESC, 34(1979), p. 267, referring to P.

Lemerle, Le premier humanisme byzantin. Notes et remarques sur
enseignement et culture A Byzance des origines au xE siecle
(Paris, 1971), passim.

108. 1. Sevcenko, 'Levels of Style in Byzantine Prose', JOB 31/1

(1981), p. 302; art.cit. (n.40), p. 88, quotingGeorge
Lapithos.

i09. Robert Browning in BMGS 4(1978), p. 40 makes a distinction

between 'functional literacy' and 'the ability to manipulate
the Atticising litErary language and its complex universe of
reference and allusion', similar to that of Malcolm Parkes,
'The Literacy of the Laity', The Medieval World, ed. D. Daiches
and A. Thorlby (London, 1973), pp. 555-578, of 'pragmatic',
'cultivated' and 'professional'. It is interesting to note
that the arguments of Parkes on cursive run parallel to thoSE
of Mango on minuscule, Byzantine Books and Bookmen, p.45. I am
grateful to Pamela Robinson for her advice on Western literacy,

llO. Entertainment is the forgotten factor. The Kataphloron piece

suggests that members of the syllogos not only amused each
other, i.e. in a theatron, but also clients in private houses,
almost as if in succession to the professional entertainers
signalled in Arethas's famous passage; see Beaton, BMGS 7(1981)
p. 21.

111, The Jeffreys in their communication to the symposium reminded

us that oral delivery with appropriate use of gesture may well
have helped comprehension.

112. For definitions, as yet still impressionistic, if based on a

lifetime's familiarity with the literature, see Sevcenko, JOB,
31/1(1981), p. 291 ff.

113. I base all that follows on my unpublished thesis, Theophylact

Through his Letters: The Two Worlds of an Exile Bishop (Birm
ingham, 1981), esp. Fig. II, pp. 814-847, which is a
preliminary sketch of such a survey, which I hope to expand
into a book on Comnenian literary society.

114. On the funeral genres, J. Soffel, Die Regeln Menanders fur die
Leichenrede in ihrer Tradition dargestellt, hrsg, ubersetzt und
komrnentiert (Meisenheim am Glan, 1974); M. Alexiou, The Ritual
Lament in Greek Tradition (Cambridge, 1974) and my thesis
UI.2.

434
 liS. J. Lefort, see above, n. 69, and A.P. Kazhdan, introduction to
W. Regel, Fontes Rerum Byzantinarum, Rhetorum saeculi XII
orationes politicae, I 1-2, repro (Leipzig, 1982), v-xviii.

116. R. Browning, 'A New Source on Byzantine-Hungarian Relations in

the Twelfth Century. The Inaugural Lecture of Michael 0 tou
Anchialou as hypatos ton philosophon', Balkan Studies, 2(1961),
pp. 173-214; J. Lefort, 'Prooimion de Michel, neveu de
l'archev~que de Thessalonique, didascale de l'Evangile', TM,
4(1970), 375-93.

117. A.P. Kazhdan, 'Bemerkungen zur Niketas Eugenianos', JOB,

16( 1967), 1-38; R. Morris, 'The Political Saint of the Eleventh
Century', The Byzantine Saint, ed. S. Hackel (Studies
supplementary to Sobornost, 5, London, 1981), pp. 43-50; my
chapter 11.2.

ilS. Kdzhdan, People and Power, pp. 26-9 and my paper to the 1983
Birmingham Symposium.

119. C. Morey and C.N.L. B;:oooke, Gilbert Foliot and his Letters
(Cambridge, 1965), p. 13.

120, For example in our period, Nicholas Kallikles, Niketas

Eugenianos, Nicephorus Basilakes, Michael 0 tou Thessalonikes,
Eustathios Nakrembolites, Basil Pediadites, Nicephoru3
Chrysoberges, Demetrios Tornikes have left less than a dozen
letters all told.

121. I hope
hope shortly to publish a methodological study on t~e
detection of relationship in Byzantine literary works.

l22. H. Zilliacus, Untersuchungen zu den abstrakten Anredeformen u~d

Hoflichkeitstiteln im Griechischen (Helsingfors, 1953); L.
Dineen, Titles of Address in Christian Greek Epistolography to
527 A.D. (Diss, Washington D.C., 1929); P. Koch, Die byzan-
tinische Beamtentitel von400-700 (Diss, Jena, 1903).

123. See Clanchy's discussion of the superiority of oral witness to

documentation in the twelfth century and in particular his
treatment of Edward I's Quo Warranto proceedings and the
hilarious story of Anselm's dispute with Henry I, From Memory
to Written Record, pp. 208 ff. and 21 ff. He reminds us, p.
97, that 'reading and dictating' were ordinarily coupled, not
'reading and wri ting'. Malcolm Parkes, op.cit. n. 109,
instances Count Baldwin of Guines in the twelfth century who
'was extremely learned and possessed a splendid library but
never learned to read: he had his books read to him by his
clerici et magistri'. To indicate the clear importance of
orality in Byzantium, Professor Kazhdan reminded me of the
priest Michael in Michael Italikos, no. 1, ed. Gautier, p. 64,
who had all the works of Theodore Prodromos by heart, and
recited them to Italikos.

124. See Kazhdan, Studies, ch. 2.

125. Sevcenko, JOB, 31/1 (1981) 312.

435
 126. Lemerle, Premier humanisme, pp. 255-7. This has become the
most attractive statistic of the decade, causing such crass
remarks as B. Baldwin, 'Literature and Society in the Later
Roman Empire', ed. Gold, Literary and Artistic Patronage, p.
79: 'This prefigures the situation in Byzantium where it has
been calculated that only 300 or so people were receiving
higher education at any given time, thus ensuring a very small
audience for anything topical. And Byzantium produced
notoriouRly little poetry after the sixth century'.

127. Kazhdan, People and Power, p. 101-2; Sevcenko, art.cit. (n.40),

pp. 69-92; H.G. Beck, Das byzantinische Jahrtausend (Munich,
1978), p. 123.

128. Sevcenko, op.cit., p. 74; see above, Magdalino, Chapter 4,

pp.58ff.

129. For all justification see my thesis, Fig. II. I stress that
these figures are preliminary and will undergo revision. My
assessment of the whole of literary society, positive and
negative, would be a little over 400 in the proportion 9:11 in
favour of the receivers.

130, The profession most acutely underrepresented is that of

teachers: since at least R. Browning, 'Unpublished correspon-
dence between Michael Italicus, Archbishop of Philippopolis and
Theodore Prodromos', Byzantino-Bulgarica 1(1962), pp. 279-97,
we have been aware that the natural summit for a teaching
career was a see. Note also G. Weiss's warning that very few
Byzantines lived by teaching and writing alone, op.cit. (n.
96), p. 7.

131. J.P. Sartre, 'Qu'est-ce que 1a litterature?', Situations II

(Paris, 1948); R. Barthes, Le Degre Zero de l'ecriture (Paris,
1953).

132. S.R. Suleiman, The Reader in the Text, p. 37.

133. J. Leenhardt, 'Towards a Sociology of Reading', ed. Suleiman,

op.cH., p. 206.

134. R. Jauss, 'Literary History as a Challenge to Literary Theory',

New Literary History, 2(1970), p. 14.

135. ibid.

136. W.C. Booth, The Rhetoric of Fiction (Chicago, 1961), p. 138; ~.

Iser, The Implied Reader: Patterns of Communication in Prose
Fiction from Bunyan to Beckett (Baltimore, 1974).

137. V. Lange, 'The Reader in the Strategy of Fiction', Expression,

Communication and Experience in Literature and Language, ed,
R.G. Popperwe11 (London, 1973), pp. 86-102.

138. G. Prince, 'Introduction a l'etude du narrataire', Poetique,

14 (19 7 3). pp. 178-96.

436
 ,
139. E.g. A. Cameroll, "Images of Authority: Elites and Icons in
Late Sixth-Century Byzantium", Past and Present, 84(1979),
p. 3-35.

140. E.g. Sevcenko, JOB 31/1 (1981), p. 302.

141. ~;ee n. 12 above.

142. Cf. Clanchy, From Memory to Written Record, pp. 195-6. This is
not to deny that over the twelfth century certain families came
to playa particular part in literary activity. Literature ran
in the family with the Tornikes, the Kamater.os and by the end
of the century the Choniates and Mesarites families, and simil-
arly, up to twenty famUies including the Hagiotheodoritai, the
Kontostephanoi, the Mesopotamitai and the Pantechnai, show a
tendency to "negative rhetoricity" over a considerable span of
time. We should also not ignore the pervasiveness of the 0 teu
relationship and the support of episcopal uncles for beigh(
nephews throughout uur period, see Darrouzis, Documents
inedits, p. 56, Stiernon, REB, 2181963), p. 185; V. Tiftixo
glou, -Gruppenblldungen innerhalb des konstantinopolitanischen
Klerus wahrend die Komnenenzeit", BZ 62(1969), p. 34, who
exaggerates the norenallty of the procedure.

143. The for.thcoming edition by Charlotte Roueche will highlight his

literary debts; he appears to be writing in the first flush of
R late eleventh-century fashion for paralnesls, otherwi.e
attested by Theophylact"s Paideia basilik~, and Alexius l"s
Mousai.

144. BMGS, 4 (1978), p. 44.

145. See for example Opheomachos, in Theophylact, epp. '1eur, ",

VlIl, ~~g..::.~, PG 126, cols. 71-4.

і 46 , Thi.s paper has gOlle tllrough SllC(!essive versions given t(l

co 11')'l,tium of t'1e Royal. Irish Academy, The New University
Ulster. the gener-ll .s€rnr.flar of the Cent.re for Byzantine StIJdi.::::.
at Birmingham Clnei ·,8 'ln informal talk at Dumbarton Oaks. I al;)
gratefnl :=Jr thp. r~.o..;pt)nses of Cyril Mango I Ell7.aheth Jeffreys t

Judith Herrin, and l\l(~xandeL Kazhdan .)n those ,)cea.slons, to

Michaf?l Mc(;ann f,-)( his critie.A.l t01er,qnc~ of each sll(':(:t-~~sL'Je
',er,lnll and 1:1) '~, h·,e.! Angold for r:"king It off my ilClOci.q at
1".s t .

437
This page intentionally left blank
PANAGIOTIS ROILOS / CAMBRIDGE MASS.

AMPHOTEROGLOSSIA: THE ROLE OF RHETORIC IN THE MEDIEVAL

GREEK LEARNED NOVEL

Although circumscribed by the strict and sometimes monolithic conventions of a

long-established tradition, rhetoric in Byzantium always remained .a potentially
"double-tongued" (aJtQ>oT(;pfrtAw(J(Jor;i) art which allowed its practitiomers to exploit
the ambivalences of language and, sometimes, even to undermine the authority of
their models2 • Rhetoric has invested Byzantine literature with a dialogic potentiality
and an allusive intertextuality whose aesthetic appeal should be estimated not on the
basis of our own criteria but according to the horizon of expectations of its original
public - to the extent, of course, that these horizons can be reconstructed. If the
appreciation of this kind of creativity demands considerable mental agility on the part
of the general reader or even the specialized scholar, this by no means should be
considered as an inherent flaw of Byzantine literature itselfi.

I' 'AP.q,OTEpO,,(AWOOOC;
' \μφ οτίρό-γ\ω ααος is a term used by Tzetzes in his Chiliads in association with a rhetor’s rhetor's ability
to manipulate the meaning of his words: EEp/3wOC; Σίρβίλιος ήν ~P ii'lrlrTOe;
ΰτα το ς καί Καίααρ των
KaL Klriolrp ' Ρωμαίων. /I Μίθόδψ
TOJP 'PwpKXiwp. MEOOO",
ie
Si OELPiiTl)TOC;, P'rfTOPLK~ τφ
δεινότητας, ρητορική T~ rpii'lr",
τρόπψ I/ έκ
fK EEp/3'rfAiwp
Σερβηλίων Tij<; "(ol'1jc; λέγω
τής "γονής Af"(W καιKlrL τον
TOP ΣEEp/3AilrP.
ερ β\ία ν. I Ό '0<; Ei'lrEP
ς еїтгер
άλλο ς ~9EAE,
&AAO<; ηθεΧε, Eep{3op ειττεν. I/ TOUTO
’ Ηλίακ EhEP.
Σίρβον 'HAilrP Τοντο "(txp P~TOPO<; άνδρος
ya p ρήτορος C.VOpO<; καί C.P.q,OTEPO,,(AW(JOOU, /I και
KlrL άμφοτερογΧώσσου, Klrl
'lrpC.,,(P.OlOL και KA~O€OL
π ρά γμ α σ ι KOll κ\ήσεσι KOll και Toie;
τοΐς AOL'lroi<;
λοιποίς op.oiw<;
ομοίως / 'lrPO<;
τ ρος f'lrOlLVOP
Ιταινον καί KOll φ&γον
,/;6"(op δέOE κεχρήσθαι
K€Xpijo90lL συμφερόντως
OUP.q,€POVTW<;
(ed. P. Ρ. A.
Α. M. L E O N E , Ioannis Tzetzae historiae, Naples 1968,7.295-301;
Μ. LEONE, 1968, 7. 295-301; on the aristocratic family
of Serblias in general, see OOB, ODB, s.v.). Tzetzes illustrates the character of rhetoric as άμφοτΐρογλωσ- C.P.q,OT€PO"(AWO-
σία (double-tonguedness) by means of his reference to the protean character of ψόγος
oilr if;o"(o<; and f1rOlLVO<;,
Ιτα ινο ς,
which correspond to ψό~/ος y,o"(o<; and f"(KWP.LOV,
έγκώμιοκ, two of the most influential progymnasmata in Byzantine
literary tradition. The ability of a rhetor to adapt the same words and ideas to different literary
contexts with totally opposite functions points to the potentially playful character not only of rhetoric,
but of literature in general. Such an amphoteroglossia could be perceived as a Byzantine equivalent
of the Bakhtinian heteroglossia. The notion of heteroglossia pervades the whole theoretical work of of
Bakhtin, especially his discussion of the novel, and is closely associated to his idea of dialogism.
Heteroglossia refers to the multiple conditions that define the semantic value of a specific utterance
on a specific occasion, at a specific moment; see especially M. BBAKHTIN, a k h t i n , The Dialogic Imagination,
Austin 1981, pp. 259-422.
22 Some authors emphasize their own innovative divergences from the established tradition. tr.adition. See, for
example, the cases discussed in A. GARZYA, Literarische und rhetorische Poiemiken
G a r z y a , Uterarische Polemiken der Kom-
nenenzeit;
nenenzeit, Byzantinoslavica 34 (1973) 1-14: 7-8.
1 In his famous article on "The Hellenistic Origins of Byzantine Literature," for example, R. Jenkins
expresses his negative view about rhetoric's impact on Byzantine literature in a vehement way,
without trying to define its broader aesthetic and ideological parameters: "We have been compelled
to pass an unfavorable judgement on Byzantine rhetoric, since it colored nearly every department of
Byzantine literature and since honesty forbids us to regard its influence as anything but disastrous"
(R. JENKINS, The Hellenistic Origins of Byzantine Literature, Oumbarton Oaks Papers 17 (1963)
37-52: 46). For a brief review of previous negative assessments of Byzantine rhetoric, see H.
HUNGER, Aspekte der Griechischen Rhetorik von Gorgias bis zum Untergang von Byzanz, Sitzungs-
berichte der Osterreichischer Akademie der Wissenschaften, philos.-histor. Klasse 1972 (277.3), pp.
6-7. Hunger proposes a systematic model for the contextualization of the function of Byzantine
rhetoric. For a comprehensive aesthetic and ideological approach to Byzantine riletoric, see also
H.-G. BECK, Antike Beredsamkeit und byzantinische Kallilogia, Antike und Abendland 15 (1969)

439
110 Panagiotis Roilos

In this paper I shall investigate the role of rhetoric in the Komnenian novels. Despite
their pivotal significance, the rhetorical aspects of the Komnenian novels have never
been studied in a systematic way before. My study will focus on the creative use of
rhetorical amphoteroglossia in Theodoros Prodromos' Rhodanthe and Dosikles4 •

1. The impact of rhetoric on the ancient Greek novel has been the subject of many
discussions. Giangrande, for example, has argued that the origins of the ancient Greek
novel should be sought in the area of prose rhetorical paraphrases of erotic stories -
in particular of Alexandrian love-elegies and epyllias. Reardon, on the other hand,
putting aside the thorny, and ultimately fruitless, question of origins, insists on the
importance of rhetorical theory and practic.e for the composition of ancient Greek
novels, arguing that "romance could be composed only as a form of rhetoric"6. He
puts special emphasis on epideictic rhetoric and progymnasmata, which, in his
opinion, had contributed a great deal to the formulation of the literary taste and the
aesthetic expectations of both the authors and the readers of ancient Greek novel7 •
The same holds true, I argue, for the Komnenian novels. The twelfth century Byzanti-
ne novelists and their audience received and appreciated the ancient as well as their
contemporary novels as predominantly rhetorical works. Such an attitude of the
Byzantine audience towards the genre of the novel is indicated by both the charac-
teristics of the Komnenian novels themselves and the overall aesthetic reception of the
ancient Greek examples of the genre by the Byzantines8 •
Unfortunately, we lack any concrete evidence about the aesthetic reception of the
Komnenian novels by their Byzantine audience. It is only by means of a close study

91-101; Beck's overall attitude towards Byzantine rhetoric is actually more positive than what
Hunger's criticism lets his readers think (Hunger op.cit.: 7, n.5). For the importance of rhetoric in
Byzantium in general, see also H. HUNGER, The importance of rhetoric in Byzantium, in M.
MULLETT AND R. SCOTT (eds.), Byzantium and the classical tradition, Birmingham 1981, pp. 35-47,
and G. KUSTAS, The function and evolution of Byzantine rhetoric, Viator 1 (1970) 55-73.
4 The references are to Marcovich's edition (M. MARCOVICH, Theodori Prodromi, de Rhodanthes et
Dosikles amoribus Jibri IX, Stuttgart 1992).
S G. GIANGRANDE, On the origins of the Greek Romance: the birth of a literary form, Eranos 60
(1962) 132-59: 152ff.
6 B. REARDON, Theform of Greek romance, Princeton 1991, p. 84.
7 REARDON, op. cit., pp. 87-9. Reardon rightly avoids, however, the one-sided analysis of the
influence of rhetoric on the novel criticized by E. B. PERRY, The ancient romances, Berkeley 1967,
p. 19, n. 7. Cf. also T. HAGG, The novel in antiquity, Oxford 1983, pp. 107-8. For the role of
rhetoric in the ancient Greek novel, see also R. F. HOCK, The rhetoric of romance, in S. PORTER
(ed.), Handbook of Classical rhetoric in the Hellenistic period 330 B.C.-A.D. 400, Leiden 1997, pp.
445-65. Hock notes that "despite the widespread acknowledgement of the important relation between
rhetoric and the romances, little has been done to clarify this relation" (ibid.: 450). Unfortunately,
Hock's article as well is too general to address this problem in detail. Rohde was, of course, the first
to study the impact of rhetoric on the ancient Greek novel (E. ROHDE, Der Griechische Roman,
Leipzig' 1914, pp. 336-60).
• A detailed study of both the rhetorical aspects of the Komnenian novels and the aesthetic reception
of the ancient Greek examples of the genre by the Byzantines has been undertaken in my PhD thesis
(Generic Modulations in the Byzantine Learned Novel, Harvard University 1999).

440
The role of rhetoric 111

of the evidence provided by the manuscript tradition, notably the marginal indications,
that the reception of these novels in the Middle Ages could be reconstructed in a
systematic way.
In Vaticanus Graecus 1390 (13th c.), which preserves Hysmine and Hysminias, the
annotations in the margins indicate that this novel might have been read as a text
consisting of several self-contained parts. Most of these parts are described according
to a terminology that recalls rhetorical terms used for progymnasmata: EKCPPCXUU; TOU
K~1I"0U, €KcppCXO"Le; cpp€CXTOe;, EKcppCXU~e; 'Yup.tV1/e; KLpIIOUU'Y/e;, EKcppCXO"Le; TWII apETwII,
€KcppCXULe; P.'Y/IIWII, DL~'Y'Y/P.CX, and so on. Some of these episodes are described with
exactly the same title in other manuscripts of the text too, as, for example, in Vatica-
nus Graecus 114 (13th c.). One can assume that these sections of the novel could
have been received by the Byzantine audience as potentially independent samples of
literary art.
There are also cases where particular passages are singled out as 'Yvwp.w, a term
that, I argue, refers us to progymnasmata and recalls Psellos' appreciation of the
'YIIWP.OTV1I"LCXL in Ethiopica9 • For example, the phrase TPOCP~ 'Yap 1I"0hUT€h~C; t'Y/Tfi KCXt
1I"0O"LII all&hO"(oll from the second book of Hysmine and Hysminias (2. 13. 2) is
characterized as 'YIIWP.'Y/ in Vaticanus Graecus 1390, f. 140V and Vaticanus Graecus
114, f. 9v •
Some other passages are assessed as particularly beautiful. as the marginal ab-
breviation for the Greek word wPCX'iOIl indicates. In Vaticanus Graecus 1390, for
example, as wpcxioll is characterized, among others, one passage that describes the
storm which eventually sank the ship that carried the two lovers:

Kcxl OA'Y/II ~II IICXUII CPLhOllfLKfL KCXTCXOVUCXL TOLe; KUP.CXO"L KCXt Oh'Y/II fie; {3U(JOII a"{cx-
'Yfill UUII cxvToiC; 1I"Awri]PUL, UUII CXVTc;, CPOPT~, UUII CXVToiC; UiP.{3AoU; wEpWTOe;, &
P.CO"TCx p.€NTOe; i:PWTLKOU ~II KCXh~1I EixEII 'Yup.ill'Y/II KCXt Tall' Y UP.LIILCXII i:p.€, K(Xv
IIOCTELOWII allT£ JLEhLTO<; a,pLII(Jiov 1I"A'Y/poiw aIlTEp.ixXETO (Hysmine and Hysminias
7.8.2)

The reason for which the scribe has drawn the attention of the reader to this specific
paragraph might have been that the narration of Hysminias here is constructed around
two main rhetorical axes, epanalepsis (UUII cxVTO'i<;, UUII cxVrc{J) and antithesis (O!/lTt
P.fALTO<; a>J;LII(Jiou). Predominant is the figure of anaphora (KCXt Oh'Y/II), whose rhetorical
effect is amplified by the use of homoioteleuta (Kup.cxuLi1l"AWri]PUL, cxvTo'ie;/uip.{3hoL<;).
The use of marginal titles and critical signs is well documented in the manuscripts
of both the other Komnenian novels and their ancient Greek models lO • In Vaticanus

9 A. DYCK, Michael Psellus - The Essays on Euripides and George of Pisidia and OR Heliodorus and
Achilles Tatius, Vienna 1986,94.55.
10 For Theodoros Prodromos and Niketas Eugeneianos, see the editions of Marcovich and Conca
respectively (F. CONCA, Nicetas Eugenianus, de Drosillae et Chariclis amoribus. Amsterdam 1990);
on Niketas Eugeneianos in particular cf. also the elucidating observations in F. 'CoNCA, Scribi e
lettori dei rornami tardo antichi e bizantini, in Atti del Primo Convegno di Studi Tardoantichi:

441
112 Panagiotis RoBos

Graecus 1390, for example, the novel of Makrembolites is followed by Heliodorus'

Ethiopica. Again here the scribe uses the same ways to single out the passages that he
finds especially interesting or possibly instructive from a stylistic point of view.
This brief review of the evidence provided by the manuscripts establishes the idea
that the Komnenian novels, exactly like the ancient Greek models of the genre, were
received by the Byzantines as texts closely related to the art of rhetoric and were
subjected, consequently, to analogous criteria of aesthetic appreciation. Their separa-
tion into particular units may suggest that they were read as narrative wholes consi-
sting of potentially autonomous episodes, a practice that recalls Psellos' praise of the
E7rEWOOLCi &1)'Y~/l-Cim of the Ethiopica 11 •

II. Rhetorical amphoteroglossia allowed the Komnenian novelists to invest their works
with a rich and ambivalent allusiveness. In what follows I shall try to illustrate the
intricate ways in which rhetorical amphoteroglossia functions in a banquet scene in
Theodoros Prodromos' Rhodanthe and Dosikles.
In his commentaries on Homer, Eustathios of Thessalonike associates the des-
cription of banquets with the rhetorical idea of apheleia l2 . More specifically, ban-
quets, as events of everyday life, could be connected with euteleia as well, which is
closely related to apheleia. Commenting on a banquet scene from the Iliad, Eustathios
underlines T~P -qPWLK~P &'¢€AELCiP Kl¥L EVT€AELCiP ~P €v 07rTo'ir; KP€CilJL To'ir; KCiTOr. TOil
'AXLAAf.Ci KCiL TOZr; AomoZr; (cf. Iliad 9.217.749). In rhetorical tradition, euteleia is
also associated with comedy 13. It is in this comic vein of rhetorical euteleia that
Theodoros Prodromos employs the traditional motif of banquet. In his use of this
motif, he draws both from ancient Greek authors and contemporary Byzantine reality.

In the fourth book of Rhodanthe and Dosikles Prodromos describes one of the most
grotesque banquets in the tradition of Greek novel. The banquet is attended by two
enemies and serves a specific political agenda. Its context is as follows: Artaxanes, an
envoy of the king of Pissa Bryaxes, arrives at the court of Mistylos, the chief of the
pirates who have arrested the two protagonists, bearing a letter from Bryaxes. In the
letter, which has the form of an ultimatum, Bryaxes disputes Mistylos' rule over the
city of Rhamnon: if Mistylos does not yield up the city, then Bryaxes will declare war
against him. Although greatly upset, Mistylos does not express his agitation. He

"Metodologie della ricerca sulla tarda antichita", Naples 1989, pp. 223-46.
II A. DYCK, as in note 9, 92.32. On the concept of epeisodia diegemata in Psellos and its importance
for the Komnenian novels see P. A. AGAPITOS, Narrative, rhetoric and 'drama' rediscovered: poets
and scholars in Byzantium interpret Heliodorus, in R. HUNTER (ed.), Studies in Heliodorus, Cam-
bridge 1998, pp. 125-156. I intend to pursue the examination of the evidence of the manuscript
tradition in a separate study.
12 See G. LiNDBERG, Studies in Hermogenes and Eustathios, Lund 1977, pp. 225-6.
IJ For the relation between euteles and apheles, see Hermogenes' De Ideis 324-5 RABE; Demetrius

speaks of the comic effect of euteles: 7"WV IJE XOl.pi7"wV OI.i Jl.fV €im JI.€irov€<;; WI C1€Jl.V07"€POl.I 01.1 7"WV
1I"011]7"WV, 01.1 {)f €im,,€i<;; JI.&""ov KOI.' KWJl.IKW7"€POl.L (Herm. 128; cf. H. LAUSBERG, Handbuch der
literarischen Rhetorik: eine Grundlegung der Literaturwissenschajt, Munich 1960,521).

442
The role of rhetoric 113

commands his satrapes Gobryas to prepare a banquet in honour of the Pissan envoy.
The purpose of the banquet is to intimidate the guest.
In the description of this banquet Theodoros Prodromos demonstrates a unique
dexterity in combining indirect references to contemporary Byzantine reality with
subtle allusions to literary tradition. As a result of this, the whole passage has been
invested with an CxJ.tcPOTEpo'YAwaaia that functions on several levels at the same time.
Parody - possibly accompanied by satire - is the main device that Prodromos em-
ploys here in order to achieve this effect. The scene of this banquet should be viewed
as an indirect glorification of the power of art in general, and Prodromos' own
literary art in particular. Culinary art, rhetoric, court poetry, religious poetry, mime,
and ceremonial rituals are all put together here and subordinated to the author's
orchestrating and parodying creative art, in order to construct a complex literary
artifact of a comic character, which only a special audience such as the one frequen-
ting the twelfth century Byzantine rhetorical theatra could have adequately apprecia-
ted 14 .
From the very beginning, the scene of the reception of Artaxanes reflects the
prescriptions of Byzantine ceremonial protocol. Simple details in Prodromos' des-
cription aquire an important allusiveness, if put in the context of Byzantine court
rituals IS •
At the arrival of the Pissan envoy Mistylos sits on an elevated throne surrounded
by his retinue. The author puts special emphasis on this detail:

aurae; D' h' oKpi(3aJJroe; fie; (}pOJJOJJ J.tE-yaJ)

v'/lou Ka(}Ea(}fLe; Kat n TaJlWOEt; (3AE7rWJl,
r~t; aarpa7rLK~e; arpanat; tie; rOJl (}pOJlOJJ
iaraJ.tEJI'l/t; KVKAW(}€JJ €UcPV€L arCxafL,
KaAELJI KEA€VEL raJ) aTaAEJ)Ta aarpCx7r1/JI (4.16-20)..

The titanic-like look of the barbarian leader contributes to the awe-inspiring effect of
his whole appearance. Artaxanes, on the contrary, bows his head in front of Mistylos'
feet and gives him the letter.
The contrast in the positions of the two enemies corresponds, I believe, to the
ceremony of the reception of foreign dignitaries at the Byzantine court. In his speech
delivered on the occasion of the visit of the Sultan KIlle Arslan II at the court of
Manuel I Komnenos in 1161, Euthymios Malakes refers to a similar difference in the

14 On "rhetorical theatra" in Byzantium see H. HUNGER, Die hochsprachliche profane Literatur der

Byzantine,., Munich 1978, I, pp. 210-1; M. MULLETI', Aristocracy and patronage in the literary
circles of Comnenian Constantinople, in M. ANGOLD (ed.), The Byzantine aristocracy, Oxford 1984,
pp. 173-201; also the excellent discussion in P. MAGDALINO, The Empire of Manuel I Komnenos,
1143-1180, Cambridge 1993, pp. 335-56; the latter contends that "theatre is in fact the key to
understanding both the aesthetic and the social function of high-style literacy in twelfth-century
Byzantine society" (ibid.: 339).
" For court ceremonial during the reign of Manuel Komnenos, see P. MAGDALINO, op. cit., pp.
237-48.

443
114 Panagiotis Roilos

position of the two leaders. Euthymios interprets this difference as indicative of the
Byzantine emperor's predominance over his visitor 16 • Interesting information regar-
ding this aspect of Byzantine court protocol is provided also by Western authors of
the same period. In his account of the reception of Amaury I, king of Jerusalem, at
the court of Manuel I in 1171, Williarn of Tyre notes that the Byzantine emperor was
seated on a golden throne wearing his imperial robes, while his visitor was seated on
another throne lower than that of the emperor17. An indirect confirmation of the
symbolic significance of the difference in seat can be drawn from Odo of Deuil's
description of Louis VII's visit to Constantinople in 1147. Odo, whose overall
account does not reveal a friendly attitude towards the Byzantines, does not spare the
detail that the two sovereigns sat merely on two chairs, thus implying that no diffe-
rence in status was imposed by the ceremony of the reception 18 •
After he was received by Mistylos, Artaxanes was led by Gobryas to a hall specifi-
cally assigned for the reception of foreign envoys, where the dinner would be served:
€/I OOP.OLC; / 7rPOC; T'Y/AtKOVTWV (sc. foreign dignitaries) do-ooxizv T€TCX:YP.EvOtC; (4.115-6).
Again this detail does not depart from the strict protocol of the reception of foreign

16 no,AC'I /'of V oVv fl'l//'orXPOITO!; b Kopivetoe;, • AAE~o,V{jpov TO frPWTOV U1I'0 TOV xpVUOVV oupOlviuKOV Ev
T<j> flOlpdou OpOVIf KOIOiuOlVTOC;, /'oE'YrXA'l/e; Eil!'EV ~{jon,e; UTEP'l/OiiVOIt TWV 'EAA~VWV TOUe; TEOV'f/KOTOIC;
1I'P!P ~ OE&OOlOOOlt TOP • AAf~OIPOpOV fV T<j> flOlPEiov OpOV,+, K0I8~l'-Epov' E'YW 0' Oev KOI' OIUTOe; o,mwoOli-
/'o'l/V i'xPTl TOUTO TO fl'l//'oOlpo,mov KOlI 1!'0AA~e; Oev </>OIi'l/v fUTEP~UOOlI ~C; ~00n,C; TWV Ul!'O TlPOIe;
•PWJ1.OIiwv tKEivove;, OOOL 801VOVTEe; ou 1f'DtPEUli 'YE vfI.v iOEiv TOV fOlVTWV {jOlOLAfOl AOIp.'IrPOV AOIp.1f'pWe;
h8&oE 1f'P08POVLI0P.EPOV KOIt TOP P.f'YOIV TWV nEpOWV apX'I/'YfT'f/V OIVTij> KOIAWe; Vl!'OfroOLIOP.EVOV. OfOlP.OI
TEP1I'POV OVTW KOIt ~fVOP KOIt olov OUOf1l'W TOUe; b</>OOlAp.OVe; TWV • PWP.OILWV EioTiOlOEV (ed. A. PAPAD0-
POULOS-KERAMEUS, Noctes Petropolitanae. St. Petersburg 1913, 167.12-21; cf. P. MAGDAUNO,
The Empire... , pp. 454, 242). It is worth noting that in this speech, Malakes refers to the sultan as
Persanax as well (ed. PAPADOPOULOs-KERAMEUS, op.cit.: 167.25), a term encountered also in
Niketas Eugeneianos' novel (5. 338 ). For Eugeneianos' use of these terms cf. also A. KAzHDAN,
Bemerkungen zu Niketas Eugenianos. Jahrbuch der Osterreichischen Byzantinischen Gesellschaft 16
(1967) 101-17: 109. For Malakes' speech cf. also P. MAGDAUNO AND P. NELSON, The emperor in
Byzantine art of the twelfth century. Byzantinische Forschungen 8 (1982) 123-83: 132-5. In his
description of the same event, Kinnamos offers the same information about the difference in the
position of the two leaders: ~PTO P.Ev {3~P.OI AOI/'ofrPOV KOI! oi</>poC; 1,11'1 1I'AEioTOV ~OEV OI'POP.EVOC;
EKEt'TO, OEOIJ1.OI A(ryov l!'OAAOV i'x~lOv ... E</>' o~ K0l8~OTO {301C1tAEVe; ... KNTlLE08Aixv Of bm~l!'Ep E'e;
J1.fUOVe; 7rOlp~ME. 8&J1.{jove; i\Ao~ ~v ... K0I8~OTO AOL'JI'OV h, XOIJ1.OItIilAOU TLPo<; KOI' ijKLUTOI ffr!
P.ETEWpOV K0I8flipOIe; (205-6, MEINEKE).
17 WILLIAM, ARCHBISHOP OF TYRE, A history of deeds beyond the sea, trans. E. BABCOCK, N. York
1948, II, p.380. In his description of the reception of Baldwin III by Manuel several years before this
event (1159), William again puts emphasis on the same detail: Manuel was seated on a throne higher
than that of his guest (ibid.: 278).
18 Granted that Odo's view of the Byzantines was rather hostile, his description here must be less
innocent than it seems at first sight, all the more since in his account of the same event Kinnamos
explicitly notes that the Byzantine emperor was seated on his throne. whereas Louis on a little stool
(Odo of Deui!, De profectione Ludovici VII in orientem. ed. and trans. V. G. BERRY, N. York 1948,
p. 59).

444
The role of rhetoric 115
envoys at the Byzantine court19 • From this point on, Prodromos, although continuing
to allude to the psychological and subsequent political effectiveness of Byzantine
ceremonial, proceeds to construct his own, more grotesque, version of it. Artaxanes
is now served a "marvelous dinner," O€'i1f"O" ... 1fPOC; TO OCXUjl.a(JLO" ~TO'jl.CX(Jjl.E"O"
(4.122-3). This characterization prepares Prodromos' audience for the unexpected
event that follows: Artaxanes is served a roast lamb. When he tries to cut it, a flock
of small sparrows come out of its belly and begin to fly above his head. Artaxanes is
dumbfounded, ~M€/I €iC; Oajl.{3oc; jl.E-yCX (4.130); later, when he hears Gobryas' ex-
aggerated interpretation of the incident, his surprise is transformed into terror
(4.173-188; cf. 5.52).
The culinary marvel performed at this banquet has been rightly paralleled to the
technological marvels demonstrated to foreign visitors at the Byzantine coureo.
Liutprand of Cremona relates his own experience of such demonstrations, whereas
Konstantinos Porphyrogennetos provides detailed prescriptions of their proper use in
court ceremonial21 • I would view this incident, therefore, as a comic, and perhaps
even parodic, example of ceremonial conspicuous virtuosity22. It has been pointed
out that the motif of the roast lamb out of whose belly a flock of live birds flies finds
a parallel in Petronius' Cena Trimalchionis. Although no evidence exists to prove that
Petronius' text was available in twelfth century Byzantium, this parallel corroborates,
at least, the comic nature of the motif. No direct or indirect link is to be assumed,
however, between the Byzantine and the Latin author. In my view, the culinary
marvel in Prodromos' novel constitutes a variation of the author's general interest in

19 Cf., for example, Konstantinos Porphyrogennetos' description of "the reception" of the Saracenes
in his De Cerimoniis, II. 584-5, REISKE. The Greek word for "reception" that Konstantinos uses,
that is oox~ (ibid.: 583 and passim). recalls the similar term used by Prodromos in this scene:
fiuoox~. Konstantinos Porphyrogennetos describes how the foreign envoys passed through several
buildings before they arrived at the dinner hall where the emperor had prepared a feast for them.
20 H. HUNGER. Die hochsprachliche ...• II.131-2; R. BEATON. The Medieval Greek romance. London
and N. York 1996. p. 75; cf. A. KAZHDAN. Certain traits of imperial propaganda in the Byzantine
empire from the eighth to the fifteenth centuries. in G. MAKDISI, D. SOURDEL. J. SOURDEL-THOMINE
(eds.), Preaching and propaganda in the Middle Ages. Paris 1983. 13-27. None of these scholars
discusses the issue in a systematic way.
21 Liutprand was very proud of having not been intimidated by the display of such technological
miracles at the court of Konstantinos Porphyrogennetos. He mentions a bronze tree with bronze birds
that were chirping, metal lions that were roaring, and the impressive so-called "throne of Solomon".
Liutprand, however, had already made enquiry about these marvels and he was not terrified (An-
tapodosis, 6.5). In De Cerimoniis. Konstantinos himself describes the throne and the other technolo-
gical marvels at his court (II. 566-70). For the throne of Solomon and Byzantine court technology
in general. see G. BRETT, The automata in the Byzantine Throne of Solomon. Speculum 29 (1954)
477-87; for the political manipulation of technological marvels at the Byzantine court, see J.
TRiLUNG. Daedalus and the nightingale: Art and technology in the myth of the Byzantine court. in
H. MAGUIRE (ed.), Byzantine court culture from 829 to 1204, Washington. D.C. 1997.217-30.
22 I am using the apt term introduced by Trilling in his discussion of the political manipulation of
technology at the Byzantine court; the second component of this term. virtuosity. refers to extraordi-
nary skill. most usually artistic. while the first component. conspicuous, emphasizes the public
display of such a skill (as in n. 21, p. 225).

445
 116 Panagiotis Roilos

the motif of food which he expresses elsewhere to023 • In any case, the studies that
have suggested the parallels of this marvelous dish with either technological marvels
or literary motifs, as encountered in Petronius, in particular, fail to underline the
comic character of this extraordinary incident and to locate it within the context of the
scene of the banquet as a whole, which, in my view, has been constructed as a comic
synthesis of different individual parodic elements 24 .

23 Prodromos does so even in contexts where one would not normally expect such an interest. In his
Life of Saint Meletios, for example, he extolls the saint's spartan life by contrasting it to the alleged
opulence of his audience. His enumeration of the dishes possibly enjoyed by his imagined reader
takes a rather grotesque character. I think it is worth quoting the corresponding passage because it
illustrates very well Theodoros Prodromos' fascination with food: ail Of p.o< 1'Ou~ 1'OIW~ expiIJp.f! KCXI
1'o,~ "(EpapOUe; KCXI1'OUe; fK ~a(nooe; OpPtOW;, KCXI1'O Ot7rAOUV, El {JOVAEt, TWP 'lrEpOiKWP "(fPoe; 7rpo07iIJE-
00, Kal OTEPOXWPEt p.fp 1'o,e; A07raOW;, 01'EVOXWPEt Of 1'0 o7r1'OlvEiov Toie; KpfaOt, Kat 7rO!KtAAf (TOU
TOUe; iY.pToue;, Kal Ota<popo,e; <WaH 7rop.a1'WV, KCXI1'OUe; 7rAaKOUVTae; hi 7OV1'Ote; 7rapa"(E Kat TOVe;
CJ'I/(Tap.oUVTW; Kat T~V AOt7r~V TWP 7rEp.p.a1'07rOtWV <pAuapiav Kal1'UpaVVEt 1'0, OTOtXEia, Kaloaop.oM-
"(Et Toi~ fowoip.otC; KCXI 7rP07rtPE <ptAiou Kal hcnPEiou ... Kal KEpowpiou KaTo, njc; faU1'OU KE<paA~~, Kal
iY.xpte; EP.<1'OU 1'oiC; (Jup.7r01'Olte; €"(XEt TOP iY.Kpa1'OP (ed. V. VASILEvSKn, 9EOOWPOU 1'ouIlpoopop.ou (J!OC;
MEAn!OU 1'OU- VEOU, Pravoslavnyi Palestinskij Sbornik 17 (1886) 40-69: 50). For a brief discussion
of this Life and the issue of hagiographic writing in the twelfth century in general, see P. MAGDAU-
NO, The Byzantine Holy Man in the twelfth century, in S. HACKEL (ed.), The Byzantine Saint,
London 1981, 51-66. The whole description here recalls similar passages from the Ptochoprodromic
poems, especially the ones on the poverty of the poet as a man of letters and the satire against the
monks (III, IV, HESSEUNG AND PERNOT). A special interest in food is documented in the works of
other twelfth century writers as well, such as Eustathios of Thessalonike, and might reflect a broader
change in the diet of the Byzantines in this period (see A. KAZHDAN AND A. EpSTEIN, Change in
Byzantine culture in the eleventh and twelfth centuries, Cambridge and Paris 1984, pp. 80-1;
Kazhdan and Epstein, however, do not take into account this description in Theodoros Prodromos'
Life).
24 Beaton, for example, fails altogether to view this episode in its broader context, while accepting

the possibility of an allusion to Cena Trimalchionis unquestioningly (Romance ... , p. 75). In his
opinion, beyond the fact that Theodoros Prodromos composed poems for ceremonial occasions at
court, "there is probably nothing in Prodromos' story that reflects the details of such ceremonies in
the twelfth century" (ibid.); see, however, my analysis above, where I discuss the description of
analogous ceremonial occasions by twelfth century Byzantine and Western writers. For the parallel
with Petronius cf. also. H. HUNGER, Die hochsprachliche ... , II.13I. It should also be noted that the
Cena Trimalchionis had a rather problematic manuscript tradition on which see M. SMITH, Petronius
Cena Trimalchionis, Oxford 1975, pp. XII-XIV. The similarity, however, between Theodoros
Prodromos and Petronius remains intriguing. Is it a mere coincidence or is it to be attributed to some
oral or another, yet unidentified, source? To the best of my knowledge, there is no study on the
possible familiarity of twelfth century Byzantine writers with specific examples of Latin literature.
Salanitro's short note on a possible echo of Horace in Eugeneianos' novel is not convincing (G.
SALANITRO, Orazio e Niceta Eugeniano, Sileno 18 (1992) 247-8; intriguing are the similarities
between Eugeneianos and Plautus identified by Giusti, who prefers to attribute them to an unidentifi-
able common source (A. GIUSTI, Nota a Niceta EugeniaI/O, Studi Italiani della filologia classica 11
(1993) 216-23: 221-3). On the other hand, as is well known, exchanges between the Byzantines and
the Westerners were frequent in the twelfth century; cf., e.g., E. JEFFREYS, The Comnenian back-
ground to the romans d'antiquite, Byzantion 10 (1980) 455-86; in this respect, it is worth noting that
in his monody on his brother, Nikephoros Basilakes refers explicitly to the deceased's perfect

446
The role of rhetoric 117

The first of these parodic elements is the demonstration of rhetorical mastery on the
part of Artaxanes' host, Gobryas. In a long display of eloquence Gobryas explains to
the dumbfounded Artaxanes that the miraculous dish was prepared thanks to Mistylos'
supernatural powers. Mistylos is so mighty that even nature succumbs to his will:

«' OpQ:e;», EA€t€, «7ra/L/LE'YLaT€ (JaTpCx7ra,

TOU O€(J7rOTOV /LOU ri71l {j(wa/LLIl MLaTvAov,
we; Ua/LEi{3€LJl iOXVEL Kat Tae; rjJva€Le;,
KWIla"ie; Ot/LoL{3aie; Kat Tpo7raie; 7rOAVTpCJ7rOLe;
TPE7rWIl €Kaam KaL /LdhaTWIl we; (JEA€L.
opQ:e; TOil OtPllOll we; Kui(JK€L aTpov(Jia
rile; rjJv(J€WC; /Ltll Ot'YIlo~aae; Tall IlO/LOll,
we; 7rT111l0ll OPllLIl 7rT111l0e; OPllLC; €KKV€L,
OtpllOe; 7rET€LIla {3AaaTCxll€L TWIl €'YKCxTWV» (4.134-43).

Gobryas concludes his interpretation of the culinary marvel with an indirect threat
against Artaxanes and his people: Mistylos could use his unbeatable power against his
enemies in the war. He can even impregnate them with ... puppies:

'H 7rOV K€AEv(Jae; KOtV /LE(Jwe; TVXOV /L&xwe;

Kat aTpaTLWme; i!xvopae;, OtOpOue; 07r ALme;,
a7rCx8me; (JUV avmie; KaL /LET' aUTWV Ot(J7rLOWV,
'YEVV~Topae; OEL~ELE 7rOAAWV (JKVACxKWV,
Kat 'Ya(JTEpac; (Jwpa~LV r,(JrjJaAL(J/LfVae;
€'YKV/LOVEill 7r€L(JELEV HI'O: (4.166-71).

The Greek word for puppy, (JKUAO:~, is invested here not so much with macabre
connotations, as has been 5uggested25 , but rather with comic sexual allusions. Hesy-
chius informs us that this word denotes a specific sexual position. EKUAO:~' OXTJ/LO:
Otc/>POOtaLO:KOV we; TO TWV c/>OLVLKL~bvTWV (Hesychius, s.v.). Gobryas, in other words,
underlines the prowess and omnipotence of his master by subtly alluding to his sexual
potency as well, which could have a rather humiliating and unorthodox effect on his
enemies26 •

knowledge of Latin (ed. A. PIGNANI, Nice/oro Basi/ace, Progimnasmi e monodie, Naples, 1983, 242,
164-6).
2S Again it is Beaton who argues that "the reader is presumambly intended to decipher Gobryas'
words as apunning allusion to worms (I1KW>-''l/KE<;), to which the bodies of fighting men could indeed
be said to 'give birth' - if they are killed" (Romance .... p.74).
26 It may not be fortuitous that the same word appears in Longus' novel as well, in the context of a
homosexual's violent pass on the hero. Fortunately, Daphnis escapes Gnathon, the wanton homose-
xual, and runs away as a puppy «(TKu>-'a~; 4.12.3, Reeve). Gnathon's desire and attack is described
as a violation of nature, another element that somehow recalls the paradoxical atmosphere in
Prodromos' scene. That Prodromos was familiar and influenced by Longus is beyond doubt. Suffice
is to point out that the name of the Pissan leader. Bpu&~'1/<;,· is almost identical with the name of the

447
118 Panagiotis Roilos

Artaxanes takes Gobryas' threat at face value and is scared to death. He asks Gobryas
to spare him from such an ignominous pregnancy. Besides being terrified, Artaxanes
is extremely bewildered. He cannot understand how men can get impregnated.

IIov 'Yap 7rCXp' ~p.i/l KCXt 'Y&ACX/(TO~ e/(xuuH~,

EL 7rOV OE~UEL cfiVUL/(i;J 7r&/lTW~ Airy~
'Y&ACXKTO~ bA/(oi~ €/(Tpcxcfi17/1CY.L Ta {3pecfi'YJ;
lxAAW~ Of /(CXt 7rW~ T~/I TOUCXU'Tf//I cxiaxv/I'YJ/I
(:xv~p UTpCXT&pX'YJ~ KCXPTEP~UH/I iUXVUH
€'YKVP.O/IW/l Cx(JAwq lxOALCX {3pecfi'YJ; (4.183-88).

Around the OtoV/lCXTO/l of the culinary marvel and Gobryas' paradoxical claim, which
is believed by Artaxanes, Theodoros Prodromos constructs a whole parodic argumen-
tation greatly influenced by rhetoric. The topic under question is whether it is possi-
ble, and appropriate, for a man to become pregnant and give birth to babies. The
whole discussion on this topic should be read as a parody of the established rhetorical
genres of anaskeue and kataskeue, which, according to Aphthonius, constitute the
quintessence of rhetorical art27 • The original form of the issue, however, as it is
expressed by Artaxanes, has a rather mixed character, since it can be taken as close
to the progymnasma of thesis as well, whose character is by no means very different
from anaskeue and kataskeut?8.
In the diction of the traditional form of thesis, as it is prescribed by Hermogenes
and Aphthonius, the question about the 7rPE7rOJl of a man's giving birth to babies
could have been formulated as "should a man give birth to babies?"29 Such a formu-
lation of the matter would correspond to the 7rONTL/(CXL Oeuw;, in Hermogenes' and
Aphthonius' terminology (25 RARE; WALZ I. 108), or to 7rPCX/(TLKCXt OEUELq, in
Theon's terms (W ALZ op.cit. 244). More specifically, in Hermogenes' system, this
problem would be. classified under the category of the theses that examine an issue in
relation to a specific category of people, CXi/(CXTa TO 7rpoq TL ACX!-'{3&/IOPTCY.L, like the
problem fi 'YCX!-''YJTEOJl {3CXatAEi (25 RABE), since Gobryas and Artaxanes discuss the
issue of giving birth only in relation with a particular group of people, that is men,
and especially soldiers! But the issue, as put by Artaxanes, has two sides: a "practi-
cal" side, which has been already mentioned, and a "theoretical" one, which is as
follows: do men give birth to babies? If yes, how can they produce milk in order to
feed them? These questions constitute the "theoretical" aspect of the thesis and
correspond to what Hermogenes calls ou
7rOAm/(CXt OEUH~ (25 RABE) or to the OEU~L~

Methynmian general in Longus, BpuootLl;.

27 Tb Of 1rP<ryujJ.vooujJ.cx TOUTO (anaskeue) 1r&ucxv tV tCXUTc;, 1rEPLfXEL ,",V Tii~ TfXV71~ iaxuv (C. WALZ,
Rhetores Graeci, Stuttgart 1832-6, I. 72); Aphthonius expresses the same view for kataskeue: i)
-yvjJ.vcxuicx Of cxu'n) 1<Ct.I100V 1fEPLfXEL ,",V Tlj, TfXV71, iaxuv (ibid.: 77).
28 In his discussion of thesis, Hermogenes speaks of how one can "refute" (avcxTpf1rELV) and "con-
firm" (KOOTOIuKEuai'ELv) the topic of the thesis (26 RABE).
29 The topic of a thesis is usually expressed with the formula "should one ... ?" (e.g. Et 'YOIjJ.'T/TEOV, Et
tXO}.'T/TEOV, Ei 'YEWP'Y'T/TfOV; see 25 RABE; WALZ I. 109; 242).

448
The role of rhetoric 119

(lEwP'YITLKI:Xt in the system of Theon and Aphthonius (WALZ 1.180; ibid. :244). The
physiological example of women's production of milk after delivery is mentioned by
Aristotle in his Rhetoric as an example of true syllogism (1357b. 15-6: TfTOKcJl, elTt
'YCtAa EXH); this example is discussed also in Stephanos' twelfth century commentary
on Aristotle's Rhetoric. The syllogism is formulated here as following: auT'Y/ TETOKC'
'}'CtAa 'Yap EXtt' TO '}'CtAa EX€tJl u'YIjL€'i6J1 fun TOU T€TOKeJlat (ed. H. RABE, Anonymi et
Stephani in Artem Rhetoricam Commentaria, Berlin 1896, 264.32-33). It is tempting
to assume that here Theodoros Prodromos is not only making fun of this kind of
syllogism in general, but he is also alluding to this specific example30 •
Prodromos, however, seems to adhere to the demands of traditional rhetoric only
to some extent since he actually undermines and plays with the conventions of pro-
gymnasmata. On the one hand, he apparently follows the traditional definition of
thesis: 8emr; fUT~JI E1riaK€,ptr; AO'}'tK~, 8€wpoujJ.EvOV TtJlbr; 1rpa'}'jLaTOr;, in Aphthonius'
definition (WALZ 1. 108) or, in Theon's definition, 8eatr; €aT~/I €1riaKc,ptr; AO'YtKiI,
CtjLrPtafJilT'Y/ut/l €1rtbEXOjLeJl'YI (WALZ I. 242). The whole discussion revolves around a
dubious issue indeed and is constructed on apparently logical arguments based on a
supposedly real event: the preservation of the live birds in the roasted belly of the
lamb!
On the other hand, Prodromos' treatment of thesis, which, according to the theore-
ticians of the genre, involves the process of antithesis and solution (AvaLr;), of Qwa-
aKcvatet/l and KaTauKEvatEt/l, takes the form of a complex anaskeue and kataskeue
indeed, but at the same time it defies an important rule of these two last progymnas-
mara prescribed more explicitly by Apththonius: 'A/laaK€VaUTeO/l [Kat KamaKEva-
aTEO/l] bE Ta jLilT€ Ata/l aarP~ IL71TE &bUvaTg 1ral'TEAWC, CtA}...' oaa jLEU'Y/I' €x€t rill'
τTa~t/l
ά ξ ι ν (WALZ 1. 72, 77)3J. What really happens, however, is that Artaxanes and
Gobryas are dealing here with an issue which is totally CtbUI'aTo/l. Artaxanes' "pro-
blem" is dealt with in an anaskeue of his views by Gobryas. As a matter of fact, this
takes the function of a kataskeue of an argument for the possibility and the ethical
basis of men's giving birth to babies. Following the usual rhetorical practice, Gobry-
as' kataskeue begins with a reprimand of Artaxanes' "irreverent" doubts, and pro-
ceeds to prove that there is nothing wrong with the whole issue. In order to sub-
stantiate his view, Gobryas adduces the example of Zeus, the god of gods, who gave

30 For Prodromos' close familiarity with Aristotle cf. his short treatise IIEpi 1'OU P.eycXAou KO!i TOU
P.'KPOU (ed. P. TANNERY, Theodore Prodrome sur Ie grand et Ie petit (a Italicos) , Annuaire de
I'association pour I'encouragement des etudes grecques en France 21 (1887) 104-19).
31 Cf. Hermogenes' similar prescription that re. DE 'lI'C,,"U >/;EUOfJ OU1'E OtPO!OKEUO!01'fOP OU1'E KO!1'O!OKEUO!-
01'<op (11 RABE) ,

449
120 Panagiotis Roilos

birth to Dionysus and Athene. If Zeus did that, who, then, can call in question the
morality of such marvelous deliveries 32 ?
The validity of the theoretical aspect of the issue, i.e. that men are able to give
birth to babies, demonstrated by Gobryas' use of these mythological exempla, has
also been already suggested by both the extraordinary dish served to Artaxanes, that
is the "pregnant" roast lamb, and Gobryas' elaborate praise of Mistylos. Culinary art
- the marvelous dinner - and rhetorical art - Gobryas' elaborate encomium of
Mistylos - function as a proleptical kataskeue of the theoretical aspect of the thesis.
The logical adynaton of the thesis - a pregnant man - is abolished by means of the
"real" dynaton of the culinary miracle just performed in front of Artaxanes' eyes. On
this apparently dynaton paradoxon Gobryas bases the pithanon of his rhetorical con-
struction, employing a long series of paradoxes, which in their turn underline Misty-
los' supernatural power. This encomium (4.134-72) does not follow the rules of
ancient Greek rhetoric, but rather recalls the Byzantine occasional poems written in
honour of the emperor, like the ones composed by Theodoros Prodromos himself and
described by Horandner as "genuin byzantinisch"33. Like this occasional poetry,
Gobryas' laudatory speech exploits traditional rhetorical figures such as polyptoton,
paronomasia, parechese, oxymoron, parallelism, repetition. On a first level, these
rhetorical devices enhance the encomiastic character of Gobryas' speech; on a second
level, however, they serve the parodic function of the whole scene, since they con-
tribute to Gobryas' paradoxical argumentation.
Rhetorical amphoteroglossia allows Theodoros Prodromos to appropriate, and even
parody, the sacred discourse of hymnography as well. Gobryas' kataskeue of the
absurd thesis under discussion contains, I suggest, some allusions to the miraculous
birth of Christ. Gobryas' emphasis on the example of Zeus, the god of gods, giving
birth to the androgyne Dionysus, and to Athene, a female god, sounds like a rever-
sed, and not very reverent, version of the mystery of the Nativity. Instead of the
theological thesis "how was the Virgin able to give birth to God," we are offered the
playful theoretical thesis "how can a man give birth to babies and, in particular, to
dogs?"
The abundance of antitheses, and especially of paradoxa and oxymora in Gobryas'
speech, may have been modelled upon several examples of religious poetry. The
Akathistos Hymnos is the best and most famous example, all the more since this hymn
deals exactly with the Virgin and the mystery of her motherhood. Important in the
Akathistos is the antithesis between rhetoric and religion which revolves around the
central issue of the birth of Christ. Mary's ~€Jlo<; TOKO<; surpasses the power of
rhetorical eloquence:

32 Eill'EP "Yap;' ZEiN;:, TWV OEWV b {3EATiwv, / ;, 'Y~V avexCTlI'wv "ex, TO lI'&.V lI'EP'TPElI'WV ". / E/l{3pVOV
apri<pAEKTOV cie; /l1}POV /lEaOV / TII.ttT€A€C1<pOP1}TOV €ppa1l'T€<v Of A€< / Kexl IWlI'VP1}OtV E~&'Y€t 7rpoc;:
~/lEPexv, / Kexl /l1}TPLKOV Tt KCXI "YVVextKWOE, 11'&00<;: I b TWV T'T&VWV /3exa'AEu, v7rofJTh€<, I d Ti)v
• A01}v&.v EK K€<pexAij<;: EKKVEt / ~i<pEt pex'Yfia1}, Kex, OtXexCTOEiCT1)<;: /lECTOV, / lI'w<;: exiuxvv1}v <pex'1}p.EV
(<Popwv 'Y1}LVWV / &. Tol, 8E01, Tip.'ex To1, ovpavio,,; (4.195-206).
3) W. HORANDNER. Theodoros Prodromos, Historische Gedichte, Vienna 1974, p. 79.

450
The role of rhetoric 121

p~TOpCXr; 7rOAUrpf)Cry'YOUr; Wr; ixf)ucxr; arpw/Jovr;

OPWP,E/J E7rt UOt 8EOTOKE.
a7rOpOVUL 'Yap AE)'H/J TO 7rWr;
KCXt 7rCXpf)E/JOr; p,E/JE~r; KCXt T€K€L/J iuxvucxr; ...
XCXLP€ rp~AOUOrpOVr; auorpour; OHK/JUOVUCX,
XCXiPE 7EX/JOAO")'OVr; aAO'Yovr; EA€")'XOVUCX.
XC/ipE on Ep,WPCx/Jf)1JUCX/J oi OH/JOL UUI1JTy)mi'
XCXiPE on Ep,CXPCx/Jf)y)UCX/J oi TW/J p,vf)W/J 7rO~1JTcxi
(ed. TRYPANIS, Fourteen early Byzantine Cantica,
Vienna 1968, p. 36, Lr 1_11)34.

Whereas in the Akathistos the miracle of the Nativity surpasses the intellectual
capacity of rhetors, in Gobryas' speech it is rhetoric, not less than culinary art, that
manages to contrive a fake miracle. Gobryas' insistence on the miraculous "womb"
of the lamb recalls an established image in hyrnnography: the yj)J'JlQ of the ap/Joc;, a
sanctioned symbol of Christ himself, functions as an allusion to the ~ of Christ's
mother, the aU/JCxc 35 • Byzantine religious poetry offers many examples of a creative,
although ultimately stereotypical, exploitation of the theme. In Romanos Melodos'
second hymn on the Nativity, for example, the Virgin addresses her baby in these
words:

Ev KCXp7rOr; p,ou, uv IW~ p,ov ...

TI,/J urppcx")'iocx T~r; 7rCXpf)E/JLW:; p,ov OPWII aKcxTCxA'Y'/7r7011
K1JPU77W UE aTpE7rTOII ACryO/J UCxPKCX 'Y€/JO/tEIIOII.
OUK orocx U7rOPCxIl, oiocx U€ AVT'Y'//J 711r; rpf)op&.r; ...
wr; 'Yap EA~7rEr; ~ €P,~/J rpVACx~CXr; uwcx/J CXUT~II
(MAAS AND TRYPANIS, 10, cx' 3-9).

One of the main points in Gobryas' eloquent encomium of Mistylos' power is the fact
that the fire did not manage to burn the birds inside the "womb" of the roasted lamb .

• Op&.r;, apwTE ucxTpmfwII 'Apm~CxI'Yf,

wr; E~OIiJ.Ei{3E~ [sc. Mistylos] KCXt TUpCX/J/J€L Tar; rpUUHr;,
1/;vxpall o/: 7rO~Ei TOU 7rvpor; TI,I' oVCJicx/J,

34 The Akathistos might also have been used as a model for the hymn for Mistylos performed later
by the clown Satyrion.
3' For the characterization of the Virgin as o,fLlI&<;, see, e.g., P. MAAS AND C. TRYPANIS, Sancti
Romani Melodi Cantica, Oxford 1963, 142.01' 1 (1'011 iOLOII aplla {*} o,fLva<; OfwpouC1a); ibid.: 290.
01' 3. The symbolism of the o,fLVO<; found its more explicit expression in religious art during the
iconoclastic period, where Christ was really depicted as a sheep. The Council in Troullo, however,
prohibited the use of this symbol in art; cf. the comments of the twelfth century canonists Zonaras
and Balsamon on this particular canon (G. A. RALLES AND A. POTLES, r.UV7C¥'YfLa 1'WV OfLWV KOli
ifPWV KOlIIQVWII, 6 vols., Athens 1852-9, II.493-4).

451
122 Panagiotis Roilos

apvou<; Of 7roLfL oTpov807r&Topa<; ~fVOV<;36

Kat jJ.~Tpav apT£</>AfKTOP E~W7rT'r/jJ.€VT/P,
(3Pf</>WP aKCiUOTWV, f.jJ.(3puwv KCiTCi7rT€PWP
"(fVV~TPLCiV O€£KVVOLV f.K p,OVOV AO"(OV (4.154-63).

The motif of the fire that burns without causing any harm recalls the miracle of the
KCiLOjJ.€P'r/ KCit jJ.~ </>Af"(OjJ.€Vq (3Ci.To<;, a traditional symbol of the Virgin Mary3? In
bymnography, the paradox of harmless fire is used to describe the mystery of the
Nativity. Again in Romanos' second hymn on the Nativity, the Virgin says about her
son:

IIiip V7r&PXwv ~K'r/Of p,OV

~v "(CiOT€pa KCit ou KCiT€</>A€~fli fjJ.t ~v TCi7rfL~V
(MAAS AND TRYPANIS 14, La' 3-4).

In Kosmas' canon on the Nativity the same image is used to describe the miracle of
the birth of God:

8aUjJ.CiTO<; V7rfp</>voii<; 17 opooo(3oAoe;

f.~fLKOVLOf K&jJ.LPO<; TP07rOP'
~ oue; f.O€~CiTO d>AhCL vEove, we; ouo~ 1!:.i)Q
Tfte 8fOr21T0e IIap8evov ~v inr€Ov 1:'.Il.M1!.
(R. CANTARELLA, Poeti Bizantini, Milan 1942, I.120, 140-3).

The same image appears also in John of Damaskos' canon on the Nativity:

~ a¢AfKTWe; EtKOV£rOvaL Kop'r/<;

oi ~e; 7rCiACiL&e; 7rVP7rOAOUjJ.€VOL VfOL
V7r€P</>vw<; Kuovoav, i:a</>pCi"(LojJ.€VT/P
(ibid.: 114, 101-WH.

~ The same adjective, tEVOe;, is used also in the Akathistos to describe the TOKOe; of the Virgin;
TRYPANIS 35. til' 1.
37 See, e.g., Romanos' second hymn on the Annunciation, where the image of the miraculous bush
is the Leitmotiv: KaTfll'Aa"(11 'Iwu~¢ TO UlI'fP ¢UULV OEWPWV I Kal EAap./3avEV Eie; vouv TOV ElI'llI'OKOV
VETOV I EV ri) Ct.UlI'OPI.j) KU~UEL UOU, OEOTOKE, I /3aTov liv lI'upl Ct.KaTa¢AEKTOV (MAAS AND TRYPANIS
289. prooemium, 1-4; cf. also ibid. 289, a' 1-2; 291, E'; 293, La'-L/3'). See also the second
troparion of the first ode in John of Damascus' canon on the Nativity: ijVf'j'KE yauT~p ';'YLaOP.El"I]
Aoyov I oa¢we; Ct.¢AEKTI.j) Iwypa¢ou/-LEvy) /3Ct.TI.j) (CANTARELLA, 1.111, 6-7).
38 Romanos offers a very powerful version of this motif in his first hymn on the Annunciation:
Joseph says to the Virgin: '0 ¢afLv~, ¢Mya opw Kal Ct.v8paKiav KUKAovuav UE' I ... I KAi/3avoe;
lI'A~PY)<; lI'UPOC; €YEVtTO &¢vw .;, &p.E/-LlI'TOe; yao~p UOU (MAAS AND TRYPANIS 287, LE' 3-5). It is
noteworthy that the same motif in association with the metaphorical image of the lamb is also
exploited in another poem by Theodoros Prodromos, in which he describes the symptoms of his
illness: ITwe; €rE¢POt'fLy)v T4> lI'upl Kai lI'We; Ct.lI'y)vOpaKOV/-LY)v· I E~Wll'[TY)/-LEVOe; T4> lI'upl 8E4> Ct.jvTEOvofLY)v
I OUX WIJ1I'EP &/-Lwp.oe; Ct./-LVoe; (Tie; ,,!o.p €"!w Kai rlvwv;), I [Ct.AA' we; KaTaparOe;] KPLO<; E~ Ct.p.apTiae;

452
The role of rhetoric 123

Here, as in Kosmas' canon, the comparison is between the fire in the furnace of the
Three Children that did not burn them, a motif taken from the Old Testament, and the
fire of God that did not burn the Virgin's womb. Prodromos gives this sanctioned
motif a most grotesque twist: Mary's womb is transformed into the belly of the
roasted lamb which, although dead, that is burnt, gives birth to the birds. Furthermo-
re, the image of the lamb's "womb" plays with the ideas of life (the live birds) and
death (the burnt belly of the lamb): life is preserved in death and the latter is followed
by the former in a paradoxical way that exemplifies Bakhtin's understanding of the
regenerative character of the grotesque39 •
The possibility of a parodic appropriation of the discourse of religious poetry by
Theodoros Prodromos here is further corroborated by some intriguing evidence
provided by Prodromos' own commentaries on the above-mentioned canons40 • These
commentaries, which have escaped the attention of the scholars of Prodrornos' novel,
elucidate the intricate intertextual allusions of the banquet scene under discussion in
the most revealing way. Prodromos' interpretation of the passages from John of
Damascus and Kosmas leaves no doubts about his conscious appropriation of the
hyrnnographic topos of harmless fire in Gobryas' speech. Commenting on Kosmas'
canon, Prodromos observes:

~ '" KCtJ.l.LPO<; ~ TI,P 7rOlp8ePLK~p TU7rOUCTOI 1I7J0l'" €Keip.,., TO KOI"iOP fxouaOl TOU<;
u7rOOex8EPTOI<; ou KOITE(j)Ae~e PfOU<;. ;;PTOIv801 oe b tiP WUTpOIJJ U7rOOVC KOit u7rOoe-
x8ei<;, €Ke"ipo<; hwp TO 7rUP, J.I.&AAOP oe 7rVP wP, tiP u7rooetOlU€JJnp OIUTOP t'xpn-
aTpOI/Jwc LdzTPOIP OUK €KOIUae (ed. H. STEVENSON, Theodori Prodromi Commen-
tarios in carmina sacra melodorum Cosmae Hierosolymitani et Ioannis Da-
masceni, Rome 1888, p. 52).

Similar are his remarks on the excerpt from the canon of John of Damascus:

P.fAOl~ CW. HORANDNER, Historische Gedichte, 46.56-9).

39 Bakhtin says that the notion of the grotesque encompasses the ideas of death, birth, and rebirth:
"The grotesque image reflects a phenomenon in transformation, an as yet unfinished metamorphosis,
of death and birth, growth and becoming. For in this image we find both poles of transformation, the
old and the new, the dying and the procreating, the beginning and the end of the metamorphosis"
(Rabelais and his world. Bloomington 1984, p, 24, and passim), The same ambivalence is also
conveyed in the scene of the fake suicide and resurrection of the clown Satyrion who appears later
at the banquet.
40 It seems that the exegesis of the canons experienced a special flourishing in the twelfth century; it
has been suggested that some of these commentaries were written as lectures for an academic
audience (PH, DEMETRACOPOULOS, The Exegeses of the Canons in the twelfth century as school
texts, Diptycha 1 (1979) 143-57 : 146-7). It should be noted that in the twelfth century similar
commentaries were written also by Gregory Pardos, Zonaras, and Eustathios of Thessalonike (see A.
KOMINES, rpTl'Y0p{ov TOU' Kop,vf){ou 'E~Tl'Y~UEI~ fi~ TOV~ I'EPOV~ AEITOUP'YIKOV~ KOlV6vOlC; TOU !!.0lP.0l-
UKTlVOU KOlI Koup.ci TOU MeA<j)5ou. Akten des XI. Intern. Byzantinistenkongresses MOnchen 1958,
Munich 1960, 248-53: 252). Prodromos' commentaries indicate that the commentators were familiar
with each other's work (ibid,: 251).

453
124 Panagiotis Roilos

Ol E7rL T'i}r; IIaAm&r; r pacp'i}r; a7rVp7rOA1!TWr; 7rVP7rOAOUJL€POL PEOL VIP p:qTeqp

Tile aEL7raeIJhov EiKopitoVat 8Eop.nTOeoe, T1JP JL€TCx TO U7rEpcpVWr; ')IEPIJ'i}am
Eacppa')lLtJJLtPYfV btaJLEivaaaV' 7rA~V EKEL JL€V EV TV KaJLiv4J oi 7raio€r; iaTaJLEvoL
a7rEipaTOL OLtJLEVOP TOU 7rVpOr;, €vmuOa Of TO EvaVTiop a7rav' €vroe hovqq TO
n
7rUe TUe 8E(m}TOe 8Eop.nTWe OU KaTEcbA€)'ETO' (mEp 7roAl.> EKEivov 7rapaOO~oTE­
pOl' (ibid.: 71).

In his commentary on another troparion from Kosmas' canon, Theodoros Prodromos

ridicules Greek mythology in order to prove the superiority of Christian theology. His
analysis of Kosmas' canon sounds as a perfect anaskeue of his own parodic kataskeue
in the novel. His main argument here is that the ancient Greek myths that speak about
a male god, like Zeus, who gives birth to a child are altogether nonsensical. The
theoretical thesis, therefore, "can a man give birth to a child?" is subjected here to a
totally different argumentation than that in the novel, and given, of course, a negative
answer:

•EK ')IaaTpOr; Of ')IE')IEv7joOm TOV MOVO')lEv7j 7rapa 8EOU Kal. IIaTpor; At')lOJLEV,
oux wr; aWJLanKar; u7rovoiar; 7rEPI. T'i}r; apapxov KaL aOWJLaTOV cpuaEWr; EXOPTEr;,
oux wr; ')IaaTEpa EJL{3pvoooxop T~ avapX4J IIaTpL KaL woivar; nvar; KaL AOXEiar;
7rpOOa7rTOVTEr;' •EAAilvwp ')lap b i)IJAor; OVTOr;, oinPEr; ToaOUTOV TOUr; faVTWV
OEOUr; a7rO')lvpmKoiJtJLV, waTE OU ')IaOTtpar; JLoPOV aAA' ~OYf 7rOV KaL KEcpaAar; KaL
JLYfpOUr; fJL{3pvoooxovr; TOUTOLr; Otooaat. KaL Tar; JL€v 7rEAEK€L OtEAOPTEr; ovaTOKOUV-
TOr; 01!7rOV TOU OEOU KaL Tar; woivar; ou aTE')IOVTOr; OV')IOITEpa EKEiOEV f~a')lOVaLV
EV07rAOV, TOU Of JLYfpOU TOV pacp€a ALVOV 7rEPLEAOVTEr; Kpm7ra}...wpm ~OYf Kat
U7ro{3E{3peYJLEPOP i:~ ailT1}e;; T1}e;; WOLPOe;; view 7rPOiOXOVUL fuopvuov (ibid.: 48)

This part of Prodromos' analysis of Kosmas' canon has been based, I argue, on
Gregory of Nazianzus' homily on Epiphany (Eie;; Ta a')lLOI cpwm; or. 39; PG
36.336-58, new edition Sources Chretiennes 358, 150-97 Moreschini-Gallay). In this
oration, Gregory dwels extensively on the miracle of Christ's birth and opposes it to
ancient Greek myths, which, as he notes bitterly, were not real mysteries, but illu-
sions and misleading twaddle (or. 39.1; SC 150). In the introduction to his speech,
Gregory enumerates those ancient Greek stories that speak about miraculous births of
gods, only in order to ridicule them:

Ou ~LOr; mum ')IOVaL Kal. KA07rai, TOU KPYfTWP TVpaPIJOV, KO/v "EAAYfPEr; Cx7rapE-
aKWVTm' ovo"E KovpilTwv ~XOL Kat KPOTOL Kal. bpxilO€Lr; EV07rAOL, IJEOV KAaioPTOr;
~x~v aV')IKaAU7rTOVam, ivOI 7raTEpa Aa01J JLLaOTEKvoV' ... OVOf CPpv/,wp fKTOJLaL
KaL av}...ol. Kat Kopu{3aVTEe;; Kal. oaa 7rEPL T1JV •PtaI' av(JpW7rOL JLOILPOVTOIL, TE-
}...OUVTEr; TV JLYfTPI. TWP (JEWV KO/t TE}...OUJLEVDt oaa TV JLYfTpt TWP TOWUTWV EiKor; ...
OuoE ~LovvaOr; mum Kat JLYfPOr;, woillwv cXTE}...i:r; KUYfJLa, Wa7rEp a}...}...o n KECPa}...~
7rPOTEPOV' KO/t (JEOr; cXVOpO')lVPOr; Kat aTpaTOr; JLEIJVOVTWP, KO/t OTpaTOe;; EKAVTOr;
(or. 39.4; SC 154-6).

454
The role of rhetoric 125

Gregory of Nazianzus was one of the most appreciated Christian authors throughout
Byzantium and his direct int1uence on Theodoros Prodromos is very probable. In my
view, however, another source of inspiration could also be detected here: Pseu-
do-Nonnos' commentary on Gregory's homilies. Pseudo-Nonnos offers a much more
detailed description of the ancient myths referred to in Gregory's homilies than
Gregory himself. In his commentary on the birth of Dionysus, Pseudo-Nonnos
narrates how Zeus took the fetus from Semele first and then sewed it into his thigh
(PG 36.1048 [Comm. in or. 5, hist. 20] and 1068-9 [Comm. in or. 39, hist. 4]; new
edition by Jennifer Nimmo Smith [CCSG 27], Turnhout 1992, pp. 195 and 222-3).
The description of the same myth in Prodromos' novel follows the same sequence.
Furthermore, Pseudo-Nonnos' detailed description of Dionysus' company, Satyrs and
Maenads, who indulge in revelry (PG 36.1068-9; Nimmo Smith 222-3), can be
paralleled with the detailed description of the depiction of Dionysus and his entourage
on a precious cup that was used during Gobryas' banquet (4.344-411).
Anotherfact should also be noted in respect to Prodromos' reference to the ancient
Greek myths of Zeus' miraculous deliveries and its possible allusions to an established
Christian topos. Four manuscripts from the eleventh and twelfth centuries preserve
several illuminations to Pseudo-Nonnos' commentary. Along with Pseudo-Nonnos'
text, these manuscripts also preserve Gregory's homilies4l . Some of these illumina-
tions present the story of Dionysus' and Athena's birth. In two manuscripts 42 , the
birth of Dionysus is illustrated in three phases: in the first phase, Zeus takes the
embryo from the womb of the dead Semele who has been struck by the thunderbolt,
(cf. Prodromos' "Ep.{3pU01J ixPTi<l>A€KTov ... ~p.mA€u<I>6pYJTOV; 4.199-200); in the second,
he sews it into his thigh (cf. Prodromos' de; j.l.YJP()1J j.I.€UOP EppenrTELV OiA€L; ibid.) and
in the last one, he delivers Dionysus as a baby (cf. Prodromos' Kat tW7rvPYJO'€P €~&"y'€L
7rpOe; ~j.I.€pa/l, / Kat JJ-YJTPLK(J/I TL Kat "y'VVaLKWO€e; 7r&Ooe; / 0 TW/I TLT&/IW/l (3CtULAEUe;
U7rOUTE"y'€L; 4.200-2). In two other manuscripts, only the third phase, that is the
delivery of the baby Dionysus, is depicted43 . Athena's birth is illustrated only in the
Jerusalem manuscript. She is depicted coming out of Zeus' head; in front of Zeus
stands Hephaestus holding an axe (cf. Prodromos' Trw' AOYJ/I&/I €K KE<I>CtA~e; EKKVeL,
/ ~i<l>€L pCt"y'€iU'f]e; KCtt OtXCtU8Ei(J'f]r; j.l.E(JO/l; 4.203-4).
It is tempting to assume that Prodromos was familiar with the iconographic vocabu-
lary of such manuscripts, all the more since Gregory's homilies and Pseudo-Nonnos'
commentary were the main sources of inspiration in his construction of Gobryas'
second speech. Perhaps Gobryas' reference to Zeus as 0 TW1J Tmx/lw/I BCtC1LAeUC
(4.202) may ret1ect such a familiarity, since in some of the above-mentioned manu-

41 These manuscripts are the Sancti Sepulchri 14 in Jerusalem (second half of the 11th c.), Vaticanus

Graecus 1947 (11th-12th c.), Panteleimon 6 (lith c.), Paris Coislin 239 (end of the 11th c.); see K.
WEITZMANN, Greek mythology in Byzantine art, Princeton 1951, pp. 9-11.
42 These are the Jerusalem and Vatican manuscripts (K. WEITZMANN, as in n. 41, pp. 47-8).
43 The Panteleimon and Paris manuscripts; see K. WEITZMANN, op.cit.: pp. 49.

455
126 Panagiotis Roilos

scripts Zeus is really depicted as a Byzantine BC¥OJA.€vc44 • Be this as it may, the

literary evidence discussed so far - Prodromos' own commentary on John of Damas-
kos and Kosmas, Gregory's homily on Epiphany, and Pseudo-Nonnos' commentary
on it - indicates that in the scene of the culinary marvel served to Artaxanes and its
rhetorical exploitation by Gobryas, Prodromos makes a parodic use of elements drawn
from religious literature and especially hymnography.
To conclude: Manuscript tradition indicates that the Byzantines perceived the genre
of the novel in rhetorical terms. As a matter of fact, Theodoros Prodromos' novel
bespeaks a close familiarity with traditional rhetorical theory and practice. Rhetoric's
inherent quality of amphoteroglossia enables Theodoros Prodromos to allude to
different domains of cultural experience at the same time: on the one hand, he draws
from the long tradition of Greek rhetoric, on the other, he plays with the conventions
of Byzantine ceremonial and of Christian literature. Amphoterogiossia should also be
viewed as the theoretical key to understanding the often double character of the
Komnenian novels in general: on the one hand, these texts follow the conventions of
their ancient Greek models, on the other, they allude to contemporary cultural
Byzantine reality. The pseudo-aristotelian ... nursing men, the suggestive image of
the aKVA(x~, and the marvelous onor; ap,,6r; at Prodromos' grotesque banquet illustra-
te the dynamics of this amphoterogiossiain the most graphic possible way.

44 These are the miniatures in the Jerusalem, Panteleimon, and Paris manuscripts; see K. WEITZ-
MANN, op.cit.: pp. 47, 49. Perhaps an association between Zeus, the god of gods, and the emperor
may also be detected in the indirect comparison of Mistylos with Zeus in the presentation of the
former as ul/loD KcxOw8ftC; KCXt TlmpWOEC (3)..f7:WP (4.17-8; my underlining points to the possibility of
a deliberate association of Mistylos' titanic-like appearance with the characterization of Zeus as ;, TWV
TtTO<vwv (3o/(n)..ftJl;).

456
 Copyright Acknowledgments

Brock, Sebastian. "From Ephrem to Romanos." StudiaPatristica 20 (1989): 139-

151. Reprinted with the permission of Editions Peeters s.p.r.I.
Cameron, Alan. "Wandering Poets: A Literary Movement in Byzantine Egypt."
Historia 14 (1965): 470-509. Reprinted with the permission of Franz
Steiner Verlag Stuttgart GmbH.
Usher, Mark D. "The Sixth Sibylline Oracle as a Literary Hymn." Greek, Roman, and
Byzantine Studies 36 (1995): 25-41. Reprinted with the permission of Duke
University Press.
Hunger, Herbert. "On the Imitation (Mimesis) of Antiquity in Byzantine
Literature." Dumbarton Oaks Papers 23/24 (1969/1970): 17-38. Reprinted
with the permission of Dumbarton Oaks Publications.
Browning, Robert. "The Language of Byzantine Literature." In S. Vryonis, ed.,
ByzantinakaiMetavyzantina, Vol. 1 (Malibu, 1978): 103-133. Reprinted
with the permission of Undena Publications.
Jeffreys, Elizabeth, and Jeffreys, Michael. "The Oral Background of Byzantine
Popular Poetry." Oral Tradition 1 (1986): 504-547. Reprinted with the
permission of Slavica Publishers, Inc.
Kustas, George L. "The Function and Evolution of Byzantine Rhetoric." Viator 1
(1970): 55-73. Reprinted with the permission of the University of
California Press. Copyright (1970) by the Regents of the University of
California.
Sevcenko, Ihor. "Levels of Style in Byzantine Literature." Jahrbuch der
Oesterreichischen Byzantinistik 31 (1981): 289-312. Reprinted with the
permission of Verlag der Oesterreichischen Akademie der
Wissenschaften.
Browning, Robert. "The 'Low Level' Saint's Life in the Early Byzantine World." In
S. Hackel, ed., The Byzantine Saint (London, 1981): 117-127. Reprinted
with the permission of Fellowship of St. Alban and St. Sergius.
Dennis, George T. "Imperial Panegyric: Rhetoric and Reality." In H. Maguire, ed.,
Byzantine CourtCulture/rom 829 to 1204 (Washington, D.C., 1997): 131-
140. Reprinted with the permission of Harvard University Press.
Kazhdan, Alexander P. "Byzantine Hagiography and Sex in the Fifth to Twelfth
Centuries." Dumbarton Oaks Papers 44 (1990): 130-165. Reprinted with the
permission of Dumbarton Oaks Publications.
Maguire, Henry. "Byzantine Descriptions of Works of Art." Dumbarton OaksPapers
298 (1974): 113-140. Reprinted with the permission of Dumbarton Oaks
Publications.
Alexiou, Margaret. "The Poverty of Ecriture and the Craft of Writing: Towards a
Reappraisal of the Prodromic Poems." ByzantineandModern Greek Studies
457
458 Copyright Acknowledgments

10 (1986): 1-40. Reprinted with the permission of Byzantine and Modern

Greek Studies.
Macrides, Ruth. "Poetic Justice in the Patriarchate: Murder and Cannibalism in the
Provinces." In L. Burgmann et al., eds., Cupido Legum (Frankfurt am Main:
Lowenklau Gesellschaft e.V, 1985): 137-168. Reprinted with permission
of the Lowenklau Gesellschaft e.V.
Magdalino, Paul. "Honour among Romaioi: The Framework of Social Values in the
World of Digenes Akrites and Kekaumenos." Byzantine and Modern Greek
Studies 13 (1989): 183-218. Reprinted with the permission of Byzantine and
Modern Greek Studies.
Mullett, Margaret. "Aristocracy and Patronage in the Literary Circles of Comnenian
Constantinople." In M. Angold, ed., The Byzantine A ristocracy (Oxford,
1984): 173-201. Reprinted with the permission of Clarendon Press.
Roilos, Panagiotis. "Amphoteroglossia: The Role of Rhetoric in the Medieval
Learned Novel." In P. A. Agapitos andD. R. Reinsch, eds.,DerRomanim
Byzanz der Komnenenzeit. Referate des lnternation:den Symposiums an der Freien
Universitat Berlin, 3. bis 6. (Frankfurt am Main: Beerenverlag, 1998): 109-
126. Reprinted with permission of Beerenverlag.